key: cord-0041839-4508tbhi
authors: nan
title: Research Abstract Program of the 20(th) Annual ACVIM Forum Dallas, TX, May 29‐June 1, 2002
date: 2008-06-28
journal: J Vet Intern Med
DOI: 10.1111/j.1939-1676.2002.tb02376.x
sha: 88076809b0feb06abd3c23dad8cbcaf2e69da5e5
doc_id: 41839
cord_uid: 4508tbhi

nan

Inflammatory bowel disease (IBD) is the most common cause of chronic vomiting and diarrhea in dogs, but the pathogenesis of IBD has not been well understood. To characterize the immunologic response in canine IBD, the mRNA expression levels of various cytokines were quantitatively analyzed in the duodenal mucosa obtained from dogs with IBD.

From 13 dogs with IBD and 10 control dogs, duodenal mucosal tissue specimens were obtained by endoscopic biopsy and subjected to total RNA extraction by using the acid guanidium-phenol-chloroform method. For the examination of the amount of mRNAs of cytokines including IL-1beta, IL-2, IL-4, IL-5, IL-6, IL-8, IL-10, IL-18, IFN-gamma, TGF-beta 1 and TNF-alpha, a pair of oligonucleotide primers and an oligonucleotide probe were designed from the previously reported canine specific sequences. The quantitative analyses of cytokine mRNAs were carried out by a real-time TaqMan polymerase chain reaction technique. The amounts of mRNA expression were expressed as values relative to the amount of beta-actin mRNA in each sample.

The expression level of IL-6 mRNA in the duodenal mucosa from dogs with IBD was significantly higher than that in the duodenal mucosa from control dogs. Expression levels of IFN-gamma and TGF-beta 1 mRNAs in the IBD group were significantly lower than those in the control group. IL-4 mRNA expression in the duodenal mucosa was detected in 2 of the 13 dogs with IBD, but not in any of the 10 control dogs. Moreover, the expression levels of IL-1beta and IL-5 mRNAs in the IBD group showed a tendency to be higher than those in the control group. In the expression level of IL-2, IL-8, IL-10, IL-18, and TNF-alpha mRNAs, there was no significant difference between the IBD group and control group.

These findings indicated that the expression levels of proinflammatory cytokine mRNAs were increased and Th2-type response rather than Th1-type response was predominant in the duodenal mucosa in dogs with small intestinal IBD. These cytokine profiles were considered to be very similar to those of human ulcerative colitis. Moreover, the decreased level of TGF-beta 1 mRNA expression might be derived from the abnormality of oral immune tolerance in the intestinal mucosa in canine IBD. Canine and feline colonic motility is regulated by serotonergic (5-HT 2 , 5-HT 4 ) neurotransmission. Depolarization of neuronal 5-HT 4 receptors stimulates gastric, intestinal, and colonic prokinesis. We have previously shown that cisapride, a 5-HT 4 agonist, stimulates contraction of feline colonic and megacolonic smooth muscle. New 5-HT 4 prokinetic agents are in various stages of drug development following withdrawal of cisapride from the American, Canadian, and western European markets. In this study, we examined the effects of prucalopride, a new 5-HT 4 agonist, on colonic smooth muscle contraction in the dog and cat.

Ascending, transverse, and descending colon was obtained from healthy adult dogs and cats. Smooth muscle strips were prepared in the circular or longitudinal muscle orientation, suspended in physiologic (HEPES) buffer solution, attached to isometric force transducers, and set to optimal muscle length (L o ) with acetylcholine (ACh; 10 Ϫ4 M) or potassium chloride (KCl; 80 mM). Muscles were treated with cumulative (10 Ϫ8 to 10 Ϫ4 M) or bolus (10 Ϫ4 ) doses of prucalopride, and isometric stress responses (magnitude and frequency) were recorded.

Prucalopride dose-dependently increased the amplitude (2.0-3.0 ϫ 10 4 N/m 2 ) and frequency of phasic contractions in longitudinal smooth muscle from the ascending, transverse, and descending colon of adult dogs and cats. Maximal responses were attained at 10 Ϫ4 M prucalopride. Maximal prucalopride responses were 75-82% of the maximal ACh response (10 Ϫ4 M), and 69-78% of the maximal KCl (80 mM) response. Prucalopride-induced contractions of longitudinal smooth muscle were of greater magnitude and frequency in the ascending and transverse colon. Prucalopride had a modest effect in increasing tonic force in circular smooth muscle from ascending, transverse, and descending colon.

We conclude that prucalopride: (1) increases the phasic activity of longitudinal smooth muscle in the canine and feline colon; and, (2) may be useful in the therapy of feline (and canine) colonic motility disorders. We have previously reported that cats with severe cobalamin deficiency, defined as a serum concentration of cobalamin less than 100 ng/L, have dramatic elevations in serum methylmalonic acid (MMA), a marker of tissue-level cobalamin availability. We have also demonstrated that serum MMA concentration normalizes with parenteral cobalamin therapy. The aim of this study was to assess the diagnostic utility of serum cobalamin concentration in cats, for the determination of tissue cobalamin deficiency.

Feline sera (n ϭ 206) with cobalamin Ͻ 290 ng/L were obtained from accessions to the GI Laboratory. Serum MMA was determined using gas chromatography/mass spectrometry. Tissue-level cobalamin deficiency was defined as being present when serum MMA was Ͼ 867.1 nmol/L (upper limit of the reference range derived from 24 clinically healthy control cats with serum cobalamin within our laboratory reference range). Data were analyzed by comparing sensitivity and specificity at varying diagnostic cut off values for serum cobalamin between 100 and 290 ng/L. Positive and negative predictive values of each cut off value were also determined.

The occurrence rate of tissue-level cobalamin deficiency in cats with a serum cobalamin less than 290 ng/L was 68%. A cut-off value of 160 ng/L serum cobalamin had the best theoretical combination of characteristics for the diagnosis of significant tissue-level cobalamin deficiency, with 74% sensitivity and 80% specificity. The negative predictive value using 160 ng/L was unacceptably low (59%), indicating that serum cobalamin greater than 160 ng/L does not reliably rule out tissue-level cobalamin deficiency in cats. Negative predictive value increased markedly for serum cobalamin greater than 280 ng/L (100%). Cats with serum cobalamin concentration less than this value had a 68% probability of tissue-level cobalamin deficiency. This deficiency may be clinically limiting.

We conclude that tissue-level cobalamin deficiency should be suspected in cats with serum cobalamin concentration Ͻ 280 ng/L. The frequent occurrence of cobalamin deficiency, in cats with serum cobalamin concentrations close to the lower limit of normal, suggests that the reference range for feline serum cobalamin may extend lower than the optimum for this vitamin.

Inflammatory bowel disease (IBD) is a frequent cause of chronic diarrhea, vomiting and weight loss in dogs. Dietary hypersensitivity is proposed as one of the possible cause of IBD. The purpose of this study was to assess, in field conditions, a hypoallergenic diet formulated with soy isolate hydrolysate as the main protein source (Royal Canin, hypoallergenic DR 21) .

Eight dogs were included in the study. Inclusion criteria were IBD confirmed by histological evaluation of intestinal biopsies and associated with at least one clinical sign (diarrhea, vomiting or weight loss). Initially, the diet was given as the only treatment. If the diet alone did not improve the clinical condition after 4 weeks, metronidazole first and then prednisolone were used. Clinical and fecal assessments were performed every 2 weeks (fecal scoring-from 1: the worst to 4: the best, number of bowel movements or vomiting episodes). Endoscopic and histological assessments of the intestinal mucosa were performed before and at least 8 weeks after inclusion in the study.

Seven out of 8 dogs were in normal body condition at inclusion. Mean fecal score dramatically improved with the diet (1.5 Ϯ 0.53 vs 3.62 Ϯ 0.52, p Ͻ 0.01, Student's t test). In 6/8 dogs in which they were initially abnormal, mean number of bowel movements improved from 4.12 Ϯ 1.73/day to 2.5 Ϯ 0.53/day (pϽ0.05, Student's t test). Two out of 8 dogs experienced severe vomiting episodes with hematemesis; they stopped vomiting within 1 to 2 months following treatment. Complete recovery was observed in all dogs between 4 and 16 weeks following introduction of the diet but 3/8 dogs required additional treatment (metronidazole 3/3 dogs and prednisolone 1/3 dogs). During endoscopic examination, no obvious improvement of the intestinal mucosa was noted except for 1 dog in the colon. General histological scores were not different before (3.8 Ϯ 0.9) and after treatment (3.4 Ϯ 0.5) but for 2/8 dogs the infiltrate in the intestinal mucosa was reduced.

These preliminary results suggest that a soy isolate hydrolysate diet could be useful in the clinical management of IBD, even in severe cases, and might be a good alternative to the use of corticosteroids. The purpose of this study was to test the hypothesis that non-challenged dogs sensitized with food allergens could be differentiated from clinically healthy control dogs using gastrointestinal permeability and mucosal function testing using a five sugar solution. Eight dogs were sensitized with 6 food allergens by subcutaneous injections. Seven dogs were used as non-sensitized, clinically healthy control dogs. All dogs were 1.5 to 2.0 years of age. The dogs were fed an elimination diet containing egg and rice flour (to neither of which any dog had been sensitized) for 6 weeks, and were fasted overnight prior to gastrointestinal permeability and mucosal function testing. A urinary catheter was used to completely empty the bladder. A solution containing methylglucose, rhamnose, xylose, sucrose, and lactulose was administered via stomach tube (200 ml for dogs less than 20 kg body wt., 400 ml to dogs over 20 kg body wt). Six hours after administration of the solution, the dogs were catheterized and the entire urine volume was collected for measurement of sugar recovery by high pressure liquid chromatography, followed by pulsed amperometric detection. Urinary recoveries were expressed as percentage recovery (%R) for each sugar. The lactulose:rhamnose recovery ratio and the xylose:methylglucose recovery ratio were calculated. Mean urinary recoveries for each sugar and mean urinary recovery ratios were compared between sensitized and control dogs using twosided t-tests.

Serum IgE analysis indicated a statistically significant difference between sensitized and control dogs. None of the sensitized dogs showed any urinary sucrose recovery. There were no significant differences between sensitized and control dogs of the mean urinary sugar recovery for methylglucose (p ϭ 0.367), rhamnose (p ϭ 0.665), xylose (p ϭ 0.286), and lactulose (p ϭ 0.676). Mean urinary lactulose/rhamnose and xylose/methylglucose recovery ratios were not significantly different between the groups (p ϭ 0.327, and 0.416, respectively). We conclude that food allergen sensitized dogs could not be differentiated from control dogs after feeding an elimination test diet containing egg and rice flour. BACKGROUND: Acute pancreatitis, hepatobiliary disease, and proximal gastrointestinal tract disorders are well recognized clinical situations where delivery of nutrients via jejunostomy tube is preferable to a feeding gastrostomy. Presently, placement of jejunostomy feeding catheters requires laparotomy. A thorough description of the percutaneous endoscopic jejunostomy (PEJ) technique and practical guidelines for its use in dogs have not been described. AIM: To adapt a simple technique of PEJ tube placement in humans for use in healthy dogs, and to evaluate the practicality and potential complications associated with short-term PEJ tube use.

METHODS: Placement of commercially-prepared PEG/PEJ tubes (Wilson-Cooke Medical Inc., Winston-Salem, NC) through a gastrostomy catheter was performed in 11 dogs using endoscopic guidance. This technique requires passage of the endoscope deeply into the small intestine where a guide wire is then passed through the accessory channel. A 12 F jejunostomy feeding catheter is next advanced over the guide wire as the endoscope is removed from the dog. Using a unique intermediary maneuver, the proximal (orad) end of the guide wire is identified, pulled through the gastrostomy tube, and the PEJ tube is advanced over the guide wire into the jejunum. The following parameters were evaluated in all dogs for 14 days following PEJ placement: attitude/activity, temperature, infection at tube site, tube migration/clogging, and nutritional status. Seven dogs were bolus-fed enteral diets via PEJ tube.

RESULTS: PEJ tube placement was performed without complication in all dogs with an approximate procedural time of 40 minutes. Correct positioning of the catheter tip, 60 cm distal to the pylorus, was best confirmed by fluoroscopy or contrast radiography. Complications associated with PEJ tubes were rare but included iatrogenic PEJ tube removal (2 dogs), localized PEJ infection (1 dog), and retrograde tube migration (pilot study-2 dogs). Bolus feeding (q 8 h) using 1 of 2 isotonic enteral diets was well tolerated in all 7 dogs. Significant alterations in body weight and body condition scores were not observed as a consequence of PEJ tube feeding. Acute nosocomial diarrhea was observed in 2 dogs.

CONCLUSIONS: This study shows that placement of a PEJ tube is an effective, non-invasive technique for providing enteral nutritional support in dogs. The success of PEJ placement is most dependent on correct distal catheter positioning and prevention of iatrogenic removal. Bolus-feeding techniques via PEJ tubes in healthy dogs are well tolerated and maintain normal nutritional status. This procedure for jejunostomy feeding may be adapted for use in routine clinical practice outside of an intensive care facility. Pepsinogen (PG) is the zymogen of the major gastric protease pepsin. Multiple isoforms of PG have been purified from the gastric mucosa of human beings and various other species. These PG isoforms differ in their electrochemical, enzymatic, and immunological behavior. In human beings serum concentrations of PG A and PG C are used as markers for specific gastric disorders. The aim of this study was to purify and partially characterize fPG as a prelude to the development of immunoassays to measure pepsinogen concentrations in cat serum.

A crude extract was prepared from gastric mucosa of cats sacrificed for unrelated research projects. PGs were purified by weak anion-exchange chromatography, size-exclusion chromatography, and strong anion-exchange chromatography. Purified isoenzymes were partially characterized by determination of molecular weights using sodium dodecyl sulfate-polyacrylamide gel electrophoresis, isoelectric points (IEP) by polyacrylamide gel electrophoresis, and the inhibitory effect of pepstatin on the activated pepsin isoforms. The N-terminal amino acid sequences of the first 25 residues were determined from 5 isoforms by the Edman degradation method. Antibodies against one of the isoenzymes of fPG were raised in sheep. Immunological cross reactivity of all purified fPG isoenzymes was tested by immunodiffusion.

At least 8 different isoenzymes of fPG A were purified, all with IEPs below 3.0 and molecular weights ranging from 37,000 to 44,820. The proteolytic activity of all isoforms was inhibited completely by pepstatin in a range from equimolar to 5-fold molar excess. The N-terminal amino acid sequence of the first 25 residues of the predominant fPG A isoform was Thr-Ile-Ile-Lys-Val-Pro-Leu-Ile-Lys-Lys-Lys-Lys-Thr-Leu-Leu-Glu-Asn-Leu-Ile-Glu-His-Gly-Leu-Leu-Asp-Asp. The sequences of 4 other isoforms showed almost complete homology. All showed a high degree of homology with PG A of other species. One isoenzyme did not show immunological cross reactivity with anti-fPG A antibodies and based on biochemical characteristics is most likely PG B (Cathepsin E).

Two different groups of fPGs were purified. PG A was the predominant group in cats. The second group consisted of one isoform that was immunologically and biochemically distinct from isoenzymes of the fPG A group. PG C could not be found and may not be present in cats. Results of this study will facilitate future development of immunoassays to measure the concentration of feline PGs in serum. There is growing evidence that Helicobacter spp. are associated with inflammation and neoplasia of the liver and biliary tract of rodents and people. The objective of this study was to evaluate archived hepatic tissue samples from cats with inflammatory liver or biliary disease for the presence or absence of Helicobacter spp.

The pathology database was searched for ''cats and cholangiohepatitis (Ch) or cholangitis'' for the period 1992-2001. Cats with a recent history of antibiotic therapy were excluded. Tissue blocks from 32 cats (16F, 16M, aged 2 to 17 yrs), 29 with cholangiohepatitis and 3 with cholangitis yielding a total of 39 liver, 2 bile duct and 5 gallblader samples were identified. Tissue blocks from a group of 19 cats (5F 14M, aged .33 to 10.5 yrs) with non-inflammatory liver disease (10 PSVA, 3 lipidosis) or normal livers were used as a control group. Tissue sections were deparaffinized in xylene, dehydrated in ethanol, and DNA extracted in accord with the Qiagen minikit. Gastric tissue blocks from Helicobacter infected and uninfected cats were used as method controls. PCR of hepatic and gastric DNA was performed using two sets of Helicobacter genus specific primers (16srRNA) designed to yield amplicons of 280 and 400bp. Positive PCR samples were gel purified, sequenced (ABI 3700) and blasted against the NCBI database. Hepatic tissue from infected cats was further analyzed by Steiner's stain and immunocytochemistry with a monoclonal antibody that cross-reacts with a variety of Helicobacter spp.

PCR of gastric tissue samples yielded sequences with close homology (99%) to H. heilmannii, bizzozeronii, felis and salomonis confirming that the extraction and amplification procedures were effective. Positive PCR results of liver samples were obtained from 2 cats with Ch, and 1 cat with PSVA. Sequence analysis indicated homology with H. nemestrinae/pylori (100%) in one Ch cat, and H. bilis (100%) in the PSVA cat. Two amplicons of different sizes (280 and 600bp), persistently detected in the third cat, were consistent with H. nemestrinae/pylori (100%) and H. fenelliae/cinaedii (100%). No Helicobacter-like organisms were identified by Steiner stain or immunocytochemistry.

This study has identified Helicobacter DNA in formalin-fixed paraffin embedded liver samples from 2 cats with cholangiohepatitis and a cat with a PSVA. The Helicobacter sequences identified are consistent with those associated with liver disease in other species. Further, prospective studies are warranted to elucidate the role of Helicobacter in liver disease in cats. Fecal alpha 1 -proteinase inhibitor (alpha 1 -PI) is a protein present in plasma, interstitial fluid and lymph. Like albumin, it is lost into the gastrointestinal tract in protein losing gastroenteropathy, but unlike albumin it remains essentially immunologically intact in feces and can be measured by immunoassay. The effect of collection of fecal samples by rectal palpation on alpha 1 -PI concentrations is unknown. We hypothesized that fecal alpha 1 -PI concentration would not be different in samples obtained via rectal palpation compared to those obtained via spontaneous defecation.

Fifteen pet dogs were studied. Samples were collected from three fresh spontaneous defecations from each dog. After these were collected, three more samples were obtained from each dog via rectal palpation using a latex exam glove and a small amount of sterile lubricant. Prior to collecting the first rectal sample, each dog was given a standard digital rectal examination. The samples were obtained over 24-36 hours at no less than 6 hour intervals. Samples were immediately frozen at Ϫ20ЊC, stored frozen at Ϫ20ЊC, and assayed for alpha 1 -PI by immunoassay. Samples of sterile lubricant and glove powder served as negative controls. All samples from each dog were assayed during the same assay run.

Mean and standard deviation were calculated for each group of three values. These means represented each dog for the remaining analysis. Concentrations of alpha 1 -PI in spontaneous samples were compared to rectally obtained samples using a Wilcoxon signed rank test.

The median of the 15 mean alpha 1 -PI concentrations was 4.23 mcg/g for spontaneous samples (range 0.83-21.97 mcg/g) and 9.20 mcg/g for rectally obtained samples (range 2.3-44.57 mcg/g). Thirteen of 15 dogs had higher mean alpha 1 -PI concentrations in samples obtained rectally. Of those 13 dogs, 7 had mean concentrations within the reference range for spontaneous samples but above this range in samples obtained rectally. The mean alpha 1 -PI concentration in rectally obtained samples was significantly different from spontaneous samples (p ϭ 0.0016).

We conclude that samples obtained via rectal palpation should not be used for alpha 1 -PI analysis. The higher concentrations observed in rectally obtained samples probably reflect trauma to the rectal mucosa causing leakage of alpha 1 -PI. Acute pancreatic necrosis (APN) is recognized as a feline gastrointestinal disorder of significant morbidity and mortality, although ante-mortem diagnosis continues to be a difficult challenge. The aim of the study was to review the ultrasound findings of cats with necropsy-and histology-confirmed APN, and to analyze factors contributing to the apparent inability of ultrasound to consistently diagnose pancreatic necrosis ante-mortem.

Cats were identified from pathology records between Jan. 1994 and June 2001. Cats were included in the study if necropsy and histopathology confirmed APN, clinical history and laboratory data were consistent with APN, complete abdominal ultrasound examination was available for review, and Ͻ72 hours had elapsed between the last ultrasound exam and necropsy. Cats with suppurative and chronic, non-suppurative pancreatitis, pancreatic neoplasia, and pancreatic nodular hyperplasia were excluded from the study. Gross pathologic, histopathologic, radiographic and ultrasonographic results were reviewed. Anatomic localization of APN was determined from the gross pathology report; duration, severity, and location were determined from review of histopathologic specimens. The final ultrasound report and the ultrasonographic permanent images were reviewed.

The pancreas was considered ultrasonographically normal in 10 cats, compatible with pancreatitis in 7 cats and not observed in 3 cats. The pancreatitis in the 7 cats diagnosed using ultrasound was multifocal in all cats on gross pathology; the majority were acute or subacute (n ϭ 5) with severe or moderate necrosis (n ϭ 6) on histopathology. The 13 cats with a normal appearing or unobserved pancreas on ultrasound were mostly multifocal (n ϭ 8) but others were focal (n ϭ 2) or normal (n ϭ 3) on gross pathology. The pancreatitis in a majority of these cats was peracute or acute (n ϭ 11) and although most had severe or moderate necrosis (n ϭ 8), the remaining had mild or minimal necrosis on histopathology. Thoracic and abdominal radiographic findings were non-specific. There was no significant difference in experience level between ultrasonologists who did (median, 2.5 years) or did not (median, 2.5 years) diagnose APN.

The ultrasonographic diagnosis of feline APN using established criteria was made in only 35% of 20 cats with confirmed APN. We conclude that: (1) the low sensitivity of ultrasound in diagnosing feline APN may be due to inherent subtleties in feline pancreatic parenchymal changes and/or inappropriate ultrasound criteria, and, (2) new ultrasonographic criteria must be established if abdominal ultrasound is to be an effective diagnostic tool in cats with pancreatitis.

BIOTICS. J Scott Weese, Maureen Anderson. Dept of Clinical Studies, Ontario Veterinary College, University of Guelph, Guelph, Ontario, Canada.

Probiotic therapy is increasing in popularity in veterinary medicine and a number of probiotics are available commercially. However, probiotics are considered food supplements, not drugs, and are not regulated. The goal of this study was to evaluate the contents of commercially available probiotics.

Seventeen commercially available probiotics were tested. Eight were labeled for use in people while 9 were intended for use in animals. All products were stored as per label recommendations and used prior to the expiry date. Serial dilutions were made in phosphate buffered saline (pH 7.2). Aliquots were inoculated onto MRS agar and incubated anaerobically for 72 hours at 37ЊC for lactic acid bacteria isolation. Aliquots were also inoculated onto blood agar incubated both aerobically and anaerobically. Colonies were counted and identified via Gram stain morphology, catalase reaction and biochemical characteristics using API 20 Strep and API 50CHL identification tests (BIOMERIEUX SA, St. Laurent, Quebec) .

Overall growth ranged from 3 ϫ 10 3 to 4.3 ϫ 10 10 CFU/g. 6 products (5 veterinary) did not list specific organisms on their label. 5 products, all veterinary, did not state numbers of organisms that were supposed to be present. Of the products containing a label claim of number of viable organisms, the actual numbers ranged from Ͻ0.001 to 215% of label claim. The best veterinary product contained only 1.8% of its label claim. Only 3 products, all intended for human use, contained the organisms and numbers stated on their labels. One or more of the listed species were not detected in 9/13 (69%) products that listed their contents, while additional organisms were detected in 7/13 (54%) products. Two products contained Staphylococcus spp. which have no probiotic effects and were likely contaminants from processing. Two products contained Enterococcus avium which is regarded to be a pathogen in people. Four products claimed to contain Bifidobacterium spp. yet none were isolated. Only 1 product, marketed for people, contained a probiotic strain with specific in vitro and in vivo efficacy testing.

Overall, the quality of commercial probiotics, particularly those intended for veterinary species appears to be poor. It is unclear whether this is due to improper formulation, loss of viability during processing or loss of viability during storage. Regardless of the quality control of these products, probiotic effects are strainspecific and likely vary between host-species, so unless specific testing of the included isolates on the intended host species is available, there is no indication whether treatment will be effective. Delayed gastric emptying is involved in the pathogenesis of many conditions in the horse and may be difficult to diagnose clinically. The 13 C-octanoic acid breath test ( 13 C-OABT), a non-invasive stable isotope technique, has been shown to correlate closely to radioscintigraphy for the measurement of gastric emptying indices in healthy horses. The purpose of this study was to determine whether this novel diagnostic modality was still valid for horses in which gastric emptying was delayed by the prior administration of low-dose atropine.

Gastric emptying rate was measured in eight healthy horses on two occasions after administration of either intravenous atropine 0.035 mg/kg or saline in random order (Texas A&M Animal Use Protocol 2001-35) . Gastric emptying indices were determined by 13 C-OABT on each occasion and also by concurrent radioscintigraphy following atropine administration. The same standard test meal of oats and bran was used on each occasion with the 13 C and 99m technetium tracers contained in baked egg yolk. Immediately after consumption of the meal, saline or atropine was given IV. Serial gastric scintigraphs and/or expired breath samples were then collected at 15-minute intervals for up to 12 hours. After processing both data sets, non-linear least squares regression analysis was used to determine the gastric half-emptying time (t1/2), duration of the lag phase (tlag) and the gastric emptying coefficient (GEC) for each technique.

No ill effects were observed following atropine administration, but paired ttests on the breath test data confirmed that the dose used here delayed t1/2 significantly (P Ͻ 0.001), with no overlap in the 99% C.I. range. A Wilcoxon rank sum test also showed that scintigraphic t1/2 after atropine administration in this study was delayed compared to our previous measurements on individuals in the same population (P Ͻ 0.01).

Following atropine administration, mean (s.d.) scintigraphic and breath test t1/ 2 values were 4.75 (1.60) and 6.76 (1.65) hours respectively. Significant correlations were found between the two techniques for t1/2 (P Ͻ 0.01) and tlag (P Ͻ 0.05). In addition, breath test GEC was correlated significantly to scintigraphic t1/2 (P Ͻ 0.01).

It is concluded that the 13 C-OABT is an effective diagnostic modality for the measurement of delayed gastric emptying, with significant correlation to the standard method of radioscintigraphy. As it is non-invasive, not radioactive and requires minimal equipment or expertise, it offers several advantages to existing techniques for clinical use. Probiotic therapy is becoming popular in veterinary medicine, however probiotics have not been shown to be effective for the treatment or prevention of any condition in horses. This may be due, in part, to a lack of proper in vitro testing of potential probiotic isolates. Further, probiotic organisms of equineorigin may be preferable for use in horses, as equine-origin organisms may be better adapted to survival in the equine gastrointestinal tract than organisms isolated from other species. This study was designed to screen equine fecal samples for lactic acid bacteria (LAB) and evaluate them for beneficial in vitro properties. These included acid and bile resistance, aerotolerance and antimicrobial factor production.

Forty-seven LAB were isolated from healthy horses and identified via Gram stain, catalase reaction and biochemical characteristics using the API 50CHL identification system (Biomerieux, St.Laurent, Quebec) . Acid and bile tolerance were evaluated via spectrophotometric determination of bacterial growth following inoculation into MRS broth containing varying levels of HCl and bile. Aerotolerance was assessed by incubating inoculated culture plates under aerobic conditions. Sterile bacterial supernatant was prepared by growing isolates in MRS broth for 48 hours, then filter sterilizing the supernatant with a 0.2 um syringe filter. The inhibitory effect of this supernatant on Salmonella growth was assessed with pH adjusted MRS broth as a control to exclude inhibition due to production of organic acids. The inhibitory factor was further characterized using a variety of techniques.

Two isolates tentatively identified as Lactobacillus salivarius WE5 and L. plantarum WE7 were demonstrated to be adequately acid and bile resistant, aerotolerant and produced an antimicrobial factor inhibitory against Salmonella isolates of equine origin. The antimicrobial factors produced by both isolates were markedly inhibitory against 3/3 Salmonella isolates, but were not self-inhibitory. Inhibitory activity was persistent for at least 1 month of storage at Ϫ70ЊC, but began to decline after 2 weeks of storage at 20ЊC and 4ЊC. The inhibitory factor produced by both isolates was determined to be bacteriostatic based on pre-and post-incubation quantitative culture. Inhibitory factor production was evident by 6 hrs of incubation of isolates and reached a plateau at 48 hrs. Further characterization of the antimicrobial factor(s) is required. This is the first report documenting in vitro probiotic properties by equine-origin lactic acid bacteria. Further in vitro and in vivo evaluation of these organisms, which may have promise for the prevention and/or treatment of equine salmonellosis, is warranted. Suppurative inflammation involving multiple joints is a recognized clinical syndrome in the dog. Chronic antigenic stimulation and immune-complex deposition are believed to be the underlying pathogenesis of this syndrome, but the syndrome has been otherwise poorly characterized.

Medical records of dogs with a diagnosis of polyarthritis were evaluated retrospectively (1995) (1996) (1997) (1998) (1999) (2000) for historical, physical examination, and clinicopathologic findings. Dogs were included in the study if they had cytologic evidence of suppurative inflammation in 2 or more joints. Suppurative inflammation was defined as joint fluid containing nucleated cell counts Ͼ3,000/microliter (or Ͼ2/ 400ϫ field) and/or neutrophils comprising Ͼ12% of the cells. Cases were excluded if bacterial culture yielded growth in any joint.

Fifty-nine cases met the criteria for inclusion, and 26 breeds were represented. Male and female dogs were equally represented. The average age was 4.9 years. Lameness or stiff gait was the primary complaint in 35/59 dogs (59%), while weakness or inability to rise was the primary complaint in 15 (25%) dogs. Signs of systemic illness included lethargy (43/59,73%), anorexia (32/59,54%), weight loss (10/59,17%) , diarrhea (9/59,15%), and vomiting (7/59,12%). Fever (Ն103ЊF) was reported in 34/59 dogs (58%). Joint involvement was frequently evident during physical examination with dogs exhibiting lameness (46/59,78%), joint swelling (45/59,76%), or joint pain (41/59,69%). Routine laboratory data were obtained in 58/59 dogs (98%), and the most consistent findings were anemia (22/58,37%) and leukocytosis (15/58,25%). One or more joints were radiographed in 42 dogs. Joint effusion was visible in 27 dogs (64%), and erosive changes were reported in 1 dog (2%). Chest (n ϭ 27) and abdominal (n ϭ 10) radiographs and abdominal ultrasound (n ϭ 22) revealed bronchopneumonia (1), prostatomegaly (1) and intrabdominal lymphadenopathy (7). ELISA testing for Borrelia burgerdorferi antibody was positive in 29/49 (59%) dogs tested. Western blot analysis for Borrelia antigen performed in 17 dogs indicated natural infection in 9 dogs (53%), and combined natural infection and vaccinal antibody in 6 dogs (35%). Additionally, polyarthritis was associated with systemic infection (9/59,15%), gastrointestinal disease (2/59,3%), neoplasia (1/ 59,2%), lupus (1/59,2%), or no identifiable underlying disease (20/59,34%).

We conclude that: (1) while some cases of canine suppurative polyarthropathy are associated with positive Lyme Western blots, most cases are not associated with underlying infectious or other systemic disease; and, (2) a small number of suppurative polyarthritis cases are associated with neoplasia, G.I. disease, or lupus.

LS Garosi 1 , SR Platt 1 , GD Shelton 2 . 1 The Animal Health Trust, Newmarket, England, 2 University of California, San Diego, La Jolla, California.

A syndrome of exercise-induced muscle hypertonicity was described several years ago in young Cavalier King Charles Spaniels (CKCS) of common ancestry from the United Kingdom. Recently, 3 CKCS with very similar clinical presentations were evaluated in the UK and 4 dogs were identified in the USA. Age at presentation ranged from 4 to 7 months and both males and females were affected. A typical history included exercise and excitement-induced ''collapse'' which was preceded by a ''deer-stalking'' action with the head held close to the ground and the bottom high in the air. An increase in extensor tone of the muscles of all four limbs was evident prior to falling over and during the period of collapse. Mentation was not abnormal throughout the episodes.

Hereditary hyperekplexia (also called startle disease), a syndrome described in humans, has a similar clinical presentation in which hypertonia may be elicited by unexpected auditory, visual, or sensory stimuli. Similar to the falling over that occurs in the CKCS, these attacks in humans may result in unchecked falling despite complete retention of consciousness. Hyperekplexia in humans responds dramatically although not completely to the benzodiazepine drug clonazepam, which enhances gamma-aminobutyric acid (GABA) neurotransmission. Low levels of CSF GABA have been reported in human patients. Following extensive neurological and neuromuscular evaluations that resulted in no specific abnormalities, a treatment trial with oral clonazepam (0.5 mg/kg TID) was performed in 2 affected CKCS dogs in the UK and 2 dogs in the USA.

All 4 dogs responded dramatically to clonazepam therapy. Episodes of hypertonicity decreased from 25-30 per week to 1 every 2-3 months. After 2 years of therapy with clonazepam, the 2 dogs from the USA were described as clinically normal with only infrequent episodes. Although the inciting stimuli differ, the response to clonazepam therapy suggests a similar underlying disorder of GABA neurotransmission in the CKCS dogs and human hyperekplexia patients. We have explored the capability of large hydrophilic polymers to seal breaches in damaged axolemmas. In particular polyethylene glycol (PEG) has been shown to anatomically and electrophysiologically seal and recover the conduction in guinea pig spinal cord injury models. Here we test the ability of PEG to recover function in neurologically complete, deep pain negative clinical cases of paraplegia in dogs.

Twenty cases were admitted to the veterinary teaching hospitals of Texas A&M University and Purdue University within less than 48 hours of onset of paraplegia. The neurological status of the dogs was determined by scoring of pain perception, proprioception and ambulation at time 0, 3 days, 1 week and 6 weeks post-surgery. In addition to a hemilaminectomy procedure and methylprednisolone sodium succinate (30mg/kg IV), we administered PEG (ϳ 3500 Daltons w/w in sterile saline; 2ml/kg IV) just prior to surgery and 24 hours later as well as a topical application of PEG (ϳ 1450 Daltons; for 2 minutes) to the durotomy site. These dogs were compared to 23 published historical controls treated in an identical manner except for PEG application. An additional 12 control dogs were available for the 3 day comparison time.

Of the 35 control animals none showed any measurable recovery by 3 days, when 4 of the 18 evaluated PEG treated animals had recovered pain perception (P ϭ 0.01; Fischer's Exact, two-tailed tests). At 6 weeks post-treatment 7 of the 20 PEG treated dogs had recovered proprioception while only 1 control dog recovered proprioception (P ϭ 0.016; Fischer's Exact, two-tailed tests). The highest significance was achieved by comparing ambulation; 15 of the 20 PEG treated and only 6 of the 23 control animals were able to ambulate (P ϭ 0.002; Fischer's Exact, two-tailed tests). At this time 6 of the 15 ambulatory PEG treated dogs (40%) remained deep pain negative, whereas all ambulatory control dogs were deep and superficial pain positive.

We conclude that injection and topical application of PEG shows beneficial effects in functional recovery of neurologically complete paraplegic dogs in the clinical setting. Cerebellar hypoplasia in cats is most commonly caused by an in utero infection with feline parvovirus. In utero viral infections also cause cerebellar hypoplasia in other species including pigs, cattle, sheep, and chickens. Cerebellar hypoplasia has been reported in dogs, but to date no viral etiology has been identified. Parvoviral DNA has been amplified previously using the polymerase chain reaction (PCR) from fecal material and from a variety of fresh and fixed tissues including mesenteric lymph nodes, thymus, and ileum. We have now amplified parvoviral DNA extracted from fresh brain tissue from 1 cat with cerebellar hypoplasia, from archival brain tissue from 8 cats with cerebellar hypoplasia, and from archival brain tissue from 2 out of 5 dogs with various congenital cerebellar lesions.

DNA was extracted from archival, paraffin-embedded cerebellar tissue from 7 dogs and 8 cats with cerebellar hypoplasia and from freshly frozen brain tissue from 1 cat with cerebellar hypoplasia. Histopathologic examination of the brains from all cats was consistent with in utero infection with parvovirus, but no etiology was identified for any of the affected dogs. The DNA extracted from each brain was first subject to polymerase chain reaction (PCR) amplification using a primer pair for the canine and feline histone (H3.3) gene as a positive control for DNA integrity. Three independent primer pairs specific for parvoviral DNA were designed and tested on each sample.

All 16 samples produced a histone PCR amplicon. Parvovirus was successfully amplified from all 9 cats with cerebellar hypoplasia. Two (littermates) out of the 5 dogs tested were positive for parvovirus. Sequence analysis of the PCR products from the two littermates confirmed the identity of canine parvoviral DNA. No parvoviral PCR products were obtained from control DNA samples extracted from the cerebella of 7 cats and 8 dogs with other causes of encephalitis.

This study shows that parvoviral DNA can be amplified from archival and fresh brain tissues from dogs and cats with hypoplastic cerebella. Furthermore, cerebellar hypoplasia may be associated with in utero parvovirus infection in some dogs as occurs in cats with cerebellar hypoplasia. Prosaposin is a precursor of the lysosomal proteins, saposin A, B, C and D. A homologue of a neurotrophic peptide sequence within saposin C, Prosaptide TX14(A) has been shown to treat effectively both peripheral neuropathy and neuropathic pain in several animal models with a high safety margin. The purpose of this study was to evaluate the potential benefit and safety of Prosaptide TX14(A) to treat canine axonal peripheral neuropathy in a multicenter clinical trial study.

Nine large breed dogs (Ͼ 60 kg) diagnosed with idiopathic distal, symmetric peripheral neuropathy had some or all of the following signs: weakness, exercise intolerance, high steppage pelvic limb gait, dysphonia, dyspnea, and muscle wasting. Definitive diagnosis was based on abnormal motor nerve conduction and/or the presence of chronic axonal degeneration in fascicular peroneal or tibial nerve biopsy in all dogs. All dogs were screened normal for underlying metabolic and neoplastic diseases.

An open label, non-blinded exploratory study without the benefit of a placebo control group was conducted. Owners injected Prosaptide TX14(A) 4 mg SC thrice weekly for 6 months. Pre-and post-treatment efficacy measurements included neurological examinations, electrodiagnostic testing, and muscle and nerve biopsies. No dogs suffered from any adverse effects. Of the 7 dogs that completed the trial, 3 dogs showed improved neurologic function, 2 dogs were ambulatory but with no change in neurologic function, 1 dog was ambulatory but with deteriorated neurologic function, and 1 dog was euthanized due to progressive weakness. Comparing pre-and post-treatment nerve biopsy specimens, 1 dog showed continued loss of myelinated nerve fibers with extensive endoneurial fibrosis, 5 dogs showed no appreciable difference from the initial biopsy specimens, and nerve fiber sprouting consistent with attempts at regeneration was seen in 1 dog.

This pilot study demonstrates promising results that Prosaptide TX14(A) may be beneficial in the treatment of canine axonal degenerative polyneuropathy. Future studies are planned to evaluate this treatment in a placebo-controlled, longer term study. Support provided by the Myelos Corporation, San Diego, CA.

ACETYLCHOLINE RECEPTORS IN CANINE DYSAUTONO-MIA. DP O'Brien 1 ; GC Johnson 1 ; J Cooper 2 ; J Lindstrom 2 ; GD Shelton 3 . 1 University of Missouri, Columbia, MO; 2 University of Pennsylvania, Philadelphia, PA; 3 University of California-San Diego, La Jolla, CA.

Canine dysautonomia is characterized by degeneration of autonomic ganglia and failure of autonomic functions. The cause is unknown. Autoantibodies directed against the postsynaptic nicotinic acetylcholine receptor (AChR) in skeletal muscle results in myasthenia gravis in both people and dogs. In ganglia, alpha3 AChRs play a postsynaptic role similar to that of muscle AChRs. Recent studies in humans with dysautonomia have demonstrated antibodies against the nicotinic ganglionic AChR. This study assessed antibodies against ganglionic AChR in serum from dogs with dysautonomia.

Serum samples were collected from 38 dogs diagnosed with dysautonomia and stored at Ϫ70ЊC. Criteria for diagnosis have been previously described (JA-VMA 218(8):1285 -1290 , 2001 . Normal controls were 21 healthy coonhounds. Serum was also analyzed from 5 dogs with other diseases affecting the nervous system or organs affected in dysautonomia. Screening assays for autoantibodies were performed using transfected human embryonic kidney (HEK293) cell lines that stably expressed human alpha3beta2 and alpha3beta4 nicotinic neuronal AChR subtypes. Solubilized AChRs were labeled in triplicate with 125 I-epibatidine and incubated overnight with serum from affected and control dogs. Bound IgG and IgM were precipitated by the addition of fixed Staphlococcal protein A and radioactivity measured in washed pellets. Positive controls included previously characterized monoclonal and polyclonal antibodies against the alpha3 and beta4 subunits. Serum samples that yielded results greater than the mean value ϩ2 SD obtained from the 21 normal control dogs were considered as having antiganglionic AChR antibodies.

Antiganglionic AChR antibodies were identified in 6/38 dogs with dysautonomia using the alpha3beta4 cell line and in 2/38 dogs using the alpha3beta2 cell line. Antibodies were detected in one dog with diffuse GI atony, but no other signs of dysautonomia. Antibodies were not detected in other related disease controls including 2 dogs with gastrointestinal disease, 1 dog with a bladder disorder, and 1 dog with cervical transverse myelopathy. Since human AChRs were used and antibodies can be species specific, repeating this assay with canine AChRs should result in higher levels of antibody detection. The human cell lines had the advantage of permitting identification of receptor subtypes. This study suggests an autoimmune basis for some dogs with dysautonomia. Computed tomography (CT) is the imaging modality of choice for head trauma patients. It provides rapid acquisition of images, superior bone detail, and better visualization of acute hemorrhage than magnetic resonance (MR) imaging. It is also less expensive and more readily available. Computed tomography images are used to determine if the patient would benefit from decompressive surgery. In some cases, surgery may be postponed pending decline in neurological status. The purpose of this study was to evaluate the most common changes seen on CT in head trauma patients, compare CT findings to surgical and necropsy findings for applicable cases, compare CT finding to MR findings if available, and to determine if any CT characteristics affect survival.

All patients that presented to the University of California Davis Veterinary Medical Teaching Hospital for head trauma that had a CT scan between January of 1990 and November of 2001 were included in this retrospective study. Computed tomography characteristics evaluated were: cranial cavity fractures (compressive/noncompressive), other skull fractures, intracranial hemorrhage, focal cerebral edema, diffuse brain swelling, extra-axial hematoma, and mass effect. These findings were compared to MR imaging, surgical findings and necropsy findings.

Traumatic events included hit by car (16), falling (4), hit with blunt object (3), dog bite (2), gunshot wound (1), rammed by goat (1), kicked by horse (1), run over by large dog (1), submerged by wave (1), and stepped on (1). There were 19 cranial cavity fractures (13 non-compressive fractures, 6 compressive fractures), skull fractures without cranial cavity involvement (4), evidence of intracranial hemorrhage (9), focal cerebral edema (9), diffuse brain swelling (2), extra-axial hematoma formation (3), and mass effect (7). Two of the 31 CT scans were unremarkable. When possible, CT findings were compared with MR findings (4), as well as surgical (4) and necropsy findings (8). Four patients had subdural hematomas, and two had epidural hematomas found at surgery or necropsy that were not seen on CT.

Evaluation of brain parenchyma changes on CT was very poor when compared to actual changes seen at necropsy. Meningeal enhancement and mass effect were the most common findings on MR imaging that were not appreciated on CT. Eight of 19 patients with cranial cavity fractures died or were euthanized. Four of 12 patients without cranial cavity fractures died or were euthanized. There did not appear to be any characteristic CT finding that correlated with survival and CT findings should not be used to determine prognosis in patients with head trauma. Diffusely infiltrating astrocytomas are a group of astrocytic tumors that may be classified on cytological features. The most malignant is the glioblastoma multiforme. There has been no attempt in the veterinary literature to classify astrocytomas using similar grading schemes although glioblastoma multiforme has been described in dogs. The purpose of this report is to describe the MR imaging features and histopathological characteristics of glioblastoma multiforme in five dogs and to compare these findings with those of people.

The MR characteristics, histological and immunocytochemical features of five dogs with progressive signs of CNS dysfunction, and MR imaging features of diffuse astrocytoma, were retrospectively reviewed. Magnetic resonance imaging sequences obtained were spin echo T1 weighted, T2 and proton density (PD) weighted and T1 weighted images after the administration of gadolinium-diethylenetriaminepenta-acetic acid (DTPA). Immunolabeling was performed for GFAP, MIB-1, cytokeratin, VEGF, PDGF-A and apoptosis (TUNEL).

Many of the MR characteristics described for glioblastoma in people were present in the 5 dogs in this series. All of the lesions were isointense to hypointense on T1 weighted, and hyperintense on T2 weighted images with significant edema and mass effect. Two dogs had diffuse heterogeneous contrast enhancement while two dogs had ring enhancement. Necrosis was present in all five tumors corresponding to the areas of ring enhancement in 2 dogs, and small nonenhancing foci in the other dogs. The major histological features were a highly cellular pleomorphic anaplastic astrocytic tumor cell population with mitotic activity. Throughout the tumors were prominent areas of irregular branching, serpiginous, acute necrosis bordered by palisading glial cells. There were other more extensive areas of acute coagulative necrosis commonly at the periphery of the tumor or near necrotic foci. There was also a very prominent microvascular proliferation mostly bordering areas of necrosis or the tumor margins. Two of the tumors also had some oligodendroglial like differentiation. All of the tumors were GFAP and TUNEL positive and MIB-1 labeling demonstrated a proliferative index of 12-25%.

We demonstrated that canine and human glioblastomas share similar MR imaging and histological features. The dog may be a good model for studying glioblastoma multiforme in people. A large number of single gene disorders affect the central nervous system. Somatic gene transfer has the potential to transfer a normal copy of the defective gene into a patient's own cells. Delivery of the gene directly into the brain will likely be necessary to circumvent the blood brain barrier. A number of methods have been developed to transfer genes into the mammalian brain, but achieving stable expression of therapeutic levels of the gene product remains a major problem.

Adeno-associated viral (AAV) vectors are capable of expressing a therapeutic gene for long periods of time in the murine brain. In mutant mice with a betaglucuronidase (GUSB) deficiency, the transfer of the GUSB gene using an AAV vector to the mouse brain resulted in the widespread diffusion of the secreted GUSB enzyme resulting in correction of pathology in the brain. However, the mouse brain is approximately 100 times smaller than the cat brain and approximately 1000 times smaller than the brain of the human neonate. In this study, we evaluated the use of AAV to achieve widespread enzyme expression in the normal cat brain. GUSB was used as a marker gene due to our ability to monitor enzyme expression using histochemical methods performed on frozen brain sections.

AAV vectors contained the human GUSB transgene driven by the human GUSB promoter (the HßH genome). The HßH genome was packaged into AAV vectors and a viral titer of 4.5 ϫ 10 12 particles/ml was used in each experiment. MRI was used to identify the brain structures of interest and 2 uL of vector was injected into each structure while the cat was anesthetized and stabilized in a stereotaxic head holder. The AAV-HßH vector was injected into several major gray matter regions (cerebral cortex, caudate nucleus, thalamus, hypothalamus, hippocampus, and cerebellar gray matter) and white matter regions (corona radiata, internal capsule, centrum semiovale, and cerebellar white matter) of 8week-old cats. Cats recovered without difficulty following surgery. Cats were euthanized at 18 weeks of age. Endogenous feline GUSB expression was heatinactived and the brains were evaluated for gene expression using enzyme histochemistry and in situ hybridization. AAV vectors resulted in significant transduction of the brain in both gray and white matter regions. This study will serve as baseline data in our continuing studies to treat lysosomal storage diseases in the cat using gene replacement therapy. Congenital hydrocephalus is a common finding in certain small breeds of dog, and can be diagnosed using transcranial ultrasonography. However, the severity of structural changes and the severity of clinical signs do not always correlate. Transcranial Doppler ultrasonography (TCDU) can be used to measure the resistance index [RI ϭ (systolic velocity-diastolic velocity)/systolic velocity] of intracranial blood flow, and intracranial RI is directly correlated to intracranial pressure. The aim of this study was to determine whether there was a correlation between basilar artery RI, neurologic status and/or severity of ventriculomegaly in dogs with congenital hydrocephalus.

Transcranial Doppler ultrasonography was performed in 47 small breed dogs. All dogs underwent a physical and neurologic examination and other appropriate testing in order to establish a diagnosis. Severity of neurologic signs was scored on a scale from 0-3. The severity of ventriculomegaly was determined from the ratio of ventricular height to cerebral mantle thickness according to previously published guidelines. Basilar artery RI was measured in all dogs. Six dogs were evaluated on more than one occasion following changes in neurologic status. Based on clinical and ultrasonographic findings, dogs were divided into 4 groups: I (n ϭ 8): normal controls; II (n ϭ 11): asymptomatic hydrocephalus; III (n ϭ 14): symptomatic hydrocephalus; IV (n ϭ 14): other intracranial disease. The mean RI and degree of ventriculomegaly were determined for each group and compared using one way analysis of variance. The correlation between neurologic score and RI, and neurologic score and ventriculomegaly were evaluated using Spearman rank correlation.

Ventriculomegaly was significantly higher in group III than the other groups, and in group II when compared to groups I and IV. Basilar artery RI was significantly higher in both groups III and IV when compared to groups I and II. Neurologic score was significantly correlated with RI and with ventriculomegaly. Changes in RI on serial measurements in six individual dogs reflected changes in neurologic status but there was no accompanying change in ventricular size.

We conclude that basilar artery RI is correlated to neurologic status in dogs with congenital hydrocephalus. Measurement of basilar artery RI is non-invasive and relatively inexpensive. Serial RI measurements could potentially be used to evaluate the response of a patient to treatment, and to further investigate the pathogenesis of congenital hydrocephalus. Spinal cord dysfunction due to spinal arachnoid cysts (SAC) has been reported in dogs. This retrospective study reviews the clinical signs, radiographic findings, and outcome of surgical intervention, in 14 dogs diagnosed with SAC.

The medical records of 14 dogs with confirmed SAC that had been subjected to surgical resection were reviewed. All dogs had been admitted to the Veterinary Medical Teaching Hospital, University of California, Davis, between January 1995 and January 2001.

Non-contrast vertebral radiographs, cerebrospinal fluid analysis, and myelography were done in all dogs. Computed tomography was done in 7 dogs, and magnetic resonance imaging in 3 dogs. Affected dogs were between 1 and 12 years of age and 8/14 were Rottweilers. Abnormalities detected on neurological examination depended on the location of the SAC. Three dogs had multiple SAC and 2 dogs had bilobed SAC. SAC were located in the cervical spine in 10 dogs and in the thoracolumbar spine in 5 dogs. In addition to the SAC, spinal cord compression secondary to intervertebral disc extrusion/protrusion was demonstrated at the same site in 2 dogs. The spinal fluid from 5 dogs showed a mild mononuclear reactivity and increase in protein content. Spinal fluid from 2 dogs showed a mild mononuclear reactivity only, and from 6 dogs an increase in protein content only. Surgical resection of the SAC was done in all dogs. Eleven dogs were available for follow-up. Six dogs showed improvement in neurological function, but had residual ataxia and paresis, 1 dog was neurologically normal 5 weeks post-operatively, and 4 dogs had a poor outcome.

It was concluded from this study that myelography, CT and MR imaging are effective in localizing SAC, and that surgical resection results in a reasonable outcome in a majority of dogs with SAC, however neurological deficits may persist. It is further noted that SAC may be multiple or bilobed. There was a higher incidence of SAC in Rottweilers in this series (PϽ0.0001). I-Cell disease (Mucolipidosis II), a lysosomal storage disease due to a deficiency of the enzyme GlcNAc-phosphotransferase, has been described in humans and one DSH cat. We report on the clinical signs, biochemical studies, radiologic findings, fundic examination, and pedigree analysis from 25 affected kittens related to the original cat, which is the only animal model of this human genetic disease. Results were compared to age-matched controls.

Clinical presentation included failure to thrive, a flat broad face with frontal bossing, and hypertelorism. Affected cats appeared dull, ataxic, and had decreased muscle tone. Neurologic examination revealed subtle signs believed to be secondary to the orthopedic changes. Preliminary serum IgG, IgM and IgA levels were low. Studies of four lysosomal enzymes revealed elevated activity in the serum, and very low levels in the fibroblasts. Inclusion bodies were seen in cultured fibroblasts, but not in white blood cells, and the urine MPS spot test was negative. Pedigree analysis confirmed the autosomal recessive mode of inheritance.

Radiologic facial and skeletal abnormalities were present at the first 2 weeks of life. Lesions of the long bones included metaphyseal flaring, radial bowing, and luxation of the antebrachial-carpal joint. Delayed mineralization, hip luxation, and fusion of the cervical and lumbar vertebral bodies developed over the first 5 months of life. Other skeletal abnormalities, such as spina bifida and hemivertebrae were also seen in severely affected cats. Intrathoracic and intraabdominal radiographs were normal.

Fundic examination revealed apparently normal development until 2.5 months of age, after which time evidence of early retinal degeneration was noted (altered tapetal reflectivity initially, followed by retinal vascular attenuation). These changes progressed to end stage retinal degeneration and complete blindness by 4 months of age. One cat also had incipient cataracts.

Because of the severe nature of the disease, affected cats died or were euthanized within one day to 7 months.

Except for the retinal changes, the clinical features, enzyme, radiographic, and pathologic findings, as well as the mode of inheritance are homologous to the I-Cell disease in humans. There are several publications about disease processes affecting the spinal cord of cats, but none provides a complete accounting of what diseases can occur or how common each diagnosis is. Also, it is not known how age, progression of disease, or concomitant clinical signs may be used to reach a definitive diagnosis.

With this as our objective we performed a retrospective study using case material from the biopsy and necropsy service of the School of Veterinary Medicine at the University of Pennsylvania from 1986 to July 2001. We identified 208 cases with a definitive diagnosis, 61 of them were samples of spinal cord submitted to the biopsy service. The cases were divided into 6 groups based on the disease process. The clinical records were reviewed to correlate signalment, history and clinical findings with the disease process.

The group of spinal cord lesions induced by inflammatory diseases represents 31% of the cases (65/208). In 33 of these cases the cause was FIP induced myelitis, 25 of these cats died at less than 2 years of age, 26 presented with systemic signs. Of 29 spinal cords examined at the autopsy, 28 had involvement of the cervical spinal cord and when the brain was available for examination, it was invariably affected. Other infectious causes were bacterial or suspected bacterial in 10 cases, C. neoformans in 6 cases and T.gondii in 2 cases. There were also 5 polioencephalomyelitis, 4 eosinophilic/histiocytic meningitis and 5 meningitis and myelitis of unknown cause.

In 28% of the cases (58/208) a neoplastic process affected the spinal cord or the tissues surrounding the cord. In 22 cases the final diagnosis was lymphosarcoma, 6 cats had other intramedullary neoplasms, in 17 cases there was a vertebral column neoplasm with compression of the spinal cord, in 7 meningeal neoplasia and in 6 other extradural neoplasia. In 13% of the cases (28/208) direct trauma or trauma secondary to a vertebral column injury or intervertebral disc herniation affected the spinal cord. In 11% of the cases (24/208) a congenital or inherited disease of the spinal cord or vertebral column affected young cats. In 9% of the cases (19/208) ischemic damage, infarcts or focal malacia of unknown cause affected the spinal cord. Among the 15 cats with spinal cord disease associated with vascular accident seven were 10 year-old or older. Finally 7% of the cases (14/208) had degenerative diseases of the spinal cord with a majority (8/14) of neuroaxonal dystrophy. 27 L-2-HYDROXYGLUTARIC ACIDURIA IN STAFFORDSHIRE BULL TERRIERS. C. J. Abramson 1 , S. R. Platt 1 , C. Jakobs 2 , R. Dennis 1 , F. McConnell 1 , L. Garosi 1 , G. D. Shelton 3 . 1 The Animal Health Trust, Newmarket, England, 2 Metabolic Unit, Dept of Clinical Chemistry, VU University Medical Center, Amsterdam, The Netherlands, 3 University of California, San Diego, La Jolla, California.

L-2-hydroxyglutaric aciduria is a rare metabolic disorder in humans which produces a slowly progressive neurological syndrome. Onset of clinical signs ranges from 1.5 years to 30 years. Clinical signs are usually associated with intellectual impairment that frequently progresses to dementia. Cerebellar ataxia and seizures have also been reported. We recently identified two Staffordshire Bull Terriers with L-2-hydroxyglutaric aciduria. Signs in one dog included severe lethargy, depression, and seizure activity. Signs in the other were exercise intolerance, stiffness in gait, and abnormal mental status. Age at presentation was 7 months and 23 months respectively. The signs were progressive in both cases, though the seizures were well controlled with phenobarbital therapy.

Complete haematology, serum biochemistry profiles, and thyroid hormone (T4) were normal in both cases. Cerebropinal fluid analyses were also normal. In the patient with exercise intolerance, electromyography and nerve conduction velocities were normal as was a muscle biopsy. Muscle carnitine concentration was abnormally low (total muscle carnitine of 2.80 nmol/mg protein; reference range 12-41 nmol/mg protein).

Brain magnetic resonance imaging in both cases revealed diffuse polioencephalopathy with hyperintensity on T2-weighted images of the cerebral, cerebellar, and brainstem gray matter. T2-weighted image hyperintensity was also noted in the thalamus and cerebellar nuclei. The regions of T2 hyperintensity were mildly hypointense with T1-weighting. There was no abnormal contrast enhancement with intravenous gadolinium.

Fresh urine samples were immediately placed into dry ice and stored in this way until analysis. Organic acid screening with gas chromatography-mass spectroscopy revealed elevations in L-2-hydroxyglutaric acid (2,957 and 1,936 mmol/mol creatinine, human control range ϭ 1.3-18.9 mmol/mol creatinine). Subsequent analysis of serum and cerebrospinal fluid utilizing the same laboratory techniques have shown elevations in L-2-hydroxyglutaric acid in these fluids as well.

The specific metabolic pathway involved in the development of this organic aciduria is unknown, though lysine catabolism has been proposed. Further studies are underway to identify other affected Staffordshire Bull Terriers and to culture cell lines for chemical and metabolic analysis. Future research efforts are proposed to focus on genetic studies and potential treatments. The purpose of this retrospective study was to compare myelography and computed tomographic myelography (CTM) findings in dogs with spinal cord compression. We hypothesized that CTM would most closely match neurological abnormalities and would most accurately predict the degree of spinal cord compression observed at surgery. The study population included 101 dogs who were presented to VMRCVM between January 1992 and October 1997 for clinical signs related to spinal cord dysfunction. Films for each modality were independently reviewed by two board-certified veterinary radiologists. Readers were unaware of the signalment or clinical history. For each dog and each modality, the location of compression relative to the spinal cord was recorded and the severity of spinal cord compression was graded (0 ϭ none, 3 ϭ severe). Diameters of the dural sac were measured at 3 vertebral locations by a third observer who was unaware of radiographic grades. Severity of neurological abnormalities was recorded using a grading scale (0 ϭ normal, 5 ϭ no deep pain). The degree of compression described in surgery reports was also graded (0 ϭ none, 3 ϭ severe). Relationships between severity of clinical signs and the degree of compression noted on myelogram, CTM and surgery were determined using Spearman's correlation. Agreement between the two modalities for compression severity rating and compression measurements was evaluated using scatterplots and agreement plots. There was a significant correlation between the two modalities for estimated severity of spinal cord compression (pϽ0.0001), although the degree of compression recorded with CTM was consistently higher compared to myelogram results. There was no correlation between the severity of neurological abnormalities or the severity of compression observed at surgery and the severity of spinal cord compression with neither imaging modality. A significant correlation was found between severity of neurological abnormalities and degree of spinal cord compression observed at surgery (pϽ0.0005). We conclude that the severity of clinical signs is still the best predictor of the degree of spinal cord compression in dogs. Both myelography and CTM are helpful for identifying the anatomical location of compression, but neither should be used to predict the degree of spinal cord damage. Our choice to use transverse CT images only may have limited our ability to appreciate changes in spinal cord diameter. In the clinical setting, sagittal and dorsal planar reformatted images are recommended. Fort Dodge Animal Health has released a conditionally approved vaccine against S. neurona. A major diagnostic concern is if the vaccine-elicited IgG response affects the ability to use the WB as a diagnostic tool in a vaccinated horse with neurologic disease.

The purpose of this study was several-fold. One goal was to determine if vaccination of horses causes seroconversion. Second, we wanted to determine if vaccination causes at least a transient increase in S. neurona specific IgG in the CSF. Finally, an evaluation of WB banding patterns pre and post vaccination was done to determine whether an IgG response induced by vaccination can be distinguished from that due to natural exposure.

Horses that were Ͼ6 months old with normal neurologic exams were considered for the study (n ϭ 27). The initial serostatus of each horse was determined by WB analysis. Horses that were seronegative were presumed to be CSF negative. Horses that were seropositive had CSF collected, which was submitted for cytology, CSF index and WB analysis. Horses were then vaccinated with the S. neurona vaccine and boostered 3-4 weeks later. On day 14 following the second vaccination, serum and CSF were collected and submitted for WB analysis, CSF index and cytology (CSF only).

Seven of the horses were seronegative by WB at the beginning of the study. While 3 remained negative or showed very weak IgG responses, the remaining 4 had seroconverted after vaccination. Six of these horses were tapped; 2 remained negative and 4 were weakly positive. Most of the initially seropositive horses (n ϭ 20) demonstrated increases in S. neurona specific serum IgG post-vaccination (80%, n ϭ 16). Fourteen of the 20 were also CSF positive initially, with 6 showing borderline immunoreactivity. All 14 demonstrated IgG increases postvaccination. Of the 6 originally CSF WB negative, 4 were retapped postvaccination with 2 testing borderline positive and 2 testing positive. Persistence of the vaccine induced response was not evaluated in the current study, but it would be of interest. Although the study size was small, no immunoreactive bands or banding patterns were noted that could distinguish the IgG response of naturally infected from that of vaccinated horses. This has the potential to confound diagnostic testing for EPM on vaccinated horses, at least on a temporary basis, and should be taken into consideration when electing to vaccinate. Equine infectious anemia virus (EIAV) induces a lifelong persistent infection of horses. Infected horses may undergo acute and chronic stages of disease characterized by recurrent episodes of viremia, fever, and thrombocytopenia, followed by an inapparent carrier stage with no clinical signs. Genetic and antigenic variation are well-recognized as mechanisms of persistence of EIAV and are thought to account for the recurrent nature of the disease. The persistence of antigenic variant virus has significantly hampered development of an effective vaccine for EIAV. Genetic variation within the surface envelope glycoprotein (SU) of EIAV occurs predominantly within 8 defined variable regions (V1 to V8); the third variable region, V3, contains the principal neutralizing domain (PND). The goal of this study was to use infectious molecular clones to better understand the specificity and development of neutralizing antibody during infection.

The virus neutralizing (VN) antibody response was examined in a pony experimentally infected with the virulent, wild-type Wyoming strain of EIAV. This pony underwent recurrent disease episodes followed by a period of clinical quiescence. The SU-V3 region of the inoculum was used to construct an infectious chimeric clone in the backbone of pSPEIAV-19R, an attenuated, laboratory-adapted strain of EIAV. Virus stocks were made from the chimeric virus (EIAVwtV3) and from the parental backbone (EIAV19R). Sera were collected throughout all stages of disease and were tested for the presence of VN antibody to each virus.

Results from the VN assays indicated that the two viruses were antigenically distinct. Neutralizing antibodies to EIAVwtV3 were detected earlier, and achieved maximum titer sooner than neutralizing antibodies to EIAV19R. Additionally, the maximum VN antibody titer to EIAVwtV3 was four-fold higher than that to EIAV19R. The type-specific neutralizing antibody to the inoculating virus appeared shortly after infection. Our results also suggest that a group-specific response develops later and may be of lesser magnitude than the type-specific response.

The results of our study conclusively demonstrate that the V3 region contains type-specific neutralizing epitopes recognized by EIAV-infected horses. Molecular infectious clones will be particularly useful in defining the specificity of virus neutralizing antibody during disease progression. Ongoing studies are using a similar strategy to characterize neutralizing antibody responses to EIAV variants isolated at later stages of infection. Characterizing the specificity of the protective antibody response to EIAV variant virus will provide important insight for development of effective vaccines. Clostridial myositis or clostridial myonecrosis is a rapidly progressive, often fatal condition with liquefactive necrosis of muscle, gas formation, and associated clinical signs and toxemia. A majority of equine cases are a result of intramuscular injection; however, it is not clear whether clostridial organisms are inoculated or are preexisting in the skeletal muscle. This study was designed to determine the presence of dormant clostridial spores in skeletal muscle. Muscle samples were obtained from 28 horses that were euthanatized for reason unrelated to infectious disease, shock or trauma. Aseptic surgical technique was used to obtain 1.5 cm 3 muscle samples from the cervical region. Immediately after the muscle samples were obtained, they were plunged into 70% isopropyl alcohol for 1 minute, placed into a sterile container, and frozen at Ϫ80ЊC until further analyzed. For analysis, muscle samples were anaerobically cultured at 37ЊC in a brain-heart-infusion broth for 4 days. After 4 days, 1 ml of broth/muscle homogenate was mixed in a sterile screw-cap tube with 1 ml of 95% ethanol. The specimen was gently mixed at room temperature for 60 min. The ethanol treated homogenate was inoculated onto blood agar and incubated anaerobically at 37ЊC for up to 5 days. Morphology and growth of colonies was assessed. Each different colony was Gram stained and subcultured onto blood agar and incubated anaerobically. Identification was performed using the API rapid ID 32A biochemical identification test (BIOMERIEUX SA, Marcy-l'Etoile, France).

Nine isolates were cultured from 5/28 horses (18.5%). Two separate isolates of Clostridium sporogenes were isolated from horse 11. Clostridium histolyticum was isolated from horses 10 and 19. Clostridium glycoliticum was isolated from horse 10. Two separate isolates of C. beijernickii/butyricum were isolated from horse 26 and a single isolate of this organism was obtained for horse 18. Clostridium clostridioforme was isolated from horse 18.

Our The purpose of this study was to determine factors that contribute to rhabdomyolysis (Rh) in horses with PSSM. Records of 700 biopsies submitted to the Neuromuscular Diagnostic Laboratory were searched to identify QHs or QH crosses diagnosed with PSSM for at least 1 yr. Seventy six QHs were selected from Ͼ200 PSSM cases. Owners had received written recommendations to gradually increase daily exercise, maximize turnout, feed grass hay and replace grain in the diet with a fat supplement such as 1-4 lbs/day of rice bran. 39/76 PSSM horseowners were available for a telephone interview and 21/76 also provided responses about another healthy horse to form a case control. Gender ratio, temperament and muscle build were similar between PSSM and controls. More PSSM horses (21% vs controls 14%) had a history of respiratory disease and 63% of these PSSM horses developed Rh at that time with muscle wasting, pyrexia and no muscle cramping. Rh began between 3 mo. and 14 yrs of age (mean 4.8 Ϯ 3.6yrs) in PSSM horses and was not present in controls. Clinical signs, in order of frequency, included muscle stiffness and pain (primarily hind limb, back and abdominal muscles), muscle fasiculations, sweating, exercise intolerance, muscle wasting, and colic. Prior to PSSM diagnosis, 87% of PSSM and 95% of controls were fed grain, and 14% of controls (0% PSSM) were fed a fat supplement. Prior to diagnosis, 44% of PSSM horses were getting little to no exercise (14% for controls). After PSSM diagnosis, 82% of owners decreased the amount of grain, added a fat supplement and/or provided more regular and more frequent exercise. Half of the owners complied with all diet and training recommendations resulting in 94% of these horses improving and 83% never having another episode of Rh. Noncompliance of the other 50% of owners, was due to 28% not adjusting the diet, 50% not adjusting training and 22% not adjusting diet and training. In this group, 50% of horses had further episodes of Rh. In all of the cases that did not decrease in severity or frequency either one or both of exercise/diet recommendation were not followed. This study indicates that PSSM can be managed successfully by increasing the fat and decreasing the starch content of the diet combined with a gradual increase in the amount of daily exercise. Diet or exercise regimes alone are not the cause of PSSM since PSSM horses were managed similarly to healthy control horses prior to the time of diagnosis. Thus, there is likely a genetic predisposition to Rh with PSSM that can be influenced by factors such as respiratory disease, dietary starch and carbohydrate content as well as amount of daily exercise. Equine protozoal myeloencephalitis (EPM) is a progressive neurologic disease of horses most commonly caused by infection with the apicomplexan parasite, Sarcocystis neurona. Factors affecting neuroinvasion and neurovirulence have not been determined. We hypothesized that infection of SCID foals or mice (incapable of specific B or T cell responses) with S. neurona would result in fulminant neurologic disease.

Three SCID foals (SCID 1-3) and three immune competent foals (IC 4-6) were infected orally with 5 ϫ 10 5 sporocysts of S. neurona. Venous blood samples were obtained twice weekly for parasite isolation. Buffy coat cells were inoculated on bovine turbinate cells for parasite culture. Foals were euthanatized at post-infection day (PID) 21-60. SCID mice were infected with 10 4 S. neurona sporocysts orally (n ϭ 2) or 10 5 -10 6 merozoites intraperitoneally (n ϭ 4). Gamma-interferon knock-out (GKO) mice and Balb/c mice were similarly infected as positive and negative controls, respectively. Mice were euthanatized when they showed neurologic abnormalities or at PID 45.

Merozoites were isolated from the blood of SCID foal 1 on PID 21. Merozoites were isolated from the blood of SCID foals 2 and 3 on every sampling day after PID 7 (11 isolates) or 8 (5 isolates), respectively. Merozoites were not isolated from the blood of any IC foal on any day. No SCID foals developed neurologic abnormalities; one IC foal developed ataxia and proprioceptive deficits at PID 42. Multiple peripheral organs from SCID foals 2 and 3 were culture or PCR positive for S. neurona; neurologic tissue was negative for S. neurona in all SCID foals. All GKO mice developed neurologic deficits and parasites were isolated from CNS tissue at necropsy. One SCID mouse exhibited unusual behavior that might have been due to neurologic damage; however, parasites were not isolated from the CNS of any SCID or Balb/c mouse.

Despite prolonged parasitemia in SCID foals infected with S. neurona, these foals did not develop neurologic abnormalities and there was no evidence of CNS infection or pathology. The lack of neuroinvasion and neurovirulence in SCID foals and mice may be due to poor sensitivity of assays for detection of parasites and lesions, insufficient time between infection and euthanasia, effective innate immune responses, and/or a critical role of specific immunity in neuropathology. Fifty percent reduction of the dorsal myelographic column (DMC) at a given intervertebral site compared to the corresponding maximum intravertebral DMC is commonly stated as the diagnostic criterion for cervical spinal cord compression. However, this has been reported to be inadequate. A reduction of the dorsal myelographic column to a width of 2 mm has also been suggested. The purpose of this study was to determine the best diagnostic rule for cervical myelography in horses with suspected cervical stenotic myelopathy (CSM). We also investigated whether another cut-off point for reduction would result in better accuracy of the test and determined the reliability of the measurements.

We selected 38 horses retrospectively in which myelography and necropsy was performed. Repeated DMC measurements were taken at the intravertebral site at its maximum and at the intervertebral site at its minimum by two different readers. The final diagnosis and site of compression based upon histopathological examination served as gold standard. 50% of the cases had compression of the spinal cord consistent with CSM. Intra-and inter-observer repeatability of the ratios was high.

Several diagnostic criteria were examined and 50% reduction was determined to be the optimal cut-point. The test was specific and therefore accurate at detecting sites without compression. The sensitivity was inadequate however, indicating a problem detecting sites of compression. Due to the low prevalence of compression at each site, the test had an even lower ϩ PV. Accuracy improves by using a combination of neutral and flexed views. Therefore ''50% reduction of the DMC'' should not be used alone to diagnose CSM, or plan surgical treatment of a site of suspected compression. An understanding of the range of natural responses during various states of arousal is essential when interpreting an electroencephalogram (EEG). Identification of artifacts and normal transients is necessary before pathologic changes can be recognized. The purpose of this study was to establish normative data for this species that can be used as a basis for comparing sedative and anesthesia effects on the equine EEG. It will also serve as a standard for the interpretation of recordings made on neurologically compromised patients.

Five neurologically normal horses of various breeds were selected from a research herd for use in this study. All recordings were performed in a specially equipped stall that allowed the animals freedom of movement. Equipment and personnel were isolated in a corner behind a barricade. Normally kept on dry lots, each horse spent a minimum of 5 days in the stall prior to the actual recording. Lighting was provided solely by the use of infrared lamps during these sessions. All recordings took place after 6:30 PM during winter and were a minimum of eight hours in length. Horses had free access to food. Each animal was instrumented with a total of 19 surface electrodes for recording EEG and 3 for EOG. Two additional pairs of needle electrodes were used for monitoring the EKG and splenius m. EMG, respiration was monitored using a Piezo sensor. Simultaneous video recordings were obtained to assess behavioral status and assist in the identification of artifacts. Visual inspection of the digitally recorded EEG was performed utilizing bipolar and referential montages. Frequency and amplitude measurements of background rhythms were determined for the various states of arousal. Additionally, transients were identified and measured. Body position and time spent in each state were noted.

Artifacts, related to ear, eye and jaw movement, obscured most of the EEG during arousal. Slow wave sleep (SWS) contained numerous spindles, which varied from 10-14 Hz, as well as, short bursts of V-waves at 8 Hz. Voltage maximum was at the frontal and central vertex. Heart block was common during this state. Non-REM sleep (SWS and drowsiness) was performed almost entirely while standing with the head up. Due to the concomitant atonia, only short bursts of REM sleep were possible in this position, with sternal recumbency longer periods were recorded. Low amplitude beta activity superimposed on 4 Hz theta waves was seen in this state.

Though not a true representation of natural sleep in these horses (as they were herd animals that had been isolated for the study), this data does provide critical information applicable to EEG interpretation. The aim of this longitudinal field-based study was to identify factors associated with the development and clinical outcome of aortic insufficiency in horses. 1159 horses and ponies were included in an auscultatory survey. 74 (6.4%) had a murmur characteristic of aortic insufficiency (AI). In a multivariant model, horses Ն 15 years of age (odds ratio [OR] ϭ 1.09, 1.06-1.21, p Ͻ 0.0001) were at increased risk, and small ponies (OR ϭ 0.295, 0.17 ϭ 0.89, p ϭ 0.031) were at reduced risk, of having a murmur of AI. Information on the clinical status of 935 of the animals was obtained by interviewing their caretakers 24 months after recruitment. 25 (33.8%) of the horses with AI were dead, 37 (50%) alive and 12 (16.2%) lost to follow-up whereas 155 (19.3%) of the horses without regurgitant murmurs were dead, 663 (78.6%) alive and 17 (2.1%) lost to follow-up. In 7 (28%) of the AI horses that were dead, this was attributed to heart disease whereas in 18 (11.6%) of the horses without regurgitant murmurs that were dead, this was attributed to heart disease. Survival analysis revealed that horses with AI had a significantly reduced survival time (p ϭ 0.0017, hazard ratio 1.93, 1.42-4.49) compared to horses with no regurgitant murmurs.

The horses with AI underwent a standardized cardiological examination including physical examination, 24 hour ambulatory electrocardiography, heart rate variability analysis, indirect measurement of coccygeal arterial pressure and assay of plasma catecholamine concentrations. The AI horses were subdivided into a stable heart disease group (n ϭ 44, i.e. those in which there was no clinical evidence of progression of the heart disease over 24 months) and a progressive heart disease group (n ϭ 13, i.e. those that had developed clinical evidence of heart failure at rest, n ϭ 6, plus those that had died or been euthanased due to cardiac failure, n ϭ 7). The value of individual criteria evaluated at entry to the study as predictors for progression of heart disease, over a 24 month period, was assessed. The presence of a loud (grade Ն 3/6) diastolic murmur (sensitivity 0.54, specificity 0.85, likelihood ratio 3.60), palpable hyperkinetic pulses (sensitivity 0.62, specificity 0.78, likelihood ratio 2.82), pulse pressure Ն 60 mmHg (sensitivity 0.62, specificity 0.81, likelihood ratio 3.26), and frequent ventricular premature depolarizations (sensitivity 0.46, specificity 0.96, likelihood ratio 11.5) appeared to be the most useful prognostic tools. Thus, these data suggest that AI is a common disease in the older horse population of UK and, of the clinical tools assessed, physical examination, measurement of coccygeal arterial pressure, and most importantly, electrocardiography are the most useful for forming a prognosis in horses with AI. Screening tests are commonly utilized by equine practitioners for identifying foals with failure of passive transfer (FPT). However, there is limited knowledge of the reliability of the test results. In this study, the sensitivity and specificity of five FPT screening tests were critically evaluated using serum from 203 foals that were between 24-72 hours of age. Two commercial ELISA tests (Cite and SNAP, Idexx Inc) and a latex agglutination test (Foalcheck, Haver Moby Corp) were run according to the manufacturers instructions. Non-commercial tests (glutaraldehyde coagulation test and modified zinc sulfate turbidity test) were also performed. Radial immunodiffusion (RID) for equine IgG (VMRD Inc.) was used as the gold standard for determining each foal's serum IgG concentration. The results of the screening tests were compared to the results of the equine RID IgG test. The sensitivity and specificity for each screening test were determined for serum IgG concentration at both breakpoints, Ͻ800 mg/dl and Ͻ400 mg/dl. All screening tests were not performed on each foal.

The prevalence of FPT at serum concentration of IgG Ͻ800 mg/dl and IgG Ͻ400 mg/dl was 37% and 22%, respectively. The sensitivity and specificity of each screening test when detecting foals with serum IgG concentrations Ͻ800mg/dl were as follows: Ulcerative dermatitis, thrombocytopenia and neutropenia is a nationally emerging syndrome of unknown etiology in neonatal foals. The purpose of this effort is to characterize it in 5 neonatal foals from diverse geographical origins. History and physical examination findings, CBC, chemistry panel, coagulation panel, skin and bone marrow biopsy results, treatments and outcomes of foals were gathered from 3 referral hospitals (1998) (1999) (2000) . Three thoroughbred (TB) and 2 warmblood/paint foals from multiparous mares presented at Ͻ 4 days of age with ulceration of the tongue and oral mucous membranes and generalized multiple small crusts. Four of the 5 foals had petechiae and 2 had bleeding tendencies. Two mares had affected foals on 2 successive years but were bred to different sires. Similar drugs were not administered, either topically or systemically, to the mares during pregnancy or to the foals. All foals had IgG Ͼ1200 mg/dL. All foals were thrombocytopenic (0-30,000/uL), leukopenic (2900-3200/uL), neutropenic (500-1768/uL) and 4/5 had a left shift (90-780/uL). The PTT (55 and 118 s) was prolonged in 2/4. A normal bone marrow biopsy was obtained (1/1). Corticosteroids were given to 3 foals before the skin biopsy was taken. Histopathology of the skin in all foals showed subepidermal clefting with subjacent vascular dilation, dermal hemorrhage and superficial papillary necrosis. All foals were given gastric ulcer prophylaxis, 4/5 broad-spectrum antibiotics, and 3/5 corticosteroids. Whole blood (1L) was administered to 2 foals and platelet rich plasma (2 ϫ 4 units) was given to another. The mucosal and skin lesions improved in 7 days (5/5). Platelet counts gradually increased to Ͼ50,000 after 12 days (3/5) whereas 2 foals had persistent thrombocytopenia and were given corticosteroids for 19 days. Foals were hospitalized for 4-33 days and were healthy 5-24 months later. The 2 mares that had 2 affected foals each, upon subsequent pregnancies had healthy foals (n ϭ 3) when an alternate source of colostrum was given.

Foals with mucosal erosions, ulcerative dermatitis, thrombocytopenia and neutropenia have a good prognosis for life with supportive therapy. Once a foal has this syndrome, the mare may be predisposed to having similarly affected foals in the future. While an etiology is unknown, this group of foals may suggest an immune-mediated thrombocytopenia and neutropenia with vasculitis that may be related to colostral antibodies, but further investigation is warranted. Repeated bouts of maximal intensity exertion or endurance exercise result in substantial declines in skeletal muscle glycogen concentration. Decreased availability of muscle glycogen stores diminishes athletic capacity of horses. We studied the effect of 3 isoenergetic diets of varying glycemic indices on rates of post-exercise muscle glycogen resynthesis.

In a randomized, semi-balanced, three-way crossover study, seven fit adult horses performed strenuous exercise that depleted muscle glycogen by at least 60% on three occasions. On each occasion, the horses received either a high soluble carbohydrate diet (grain, HCO) or an isoenergetic low soluble carbohydrate diet (hay, LCO) or an isoenergetic mixed diet (grain and hay, M, control group) every 8 hours for 72 hours beginning immediately after the glycogen-depleting exercise (GDE). Glycogen concentration of middle gluteal muscle was measured before GDE, and 0h, 3h, 6h, 12h, 24h, 48h and 72h after the end of GDE. Blood was collected for measurement of plasma glucose, insulin and total protein concentrations, and hematocrit before GDE, every hour for the first 12 hours and at the time of each muscle biopsy. Statistical analyses were performed using analysis of variance for repeated measures. Values are reported as mean Ϯ SEM.

Strenuous exercise resulted in substantial depletion of muscle glycogen stores (from 129.0 Ϯ 3.6 to 28.0 Ϯ 3.6 mmol/kg wet weight before and after GDE, PϽ0.001). Post-exercise feeding of HCO diet resulted in higher plasma glucose concentration (up to 4 hours after feeding) compared to LCO and MCO diets (P ϭ 0.005). Plasma glucose concentration was similar after MCO and LCO diets (4.7 Ϯ 0.1 and 4.8 Ϯ 0.1 mM, respectively). Feeding HCO meal resulted in greater muscle glycogen concentration at 48h and 72h after exercise compared to MCO and LCO meals (133.6 Ϯ 3.6, 106.8 Ϯ 3.6 and 102.6 Ϯ 3.6 mmol/kg, respectively, 72h after exercise; PϽ0.001). Muscle glycogen concentrations similar to baseline were obtained 72 hours after exercise in horses fed HCO diet (123.3 Ϯ 3.6 and 133.6 Ϯ 3.6 mmol/kg before and 72h after exercise; P ϭ 0.12). Rate of glycogen resynthesis was higher for horses fed HCO diet, compared to horses fed MCO and LCO diets (1.51 Ϯ 0.15 vs. 1.12 Ϯ 0.11 vs. 0.97 Ϯ 0.10 mmol/kg/h, respectively; PϽ0.001). No treatment effect was observed on hematocrit and plasma total protein concentrations.

We concluded that at least 72 hours is required after strenuous exercise for restoration of muscle glycogen stores in horses. Post-exercise feeding of high-glycemic index (HCO) meals hastened muscle glycogen replenishment compared to LCO and MCO diets by increasing blood glucose availability to the skeletal muscle. Volumetric urine collection is reported to be more accurate than single sample urine collection for determining fractional excretion (FE) of electrolytes and minerals (E&M) in horses. DCAB alters urinary excretion of E&M in horses. The purpose of this study was to compare FE of E&M from single urine samples to volumetric urine collection, as well as to determine daily variation of FE in horses consuming different DCAB diets.

Eleven mares were used in the study, 5 had a history of Recurrent Exertional Rhabdomyolysis (RER). An anionic diet (low DCAB, ϩ87) was fed initially, followed by a medium DCAB diet (ϩ187) and then a cationic diet (High DCAB, ϩ400) each for 2 weeks. On the last 3 days of each period total urine collection was performed on 6 mares and a urine sample was taken from the remaining 5 mares. Concurrent blood samples were taken for analysis of Na, K, Cl, Ca, P, Mg and creatinine (Cr). Urine E&M were measured in acidified samples using emission spectrometry. Results were analyzed using ANOVA and coefficients of variation (CV) calculated to determine within-horse variation in clearance and FE values between 24 h periods.

Varying DCAB caused significant changes in plasma and urine E&M. No differences were found between RER and control horses. Volumetric and single sample collection detected similar patterns of change in E&M across diets with the exception of Mg which was higher in volumetric collection vs catheterized samples on the high DCAB diet. Large CVs for the clearance of Cr, E&M between 24-h periods were found within individual horses (17.3%-92.8%) across diets. The CV within-horses, between 24-h periods, for the FE of E&M ranged from 14.5%-85.1% with volumetric collection and from 17.6%-97.3% with catheterization. The excretion of Na, P and Ca displayed the greatest fluctuation between 24-h periods.

FE of E&M from single catheterized samples reflected FE values from 24hour volumetric urine collection over a range of DCAB. However, great variation within individuals can occur in urinary E&M excretion between 24-hour periods despite standardization of diet, routines and environmental conditions. Isoflupredone is used for the treatment of horses with inflammatory airway disease, but its efficacy and innocuity have not been critically evaluated.

In a parallel study, we compared a 14 days course of treatment with dexamethasone (0.04mg/kg IV SID, n ϭ 6) and isoflupredone (0.03mg/kg IM SID, n ϭ 6) in heaves affected horses. Lung function, clinical exams, blood cortisol before and after adrenocortical stimulation, serum electrolytes and CBC were studied.

There were no significant differences between groups at baseline and following a 14 days control period for any of the parameters studied. Both drugs were well tolerated clinically and resulted in a significant improvement in lung function starting on day 3 and lasting for the duration of the study. Blood cortisol levels were significantly decreased during the treatment period in both groups, but remained below baseline values 1 week post-treatment in the isoflupredone treated group only. Blood cortisol response following ACTH stimulation was normal following drug administration, however. Serum electrolytes were not affected by treatment with dexamethasone but horses treated with isoflupredone demonstrated a significant decrease in serum potassium level. Both treatments induced a significant increase in peripheral blood neutrophil count and a decrease in lymphocyte and eosinophil counts.

In conclusion, treatments with isoflupredone and dexamethasone improved lung function of heaves affected horses to a similar degree. Hypokalemia was associated with isoflupredone administration, but hypokalemic myopathy was not observed. The frequently observed phenomenon of exercise-induced arterial hypoxemia (EIAH) in human athletes is thought to occur in response to capillary stress failure related pulmonary injury causing release of histamine (and possibly other inflammatory mediators). Inflammatory mediators are felt to contribute to the generation of this phenomenon because administration of nedocromil sodium to stabilize airway inflammatory/mast cells mitigates EIAH. Presently, it is not known if this same type of mitigation would occur in racehorses.

Thus, in order to elucidate the role of pulmonary injury related release of inflammatory mediators in causing EIAH in horses, we examined the effects of pretreatment with a potent antiinflammatory agent-dexamethasone. Seven healthy, sound, exercise-trained Thoroughbred horses were studied in control (no medications) experiments, followed in 7 days by intravenous (IV) dexamethasone (0.11 mg/kg/day for 2 consecutive days) studies. Arterial and mixed-venous blood-gas measurements were made at rest and during incremental exercise leading to maximal exertion at 14 m/s on a 3.5% uphill grade.

Galloping at this workload elicited maximal heart rate, and exercise-induced pulmonary hemorrhage in all horses in both treatments, thereby, indicating that stress failure of pulmonary capillaries/pulmonary injury had occurred. In both treatments, significant arterial hypoxemia, desaturation of hemoglobin, hypercapnia, metabolic acidosis and hyperthermia developed during maximal exercise, but statistically significant differences between the control and dexamethasone studies could not be discerned.

The failure of pretreatment with dexamethasone to significantly affect EIAH suggests that pulmonary inflammation/injury-induced release of inflammatory mediators may not play a major role in bringing about the phenomenon of EIAH in racehorses. However, our observations in both treatments that EIAH developed quickly-being evident by 30 s of exertion, and that its severity remained unaffected by increasing exercise duration (to 120 s), suggests that EIAH more likely has a functional basis, probably related to the significant shortening of the transit time for blood in the pulmonary capillaries as cardiac output increases dramatically. Increased concentrations of volatile fatty acids in gastrointestinal contents due to fermentation of carbohydrate-rich feeds have been reported to inhibit gastrointestinal motility in ruminants. We describe the effects of an abrupt increase of concentrates in the diet of dairy cows on myoelectric activity of the spiral colon and on fermentation patterns in the rumen and large intestine.

Six healthy, lactating Simmental ϫ Red-Holstein cows were instrumented with bipolar electrodes in the spiral colon. The diet was changed from hay only to a ration of 50% hay and 50% starch-rich concentrates within 60 hours. Myoelectric activity of the spiral colon, concentrations of absolute and undissociated volatile fatty acids (VFA) and pH values in ruminal and large intestinal contents were monitored during four days before (period 1), six days during and immediately after the dietary change (period 2), and four days after adaptation to the new diet (period 3). Myoelectric activity was analyzed using computer-assisted methods, and volatile fatty acid concentrations were determined by gas chromatography.

Disruption or disorganization of the myoelectric activity patterns of the spiral colon was not observed during or after the diet change. Statistically significant changes in parameters of myoelectric activity of the spiral colon were of small magnitude and restricted to phases III and IV of the bovine colonic migrating myoelectric complex (bcMMC) and to propagation velocity of the bcMMC. Duration of phase III increased slightly from 317.8Ϯ77.3 seconds during period 1 to 321.2Ϯ76.5 seconds on the first day of period 2, decreased to 302.2Ϯ57.9 seconds on the next day before returning to original values by the end of period 2 and during period 3. Changes in duration of regular spindles of phase III, and duration and number of hyperactivity complexes of phase IV followed a similar pattern. Propagation velocity of the bcMMC increased from 4.4 cm/min (period 1) to 5.2 cm/min (period 3). No significant alterations were observed in pH or VFA concentrations of ruminal fluid. In large intestinal contents, pH values decreased significantly from 7.22 to 6.71 (median), and VFA concentrations increased significantly from 19.2 to 35.1 mmol/L (median of absolute total VFA concentration) after the change of diet.

We conclude that increased concentrations of VFA and low pH in large intestinal digesta have a minimal influence on myoelectric activity of the bovine spiral colon. Thus, increased luminal VFA concentrations are unlikely to play an important role in the etiopathogenesis of motility disorders of the bovine large intestine.

KALEMIA SYNDROME. N. Sattler, G. Fecteau*, J. Pareϩ. ENV Alfort, Maisons-Alfort, France; * Université de Montréal, Saint-Hyacinthe, Canada; ϩ Canadian Food Inspection Agency, Saint-Hyacinthe, Canada.

The first objective of this study was to reproduce bovine hypokalemia syndrome experimentally using isoflupredone acetate (IA) in food deprived dairy cattle. A second objective was to investigate the external and internal potassium balances modifications induced by IA administration when combined to food restriction.

Nine lactating holstein cows were randomly allocated to one of the following experimental groups: Control group: food deprivation (5 days); Treatment group 1: food deprivation (5 days) and 20 mg IA, on days 2 and 3; Treatment group 2: food deprivation (4 days) and 20 mg IA on days 2, 4 and 6. Serum (SK), red blood cell, urine and milk potassium concentrations as well as serum and urine creatinin concentrations were measured once daily. Bovine hypokalemia syndrome (BHKS) was defined as the presence of 5 out of 9 clinical signs (twisted neck, delay to standing, standing duration Ͻ 10 min, decubitus, flaccid paralysis, tachycardia, arrhythmia, ruminal atonia, stupor) AND SK Ͻ 2.5 mmol/L. Survival analysis was used to compare on time to a reduction to SK Ͻ 2.5 mmol/ l and on time to the development of BHKS.

One cow in the control group died from unrelated disease (disseminated calcinosis). A SK Ͻ 2.5 mmol/L was observed in 1 cow of Group 1 and all cows of Group 2 (p ϭ 0.07). BHKS was not observed in control group, nor group 1, but was observed in 2 cows of Group 2 (p ϭ 0.08). All cows except one returned to normal production.

Administration of IA to food deprived lactating dairy cattle is a good model to investigate BHKS. Further studies using more animals will be required to precise the pathophysiology of this syndrome. The purpose of this study was to determine if significant and rapid loss of bone mineral density (BMD) could be induced using a diet low in cation-anion balance (DCAB) compared to a grass hay diet in skeletally mature ovariectomized ewes.

52 skeletally mature (4-7 year old) Columbia-Rambouillet cross ewes were used in the study. 38 sheep underwent ovariectomy (OVX) and 14 sheep underwent sham surgery (ShOVX). Following surgery sheep were assigned to consume a low DCAB diet (n ϭ 24, OVX), or a Hay diet (n ϭ 14 ShOVX and n ϭ 14 OVX). All sheep underwent dual energy X-ray absorptiometry (DEXA) of the lumbar vertebrae on day 0. All 24 low DCAB sheep had DEXA at 90 days and 14 Hay diet sheep (7 OVX and 7 ShOVX) were euthanized and DEXA scanned at 90 days. The balance of the sheep are being held for future study. The Hay diet consisted of free choice grass hay (DCAB 300 Meq/kg DM). The low DCAB diet consisted of grass hay and a low DCAB pellet, such that the total DCAB consumed per sheep per day was approximately Ϫ350 Meq. Days 0 and 90 DEXA scans were compared and data was analyzed using an ANOVA. Percent change in BMD (%BMD) was compared using a standard of significance of pϽ0.05.

There was no difference in %BMD within the Hay diet group between OVX and ShOVX (p ϭ 0.35) therefore these sheep were grouped together for comparison with the low DCAB group. Mean %BMD for the low DCAB group was significantly greater than for the Hay diet group. Sheep fed the low DCAB diet lost 10.19% Ϯ 1.27% of their BMD compared to 1.95% Ϯ 0.78% BMD loss in the sheep fed the Hay diet (p ϭ 0.00004).

Sheep are often used as a model for human osteoporosis. However, one drawback of this model is that 6-12 months post OVX (to induce surgical ''menopause'') sheep fed hay alone typically fail to lose significant amounts of bone mineral density when compared to humans with osteoporosis. This pilot study shows that a low DCAB diet can induce a greater decrease in bone mineral density over a shorter period of time when compared to the conventional model of a hay diet in conjunction with OVX. This model shows considerable promise for use in the study of human osteoporosis. This study determined the frequency of deficient and marginal selenium status in 532 spring-born Missouri feeder calves. Whole blood selenium concentration was determined by HPLC. In conjunction with sample collection, cooperating practitioners completed a survey which summarized exposure to a variaety of potential risk factors. Selenium deficiency was widespread in the study population. Calculated rates of deficient, marginal and replete selenium status were 0.165, 0.630 and 0.205. Multivariate models were developed predicting both selenium status and actual blood selenium concentration. Factors significantly associated with selenium deficiency included geographic area, pasture application of either manure or lime, specific pasture plants and provision of trace meneralized salt. Factors significantly associated with blood selenium concentration included geographic area, cow body condition, creep feeding, pasture application of commercial fertilizer, manure or lime and specific pature plants. Lactoferrin is a non-heme associated iron-binding apoprotein, present in many body secretions, colostrum, and neutrophil (PMN) granules. Biological effects ascribed to LF include iron binding, LPS binding, and alteration in PMN and lymphocyte function. LF prevents LPS binding to leukocytes and limits PMN priming or cytokine release. Recent work in our lab indicates LF increases PMN superoxide production ex vivo. Studies of mammary leukocytes indicate a distinct role of LF in phenotype, tissue persistence and proliferative ability of leukocytes depending upon concentrations of LF. The purpose of this study was to examine the ability of LF to alter proliferation of peripheral blood mononuclear cells (PI) exposed to LPS. We also examined the role of exogenous iron (FE) on proliferative responses of mononuclear cells in the presence or absence of LF.

Leukocyte isolation from buffy coat cells of healthy 2-month old calves was performed first by cold lysis of RBC using Tris-NH 4 Cl. Mononuclear cells were then centrifuged over Ficoll-sodium diatrazoate. Cells were enumerated, and viability determined and approximately 4 ϫ 10 6 cells were placed into individual tissue culture plate wells. Cells were treated with LF (200 ng/ml) or medium alone followed by LPS stimulation (1 g/ml) and allowed to incubate for 72 hours after which time 3 [H]-thymidine was added. Cell labeling was allowed for 24 hours. The cells were then harvested unto a filter mat and 3 [H]-thymidine incorporation was determined in a gamma counter. In several additional experiments, the effects of LF on lymphocyte proliferation were further studied by examining the role of exogenous iron in lymphocyte proliferation. LF in vitro, reduced LPS-induced cell proliferation index (PI) significantly (LF ϩ LPS ϭ 1.38 ϩ 0.9; LPS ϭ 14 ϩ 13; pϽ0.001). LF by itself reduced incorporation of label in vitro (PI ϭ 0.73 ϩ 0.42). To determine if the effects of LF were due to iron sequestration in the medium, we added Fe at two concentrations (1mM and 10 nM) and repeated the above experiments. Iron at the higher concentration significantly increased the PI of all treatments suggesting that Fe becomes limiting in vitro, or induces some oxidative stress leading to DNA repair, which increased incorporation. The lower concentration produced mild increases in PI compared to no additional Fe treatments. Our study suggests that LF reduces mononuclear cell proliferation associated with LPS exposure. This reduction in PI suggests LF prevents LPS interaction with the cell and may also sequester Fe, a necessary factor for cell proliferation. The objective of this study was to estimate the effect of clinical and laboratory covariates (age, infection site and dehydration) on the relationship between failure of passive transfer (FPT) assays (serum protein, plasma protein, percentage of optical density measured from the zinc sulphate (ZnS) turbidity test, semiquantitative interpretation of the ZnS turbidity test and the serum gamma-glutamyl transferase activity (GGT)) and RID.

Serum and plasma samples were collected from one hundred calves between 1 and 19 days of age. Linear regression models were built to predict IgG with each FPT assay. Studied covariates were divided in clinical and laboratory categories. Eight different models were built. Clinical models included age, hydration and infection site as covariates. For the serum protein model, hydration, infection and an interaction term were included. For the plasma protein model hydration and infection were also included. Hydration and age were included in the GGT model. For ZnS models (semi-quantitative or optical density reading) only age was included. Laboratory models included PCV, age and fibrinogen concentration as covariates. PCV and fibrinogen concentration were included in the serum protein model and fibrinogen only was included in the plasma protein model. Fibrinogen concentration and age were included in the GGT model. Adjusted R2 for the 8 models ranged from 44 to 62%.

The multiple regression analyses of the serum IgG concentration using different predictors provides an insight into the important covariates to take into consideration when evaluating passive transfer of immunity through an indirect method. As an example, for the same serum protein concentration, the predicted value of IgG concentration may vary enough to influence the actual status of a calf, if one was to use categorized classification for IgG. The results presented suggest that it would be hazardous to use the different FPT assays without considering identified significant covariates. In cattle with Bovine Pneumonic Pasteurellosis (BPP), the host immune response has been implicated to play a significant role in causing much of the pulmonary injury that occurs. Traditionally, antibiotics are used to treat BPP, although some producers also administer corticosteroids in an attempt to reduce pulmonary inflammation by limiting the host immune response. The objective of this study was to assess the clinical efficacy of a corticosteroid, isoflupredone acetate, in the treatment of experimentally-induced bronchopneumonia when administered in conjunction with antibiotics.

Ninety-six crossbred beef weanling heifers were randomly allocated to one of four possible treatment groups: controls (CON; n ϭ 24), infected/no treatment (NTX; n ϭ 24), infected/antibiotic (ABX; n ϭ 24), or infected/antibiotic/steroid (SAB; n ϭ 24). On day 0, heifers in the NTX, ABX and SAB groups were inoculated intrabronchially with M. haemolytica by endoscope. Heifers in the CON group underwent a similar sham procedure using sterile saline. Animals in the CON and NTX groups received no subsequent treatment. Oxytetracycline (Bio-Mycin 200; 20 mg/kg SQ) was administered to the ABX and SAB groups on days 1 and 4 post-infection (PI). Isoflupredone acetate (Predef 2X; 0.05 mg/ kg IM) was also administered to the SAB group on days 2 and 4 PI. Rectal temperature (RT) was recorded daily for each animal, and serum cortisol concentration (SC) was measured daily for 4 heifers within each group.

Data were analyzed for significant differences between the four treatment groups using ANOVA (PϽ0.05). Significant differences were noted for RT on days 1, 2, 3 and 5 PI. On days 1 and 2, all three infected groups had a significantly higher RT than the CON group, as expected following experimental inoculation. This elevation in RT had resolved by day 3 for both the ABX and SAB groups, which were no longer different from the controls. The RT of the SAB heifers remained normal throughout the rest of the trial period, however RT of the ABX group increased again and was statistically higher than both the CON and SAB animals on day 5. The SC was significantly lower in the SAB group compared to all other groups on days 3, 4, 5 and 6 PI.

The results of this study demonstrate that isoflupredone acetate in conjunction with oxytetracycline is effective in resolving pyrexia associated with M. haemolytica infection. Although RT initially improved in both the ABX and SAB groups, a second phase of pyrexia was observed in the ABX group, which was not evident when steroids were also administered. The marked decrease in SC in the SAB group demonstrates that the dose of isoflupredone acetate used in this study was sufficient to cause adrenal suppression. Long-acting formulations of oxytetracycline are some of the most widely used antimicrobials for the treatment of bovine respiratory disease complex (BRD). The long-acting nature of these products depends on their formulation. Most veterinarians and producers regard these formulations as equivalent. Ltd.)(BM) contains 200mg/ml oxytetracycline in polyethylene glycol. It is approved for intramuscular (IM) or subcutaneous (SC) injection at 20mg/ml. Tetradure LA-300 (Merial Canada Inc.)(TD) contains 300mg/ml oxytetracycline in glycerol formal. It is approved for IM injection at 20mg/kg, but is used at 30mg/kg IM on the basis of a manufacturer sponsored feedlot trial.

A double blind cross-over study was performed using 6 pregnant heifers treated with BM at 20mg/kg SC and TD at 30mg/kg IM. A two week washout period was waited between treatments. Blood samples were collected and the concentration of oxytetracycline in the plasma was determined using a proven HPLC assay.

According to the FDA's recommended evaluation method, the two formulations at the dosages used are not bioequivalent. The AUC values indicate that TD resulted in significantly more oxytetracycline being absorbed into the body, however this did not result in significant increases in either the maximum concentration of oxytetracycline (C max ) or the half life (T 1/2 ) of the drug in the plasma. Therefore, this pharmacokinetic study does not support the extra label dosage of TD.

CALVES WITH A NUCLEAR SCINTIGRAPHIC PROCEDURE. G Nappert*, JC Lattimer. *University of Missouri. Lachute Veterinary Hospital, Quebec.

The purpose of the present study was to demonstrate that nuclear medicine technology allows observation of the effect that milk clotting has on abomasal emptying in the living neonatal calf. Scintigraphic evaluation of abomasal emptying was carried out in six healthy male Holstein calves. The calves were fed 10% of their body weight daily as whole cow's milk that was divided equally and consumed as two feedings via a nipple bottle. One day before the nuclear scintigraphic procedure, the calves were randomly fed whole cows milk or an oral rehydration solution containing bicarbonate and high levels of soluble fiber was fed for three consecutive feedings an hour before the portion of milk. For each calf, both feeding programs were repeated twice at a one-week interval. Immediately following administration of the 99mTC-sulfur colloid containing milk, the calves were imaged with the gamma camera positioned lateral and ventral to the abomasum. Additional right lateral and ventral views of the abomasum were collected at 15, 30, 45, 60, 90, 120, 150, 180, 210 and 240 minutes after administration of the radionuclide. Blood glucose determination were performed at one-hour intervals for seven hours after feeding milk to evaluate milk digestibility in both feeding programs. Baseline glucose values during the first three hours of the experiment were not significantly different between both feeding programs. Baseline number of radionuclide counts from the lateral and the ventral view of the abomasum were not significantly different between both feeding programs. In addition, there were no significant treatment effects on the number of radionuclide counts from both views during scintigraphic evaluations. We successfully developed an in vivo method of measuring abomasal emptying and showed that an oral electrolyte containing a small amount of bicarbonate did not interfere with milk clotting. This is the first time nuclear scintigraphy has been used to study milk clotting and abomasal emptying in calves. The increasing value of replacement dairy heifers has placed greater emphasis on colostrum feeding, a key element for neonatal calf health. While colostrum immune factors are essential for health, bacterial contamination may negate some benefits. From our calf health problem investigations, 82% of colostrum samples exceeded the industry standard of 100,000 cfu/ml. Many of these colostrum samples grew more than 1,000,000 cfu/ml but there is limited data on the impact of these excessive colostrum bacterial counts on calf health. The purpose of this study was to determine the impact of bacterial contamination (total, fecal coliform, other gram-negative bacteria, mastitis and environmental Staphylococcus and Streptococcus species) on neonatal calf health.

Total bacteria, fecal coliform, mastitis and other environmental bacteria were quantified in 82 colostrum samples fed to 101 newborn calves. Calf passive transfer status and daily health scores based on 14 days of temperature, fecal consistency, appetite and attitude scores were correlated with the level of bacterial contamination in their single colostrum meal.

Bacterial contamination had a negative impact on acquisition of passive transfer. Total bacterial and fecal coliform counts were positively correlated with abnormal fecal consistency. The impact of increasing total bacteria and fecal coliform counts in colostrum had a greater negative influence on the health of calves with failure of passive transfer of immunity. Other gram-negative bacteria, environmental Streptococcus and Staphylococcus species have little impact on neonatal calf health. Factors other than colostrum bacterial contamination create farm-to-farm variability in calf health scores.

From this study, we conclude that excessive bacterial contamination of colostrum negatively impacts calf health. Given this information, determination of total bacteria and fecal coliform bacterial counts in colostrum is an important aspect of calf health problem investigations. New infections by N. caninum in cattle occur primarily following vertical transmission and there is no proven way to prevent it. The embryo transfer procedure proposed by the International Embryo Transfer Society (IETS) has been proven effective against transmission of many pathogens in cattle. The objective of this study was to determine the efficacy of the embryo transfer procedure defined by IETS into seronegative recipients to prevent vertical transmission of N. caninum in cattle.

Eighty-seven recipient cows and heifers and their embryo transfer calves from 22 donors originating from 9 different dairy herds provided data for this study. N. caninum serologic status of donors and recipients was determined before collection and transfer of embryos. Viable embryos after washing and trypsin treatment were either transferred fresh (n ϭ 61) or frozen and transferred after thawing (n ϭ 113). Pregnant recipients in experimental groups A (n ϭ 50) and B (n ϭ 29) were seronegative and received embryos from seropositive and seronegative donors, respectively. Pregnant recipients in group C (n ϭ 8) were seropositive and received embryos from seronegative or seropositive donors. Antibodies against N. caninum were determined monthly during pregnancy in recipients and in calf blood samples collected at birth. Tissues collected from stillbirths and aborted fetuses were analyzed by use of histology and immunohistochemistry (IHC).

Seventy-six calves and 11 fetuses and stillborn calves were sampled at birth. All calves from groups A and B were seronegative (n ϭ 70) or lacked evidence of infection by use of tissue analysis (n ϭ 9). In group C, 5 of 6 calves were seropositive at birth and IHC results were positive for 1 of 2 calves. Vertical transmission rate was significantly lower in groups A and B (0%) than in group C (75%). Embryo transfer using the procedure proposed by IETS, into seronegative recipients is an effective way to prevent vertical transmission of N. caninum. Results provide support for pre-transfer testing of all embryo transfer recipients. We have previously demonstrated excellent in vitro antiviral activity and lack of cytotoxicity for human recombinant alpha interferon-2A against selected bovine viruses, including type 1 non-cytopathic BVDV. This study was designed to examine the in vivo antiviral efficacy and safety of this product in cattle persistently infected with BVDV.

The study utilized 5 Holstein heifers between 4 and 12 months of age that were persistently infected with a non-cytopathic type 1 BVDV, as confirmed by monoclonal antibody and repeated microplate ELISA testing. Human recombinant interferon-2A was administered by deep intramuscular injection three times weekly for 12 weeks at a dose of 10000 IU/kg. Serum biochemical profiles and complete blood counts were performed during the 3-month period prior to the onset of the study, for the 12 weeks of treatment and monthly for 12 weeks following withdrawal of therapy. Cattle were examined daily and underwent comprehensive physical examination weekly throughout the study. Viral loads were measured by a standard immunohistochemical titration assay. Treatment animals acted as their own historical controls with viral loads, hematologic and biochemical parameters being compared to pretreatment values for each animal using the paired T test and Wilcoxon's rank sum tests for Gaussian and non-Gaussian data respectively. Significance was set at p Յ 0.05.

There was no significant reduction in viral load documented in any animal during the 12-week treatment period. Injections were well tolerated and no treatment related physical or serum biochemical abnormalities were noted. Microcytic anemia was noted in each animal towards the end of treatment that persisted following drug withdrawal. Bone marrow analysis revealed normal erythroid precursors at all stages of maturation, and myeloid:erythroid ratios varying from 0.7:1 to 1.4:1 during the period of anemia. Serum iron, transferrin, and ferritin levels were also normal during this time period.

Human recombinant alpha interferon-2A demonstrated no significant antiviral effect against type 1 non-cytopathic BVD in persistently infected cattle. Intramuscular administration of this product at a dose of 10000 IU/kg three times weekly for 12 weeks was well tolerated but associated with the development of microcytic anemia. No other biochemical, hematologic or clinically significant side effects were noted. The objective of this study was to estimate the success rate of bacterial culture of clinically affected joints with septic arthritis in cattle, and to establish the type of bacteria most commonly cultured.

Medical records from all cattle diagnosed with septic arthritis at the Faculty of Veterinary Medicine from 1983 to 2000 were studied. Animals with clinical signs of septic arthritis confirmed by cytology of synovial fluid, radiology and/ or bacteriology of synovial fluid were included. Bacteriological culture results of synovial fluid or synovial membrane were studied. Aerobic culture was performed on all specimens. Anaerobic culture and mycoplasma culture were performed only when requested. Comparisons between groups of age were made using chi square test. A logistic regression was used (PϽ0.05) to evaluate influence of age, type of joint, duration of arthritis or use of antibiotics before sampling.

One hundred seventy-two cases were included in the study (153 dairy cattle and 19 beef cattle). Ninety-one animals were less than 6 months of age. Carpal and stifle joints were the most frequently affected joints in young animals, whereas fetlock and tarsal were more frequently involved in adults. Bacteriological culture was negative in 40% of cases, while 47% had one type of bacteria, 9% had two types of bacteria, and 4% had three or more types of bacteria. Age, joint, duration of arthritis or use of antibiotics before sampling did not influence the culture result. Arcanobacterium pyogenes was the most common bacteria isolated, (35% in young animals and 48% in adults). Streptococci were isolated in 14%, staphylococci in 12%, enterobacteriaceae in 11%, anaerobacteriaceae in 11%, mycoplasmas in 4%, and pasteurellaceae in 4%.

Sixty percent of cases had a positive culture. Anaerobacteria and mycoplasma were probably underestimated in this study. To increase the rate of positive culture, sampling techniques should be standardised. Thus, hemoculture bottles may be used, as was demonstrated in others species and anaerobic and mycoplasmas culture must be performed. In adult cattle, it appears important to use an antimicrobial effective against Arcanobacterium pyogenes, such as one in the beta lactam family. In young animals, an anti-microbial effective against Gram ϩ and Gram Ϫ should be used. However, knowing the primary cause of haematogenous spread could help to determinate with more precision which organisms are involved and therefore exactly which anti-microbial to use. Antimicrobials are commonly utilized to treat salmonellosis in bovine neonates. The objectives of this study were to evaluate the therapeutic efficacy of ceftiofur used at a high, extralabel dose, for treatment of salmonellosis in a challenge model in neonatal calves and to determine the effect of such ceftiofur treatment on salmonella fecal shedding.

Forty-two Holstein bull calves were collected from a single dairy over a 96hour period. Calves were administered a colostrum supplement (Lifeline) in lieu of colostrum, providing 135g of IgG. Thirty-six calves were orally challenged with 6.5 ϫ 10 8 Salmonella enteritica serovar Typhimurium. All calves were between 24 and 108 hours of age at the time of challenge. Four days following Salmonella challenge, surviving calves were randomly allocated to either ceftiofur treated (Excenel RTU, 5 mg/kg by IM injection SID) or nonmedicated control groups. Calves assigned to the treated group were medicated daily for 5 days starting on day 4 post challenge.

Calves were monitored for 18 days following Salmonella challenge. Calves that became ill and were unable to maintain sternal recumbency were euthanized. Outcome assessments included clinical parameters (attitude, appetite, fecal characteristics, and rectal temperature), mortality, quantitative Salmonella fecal cultures, and Salmonella culture of mesenteric lymph nodes and cecal contents at necropsy.

Prior to the initiation of treatment on day 4, 22 of the remaining 29 challenged calves had had an episode of diarrhea, 8 had had a fever and 7 had died. Ceftiofur treatment was associated with a significant reduction in rectal temperature and diarrhea (P Ͻ 0.05). Twenty nine percent of non-medicated and 20% of medicated calves died following initiation of treatment. A significant reduction in Salmonella fecal shedding was observed in treated calves (P Ͻ 0.05). Salmonellae were isolated from all of the non-medicated calves at necropsy. No salmonellae were isolated from 41.7% of medicated calves at necropsy. No change in the MIC of the Salmonella challenge strain was observed during the course of the study.

The results of this experiment demonstrate that high dose, extralabel ceftiofur treatment of salmonellosis in neonatal calves attenuates the severity of disease, reduces Salmonella fecal shedding, and promotes clearance of Salmonella from the tissues of infected calves. Systemic hypertension is a common problem in cats. While previous studies suggest the classic renin-angiotensin system is not activated in all cats with elevated blood pressure, the intrarenal or other local renin-angiotensin systems may be involved in the genesis of this hypertension. As converting enzyme inhibitors appear to be only modestly effective antihypertensive agents in this species, it may be that another approach to inhibition of the renin-angiotensin system will prove efficacious as antihypertensive therapy. We evaluated the usefulness of the angiotensin II type-1 (AT1) receptor antagonist, losartan (Cozaar), as an oral antihypertensive agent in cats. This drug requires hepatic bioactivation in people and dogs but little is known about its metabolism in cats.

In initial dose-titration studies, we developed an angiotensin I (AI) challenge test to determine the optimum dosage of AI that produced a rise in systolic blood pressure of 25-40 mmHg in cats. The pressor response to AI was presumably mediated by angiotensin II binding to the AT-1 receptor as in other species. Subsequently, AI was delivered IV following varying dosages of losartan to 6 cats implanted with radiotelemetry blood pressure monitoring devices (Model TA11PA-C40, Data Sciences International) to determine if losartan could block this pressor response.

Conscious cats were dosed orally with 0-150mg Cozaar daily and then given 1 g IV injections of AI 4-6 hours after dosing. In conscious, restrained cats given 0mg of Cozaar, 1 g of AI resulted in an average of 36.7mmHg rise in systolic blood pressure. Cats dosed with 5mg of Cozaar and injected with 1 g of A1 4 hours after the dosing of Cozaar averaged a 39mmHg rise in systolic blood pressure. The cats demonstrated a similar increment in systolic blood pressure following dosing with 10 and 50mg Cozaar. The latter dosage is similar to the usual recommended daily dosage for human beings. Further dosing of losartan at 150mg for 3 consecutive days yielded no significant decrease in baseline blood pressure or in the pressor response to angiotensin I injection. The results indicate that losartan did not have a significant or consistent effect on feline blood pressure or pressor response to AI administration. Our study failed to provide evidence of potential for antihypertensive efficacy of losartan in cats. Valvular pulmonic stenosis (PS) reportedly is the third most common congenital heart defect in the dog. The severity of PS can be graded as mild (16-49 mmHg), moderate (50-79 mmHg), or severe (Ͼ 80 mmHg) based on Doppler derived estimates of transpulmonic pressure gradients (DEG). Additionally, PS may be considered severe with a DEG Ͻ 80 mmHg if marked right ventricular hypertrophy or significant clinical signs such as collapse, exercise intolerance or ascites are present. Mild and moderate PS is not thought to cause a reduction in either quality or quantity of life. Dogs with severe PS are more likely to present and/or develop clinical signs including sudden death. Current recommended therapy for severe PS includes balloon valvuloplasty (BV) or surgical vavulotomy. There are few reports of the benefits of BV in the dog and current recommendations are based on the positive results of BV in human patients with PS.

This study reports short and long-term results of BV on DEG in 50 conscious dogs with severe PS. All cases of severe PS with complete medical records that underwent BV at TAMU Veterinary Medical Center between 01/91-12/01 were included. Baseline DEG were determined in dogs prior to BV. Short-term BV results were based upon DEG obtained within 7 days of BV. Long-term results were based upon DEG at least 1 month following BV (med ϭ 8, mean ϭ 16, range ϭ 1-88). Successful BV was defined as Ͼ50% reduction in DEG from baseline or reduction to Ͻ 80mmHg.

The mean pre-BV DEG was 128mmHg (med ϭ 120, range ϭ 65-206). Thirty six percent (36%) of dogs showed clinical signs including syncope, exercise intolerance or ascites prior to BV. Short-term follow up was available for 45/50 dogs. The mean DEG was 50mmHg (med ϭ 50, range ϭ 15-158) representing a mean percent gradient reduction of 58% (med ϭ 58, range ϭ 15-87). Based on short-term DEG results 87% (39/45) of dogs had a successful BV, 9% failed (4/45) and 4% (2/50) died during BV. Long-term DEG data was available for 23/48 dogs and the mean DEG was 75mmHg (med ϭ 58, range ϭ 36-185). Based on long-term DEG results, 78% (18/23) of dogs had a successful BV and 22% (5/23) failed. Seventy one percent of symptomatic dogs became asymptomatic by their long-term reevaluation. These finding suggest that BV can successfully decrease the DEG and that gradient reductions tend to be maintained in the majority of dogs. Additionally, BV may decrease symptoms and thus improve quality of life in dogs with severe PS. Artificial cardiac pacing is a well-recognized treatment for symptomatic bradycardia in dogs. Ventricular demand pacemakers (VVI) that serve to pace the ventricle during periods of ventricular asystole are most commonly employed. This pacemaker system is unable to maintain atrioventricular (AV) synchrony and is therefore considered non-physiologic. A newer physiologic, single lead pacemaker system (VDD) utilizes two floating atrial electrodes (FAE) to sense native atrial activity. After detection of atrial depolarization and pausing for a programmable AV delay, the VDD system delivers a pacing stimulus to the ventricle thereby simulating the normal sequence of cardiac depolarization. The purposes of this study were 1) to determine the feasibility of physiologic single lead pacing in dogs with third degree AV block and 2) to determine if VDD pacing, in comparison to VVI pacing, displays improvements in acute hemodynamic variables.

Six dogs with third degree AV block have undergone successful permanent VDD pacemaker implantation with acute measurements of cardiac output, pulmonary and systemic arterial, right atrial and pulmonary capillary wedge pressures (PCWP). All dogs were sedated with intravenous acepromazine and atropine followed by placement of a temporary transvenous pacemaker. Following stable temporary pacing dogs were induced with intravenous ketamine and diazepam and maintained with isoflurane. A Swan-Ganz catheter was advanced into the pulmonary artery for hemodynamic measurements and a permanent VDD pacemaker lead was advanced into the right ventricular apex. The FAE were positioned in the mid to high right atrium and the permanent VDD pulse generator was attached. After ten minutes of stable pacing hemodynamic variables were measured. The pacemaker was re-programmed to rate-matched VVI and following ten minutes of stable, non-physiologic pacing the hemodynamic variables were again measured.

VDD pacing resulted in a 29% reduction in PCWP (4.00 Ϯ 2.17 vs 5.67 Ϯ 2.25 mm Hg; p ϭ 0.022) and a 13% reduction in pulmonary arterial systolic pressures (18.8 Ϯ 3.97 vs 21.7 Ϯ 5.01 mm Hg; p ϭ 0.030) compared to VVI pacing. There were no significant differences in cardiac output, systemic arterial, pulmonary arterial diastolic, mean pulmonary arterial or right atrial pressures.

Initial results from this ongoing study indicate single lead physiologic VDD pacing is feasible in dogs and maintenance of AV synchrony decreases both pulmonary arterial systolic and pulmonary capillary wedge pressures. Systemic arteriovenous fistulas (AVF) in limbs may cause pain, swelling, lameness, and high output CHF. Since surgery is frequently associated with high morbidity, we attempted transcatheter coil occlusion for AVF repair.

AVF was diagnosed in 3 cats and 1 dog presented for extremity swelling or lameness, and 1 dog with acute carpal pad bleeding (duration of clinical signs, 24 hrs-2 yrs). Affected limbs were R front (2 dogs, 2 cats) and L pelvic (1 cat). Physical findings were brisk arterial pulses (n ϭ 2), warm limb extremity (n ϭ 4), Branaham's sign (n ϭ 1), warm skin overlying AVF (n ϭ 4), continuous AVF bruit (n ϭ 5), systolic heart murmur (n ϭ 2), and volume overload (n ϭ 1). AVF site and aneurysmal lesion was visualized by CFD and spectral Doppler recorded continuous high velocity flow (n ϭ 3). Etiology was idiopathic (n ϭ 3), gunshot (1 dog) and hemangiosarcoma (1 cat). Selective arteriography identified feline shunt locations as R brachial, R axillary, L saphenous. Vascular access utilized 4Fr, 4cm introducers in R femoral a. via percutaneous or arterial cutdown (2 cats) or carotid a. cutdown (1 cat); .018in embolization microcoils were used for occlusion. For R front AVF's: 1 cat received two 3mm diam ϫ 3cm long coils; the other received one 2mm ϫ 2cm plus two 3mm ϫ 3 cm coils. All were delivered via 3 Fr catheter (.018in. guide wire). For saphenous a. AVF: three .038in. coils were delivered (one 5mm ϫ 3cm plus two 3mm ϫ 5cm) via 4Fr catheter (.035in guide wire). In the 2 dogs shunt locations were R median a. and R axillary a. Vascular access utilized 5Fr, 4cm introducer via R median or R femoral a cutdown. Six to 8 .038in. coils (from 1mm ϫ 3cm to 8mm ϫ 6cm) were deployed in each via 5Fr catheter (.038in. guide wire). Coil slippage/embolization occurred in 1 dog, 1 cat without problems. In all-catheterization time was 2-3.5 hrs; complete occlusion resulted in 3 cats and 1 dog (1 dog retained small residual shunt); animals were discharged Ͻ24 hrs later; coil occlusion resolved the bruit, reduced extremity swelling and warmth, and decreased varicosity size; 1 cat was euthanized at 30 days for acute R front lameness (hemangiosarcoma) and 2 were asymptomatic up to 24 mo. Coils initially closed AFV in both dogs but each developed recurrent signs and shunt flow within 2 months despite stable coil placement.

Transcatheter embolization of AVF's using vaso-occlusive coils was accomplished safely, with good technical success and short hospital stay. Long-term outcome was favorable in 2/3 cats. However, durability of repair was discouraging in both dogs. To determine the ideal balloon annulus ratio (BAR) for dilation of the pulmonic valve (PV), we prospectively evaluated the immediate and long-term results of valvuloplasty (BV) in 25 beagles with congenital PV stenosis. Murmur class, electrocardiogram, echocardiogram, Doppler, and direct right ventricular (RV) pressure (PR) were determined before and after BV. Dogs were randomly assigned to be treated with single or double balloons and with a BAR of less or more than 1.25. The PV annulus size was determined from the echocardiogram and angiocardiogram. RV and systemic PR were monitored in the anesthetized dogs. The femoral vein(s) were accessed percutaneously with introducers to allow access for the dilation catheters. Dogs (n ϭ 11) were reexamined immediately post BV and on days 1 and 90. Ten dogs were adopted for long-term follow-up to day 120. Postmortem examinations were done on day 1 in 13 dogs and day 90 in 2 dogs. Wilcoxon's rank-sum and Spearman Rank correlations were used for analysis.

Intra-and inter-observer variability demonstrated agreement for annulus size (pϽ0.01) but the R value was low (Ͻ0.55). Thus, a variation of 2 mm occurred 25% of the time and this could have changed the BAR. To avoid this problem concensus measurements were used in the calculation of the BAR. Echo and angiocardiogram measurements were similar (pϽ0.001, r ϭ 0.71). There were no differences in the initial RVPR between dogs treated with 1 or 2 balloons. All dogs had decreases in RVPR and Doppler gradient, and increased stenotic valve area (SVA) (pϽ0.0001) after BV. The grade of pulmonic regurgitation increased after BV (pϽ0.001), but was not different with 1 or 2 balloons. Immediate drops in RVPR were greater with 2 balloons and higher BAR; however, evaluations on day 1 showed no differences in treatment effect between 1 or 2 balloons or based on BAR. The RVPR decreased and SVA increased further on day 90 in 60% of the dogs and most dramatically in dogs treated with higher BAR. Murmur and pulmonic regurgitation significantly decreased on day 90 (pϽ0.05). Postmortem examination on day 1 revealed marked valvular edema, which was not apparent at day 90. Tearing/leaflet avulsion (anterior ϭ 0, left ϭ 4, right ϭ 5) and tears of the commissures (anterior ϭ 11, left ϭ 11, right ϭ 4) represented the results of the BV.

It was concluded that important variability can occur in the measurement of the pulmonic valve area such that actual BAR changes by the size of the balloon used. In some dogs this may alter the long-term result; however, a good result can be expected with a wide range of BAR. Furthermore, in some dogs long-term followup reveals further decrease in RVPR due to resolution of valvular edema. Measurement of plasma cardiac troponin I concentration ([cTnI]) is a sensitive and specific means for detecting myocardial damage in many mammalian species. Studies have shown that [cTnI] increases rapidly after cardiomyocyte injury. The molecular structure of cardiac troponin I is highly conserved across species, and current assays developed for its detection in humans have been validated in many species.

In this study, [cTnI] was quantified using a two-site sandwich assay in plasma of control healthy cats (n ϭ 33) and cats with moderate to severe hypertrophic cardiomyopathy (HCM) (n ϭ 20).

Plasma cTnI concentration was significantly elevated in cats with HCM (median, 0.66 ng/ml, range 0.05-10.93 ng/ml) as compared to normal cats (median Ͻ0.03 ng/ml, range Ͻ0.03-0.16ng/ml) (pϽ0.0001). It was also highly sensitive (sensitivity ϭ 95%) and specific (specificity ϭ 97%) for differentiating normal cats from those with HCM. Cardiac troponin I concentration was weakly correlated with diastolic thickness of the left ventricular free wall (r 2 . ϭ 0.354; p ϭ 0.009), but not the diastolic thickness of the interventricular septum (p ϭ 0.8467), or left atrium/aorta ratio (p ϭ 0.0652). Furthermore, cats with congestive heart failure at the time of cTnI analysis had a significantly higher [cTnI] than cats that had never had heart failure and those whose heart failure was controlled at the time of analysis (p ϭ 0.0095 and p ϭ 0.0201, respectively).

Although further studies are necessary, determining plasma cTnI concentration may be an accurate, non-invasive method for differentiating cats with moderate to severe HCM from normal cats. Further research is needed to determine if measuring [cTnI] concentration might be useful for differentiating HCM from other cardiomyopathies in cats, or useful as a screening tool for detecting less severe forms of HCM in cat populations. Furthermore, it may be useful as an indicator of disease severity, helpful in determining therapeutic efficacy of medical management, or related to prognosis.

ACUTE LIMB ARTERIAL THROMBOEMBOLISM. Smith SA, Tobias AH, Jacob KA, Fine DM. Department of Small Animal Clinical Sciences, University of Minnesota, St. Paul, MN.

Medical records of cats diagnosed with arterial thromboembolism (ATE) at the University of Minnesota from January 1992 through October 2001 were reviewed. Outcome data were compared for only those cats presenting within 48 hours of ATE, and where therapy was attempted. Cats that died or were euthanized during hospitalization were classified as non-survivors. Discharged cats were classified as survivors unless discharged against medical advice and dead within 3 days (n ϭ 2).

Eighty-seven cats were treated for acute limb ATE of which 39 (45%) survived to discharge. Survival gradually improved over the nine years reviewed, with 73% of cats presenting in the year 2001 surviving when treatment was attempted. Median length of hospitalization among cats that survived to discharge was 2 days (range 0-10 days). Significant differences between survivors and non-survivors were identified for rectal temperature (T o , pϽ0.00001) and heart rate (HR, p ϭ 0.038). Serum phosphorus (P) was significantly higher among non-survivors (p ϭ 0.024). (see table) The presence of motor function (p ϭ 0.008) and having only one limb affected (p ϭ 0.001) were also more frequent among survivors. There was no significant difference between survivors and non-survivors when compared by age, respiratory rate, albumin, glucose, BUN, creatinine, sodium, potassium, calcium, ALT, AST or concurrent congestive heart failure. Medical records of 127 cats diagnosed with arterial thromboembolism (ATE) seen at the University of Minnesota from January 1992 through October 2001 were reviewed. Males (67%) were over-represented (odds ratio 1.75, p ϭ 0.003). The majority (81%) were mixed breeds. Over-represented breeds were Abyssinian (5), Birman (4), and Ragdoll (3) with odds ratios (p values) of 6.03 (0.0019), 10.52 (0.0001), and 14.40 (0.0016) respectively. Ages ranged from 0.1 to 18.3 years (8.58 Ϯ 4.16). Disorders identified prior to the ATE were specific diagnosed cardiac disease (9), congestive heart failure (CHF) without known specific cardiac diagnosis (4), asymptomatic murmur (6), asymptomatic arrhythmia (1), asymptomatic gallop (1), hyperthyroidism but currently euthyroid due to treatment (7), and neoplasia (2). In 76% of cats, the ATE episode was the first indication of any abnormality potentially predisposing to ATE.

Areas affected were both rear limbs plus one forelimb (3), both rear limbs (91), right rear limb (8), left rear limb (8), right forelimb (9), left forelimb (6), brain (1), and intestine (1). Motor function was present in 34% of limb cases. Hypothermia (66%) and tachypnea (91%) were common. Of cats with forelimb thrombosis only, 42% were hypothermic. Abnormalities on cardiac auscultation were detected in 72 cats (57%).

Of 90 cats with ante-mortem diagnostics performed, diseases were hyperthyroidism (12), dilated cardiomyopathy (CM,8); unclassified CM (33); hypertrophic obstructive CM (5); hypertrophic CM (19), neoplasia (6), idiopathic atrial fibrillation (1), hypertension secondary to renal failure (1), supravalvular mitral stenosis (1), congenital cardiac malformation (1), and no disease identified (3). Common abnormalities on work-up were left atrial enlargement (93%), CHF (44%), and arrhythmias (44%). Of cats without CHF, 89% were tachypneic. Median (range) respiratory rate (RR) for cats without CHF was 60 bpm (20-160); Electrocardiographic evaluation of heart rate variability (HRV) is a non-invasive technique that measures the inherent irregularity of heart rate, and serves as an indirect assessment of autonomic efferent activity. In people, HRV is a predictor of sudden death in several cardiovascular diseases, with high sympathetic tone predisposing to the development of fatal arrhythmias, and increased parasympathetic tone demonstrating a protective effect. Arrhythmogenic cardiomyopathy in boxer dogs is an inherited disease characterized by ventricular arrhythmias, syncope, sudden death, and, less frequently, congestive heart failure (CHF). The objective of this study was to perform HRV analysis in boxer dogs, and study the hypothesis that a reduction in HRV indices could be identified in dogs with cardiomyopathy.

Study subjects included 25 boxer dogs divided into 3 categories: Group 1 dogs (n ϭ 10) were considered unaffected with 0-1 VPCs per 24 hours; Group 2 dogs (n ϭ 11) were considered affected having either Ͼ1000 VPCs per 24 hours or episodes of syncope; Group 3 (n ϭ 4) dogs were in CHF. Ten nonboxer dogs served as a control group. Measured and calculated indices of HRV included: mean RR interval (mean RR); standard deviation of normal RR intervals (SDNN); the mean standard deviation of normal RR intervals (ASDNN) of 5 minute periods within the 24 hrs; and the standard deviation of the mean of normal RR intervals (SDANN) for each 5 min period within the 24 hours. For each HRV parameter, group means were compared by one-way ANOVA across the four groups. Tukey's test was performed when significant differences (pϽ0.05) were identified.

Dogs in CHF (Group 3) demonstrated significant reduction in HRV compared to the other groups. This was evident in each of four parameters tested. No differences were identified between unaffected boxers in Group 1 and affected boxers in Group 2 for any of the tested parameters. Compared to the non-boxer controls, reduced HRV was identified in both Group 1 and Group 2 boxers, but only for mean RR and SDANN. For SDNN, a significant reduction was identified between controls and unaffected boxers, but not between controls and affected boxers. Although ASDNN was reduced in Group 1 and 2 dogs compared to nonboxer controls, the difference was not statistically significant. We conclude that (1) HRV is significantly decreased in boxer dogs with CHF compared to non-CHF dogs; (2) HRV does not appear to distinguish unaffected boxer dogs from boxers dogs with ventricular arrhythmias; and (3) boxers have evidence of reduced HRV relative to normal controls, however, these differences are inconsistent and of uncertain clinical significance. Additional studies of larger populations are needed to address these issues more completely. Doppler echocardiography may be used to noninvasively evaluate left ventricular diastolic function. This study was done to determine the influence of age and heart rate on diastolic function parameters obtained using transthoracic Doppler echocardiography in healthy, non-anesthetized cats of varying ages.

Transmitral and pulmonary venous flow measurements were obtained in 51 conscious, healthy cats. A minimum of 12 cats were studied from each of the following four age groups: (1) 1-3 years, (2) 3-6 years (3) 6-12 years, and (4) Ͼ12 years. The cats were determined to be clinically normal on the basis of history, physical examination, biochemical profile, complete blood count, serum total thyroxine assay, 6-lead electrocardiogram and Doppler-derived blood pressure. Two-dimensional, m-mode and Doppler echocardiography were performed on all cats.

Fourteen indices of diastolic function were analyzed. The correlation of these variables to age and heart rate was assessed based on a Pearson correlation coefficient and a significance level of 0.05. Age positively influenced isovolumic relaxation time (p ϭ 0.010), peak velocity of late transmitral flow (p ϭ 0.002), and the ratio of peak velocity of systolic pulmonary venous flow to peak velocity of diastolic pulmonary venous flow (pϽ0.001). Age negatively influenced the ratio of peak early to peak late transmitral flow velocities (pϽ0.001) and peak velocity of diastolic pulmonary venous flow (pϽ0.001). Heart rate negatively influenced the ratio of peak early to peak late transmitral flow velocities (p ϭ 0.008). The following parameters were not significantly influenced by age or heart rate: propagation rate of early left ventricular inflow, left atrial shortening fraction, peak velocity or deceleration half-time of early mitral inflow, peak velocity of systolic pulmonary venous flow, peak velocity or duration of pulmonary venous systolic atrial reversal flow.

These results indicate that some indices of diastolic function based on transmitral and pulmonary venous flow are influenced primarily by age and less so by heart rate in normal cats. The effect of age on these parameters suggests a deterioration of diastolic function in the normal aging feline. Myocardial Velocity Gradient (MVG) and Mean Myocardial Velocity (MMV) measured by colour M-mode Doppler Tissue Imaging (DTI) are two ultrasonic variables which describe the spatial distribution of intramural velocities. MVG is defined as the slope of the linear regression of myocardial velocity estimates between endocardium and epicardium; it is considered to be independent of the overall heart motion. MMV is defined as the mean value of velocity estimates across the myocardium. The aim of this study was to assess the differences of peak MVG and MMV in the free wall of normal cats and cats with hypertrophic cardiomyopathy (HCM).

The study included 17 Normal cats (mean age: 6.18 Ϯ 3.76 years, HR: 153 Ϯ 28 bpm) and 13 cats with HCM (mean age: 6.85 Ϯ 3.51 years, HR: 150 Ϯ 31 bpm). An ATL 5000 ultrasound machine with a 7.4 MHz probe were used and recordings obtained from the right parasternal long axis view (mitral valve level).

A biphasic shift (E1 & E2 peaks) was recorded during early diastole in tracings of both MVG and MMV. Late diastole exhibited a monophasic shift (A wave). Early systole was represented by a major excursion (Se) followed by a less prominent shift (Sl) during late systole. Biphasic, oppositely directed shifts were recorded during both isovolumic contraction (IVCTr and IVCTb) and relaxation (IVRTr and IVRTb) phases.

The peak MVG (mean Ϯ SD sec Ϫ1 ), measured during all three phases of diastole, was lower in HCM than in Normal cats (E1: 5.12 Ϯ 2.62 vs 10.57 Ϯ 3.79, pϽ0.001; E2: 3.53 Ϯ 1.74 vs 7.61 Ϯ 3.1, pϽ0.001; A: 6.29 Ϯ 2.48 vs 9.02 Ϯ 3.37, pϽ0.05). Similarly, Se, Sl, IVCTb and IVRTb peak MVG were lower in HCM compared with Normal cats (5.07 Ϯ 1.94 vs 9.37 Ϯ 2.94, pϽ0.001; 2.51 Ϯ 1.08 vs 4.55 Ϯ 2.06, pϽ0.01; 2.1 Ϯ 3.62 vs 6.92 Ϯ 4.15, pϽ0.01 and 0.61Ϯ 0.75 vs 2.78 Ϯ 2.77, pϽ0.05). Interestingly, there were only significant differences between HCM and Normal cats in peak MMV (mean Ϯ SD, mm/sec) during the IVCTb and IVRTr times (22.34 Ϯ 13.06 vs 40.5 Ϯ 11.2, pϽ0.001 and 5.98 Ϯ 7.92 vs 12.05 Ϯ 5.38, pϽ0.05 respectively).

The above findings show that differences between the hypertrophied and the normal feline myocardium are predominantly due to decreased velocity gradients between endocardium and epicardium, rather than differences in mean myocardial velocity. MVG is proving to be a promising tool in the investigation of diastolic dysfunction in feline cardiac diseases. Ticlopidine (Ticlid) is a thienopyridine derivative with potent antithrombotic properties and is commonly used for the prevention of cerebrovascular accidents, myocardial infarction and stenosis of coronary arterial stents in human medicine. Ticlopidine prevents primary and secondary platelet aggregation via irreversible antagonism of ADP-sensitive receptors on the platelet membrane. The ADPinduced conformational change of the platelet membrane GPIIb/IIIa complex is also antagonized preventing binding of vWF and fibrinogen. Ticlopidine does not appear to have any direct effect on platelets but undergoes extensive hepatic metabolism to form one or more active metabolites. Therefore, ex vivo and in vivo methods are used to determine the pharmacodynamics and effective dosing protocol in the species of interest. Ticlopidine has never been evaluated in the cat. Feline systemic arterial thromboembolization (FATE) is commonly associated with myocardial disease and platelets have been suggested to play a key role in the pathogenesis. The goals of the study were to determine the pharmacodynamics, platelet responses and any possible acute toxic effects of ticlopidine in the cat.

Ticlopidine was administered to 8 adult male cats at 50 mg and 100 mg orally once a day for 10 days. Oral mucosal bleeding time (OMBT) and platelet aggregation studies were performed under ketamine and acepromazine sedation at baseline and on days 3, 7 and 10 of each treatment period. There was no significant difference in OMBT from baseline on any treatment day at 50 mg (P ϭ 0.93, 0.55, 1.00) or 100 mg (P ϭ 1.00, 0.08, 0.71). There was also no significant difference in platelet aggregation with ADP (PAA) or collagen (PAC) from baseline on any treatment day at 50 mg (PAA, P ϭ 1.00, 0.99, 1.00, PAC, P ϭ 1.00, 0.64, 1.00). Although there was a significant difference in PAA at 100 mg on day 7 (P ϭ 0.01), this was not seen on days 3 or 10 (P ϭ 0.38, 0.14) and does not appear to be a clinically significant finding. There was no significant difference in PAC on any treatment day at 100 mg (P ϭ 1.00, 1.00, 1.00). There were no adverse reactions noted during the study period.

We conclude that ticlopidine dosed up to 100 mg once a day is ineffective in altering platelet function in the cat. The lack of response may be due to an inadequate dose or failure of production of an active metabolite by the feline liver. The lack of significant changes precluded pharmacodynamic estimations. Higher dose trials are currently ongoing. Hyperhomocysteinemia (HHcy) has been causally associated with thrombovascular disease in man. We previously reported fasting HHcy and low plasma folate in a cohort of cats with arterial thromboembolism and cardiomyopathy. Because humans with suspected HHcy may have normal homocysteine (Hcy) conentrations during fasting, Hcy measurement following oral methionine challenge (OMC) has been advocated to identify postprandial Hcy surges resulting from impaired Hcy metabolism.

To evaluate OMC we studied 12 clinically healthy cats (10 DSH, 2 DLH; 6 m, 6 f; 21-64 months [mean, 38.7Ϯ12.8]; 2.5-8.5kg [mean, 4.4Ϯ1.9]) fed a similar diet. Following 24 hr fast, methionine was administered (100 mg/kg) by oral gavage. Hcy, folate, and methionine levels were measured before oral methionine (baseline, time 0), and then at 1, 2, and 4 hrs following oral methionine loading. Impaired Hcy metabolism was considered to be present if the Hcy concentration increased Ͼ 2 SD above the mean after oral methionine ingestion. Mean baseline HcyϮSD for the 12 cats was 14.2Ϯ4.3 mcmol/L. In 6/12 cats elevated Hcy Ͼ2 SD occurred 4 hrs following OMC. In this cohort with impaired Hcy metabolism, baseline Hcy was 17.2Ϯ3 mcmol/L vs 4 hr Hcy, 49.1Ϯ9.8 mcmol/L [pϽ.001]. In the remaining 6 cats Hcy metabolism appeared normal (baseline Hcy was 11.1Ϯ3 mcmol vs. 4 hr Hcy, 13.7Ϯ3.4 mcmol/L). The lowest 4 hr Hcy value in the impaired Hcy metabolism cohort Ϫ 40.2 mcmol/L-was 211% greater than the highest 4 hr Hcy value (19 mcmol/L) in the cohort with normal Hcy metabolism. Mean serum methionine for all 12 cats was 103Ϯ28 mcmol/L at baseline and 1241Ϯ581 mcmol/L at 4 hours. There was no difference in serum methionine between the cohort with impaired Hcy metabolism vs. the cohort with normal Hcy metabolism at baseline (105Ϯ14 mcmol/L vs 101Ϯ 39mcmol/L) [p ϭ .825], or at 4 hours (1310Ϯ614 mcmol/L vs 1173Ϯ 595 mcmol/L [p ϭ .705]), respectively. Folate was lower in the cohort with impaired Hcy metabolism (14.5Ϯ 2.9 mcg/L) vs the unaffected cohort (18Ϯ 3.4 mcgL) at 4 hrs [p ϭ 0.08]. No side effects occurred following oral methionine challenge.

Oral methionine challenge was well tolerated, safe, and easy to perform. Impaired homocysteine metabolism was detected in 6/12 cats at 4 hrs. Serum folate was decreased (NS) in this cohort. Thus, the pathway associated with postprandial Hcy metabolism may be limited in response to a large oral methionine load in some cats, predisposing to HHcy. This metabolic step might be limited by folate availability in certain individuals. Cardiomyopathies represent a significant cause of morbidity and mortality in cats presenting to the small animal referral hospital. This study was designed to evaluate the prevalence of the various types of cardiomyopathy, differences in their presenting clinical findings and prognosis.

The case records of cats undergoing a cardiac investigation at the Feline Unit of the University of Bristol between September 1994 and September 2001 were reviewed retrospectively. A diagnosis of idiopathic cardiomyopathy was made in 72.1% of the cases based on echocardiographic changes. The prevalence of hypertrophic cardiomyopathy (HCM) was highest (41.5%), followed by restrictive cardiomyopathy (RCM) (15%), dilated cardiomyopathy (DCM) (7.5%) and unclassified cardiomyopathy (UCM) (7.5%). One cat showed echocardiographic changes compatible with a moderator band cardiomyopathy (MBCM). Thirteen different breeds were represented, the most common being domestic shorthaired (57.5%). The mean (ϮSD, range) age of cats with cardiomyopathy at presentation was 6.8 (4.3, 0.5 to 16) years, with an equal distribution between males and females. Clinical findings, electrocardiographic changes and radiographic abnormalities were also reviewed. Dyspnoea was a major presenting complaint in all forms of cardiomyopathy but was most prevalent in DCM cases (82%). Cats with HCM were least likely to be dyspnoeic but most likely to have a heart murmur. ECG abnormalities were present in 30% of cases. All cats with DCM showed generalised cardiomegaly on radiography compared to 59% of cats with HCM. 90% of DCM cats had a pleural effusion compared to 7% of HCM cases. The median survival time for 73 cats was 300 days. Significantly greater survival time was observed for cats with UCM when compared with those with RCM (PϽ0.05) or DCM (PϽ0.05). Cats with HCM that survived longer than 30 days after diagnosis showed a significantly greater survival time when compared to cats with DCM (PϽ0.001). Evaluation of the success of various treatment regimes used was not possible due to variations with time and the number of clinicians involved. Chronic rhinitis in cats is often refractory to conventional therapy and can pose a significant threat to cats with refractory disease due to anorexia and lethargy. There is circumstantial evidence that the chronic rhinitis syndrome may represent an ineffective immune response to persistent viral infection in cats. Studies in mice and dogs have shown that systemic administration of lipid-DNA complexes (CLDC) is a potent stimulus for activation of innate immune responses and release of Th1 cytokines. The Th1 type cytokines released (including IL-12, IFN-gamma, and IFN-alpha) have all been shown to have antiviral activity and thus might be expected to improve the clinical signs in cats if a viral etiology underlies the chronic rhinitis syndrome. We hypothesized therefore that administration of CLDC encoding the feline IL-2 gene to cats with chronic rhinitis would induce a Th1-type immune response that would in turn decrease clinical signs of disease.

A single blinded, randomized, placebo controlled, crossover study of cats with chronic rhinitis that had failed to respond to traditional therapy has been initiated. Cats are treated weekly for four weeks with intraperitoneal injections of CLDC (treatment group) or liposomes only (placebo group). Cats initially assigned the placebo group are crossed over into the treatment group at the end of the fourweek treatment period. Clinical responses to treatment were assessed using a standardized scoring questionnaire completed by the owners at each treatment and one week following completion. Immune responses evaluation included assessment of humoral responses to a neoantigen (KLH), which was administered SC at the first and third treatments. Lymphocyte (CD4ϩ and CD8ϩ), memory phenotype (CD44), monocyte activation status (MHC class II expression), and T and B cell subset percentages were assessed by flow cytometry.

To date, 7 of 11 owners (63.6%) thought sneezing had improved 25%-75% and 5 of 9 owners (55.5%) thought nasal discharge had improved 25%-50% at the end of the treatment period. In contrast, 1 of 5 owners (20%) felt that there was 0-25% improvement in sneezing or nasal discharge by the last placebo injection. Side effects to date (fever, lethargy, and vomiting) have been mild or transient. Flow cytometry revealed marked upregulation of MHC class II expression on monocytes of most treated cats compared to placebo cats. Changes in T and B cell subsets were less consistent. Treated cats had numerically higher KLH titers than placebotreated cats. These preliminary results suggest that administration of this novel immunotherapeutic is safe in cats, triggers systemic immune activation, and may improve clinical signs of chronic rhinitis in some cats. Haemobartonella felis, the causative agent of feline infectious anemia, has recently been reclassified as two separate species of Mycoplasma. The large form, Mycoplasma haemofelis (Hflg), seems to be the most pathogenic. Tetracyclines have been used most frequently for the treatment of haemobartonellosis but are sometimes ineffective and can induce toxicity in some cats. Fluoroquinolones have been shown to be effective against other Mycoplasma and anecdotally have been suggested to be effective for the treatment of haemobartonellosis. The objective of this prospective study was to compare enrofloxacin or doxycycline treatment to no treatment in cats experimentally infected with Hflg.

Sixteen cats were infected with Hflg and divided into four treatment groups: doxycycline (5.0 mg/kg [2.3 mg/lb], PO, q 12 h), enrofloxacin (5 mg/kg [2.3 mg/lb], PO, q 24 h), enrofloxacin (10 mg/kg [4.5 mg/lb], PO, q 24 h) and an untreated control group. Clinical signs, PCV, blood smears, and polymerase chain reaction (PCR) were used to monitor the cats during the eight-week study period. Cats that were PCR negative at the end of eight weeks were administered methylprednisolone (20 mg/kg, IM, weekly for three doses) and evaluated weekly for three weeks and then again six months later.

All cats were confirmed positive for Hflg via blood smears and PCR. Drug toxicities were not recognized. Treatment had no effect on PCV during antibiotic therapy, but PCVs were significantly greater in the 5 mg/kg enrofloxacin group compared to the control group during the post-treatment period (P ϭ 0.022). During treatment, H.felis organism counts/1000 RBCs in the doxycycline treatment group (P ϭ 0.03) and the 10 mg/kg enrofloxacin treatment group (P ϭ 0.025) decreased at a faster rate than those in the control group. In the posttreatment period, organism counts in the doxycycline treatment group (P ϭ 0.016), the 5 mg/kg enrofloxacin treatment group (P ϭ 0.026), and the 10 mg/ kg enrofloxacin treatment group (P ϭ 0.05) decreased at a faster rate than counts in the control group. There was no significant effect of treatment on the number of positive PCR results. However, two cats in the 10 mg/kg enrofloxacin group and one cat in the doxycycline group were persistently PCR negative in whole blood after treatment despite presumed immunosuppression with glucocorticoids.

To our knowledge, these are the first documented cases of sustained clearance of Hflg in experimentally infected cats. In this study, enrofloxacin was as effective as or more effective than doxycycline for the treatment of feline haemobartonellosis.

WITH 3-AZIDO-2,3-DIDEOXYTHYMIDINE AND HUMAN AL-PHA-INTERFERON. K. Hartmann, K. Brunner, and H. Lutz. University of Georgia, Athens, GA, USA; I. Medizinischen Tierklinik der Ludwig-Maximilians-Universität München, Germany; Veterinärmedizinisches Labor, Departement für Veterinärmedizin der Universität Zürich, Switzerland.

FeLV infection is still considered to account for most infection-related deaths in pet cats. Different treatment attempts with various drugs were performed in the past but none resulted in cure or complete virus elimination. The purpose of this study was to investigate the efficacy of 3-azido-2,3-dideoxythymidine (AZT) and human alpha-interferon (huIFN) in naturally feline leukemia virus (FeLV) infected cats in a placebo-controlled double-blind clinical trial.

The study included 44 naturally FeLV-infected cats which were randomly classified in 4 treatment groups. Cats were treated for a period of 6 weeks with AZT, huIFN, huIFN in combination with AZT, or placebo. All patients were private-owned and stayed in the hospital during the 6 week treatment period. AZT was administered orally at 5 mg/kg every 12 hours. HuIFN was administered subcutaneously at 100000 IU/kg every 24 hours. To determine efficacy of AZT and huIFN, clinical, laboratory, virological, and immunological parameters were closely followed over the 6 week period.

None of the cats became FeLV p27 antigen negative during the treatment period. The average p27 antigen concentration of those cats receiving compound, however, decreased in contrast to the cats in the placebo group. In cats treated with compound, data analysis showed a tendency of improvement in the general condition of the cats. Some clinical signs (e. g. the parameter ''stomatitis'') appeared to improve. However, improvement was not statistically significant. Statistical analysis of CD4ϩ and CD8ϩ cell counts did not reveal significant changes. Combination of drugs did not show synergistic or additive effects when comparing results of cats treated with a combination of drugs to those receiving single drug medication. Most of the patients did not show side effects during the treatment with AZT and/or huIFN. In one cat AZT treatment had to be discontinued because of development of severe anemia in the third week. No side effects were observed due to administration of huIFN.

Treatment with AZT and/or huIFN did not lead to a statistically significant improvement of clinical, laboratory, virological, or immunological parameters in FeLV-infected cats, although there was a tendency for improvement in the clinical signs. A longer treatment period might increase efficacy of this drug combination. As cats may develop antibodies after subcutaneous high dose human interferon treatment after 7 to 8 weeks, a longer treatment period with this protocol may not be possible. Use of feline interferon may be an option for treatment trials in the future. Purpose: InPouch TF (Biomed Diagnostics) consists of a self-contained plastic pouch designed for combined transport, culture and microscopic observation of T. foetus from bovine prepucial or vaginal samples. We optimized and examined the efficacy of InPouch TF for the specific and sensitive bench-top diagnosis of T. foetus in feline feces.

Methods: Sensitivity of the InPouch TF was determined by inoculating pouches with 200 ul of serially diluted cultured T. foetus (10,000-0.01 organisms) in the presence and absence of feces (0.05 g) at either room temperature (RT) or 37ЊC. Test specificity was determined by inoculating pouches with 100 ul of log phase feline Giardia lamblia (ATCC50163) or Pentatrichomonas hominis (ATCC30098) and incubating at RT or 37ЊC (n ϭ 3 each). The optimal amount of fecal inoculum was determined by culture of 0.025, 0.05, 0.1, or 0.2 g/pouch of feces from a cat with T. foetus diarrhea at either RT or 37ЊC (n ϭ 2 each).

InPouch TF was field-tested against the gold standard culture method (Remel's Modified Diamonds media) using freshly voided fecal samples taken from 117 purebred cats attending an international cat show (CFA International, Houston 2001). DNA was extracted from each positive culture (Qiagen) and tested for T. foetus ITS DNA by single-tube nested PCR. DNA sequencing was performed by a commercial laboratory. Results: The absolute detection limit of the InPouch TF system was 100 organisms/200 l saline at 37ЊC. The practical detection limit of T. foetus in feces was 1000 organisms/0.05 g feces at RT. After inoculation with feces from naturally infected cats, InPouch TF became positive within 1-11 days (median 3d; n ϭ 20) and contained viable T. foetus for up to 4-mos at RT. At 37ЊC cultures from naturally infected cats were all positive within 24-hrs and contained viable T. foetus for only 1-6 days (median 2d; n ϭ 7). All pouches inoculated with Ն0.1 g feces overgrew with gas-producing bacteria within 24-hrs regardless of temperature and were not diagnostic. Neither Giardia lamblia or Pentatrichomonas hominis organisms survived in culture for longer than 24-hr at either RT or 37ЊC. Of purebred cats having voided feces at an international cat show, 17% (20/117) were diagnosed with T. foetus infection on the basis of InPouch TF culture (n ϭ 19/20), Modified Diamonds media culture (n ϭ 9/20) and/or direct smear (n ϭ 5/20). Each isolate recovered in culture (19/20) was positively identified as T. foetus by PCR.

Conclusions: InPouch TF is sensitive and specific for detection of T. foetus in feline feces. We recommend inoculating pouches with 0.025-0.05g of fresh feces, incubating at RT and examining pouches for trophozoites at 20ϫ objective EOD for 12d (50%ϩ in 3d). Purpose: To develop a sensitive and specific PCR test for the diagnosis and differentiation of Babesia gibsoni and Babesia canis. To use this test to determine the prevalence of babesiasis in kennels where B. gibsoni had been diagnosed and to detect babesiasis in North Carolina stray dogs.

Methods: For the initial reaction, PCR primers were developed to amplify a ഠ340bp segment of the 18S rDNA from both B. gibsoni and B. canis, which spans a hypervariable region. Specific forward primers were designed for the secondary reaction, to specifically amplify B. gibsoni, B. canis vogeli, B. c. canis, or B. c. rossi when paired with the reverse primer from the initial reaction. The absolute limit of detection of the PCR was determined by serially diluting plasmids containing B. gibsoni and B. canis 18S rDNA gene inserts into Tris-EDTA buffer, and using these dilutions as template for the outer primer pair. The practical limit of detection was determined by serially diluting B. gibsoni or B. canis infected blood of known parasitemia into non-infected canine blood, and using these dilutions as template for the PCR. Specificity was determined by using canine, Toxoplasma, Cryptosporidium and Neospora DNA as template, and testing the specific primers against the other Babesia species and subspecies. 149 dogs from kennels where B. gibsoni had been diagnosed and 359 dogs from animal shelters were tested for anti-Babesia antibodies by IFA. Dogs with reciprocal antibody titers Ն32 (n ϭ 53, 25 kennel and 28 stray) and dogs that appeared anemic (n ϭ 4) were tested by light microscopy and the PCR.

Results and conclusions: The absolute limit of detection of the PCR test was 1-10 plasmids/reaction. The practical limit of detection was 50 organisms/ml of blood. Canine, Toxoplasma, Cryptosporidium or Neospora DNA was not amplified by the PCR test, and there was no cross amplification by the specific primers. The majority (18/25) of kennel dogs had antibodies against B. gibsoni while the majority (18/28) of the stray dogs had antibodies against B. canis. Babesia gibsoni DNA was detected in 14 of the kennel dogs and 2 of the stray dogs. Babesia canis vogeli DNA was detected in one stray dog. Babesia c. canis or B. c. rossi DNA was not detected in any dogs. Discordant results between the specific PCR and IFA tests were found in 6/16 PCR positive dogs. The PCR test is sensitive specific and detected babesiasis more frequently than light microscopy. The PCR may be more specific than IFA. Giardia spp. infection of cats is thought to be common; prevalence in cats with diarrhea residing in north central Colorado is 3.9%. Fecal flotation is indicated for the evaluation of all cats with diarrhea to look for parasite eggs, oocysts, and cysts; microscopic examination of feces after use of the zinc sulfate centrifugation technique (ZNSO 4 ) is considered optimal by many for detection of Giardia spp. cysts. ZNSO 4 is thought to occasionally result in false-negative results when Giardia spp. cysts are overlooked due to their small size and falsepositive results when other material like yeast bodies and plant material are confused with Giardia cysts. To aid in the diagnosis of giardiasis, assays that utilize antibodies to detect either free antigen or Giardia cysts in feces have been developed.. The purpose of this study was to compare results of ZNSO 4 flotation, a Giardia-antigen assay, and an immunofluorescence assay (IFA) for the detection of Giardia infection using feces from experimentally inoculated cats.

Eight, DSH, mixed sex, 1-2 year old, individually housed, Giardia-naïve cats were inoculated with a Giardia spp. initially isolated from a human. Normal feces collected on weeks 0, 3, 6, 9, 12, and 15 were stored at 4ºC until assayed 4-8 months later. ZNSO 4 was performed according to published techniques. The Giardia-antigen assay (Heska Corporation, Fort Collins, CO) and IFA (Meridian Diagnostic, Cincinnati, OH) were performed following the manufacturer's instructions. Each assay was performed and interpreted by one individual that did not know the results of the other assays or the sample collection schedule. Results of each assay were scored as positive or negative and sensitivity and specificity of the fecal flotation and fecal antigen assay calculated using results of the IFA as the reference test.

According to IFA results, all cats were negative for Giardia on week 0 and positive for Giardia in all other samples. Relative to the IFA results, the sensitivities and specificities of the fecal antigen and ZNSO 4 assays were 95% and 100%, and 87.5% and 75%, respectively.

When used with stored, normal, feline fecal samples containing one Giarida isolate, results of IFA and fecal antigen assay were comparable and superior to ZNSO 4 . False-negative and false-positive ZNSO 4 results were attributed to distortion of cysts during storage and presence of large number of yeast bodies that grew in the samples during storage, respectively. Purpose: To determine whether or not Haemobartonella canis and H. felis (large form) can be differentiated by the comparative sequence analysis of 2 phylogenetically informative genes.

Methods: Partial 16S rRNA and ribonuclease P RNA (RNase P) genes were amplified by PCR, cloned, and sequenced from three dogs infected with H. canis and two cats infected with H. felis (large form). The sequences were aligned and compared to each other and the existing Haemobartonella gene sequences in Genbank using the Clustal V multiple sequence alignment program (DNAS-TAR, USA). The 16S rRNA alignment compared 1357 nucleotide positions, and the RNase P alignments compared 177 nucleotide positions. Sequence differences were evaluated based on the 16S rRNA and RNase P RNA secondary structures.

Results: The analyses are presented as sequence identity tables.

Conclusions: Due to its faster ''molecular clock'', the differentiation of H. canis and H. felis (large form) was more clearly defined by comparative RNase P RNA gene sequence analysis than by comparative 16S rDNA gene sequence analysis. Mast cell tumors (MCTs) are one of the most common malignant tumors of the dog, representing between 7-20% of all cancers in this species. The presence of internal tandem duplications (ITDs) in the juxtamembrane domain of Kit leading to autophosphorylation in the absence of ligand binding was previously reported in a small number of MCTs. The purpose of the following study was to determine the true prevalence of Kit ITDs in canine MCTs, and to correlate the presence of these mutations with prognosis. We evaluated 157 MCTs and identified ITDs in 1/12 (8%) Grade I, 42/119 (35%) Grade II, and 9/26 (35%) Grade III tumors, for an overall prevalence of 33%. Logistic regression analysis demonstrated that the odds of Grade II and III tumors possessing ITDs were approximately 5 times greater than Grade I tumors (P ϭ 0.10 and P ϭ 0.18, respectively). Moreover, the presence of an ITD was associated with a higher likelihood of local recurrence post surgical excision (odds ratio ϭ 2.19, P ϭ 0.22) and the development of metastatic disease (odds ratio ϭ 2.04, P ϭ 0.13). These studies underscore the role of Kit activation in MCT pathogenesis and provide the first evidence that Kit ITDs are associated with those MCTs that behave in a biologically aggressive fashion. Moreover, the high prevalence of Kit ITDs in canine MCTs combined with the relative abundance of mast cell disease in the dog provide an ideal spontaneous tumor model in which to test the safety and efficacy of novel small molecule kinase inhibitors. Harmonic ultrasound has been used to image bubble contrast agents in human medicine for many years. Recently the principles of filtering the received signal to detect harmonic frequencies has been applied to noncontrast tissue ultrasound. Tissue harmonic ultrasound (THU) has been shown to improve conspicuity of certain lesions and of normal organs in many human patients. The goal of this project was to evaluate the usefulness of THU for characterizing normal organs and lesions in cats and dogs.

Normal cats (n ϭ 38) and dogs (n ϭ 40) were imaged with fundamental ultrasound (FU) and THU using a General Electric Logic 700 ultrasound machine system. Images of the liver, gall bladder, spleen, left kidney, urinary bladder and jejunum were collected in all animals. Images of the left adrenal gland were collected in all dogs. Cats (n ϭ 12) and dogs (n ϭ 16) with a total of 40 lesions noted during routine ultrasound examinations were entered into this project. Image quality was subjectively optimized by adjustments of the overall gain and number and location of focal zones. Conspicuity was subjectively assessed by 3 readers independently as the image with the best detail, contrast and spatial resolution, and least noise. Nonparametric statistical analysis was performed on the data.

All normal cats and dogs had improved imaging with THU. The number of organs with improved conspicuity ranged from one to all organs imaged. The most common organ to demonstrate improved conspicuity was the jejunum (100% of dogs and 89% of cats). No significant association with body weight was seen in cats. In dogs the only significant improvement was seen in images of the liver in dogs weighing greater than 16 Kg.

Overall, the most common lesions were liver nodules (n ϭ 14), splenic nodules (n ϭ 8) and renal cysts (n ϭ 4). Improvement of image conspicuity was seen in all categories of lesions.

These data suggest that the improvement of the ultrasound image is consistent with THU. However, with the exception of the jejunum, the improvement was not predictable based on organ or body weight criteria. Conspicuity is a subjective assessment of image quality and may be influenced by many factors. Biases not withstanding, improved conspicuity of lesions over a wide range of relative echogenicities indicates the usefulness of this modality. Sonographers should be cognizant of the potential benefits of THU and consider the use in all dog and cat patients. Immunohistochemistry was used to examine feline lymphoid tumors for bcl-2 expression and tumor cell proliferation in an effort to distinguish lymphomas in cats that responded to chemotherapy from those cats that failed to respond to therapy.

Tumor tissues from 29 cats were selected to represent two groups-those cats that responded to chemotherapy (partial or complete responses to chemotherapy and survival 6 months or longer after the onset of therapy) and those cats that did not respond to chemotherapy (defined as progressive tumor growth despite chemotherapy, or death less than 6 months following onset of therapy). Bcl-2 and MIB1 expression and T-and B-cell immunophenotypes were detected by immunohistochemistry.

There were no significant differences in MIB1 or bcl-2 immunoreactivity between cats that were FeLV test positive and those that were negative. Mean (Ϯ SEM) bcl-2 immunoreactivity was 53.1% Ϯ 7.5%. The mean bcl-2 immunoreactivity was significantly (p ϭ 0.0042) greater in T cell lymphomas (66.8%) when compared to B cell lymphomas (22.8%). There was no correlation between bcl-2 immunoreactivity and survival time. Mean (Ϯ SEM) MIB1 immunoreactivity was 45.3% Ϯ 5.0%. The mean MIB1 immunoreactivity was significantly (p ϭ 0.0052) greater in B cell lymphomas (62.5%) when compared to T cell lymphomas (36.4%). There was no correlation between MIB1 immunoreactivity and survival time. MIB1 immunoreactivity was negatively correlated (p Ͻ 0.0001; r2 ϭ 0.47) with bcl-2 immunoreactivity.

There were 13 (45%) cats classified as responsive to therapy and 16 cats (55%) that were classified as non-responsive to therapy. Median survival time for all cats was 5 months (range, 0 to 28 months). There was no significant difference in the mean bcl-2 immunoreactivity between the responsive cats and the nonresponsive cats. Likewise, there was no significant difference in the mean MIB1 immunoreactivity between the responsive cats and the non-responsive cats.

An inverse correlation between expression of bcl-2 and growth fraction is observed in human low-grade non-Hodgkin's lymphoma suggesting a similar regulatory role for the anti-apoptotic protein in feline lymphocyte growth and differentiation. Because bcl-2 expression confers resistance to chemotherapy in some patients with human non-Hodgkin's lymphoma, a similar result may have been anticipated in feline lymphomas. There are contradictions in the human medical literature, however, regarding the prognostic significance of bcl-2 expression in non-Hodgkin's lymphoma. Vinblastine (VBL) and prednisolone (PRED) chemotherapy has been described as effective against canine mast cell tumour (MCT). Our study aimed to evaluate the use of VBL and PRED as adjuvant chemotherapy in dogs with microscopic MCT where radiotherapy was unavailable.

Inclusion criteria were: grade II or III MCT with inadequate surgical margins; treatment with VBL and PRED following surgery; and no concurrent chemotherapy. Dogs with multiple MCT, known metastasis at the time of surgery, or gross disease were excluded. Treatment comprised VBL at 2mg/m 2 IV, weekly for four weeks then every two weeks for eight weeks. PRED was given concurrently at a starting dose of 2mg/kg/day, tapering to 0.5mg/kg/day. Patients underwent full physical examination, surgery site evaluation, buffy coat cytology, and fine needle aspiration of enlarged lymph nodes or cutaneous masses at the time of follow-up.

Fifteen dogs were eligible for inclusion. Nine dogs were available for followup: six (66%) were in complete remission, one (11%) had local recurrence of MCT, and two (22%) had MCT at distant sites. Follow-up was 460 to 730 days (mean 611, median 615). Two had died: one due to sepsis, and one of liver disease. A total of 114 doses of VBL were administered; 18 (16%) were associated with toxicity, with 1 (0.9%) requiring hospitalisation. Toxicity comprised neutropenia (n ϭ 13), vomiting (n ϭ 8), and diarrhoea (n ϭ 2).

We conclude that VBL and PRED is effective as adjuvant therapy for MCT, with 88% local control in our study group. The protocol is generally well tolerated. Feline mammary carcinoma tends to be biologically more aggressive than the canine and human (at least 80% are malignant) counterpart. Bilateral mastectomy and adjuvant chemotherapy helps prolong survival, but early regional lymph node and distant pulmonary metastasis is still a challenge in feline patients. Combination chemotherapy using doxorubicin with or without cyclophosphamide has been shown to induce short-term response. The objective of this retrospective study was to evaluate the clinical efficacy of carboplatin for treatment of feline mammary carcinoma.

Fourteen cats with the histologic diagnosis of mammary carcinoma were evaluated. Clinical staging was based on the TNM classification. Carboplatin was administered in adjuvant settings at a dose rate of 150-200 mg/m 2 . A range of 2 to 6 cycles were administered. Eleven of the 14 cats had bilateral radical mastectomy and three of the 14 cats had unilateral or local enbloc excision of the primary mass.

The mean age was 11 years with a median age of 11.5 years (range 5-17 years) at the time of diagnosis. Mammary tubular adenocarcinoma was the most common histologic type. Nine of the 14 cats were diagnosed at clinical stage I, two at stage II, two at stage III and one at stage IV. Gross lymph node metastasis was observed in two cases at presentation. Excluding the cat with radiographic evidence of a pulmonary nodule at presentation (Stage IV) the remainder of the 13 cats achieved local CR after surgery. Local recurrence, distant metastasis and concurrent local and distant metastasis were noted in 1, 5 and 3 cases respectively. Four cats were still alive at the time of analysis with a mean follow up of 676 days. The median duration of first remission was 421 days. The overall mean and median survival time in this study was 676 and 449 days respectively. A total of 44 cycles of carboplatin were administered with an average dose of 36.8 mg (range 19.2-62 mg), approximately 147 mg/m2. Hematological and gastrointestinal toxicities observed were mild to moderate. No treatment related fatal sepsis was noted and chemotherapy was delayed by a week in small number of cases.

We concluded that adjuvant carboplatin might be efficacious in prolonging the overall survival of cats with mammary carcinoma. Randomized case control studies are warranted. Cavalier King Charles Spaniels (CKCS) have a high prevalence of inherited macrothrombocytopenia with no apparent clinical signs of bleeding. In a study using platelet rich plasma and aggregometry thrombocytopenic CKCS did not have increased platelet responses. Recently a more physiologic method, the PFA-100 (Dade-Behring), has become available for evaluation of primary hemostatic function. Citrated whole blood under high shear flows through a 150 m aperture in a biologically active membrane coated with collagen and epinephrine (CEPI) or collagen and ADP (CADP). The time in seconds to formation of a stable platelet plug and cessation of blood flow is recorded. Thrombocytopenia and thrombopathia results in prolongation of the PFA-100 closure time (CT). To further investigate primary hemostatic function in CKCS, PFA-100 CEPI and CADP CT's of CKCS with platelet counts above and below the reference range were compared to those of a group of control dogs before and after in vitro manipulation of the platelet count.

Citrated whole blood was collected from 37 CKCS and 7 dogs of other breeds and PFA-100 CEPI and CADP CT's measured. CKCS were divided in 3 groups based on automated cell counts: pltϾ 160.000/l (189.000Ϯ 64.000/l), n ϭ 24; 50.000/lϽpltϽ160.000/l (112.000Ϯ33.000/l), n ϭ 5; and pltϽ50.000/ l (18.000Ϯ8.000/l), n ϭ 8. The control dogs had a starting mean platelet count of 241.000Ϯ53.000/l, n ϭ 7. By in vitro manipulation of the blood two thrombocytopenic control groups were obtained with mean platelet counts of 100.000ϩ12.000/l and 50.000Ϯ11.000/l respectively. The CKCS pltϽ50. Risk factors associated with hypercoagulability and thromboembolism (TE) include imbalances of pro-and anticoagulant factors, abnormalities of blood vessel integrity, and abnormalities of blood flow. Canine protein-losing enteropathies (PLE) are associated with TE, speculated to be the result of imbalanced procoagulant, anticoagulant or fibrinolytic activity. However, the coagulation abnormalities associated with canine PLE have not been well characterized. We hypothesized that dogs with PLE would have increased fibrinogen, and decreased antithrombin (AT) and plasminogen activity, a combination that would favor the development of TE.

Twelve dogs diagnosed with PLE based on clinical signs, physical examination, chemistry values (panhypoproteinemia), urinalysis (no proteinuria), and intestinal histopathology, were evaluated. Histological diagnoses included lymphocytic/ plasmacytic enteritis (n ϭ 5), lymphangiectasia (n ϭ 4), neoplasia (n ϭ 2), and partial intestinal obstruction (n ϭ 1). Citrated plasma samples were obtained from all dogs prior to placement of intravenous catheters or administration of drugs. Samples were stored at Ϫ70ЊC until factor analysis.

Hyperfibrinogenemia (median 570.5 mg/dl, range 246-1073 mg/dl [reference range 150-470 mg/dl]) was documented in 10/12 dogs. Decreased AT activity (median 60%, range 38-136% [reference range 70-126%]) was detected in 10/12 dogs, and 6/12 dogs had hypoplasminogenemia (median 83%, range 32-152% [reference range 70-125%]). Concurrent hyperfibrinogenemia and reduced AT activity were observed in 9/12 dogs; 3 of these 9 dogs also were hypoplasminogenemic.

These results suggest that, as previously speculated, patients with PLE, caused by various underlying etiologies, may exhibit imbalances of pro-and anticoagulant factor activity. A reduction of AT activity alone is considered to be a risk factor for TE. In addition, the combination of hyperfibrinogenemia and decreased AT activity in the same patient is characteristic of a hypercoagulable state. Hypoplasminogenemia has also been associated with thrombosis. The majority of the patients in this study would thus be considered at risk of development of TE based on abnormalities of one or more of the coagulation parameters assessed. These findings may have therapeutic and prognostic implications for dogs with PLE. A more comprehensive assessment of coagulation parameters in a larger group of dogs with PLE, and documentation of the incidence of thromboembolic complications in these patients, would appear warranted.

PORTS FOR BLOOD COLLECTION IN FELINE DONORS. I Aubert, A Abrams-Ogg, I Johnstone, D Dyson, D Allen. Ontario Veterinary College, University of Guelph, Guelph, ON, Canada.

Most cats require moderate to heavy sedation to allow collection of 1 unit (ഠ55 ml) of blood. The use of sedative agents can exacerbate the hypotension that follows blood loss and is associated with potential complications such as cardiopulmonary arrest and aspiration pneumonia. We hypothesized that vascular access ports could be used to collect blood in feline donors resulting in a reduced need for sedation and a reduction in the associated hypotension.

Regular blood collections using jugular venipuncture were performed in 8 cats prior to having vascular access ports surgically placed. The ports were sutured in the interscapular space and 5 or 7 Fr catheters were threaded in the right jugular vein to the right atrium. Using the vascular access port, blood was collected every 6-8 weeks over the course of 6 months. Regular collections were repeated to complete the study. Systolic pressures measured by ultrasonic Doppler technique were recorded during each collection and units of blood were cultured before and after a 30-day storage period.

All ports remained patent and were used successfully to collect blood during the study period. Complications were rare and consisted of: self-resolving seroma (3) and mild to moderate dermatitis at implantation site (3), infection (1) and breakage of the port-catheter junction (1). The infection was controlled with systemic antibiotics and by locking the port with a mixture of antibiotics and heparinized saline. Problems associated with collection using vascular access ports consisted primarily of partial catheter obstruction which was corrected with repositioning of the cats' neck (4/8 cats). Most cats required mild sedation for these manipulations to be performed (ketamine-valium administered through the port). Three of these cats were Ͻ 5 kg and 1 was Ͼ 5 kg. All had 5 Fr catheters. Transient distress (2), pale mucous membranes (1), vocalization (2) and defecation (1) occurred once in 2 different cats during blood collection. Regular collections resulted in a mean drop in systolic blood pressure of 65 mm Hg and a mean minimal systolic pressure of 60 mm Hg versus a mean drop in systolic pressure of 20 mm Hg and a mean minimal systolic pressure of 120 mm Hg when blood was collected through vascular access ports. All blood cultures were negative.

We conclude that vascular access ports can be a safe and efficient method of collecting blood in feline donors and that their use is associated with less hypotension than that observed during regular collection. The use of heavier cats and catheters Ն 7Fr were associated with less positional obstruction and therefore a reduced need for sedation. Human and experimental animal studies have shown that sepsis results in activation of the coagulation cascade with concomitant reduction in the natural anticoagulant proteins, antithrombin (AT) and protein C (PC). These alterations are believed to play a role in the high morbidity and mortality of sepsis, however, such changes have not been documented in naturally occurring sepsis in dogs. The objective of this study was to measure hemostatic parameters in septic dogs, with specific comparisons of the concentrations of anticoagulant proteins in dogs with sepsis and healthy controls. In this study, sepsis was defined as histologic or microbiological confirmation of infection and the presence of two or more of the following criteria: hypo-or hyperthermia; tachycardia; tachypnea; or neutropenia, neutrophilia, or greater than 3% bands. Control dogs were enrolled based on a normal physical examination, CBC, biochemistry profile, and coagulation profile. Plasma samples were collected and analyzed for prothrombin time (PT), partial thromboplastin time (PTT), fibrin(ogen) degradation product (FDP) and D-dimer (DD) concentrations, and AT and PC activities. Data were compared between groups using chi-square or independent t-tests. Twenty dogs with sepsis and 28 healthy controls were enrolled in the study. Mean age was 7.5Ϯ3.7 yrs for septic dogs and 6.0Ϯ3.9 yrs for controls (P ϭ 0.206). Hemostatic results are shown below: PC and AT concentrations were significantly correlated (r ϭ 0.75; PϽ0.001). FDP (PϽ0.001) and DD (P ϭ 0.005) were significantly higher in dogs with sepsis than in controls. Ten of the 20 septic dogs (50%) died but there was no association between any of the measured variables and outcome. In conclusion, this population of dogs with sepsis had decreased AT and PC activities, and increased PT, PTT, DD, and FDP when compared to healthy controls. These findings are consistent with prior studies in experimental animals and human clinical studies. Based on these results, further investigation of the role of PC in canine sepsis is warranted. Dogs with immune-mediated hemolytic anemia (IMHA) have multiple hemostatic abnormalities and a high incidence of pulmonary thromboembolism. Plasma transfusion has been proposed for the treatment of IMHA to restore hemostatic balance and correct the thrombotic tendency. A prospective study was therefore undertaken to determine whether administration of fresh-frozen plasma, a source of both coagulation factors and natural anticoagulants, decreases the incidence of pulmonary thromboembolism (PTE) in canine patients with IMHA.

Twelve dogs were recruited to the treatment group over a two-year period. A historical control group of 20 dogs with identical selection criteria (PCVϽ20% with autoagglutination or spherocytosis) was used for comparison with the study group. Dogs with infectious or neoplastic disease were excluded. The dogs received a transfusion of fresh frozen plasma (10ml/kg) over a 2-hour period within 12 hours of admission. All dogs were also treated with immunosuppressive and anticoagulant therapy (prednisone 2mg/kg PO q12h and heparin 100U/kg SQ q6h). Packed red cell transfusions and azathioprine were given at the discretion of the clinician. Blood samples for determination of hemostatic parameters (activated partial thromboplastin time and prothrombin time, antithrombin activity, fibrinogen and D-dimer concentration) were collected prior to plasma transfusion, 30 minutes following transfusion, and at 24 and 48 hours after cessation of plasma transfusion.

Two of the twelve dogs (17%) in the study group met the criteria for DIC compared with nine of the twenty (45%) dogs in the control group. No other significant differences in signalment or presenting clinical data could be identified between the study and control groups. Three patients died within the first week after admission. Nine patients were discharged from hospital and were available for follow-up but only five patients (38%) survived longer than 30 days. Necropsies were performed on 7/8 dogs that died within twelve months of admission. Pulmonary thromboembolism was identified at necropsy in five of these dogs. There was no difference in incidence of pulmonary thromboembolism between the study and control groups. In addition, plasma antithrombin activity (p ϭ 0.84) was not significantly increased following plasma transfusion.

The results of this study do not support the use of plasma transfusion to decrease morbidity or mortality associated with pulmonary thromboembolism in dogs with IMHA. Thrombocytopenia and macrothrombocytes have been reported to occur in Cavalier King Charles Spaniels (CKCS), but the clinical characteristics of a thrombocytopathy in CKCS are unknown. The purposes of this study were to describe quantitative, functional, and morphological characteristics of platelets in the CKCS.

Blood (25 ml) from 42 clinically normal CKCS was collected from the jugular vein and analyzed within 1 hour of collection. Automated platelet counts were obtained using citrate and EDTA as anticoagulants. Manual platelet counts were performed on EDTA blood in hemacytometer chambers, in duplicate, using phase contrast microscopy. Percent platelet aggregation in response to increasing concentrations of ADP (2, 4, 8, 16, 32 uM) was determined on platelet-rich plasma prepared from citrated blood. Standard, whole mount, and scanning electron microscopy (EM) was performed to examine internal morphology, distribution of dense granules, and platelet size. A cardiologist (REG) recorded the grade, character and site of murmurs.

Thrombocytopenia (Ͻ 150,000/ul) using manual platelet counts was present in 29/42 dogs (69%). Average manual platelet count was 119,817/ul. One dog had platelet clumping and was excluded from analysis. There was a good correlation (r 2 ϭ 83%, pϽ .001) between manual and automated platelet counts using EDTA, and a good correlation (r 2 ϭ 82, pϽ .001) between manual and automated platelet counts using citrate. There were no significant differences in factors causing variation in platelet counts (specimen collection or handling, methodology, mean RBC volume, or PCV) in dogs that were thrombocytopenic. Twenty of forty-two dogs (48%) had cardiac murmurs. There was no correlation between thrombocytopenia and murmurs, sex, age, or hair coat color. There was no significant difference in percent aggregation in response to varying concentrations of ADP in CKCS with thrombocytopenia. There was also no significant difference in percent aggregation in CKCS with murmurs compared with those without murmurs.

We conclude that EDTA or citrate can be used as anticoagulants for automated platelet counts in CKCS. There was no association of thrombocytopenia with murmurs, and no association of abnormal platelet aggregation with murmurs or thrombocytopenia. Further studies are underway to identify structural abnormalities in platelet morphology, granule content, or size in CKCS with thrombocytopenia. Medetomidine (Domitor) is a potent ␣ 2 adrenoreceptor agonist which produces sedation and analgesia. The anesthetic effects of medetomidine can be rapidly reversed by atipamezole (Antisedan) making it an attractive option for the pharmacologic restraint of veterinary patients and research animals when prolonged anesthesia is not desired or warranted. During the collection of baseline platelet aggregation responses in cats sedated with medetomidine prior to administration of an anti-thrombotic drug, alterations in ADP mediated platelet aggregation were detected. The purpose of this study was to determine the impact of medetomidine sedation on ADP mediated platelet aggregation in healthy cats.

Platelet aggregation in cats was measured in whole blood by monitoring a change in electrical impedance after stimulation with ADP or collagen using a whole blood lumi-aggregometer (ChronoLog Corporation, Havertown, PA). Seven cats sedated with 30g/kg medetomidine IM had markedly diminished platelet aggregation responses to 5M ADP (0.57 Ϯ 0.9 ohms) and 50M ADP (7.71 Ϯ 6.0 ohms) when compared to 5M ADP (14.25 Ϯ 5.6 ohms, pϽ.01) and 50M ADP (20.6 Ϯ 5.5 ohms, pϽ.05) platelet aggregation responses in the same cats alternatively sedated with ketamine (Ketaset, 15mg/kg IM) and acepromazine (0.2mg/kg, 1mg maximum). Platelet aggregation responses to ADP in medetomidine sedated cats were also significantly decreased compared to platelet aggregation responses in these same cats when not under the influence of any sedative agent (n ϭ 3, pϽ.05). Ketamine/acepromazine sedation did not alter platelet responsiveness to ADP when compared to non-sedated cats (p ϭ 0.770). Platelet responsiveness to collagen (1g/ml) was similar in medetomidine sedated, ketamine/acepromazine sedated and non-sedated cats suggesting a selective ADP antagonist effect by medetomidine. Dose dependency studies of ADP initiated platelet responsiveness in blood from medetomidine sedated cats showed some platelet responsiveness at ADP concentrations of 50M and above although the magnitude of the platelet response was decreased compared to blood from ketamine/acepromazine sedated cats. Oral mucosal bleeding times were no different in medetomidine or ketamine/acepromazine sedated cats.

These studies indicate that medetomidine sedation in cats diminishes ADP induced platelet aggregation as detected by whole blood aggregometry. While this effect is likely not clinically significant, medetomidine should be avoided as a sedative agent in cats undergoing evaluation of platelet responsiveness to ADP. Erythropoiesis is compromised in patients with naturally-occurring chronic renal failure. We postulated that cats with surgically reduced renal mass would have a different hematopoietic response to acute phlebotomy than normal cats. In phase 1 study, 6 normal, non-remnant kidney (NRK) cats and 7 remnant kidney (RK) cats had complete blood counts (CBC) performed (day 0), hematocrit (hct) 37.7 Ϯ 2.27 (NRK) and 32.3 Ϯ 2.10 (RK), and were then phlebotomized (day 0 to 2) to reduce hct to approximately 20% [day 3, hct 19.4 Ϯ 1.15 (NRK) and 19.9 Ϯ 1.06 (RK)]. CBC were then performed weekly until day 58 to monitor recovery. A mixed linear model with repeated measures was used to analyze the data. Comparison of pre with post-phlebotomy hct within groups indicated that NRK cats no longer had hct values significantly reduced from pre values on day 23 (P ϭ 0.1678, hct 34.7 Ϯ 2.29). However, RK cats did not recover until day 37 (P ϭ 0.2053, hct 29.7 Ϯ 1.69).

In phase 2 study, 6 NRK and 15 RK cats were sampled for CBC and plasma erythropoietin concentration (EPO) on day 0, and were then phlebotomized (day 0 to 2). In this phase a proportional reduction in hct in the 2 groups was attained pre and post phlebotomy respectively. Hematocrit 37.6 Ϯ 1.56 (day 0) to 22.3 Ϯ 1.51 (day 3), a 40% reduction (NRK), and from 32.3 Ϯ 0.99 (day 0) to 19.7 Ϯ 0.97 (day 3), a 39% reduction (RK). CBC were performed weekly until day 58 and EPO was measured on days 2, 9 and 16. Comparison of pre with postphlebotomy hct within groups indicated that in NRK cats recovery occurred on day 16 (P ϭ 0.6753) compared to RK cats recovery on day 23 (P ϭ 0.3403). Recovery hct was 38.2 Ϯ 1.08 in NRK cats and 31.1 Ϯ 1.37 in RK cats. EPO concentrations (mIU/ml) were significantly different (P ϭ 0.002) between groups on day 2 [77.7 Ϯ 8.82 (NRK) and 30.0 Ϯ 5.58 (RK)] but were not different (P Ͼ 0.05) on day 0 [4.2 Ϯ 0.53 (NRK) and 5.1 Ϯ 0.34 (RK)], day 9 [19.3 Ϯ 3.44 (NRK) and 14.8 Ϯ 2.17 (RK)], or day 16 [6.5 Ϯ 1.79 (NRK) and 9.9 Ϯ 1.14 (RK)].

These results suggest that RK cats have a compromised capacity for erythropoiesis which may be related to an attenuated ability to generate EPO soon after phlebotomy. Anemia is a common clinical sign and cause for serious morbidity associated with chronic renal failure in cats. Allogeneic renal transplantation is an effective therapy to restore renal function. This study investigated the anemia of 10 feline patients with chronic renal failure and the effects of successful renal transplantation. In addition to the regular pre-and postoperative diagnostic evaluation, reticulocyte counts (RC) as well as serum erythropoietin (EPO) and iron parameters were determined for up to 6 months following surgery.

In all cats, the preoperative azotemia (creatinine 3.1-10.6mg/dl, normal Ͻ2mg/dl) resolved within 6 days following surgery, with 2 cats subsequently experiencing a temporary rise in creatinine secondary to an acute graft rejection episode. The preoperative Hct ranged from 14-32% (mean 22%) with normal erythrocyte indices. Perioperatively, all cats required blood type and crossmatch compatible whole blood or packed red cell transfusions. One day postoperatively, the Hct ranged from 17-30% (mean 23%). In all but 2 cats, resolution of the anemia occurred within 30 days following surgery. In 2 cats that experienced acute rejection, resolution of the anemia was delayed. The EPO (1-6mU/ml, normal 1-15mU/ml) concentration and RC (Ͻ0.7%) were both low preoperatively, and their post-transplant response was highly variable. Mild to marked EPO rises from 8-374mU/ml were observed in 7/10 cats within a day postoperatively which may reflect the release of accumulated EPO in the ischemic kidney. Subsequent rises in serum EPO were less pronounced except for 1 cat that appeared to have a sustained high EPO level (26 to Ͼ400mU/ml). Despite these EPO responses, the observed rise in the RC was mild or missed. Interestingly, serum iron concentrations were generally low pre-and post-transplantation, which may have blunted the erythroid response in these patients. In conclusion, the course of erythropoietic response in the 10 cats studied was highly variable, although all reached normal values within 4 months post renal transplantation. Medical records of cats diagnosed with IMHA, and laboratory records of cats with a Coombs test from 1991 to 2001 were reviewed. Patients with 3 or more of the following were included: regenerative anemia, autoagglutination, positive Coombs test, hyperbilirubinemia, erythroid hyperplasia on bone marrow examination. Cats with a positive test for FeLV, FIV, FIP, neoplasia, feline infectious anemia, hypophosphatemia, Ͼ10% Heinz bodies, or exposure to hemolytic toxins were excluded.

Mean (SD) age was 6.24 years (4.13). Gender distribution was FS (9) and MN (16) which was similar to the hospital population. Breeds were Himalayan (3), Siamese (2), Birman (1), Persian (1), and mixed breed (18). Himalayans were over-represented (OR 9.8, p ϭ 0.007). There was no association with recent vaccination. Common complaints were anorexia (17) and lethargy (17). Common physical exam findings were pale mucus membranes (15), murmur (11), icterus (5), underweight (3), and gallop (3).

Mean (SD) Hct on admission was 12.1% (4.1). Metarubricytosis (19) and autoagglutination (10) were common. Median absolute aggregate and punctate reticulocyte counts were 72,600 and 248,070/mcl respectively. Inflammatory leukograms were uncommon [neutrophilia (3), left shift (5), monocytosis (2)] but relative lymphocytosis (Ͼ2000/mcl) was present in 13 cats (median 2175/mcl, range: 220-8490). Thrombocytopenia was present in 6 of 18 cats with platelets counted (range 13,000-100,000/mcl). Median serum bilirubin was 0.80 mg/dL (0.3-9.9).

Nineteen cats (76%) survived to be discharged. Mean (SD) length of hospitalization for survivors was 2.8 (2.1) days. Oxygen-carrying capacity was administered to 15 cats [packed RBC's (12), fresh whole blood (4), Oxyglobin (4)]. Acute complications were pleural effusion and/or pulmonary edema (5) and disseminated intravascular coagulation (2). All surviving cats were discharged on oral prednisone. Mean (SD) dose was 15.1 (5.9) mg/cat/day [3.3 (1.1) mg/kg/day]. Three cats never achieved a normal Hct. Median time to normal Hct was 33 days (range 8-228) for responding cats. Due to lack of control or relapse, 7 cats later received cyclophosphamide (4), cyclosporin (3), dexamethasone (1), and/or splenectomy (1). Eight cats were still receiving prednisone at last evaluation. Three cats experienced a relapse after being taken off therapy. Lymphocytosis persisted (2) or recurred with relapse anemia (3) despite prednisone therapy, leading to a misdiagnosis of chronic lymphocytic leukemia in one case. Other complications of chronic therapy were myelosuppression (1), pancreatitis (1), diabetes mellitus (1) ) supportive of DIC were included. Dogs with reported or laboratory confirmed exposure to anticoagulant rodenticides were excluded. Data were collected on 252 cases.

Mean (SD) age was 7.72 years (3.51). Gender distribution was FI:7%, FS: 47%, MI:14%, MC:32% which was not different from the hospital population. Seventy-four different breeds were included, with 17% of mixed breed heritage. Cocker spaniels were markedly over-represented (OR 2.4, pϽ0.0001). The most common underlying diseases were hemangiosarcoma (39), hepatic disease (39), sepsis (29), immune-mediated hemolytic anemia (IMHA,22), renal disease (21), misc. neoplasia (20), pancreatitis (16), lymphosarcoma (14), trauma (12), immune-mediated thrombocytopenia (10), heat stroke (7), non-septic peritonitis (5), and benign splenic disease (5). Patients with trauma (mean 5.02 years) and sepsis (mean 5.98 years) were statistically younger, and patients with hemangiosarcoma (mean 10.58 years) and other neoplasia (mean 9.41 years) were older.

Schistocytosis was only noted in 10% of cases. See table for coagulation changes.

Median (range) length of hospitalization for survivors was 3 days (0-22). Plasma was administered to 157 (62%) patients with a median (range) dose of 21.3 ml/kg (4.5-169.5 ml/kg). Heparin was administered to 32 (13%) patients with a median dose of 400 U/kg/24 hours (50-250 U/kg SQ q6-8h). Administration of neither plasma nor heparin was statistically associated with survival at discharge.

Survival at discharge, 30 days, and 90 days was 49%, 34%, and 28% respectively. Trauma and benign splenic disease were associated with better survival at all three time points. IMHA was associated with worse survival at discharge and at 30 days. A DIC scoring system based on severity of coagulation abnormalities was developed. Median (range) DIC score was 33 (17-91) among survivors and 42 (17-100) among-non-survivors. The DIC score was predictive of survival at discharge (p ϭ 0.0012). Due to increased availability and improved transfusion skills, blood transfusions have become more important in the management of anemic and bleeding cats. This retrospective study analyzes the feline transfusion practices at the University of Berlin from 1998 to 2001.

Besides 7 clinic-owned cats that donated a total of 41 times, there were 127 staff-or client-owned cats donating once. All donors had type A blood except for 5 B cats. Donors were screened routinely for their health status and were sedated for collection. A total of 10-50 ml of blood (2-10 ml/kg body weight, median [m] ϭ 6) was collected using an open system and was transfused mostly as fresh whole blood. The donations were well tolerated, except for one cat with an occult cardiomyopathy that died.

Over a 3 year period 91 cats were transfused because of anemia (n ϭ 87), hypoproteinemia (2) or coagulopathy (2). The 40 cats with blood loss anemia received 62 transfusions, each cat receiving 1.7-16.3 ml/kg blood (m ϭ 6). Their pre-transfusion Hct ranging from 8-20% (m ϭ 14) changed by Ϫ5 to ϩ12% (m ϭ ϩ4) 16-24 hours post-transfusion. The 13 cats with hemolytic anemia and a pre-transfusion Hct of 6-17% (m ϭ 13) received 21 transfusions or 3.5-12.5 ml/kg (m ϭ 7) which resulted in a Hct rise of 1-9% (m ϭ ϩ3). 35 cats with ineffective erythropoiesis and a Hct of 5-20% (m ϭ 12) received 76 transfusions or 3.3-16 ml/kg (m ϭ 7), the Hct changed by Ϫ4 to ϩ19% (m ϭ ϩ4). The large range in Hct changes may be due to blood volume changes and active losses or lysis of transfused red blood cells.

The first 10 day survival rate was 63%; 10 cats with blood loss anemia, 3 with hemolysis, 18 with ineffective erythropoiesis, and 3 with hypoproteinemia or coagulopathy died or were euthanized; none of these deaths were related to a transfusion reaction.

All patients had blood type A except for 4 B and 1 AB cat and received blood typed transfusions (the AB cat received type A blood). The major crossmatch, done prior to 117 transfusions, was incompatible for 8 cats, all except for one had previously received blood. Furthermore, 1-3 weeks after transfusion a major crossmatch with the same donor was incompatible in 8 cases suggesting sensitization against the previously transfused erythrocytes. Transfusion reactions were noted in 2 cats during the second or third transfusion and included pyrexia, tachypnea, and hyperbilirubinemia.

In conclusion, beside the various risks for infectious disease transmission occult cardiomyopathy may limit the use of feline client-owned donors. With appropriate initial blood typing and subsequent crossmatching, blood transfusions are well tolerated and may increase survival, but do not always result in the expected rise in Hct. Gastrointestinal (GI) hemorrhage is a common cause of blood loss. Hemorrhage may be due to a local GI abnormality, such as neoplasia or ulceration, or diffuse coagulopathy, such as thrombocytopenia. The relative importance of different etiologies of GI hemorrhage and transfusion requirements have not been documented in dogs. The purpose of this study was to describe the signalment, etiology, transfusion requirements, length of hospitalization and outcome in dogs with severe GI hemorrhage.

The transfusion log was reviewed for dogs receiving packed red blood cell (PRBC) transfusions for GI hemorrhage. The medical records identified were further reviewed for signalment, historical NSAID or glucocorticoid use, underlying etiology of hemorrhage, transfusions received (ml/kg), length of hospitalization, and outcome. Hemorrhage was attributed to either primary intestinal causes (neoplasia, IBD, organ failure), drug-related (NSAID or Glucocorticoids), immune-mediated thrombocytopenia (IMT) or other if evaluation was insufficient to attribute the blood loss to a specific category.

During the 26 month study period, 828.5 units of canine PRBC transfusions were administered. Of these, 81.75 units (10%) were given to 55 dogs with GI bleeding. Forty-two purebreds and 13 mixed breeds were represented. Sixty-two percent were male and 38% were female. The median age was 9 years with a range of 6 months to 16 years. The median weight was 25 kg, with a range of 5.2-64 kg.

In conclusion, GI hemorrhage accounted for 10% of PRBC transfusions in the study period. IMT is a common cause of blood loss, and dogs with IMT received a significantly higher volume (p Ͻ0.01) of PRBC. NSAID and glucocorticoid use plays an important role in GI hemorrhage. Immune-mediated hemolytic anemia (IMHA) is a common acquired hematological disorder of dogs. Prognosis is guarded, with a reported survival rate of 29-70%. Human intravenous immunoglobulin (IVIG) has been recommended for use in dogs with a variety of immune-mediated diseases including IMHA. The mechanism of action of IVIG is currently not known, but is thought to in part result in the blockade of Fc receptors on the mononuclear phagocytic cells. The purpose of this study was to retrospectively evaluate the early use (first 7 days) of IVIG for treatment of dogs with IMHA.

Dogs with IMHA who received IVIG during the study period were included. Medical records were reviewed for signalment, presenting laboratory values, dose of IVIG received, other immunosuppressive therapies used, complications of treatment, length of hospitalization and outcome.

Eleven dogs of various breeds were included in the study. All dogs were Coomb's positive or autoagglutinating. The median age and weight was 6 years and 19 kg. The median PCV, WBC and total bilirubin concentration at admission were 16%, 22,000 and 4.5 mg/dl respectively. Eight dogs received 0.5g/kg once, 2 dogs received two doses 24 hours apart, and one dog received IVIG daily for 3 days. No complications associated with IVIG were detected. All dogs received immunosuppressive therapy with glucocorticoids. Ten dogs also received azathioprine, 8 dogs also received cyclophosphamide, and 2 dogs additionally received cyclosporine. All dogs were transfused with packed red blood cells (range 17.5-80.2 ml/kg). Ten dogs survived to discharge, with a median hospitalization of 9 days (range 5-18 days). Two dogs died after discharge at 2 days and 6 weeks, the other 8 (73%) are currently alive and in remission.

These data suggest that further investigation of IVIG in dogs with IMHA is warranted. Chronic renal failure is a common cause of death in cats. Lymphocytic/plasmacytic interstitial nephritis is common histopathologically, suggesting immunemediated reactions may play a role. Feline herpesvirus 1, calicivirus, and panleukopenia virus for use in feline vaccines (FVRCP) are commonly grown in Crandall-Reese Feline Kidney (CRFK) cells. As a consequence, commercially available FVRCP vaccines contain CRFK proteins. The objectives of this study were to determine whether cats inoculated with FVRCP vaccines develop antibodies against CRFK cell extracts and if so, to determine if these antibodies reacted with extracts of feline renal tissue (FRT).

Fourteen age-matched, mixed-sex, unvaccinated kittens were divided into seven pairs. To each pair of kittens, one of the following was administered: 10g of CRFK protein SQ; 50g of CRFK protein SQ; 50g of CRFK protein plus an aluminum adjuvant SQ; a FVRCP vaccine for intranasal administration, or one of three FVRCP vaccines for SQ administration. The concentration of CRFK protein used was comparable to the range detected in the vaccines. Kittens receiving CRFK proteins were inoculated every two to four weeks for a total of eight times during the study period and kittens receiving vaccines were inoculated every three weeks for three inoculations. Serum samples were collected prior to inoculation and six months later. ELISAs to detect feline antibodies that bind to CRFK cell extracts or FRT extracts were optimized. All sera were assayed in both ELISAs and absorbance values calculated. An individual cat was considered positive for antibodies against either CRFK cell extracts or FRT extracts if the mean absorbance value of duplicate post-inoculation wells was greater than the mean plus three standard deviations of the 14 pre-inoculation sample absorbance values.

None of the cats was positive for antibodies against CRFK or FRT extracts prior to inoculation. All six kittens inoculated with CRFK proteins were positive for anti-CRFK antibodies in the post-inoculation sample; five of these six kittens were positive for anti-FRT antibodies. Neither cat inoculated with the intranasal FVRCP vaccine was positive for anti-CRFK or anti-FRT antibodies post-inoculation. Of the cats inoculated with FVRCP vaccines SQ, five of six and four of six were positive for anti-CRFK antibodies or anti-FRT antibodies in the postinoculation sample, respectively.

Administration of FVRCP vaccines SQ to cats can induce antibody responses to CRFK proteins and feline renal tissues. Further research will be needed to define the role of these autoantibodies in the development of chronic renal failure in cats. Cats with spontaneous renal insufficiency frequently develop severe hypertension. The most frequently studied model of renal insufficiency in cats, the remnant kidney model, is characterized by moderate systemic hypertension but blood pressure often declines over time (Brown et al, Am J Vet Res 62:375, 2001) . It has recently been reported by Finco et al. that renal wrapping can increase the magnitude of systemic hypertension in the remnant kidney model in dogs. We hypothesized that a combination of renal wrap plus ablation would lead to sustained systemic hypertension in cats. Twenty-two cats had either two intact kidneys (Group C; n ϭ 7), left partial renal infarction with right renal wrap/ablation (Group W; n ϭ 7) or left partial renal infarction with right nephrectomy (Group RK; n ϭ 8). Arterial blood pressure (BP), was measured by implanted radiotelemetric device continuously for 91 days. Compared with values for group C, SBP was elevated (PϽ0.05) in group RK although SBP declined in this group over time. In contrast, SBP was markedly and persistently elevated (PϽ0.05) in group W cats.

On day 75, GFR (ml/min/kg) significantly (PϽ0.05) reduced in group RK (1.51 Ϯ 0.25) and group W (1.34 Ϯ 0.15) compared to group C (3.55 Ϯ 0.35). We conclude that the renal wrap model of hypertensive renal insufficiency in cats exhibits marked, sustained hypertension with reduced renal function. This new model has utility for studies to evaluate the effects of feline hypertension on target organs and to characterize beneficial effects of anti-hypertensive therapy. Hemodialysis (HD) is a standard procedure for the treatment of severe acute renal failure (SRF) in humans. Its use in dogs with ARF has received only limited review, thus we reviewed the case records of all dogs that received HD treatments for the management of ARF from January 1990 to February 2001. The diagnosis of ARF was based on conventional criteria including clinical, laboratory, imaging and histopathology. All HD treatments were performed using transcutaneous vascular access, bicarbonate-based dialysate, and Cobe Centrysystem 3 dialysis delivery system. A total of 679 HD treatments were delivered to 124 dogs (6 Ϯ 7.5 treatments per patient, range 1-43). Sex distribution was F ϭ 15, FS ϭ 41, M ϭ 35, and MC ϭ 33; age was 6.6 Ϯ 3.4 years (range 0.5-17.2); and body weight was 27.2 Ϯ 11.7 kg (range 4.6-64). The etiology was categorized as: toxic (45.2%), infectious (34.7%), hemodynamic/metabolic (7.2%), inflammatory (3.2%), and undetermined (9.7%). Ethylene glycol toxicosis and leptospirosis were the leading etiologies at 33.9% and 30.6% of the cases, respectively.

Specifics for all HD treatments are as follows:

Overall survival rate was 41%. Only 18% of the dogs with toxic etiologies (12% for ethylene glycol poisoning) survived, whereas 70% of those with infectious etiologies (76% of the leptospirosis), and 56% of those with metabolic and hemodynamic causes survived. These data support the efficacy of HD for the management of azotemia in severe ARF. Survival was consistent with human data and was notable considering that these patients had exhausted all possibilities of conventional therapy. Glucocorticoid excess can cause glomerular lesions and proteinuria in dogs, but effects of glucocorticoid therapy on proteinuria in dogs with preexisting glomerular diseases of other causes are not well characterized. Heterozygous (carrier) female dogs with X-linked hereditary nephropathy (XLHN) have proteinuria attributable to glomerular lesions caused by defective type IV collagen in their glomerular basement membranes (GBM). The goal of this study was to determine the effects of short-term prednisone therapy in such dogs.

Six XLHN-carrier females were studied. Results of a CBC, serum chemistry profile, urinalysis, urine culture and fecal flotation were negative or within normal limits, except for proteinuria in each dog. Dogs were studied for 12 weeks; 4 weeks for pre-treatment (baseline) observations, 4 weeks of prednisone therapy (2.2 mg/kg PO q24h), and 4 weeks of post-treatment observations. Each 4-week treatment period included two 2-week study intervals. Indirect blood pressures (oscillometric method) were measured weekly. Serum chemistries, including SCr, were repeated at the end of each 2-week study interval. Urine protein:creatinine ratio (UPC) was determined for a sample obtained by cystocentesis on each of the last 3 days of each study interval. The 3-day average UPC value was used as an index of each dog's magnitude of proteinuria at the end of each study interval for data analyses. Observations at the ends of 4-week treatment periods were compared to assess treatment effects, and data for 2-week study intervals within treatment periods were used to assess the time-course of responses.

During prednisone treatment, UPC increased significantly (p ϭ 0.0003) from 1.5 Ϯ 0.87 (mean Ϯ SD) at the end of week 4, to 5.6 Ϯ 2.31 at the end of week 8, but UPC returned to baseline (1.6 Ϯ 0.46 at the end of week 12) after prednisone administration ceased. Both onset and offset of treatment effects on UPC were rapid, with the full effect evident by the end of the initial 2 weeks of the 4-week treatment period. Prednisone therapy did not have significant effects on blood pressures or SCr, which were normal throughout the study.

We conclude that short-term prednisone administration rapidly causes a substantial, reversible increase in the magnitude of proteinuria in carrier female dogs with XLHN. The mechanisms underlying this phenomenon are unknown, but a direct effect on the primary defect (i.e., abnormal type IV collagen content of the GBM) is an unlikely explanation. If the effect of glucocorticoid therapy on glomerular proteinuria in these dogs is nonspecific, a similar phenomenon should be expected during prednisone administration to dogs with other types of glomerular disease causing proteinuria. The presence of microalbuminuria (MA) has been shown to be an accurate predictor of subsequent renal disease in human beings with both systemic hypertension and diabetes mellitus. MA has also been observed in human beings with systemic diseases that are associated with glomerulopathy. Previous studies in dogs have shown the prevalence of MA in apparently healthy dogs and soft coated wheaten terriers genetically predisposed to developing glomerular disease to be 19% and 76%, respectively. The purpose of this study was to determine the prevalence of MA in dogs with experimentally infected heartworm disease. Twelve six-month male beagles were randomly divided into two groups of six. Group A was fed a diet with a 50:1 n-6: n-3 fatty acid ratio and group B was fed a diet with a 5:1 n-6: n-3 fatty acid ratio. All of the dogs were infected subcutaneously with 75 Dirofilaria immitis infective larvae. Urine collections via urethral catheterization were performed monthly for 14 to 23 months post infection. Urine albumin concentrations were measured using an antigen capture ELISA. To account for varying urine concentrations, results were normalized to a specific gravity of 1.010. MA was defined as urinary excretion of albumin greater than 1.0 mg/dl but less than 30.0 mg/dl. All dogs developed MA; the initial month of detection of MA was different between groups as measured by ANOVA (Group A 7.5Ϯ1.9 vs Group B 9.7Ϯ1.0, P ϭ 0.032). After the initial episode of MA, 67/82 (82%) of samples from Group A dogs had MA and 4/82 (5%) samples had overt proteinuria (Ͼ 30 mg/dl). The average concentration of urine protein in these 82 samples was 7.7Ϯ11 mg/dl. In comparison, after the initial episode of MA, Group B dogs had 45/68 (66%) samples with MA and 7/68 (10%) samples with overt proteinuria. The average concentration of urine protein in these 68 samples was 8.3Ϯ14 mg/dl. In both groups MA increased over time and MA preceded overt proteinuria when it occurred. The overall number of proteinuric urine samples (Ͼ 1.0 mg/dl) in both groups after the initial onset was 123/150 (82%) with 75% of the samples having MA and 7% having overt proteinuria. Eleven of these 12 dogs had histologic evidence of glomerular lesions on light microscopic and/or immunocytochemical evaluation.

This study shows that the prevalence of MA in dogs with experimentally infected heartworm disease is higher than that observed in apparently healthy dogs and similar to that observed in soft coated wheaten terriers. Further study is necessary to determine if all heartworm-infected dogs with MA will progress to develop overt proteinuria.

NEPROPHATHY OF SOFT COATED WHEATEN TERRIERS. S Vaden, 1 U Giger, 2 K Spaulding, 1 R Sellon, 3 M Littman, 2 T Harris, 1 M Afrouzian, 4 J Jennette, 4 D Williams, 5 S VanCamp. 1 1 North Carolina State Univ, Raleigh, NC; 2 Univ of Pennsylvania, Philadelphia, PA; 3 Washington State Univ, Pullman, WA; 4 Univ of North Carolina, Chapel Hill, NC; 5 Texas A&M Univ, College Station, TX.

A common syndrome of protein-losing enteropathy (PLE) and/or protein-losing nephropathy (PLN) has been documented in male and female soft-coated wheaten terriers (SCWT). PLE is associated with eosinophilic or lymphocytic/ plasmacytic inflammatory bowel disease and is usually diagnosed at an earlier age (mean 4.5 years) than PLN (mean 6 yrs), which is the result of focal mesangioproliferative and sclerosing glomerulonephritis. Although analysis of pedigrees from SCWT at large revealed a common male ancestor, the exact mode of inheritance could not be determined.

The purpose of this study was to further characterize the mode of inheritance of PLE and PLN in SCWT by three types of breedings: Group 1 consists of a litter of 6 SCWT (3 M, 3 F) born to an affected female ϫ affected male SCWT. Group 2 represents a litter of 4 SCWT (1 M, 3 F) born to a brother ϫ sister mating of Group 1 dogs. Group 3 involves an outcross of an affected male SCWT from Group 1 to a healthy Beagle and resulted in 8 (4 M, 4 F) offspring. These dogs were maintained at the North Carolina State University, fed a standard diet, and were serially evaluated through serum biochemical profiles, CBCs, urinalyses, urine protein:creatinine ratios (normal Ͻ1) and fecal ␣1 proteinase inhibitor (␣ 1 -PI; normal Ͻ5.67 g/g) every 3 months, small intestinal biopsy and renal ultrasound every 6 months and renal biopsy every 12 months. In order to make a diagnosis of PLE and/or PLN, dogs needed to have both laboratory and histopathologic abnormalities that were consistent with the diseases.

In Group 1 SCWT, 6 of 6 developed PLE (n ϭ 1), PLN (3) or both (2); the present age of 2 survivors (PLE/PLN, 1; PLN, 1) is 7 yrs. Both of the parents of Group 2 but none of the 4 dogs in Group 2, developed PLN at 4.5 yrs of age, however, 3 of them have PLE (1 dog appears to have no evidence of disease, but has just reached the mean age to develop PLE). Of the 8 outcross dogs in group 3 (present age 5.5 yrs), 1 has PLE/PLN and 1 has PLN.

Although the breeding of affected to affected SCWT in Group 1 and 2 producing primarily affected SCWT is consistent with an autosomal recessive mode of inheritance, the development of PLE and PLN in the SCWT ϫ Beagle outcross suggests an autosomal dominant trait. The risk for SCWT to develop PLE and PLN is clearly inherited, however, further breeding and clinical follow up is needed to reach a definitive conclusion. Anemia of chronic renal failure is responsive to exogenous erythropoietin administration, corroborating the belief that bone marrow suppression is a major factor in genesis of the anemia. Adverse effects of renal failure on the integrity of circulating RBC is less clearly defined. The purpose of this study was to examine some characteristics of circulating RBC in cats with chronic, induced renal failure.

Eight young adult cats (group 1) had renal mass reduced (remnant kidney model of renal failure) and were fed a commercial ''renal'' diet. Seven normal cats (group 2) were fed a commercial maintenance diet and were housed under the same conditions as group 1. All cats received 50 mg of iron dextran IM monthly. After 6 months cats were fasted overnight and blood was obtained for analysis for RBC osmotic fragility, serum creatinine concentration, and CBC.

The NaCl concentration at which 50% of RBC were hemolysed was computed for each cat from the slope of a truncated line derived from plotting NaCl conc vs % hemolysis. A one-way ANOVA was used to detect differences between groups in data collected; mean and SD values are reported.

Group 1 had significantly (P ϭ 0.007) higher serum creatinine concentration (3.3ϩ1.6 mg/dl) than group 2 (1.3ϩ0.2 mg/dl). Hematocrit values were significantly (P ϭ 0.004) lower in group 1 (29.9ϩ5.3%) than in group 2 (37.9ϩ3.0%). The NaCl concentration causing 50% hemolysis was significantly (P ϭ 0.005) higher in group 1 cats than in group 2 cats, but the slope of the lines did not differ (P ϭ 0.792). Incubation of washed RBC from group 1 with pooled plasma from group 2 for 24 hr at 4 C prior to conducting osmotic fragility tests did not alter the values for 50% hemolysis. Likewise, incubation of washed cells from cats of group 2 with pooled plasma from group 1 did not affect osmotic fragility of RBC from group 2.

These studies indicate that RBC from cats with chronic azotemia are osmotically more fragile than normal. The osmotic fragility may be due to changes in the cells rather than a consequence of a plasma factor. Increased RBC fragility could contribute to anemia of cats with chronic renal failure. The limited availability and the high cost of renal replacement therapy (dialysis and transplantation) necessitate less invasive and costly means to circumvent the limitations of conventional treatments to correct the azotemia and fluid overload of ESRD in dogs. Oral sorbents have been studied as adjuncts to therapy but have not gained acceptance due to lack of efficacy and narrow solute specificity. Recent developments in polymer technology provide renewed interest in the use of sorbents to ameliorate azotemia and improve fluid regulation in ESRD.

We evaluated the effects of an acrylate-based CLP (Dow) on azotemia, need for dialysis, and fluid management in a hemodialysis-dependent dog with ESRD (Ccreat ϭ 0.015 ml/kg/min). After a 2 week baseline period (B) CLP was given incrementally at 1 (Rx-1), 1.5 (Rx-1.5), or 2 (Rx-2) g/kg/d via a PEG tube in 3 divided doses. Hemodialysis (Kt/V ϭ 2) was performed twice weekly in B, then as needed to maintain a time-averaged urea concentration (TAUC) equivalent to B. Effects of CLP were assessed by the rate of postdialytic increase in urea (urea) and creatinine (creatinine), dialytic interval, and modeled dialytic interval to obtain TAUC ϭ 40mg/dl. Systemic blood pressure (BP), daily fluid retention, and bioimpedance spectroscopy (BIS) were used to evaluate fluid balance.

CLP at doses Ն 1.5 g/kg/d delayed the appearance of urea and creatinine and reduced dialysis dependency. Systolic BP decreased but ECF volume and daily fluid retention remained constant. The purpose of this study was to evaluate simplified methods for estimation of plasma clearance of iohexol in dogs and cats.

Plasma clearance was determined in 38 dogs and 23 cats following a bolus IV injection of iohexol (300 mgI/kg in dogs and 450 mgI/kg in cats). Iohexol plasma concentration was determined using x-ray fluorescence and clearance was calculated using a 2-compartment model based on a 10 point curve as a reference method. Plasma clearance was normalized using the body surface area. A 2sample method based on a single compartment model and a single sample method based on a linear-quadratic model were investigated. Simplified methods were evaluated by calculating the standard deviation of the difference (SDD) between the clearance obtained with the simplified method and the 10-point reference method. All combinations of sampling time for the 2-sample method and all sampling times for the single sample method were evaluated. The best sampling times were chosen for dogs and cats as the ones yielding the lowest SDD. Linear regression analysis was performed between the reference method and the simplified methods and coefficient of determination (R 2 ) was calculated.

The best combination of time for the 2-sample method was 5 and 120 min (R 2 ϭ 0.8754) in dogs and 20 and 180 min (R 2 ϭ 0.9805) in cats. The best time of sampling for the single-sample method was 60 min (R 2 ϭ 0.813) in dogs and 80 min (R 2 ϭ 0.9551) in cats.

It was concluded that iohexol plasma clearance can be estimated in dogs and cats with one or two blood samples with a reasonable margin of error compared to plasma clearance calculated using a 2-compartment model and 10 blood samples. Testing for excess urine albumin content holds promise as a method for early detection of dogs with progressive renal disease. Dogs with progressive nephropathies, particularly those involving damage to the glomerular filtration barrier, might be identified by finding excessive albuminuria even before they develop overt proteinuria. To test this theory, we examined the temporal relationship between development of albuminuria and the onset of overt proteinuria in male dogs with Xlinked hereditary nephropathy (XLHN).

Male dogs with XLHN have a rapidly progressive glomerular disease caused by a genetic defect in type IV collagen, which is a glomerular basement membrane (GBM)component. Early in life, renal structure and function are normal, but GBM lesions begin to develop by 2 months of age. Onset of overt proteinuria occurs between 3 and 6 months of age, and progression to end-stage renal failure happens by the time the dogs are 6 to 18 months old. For this study, we evaluated serial urine samples obtained from 36 males with XLHN and 10 normal male littermates while they were 2 to 7 months (8 to 30 weeks) of age.

Voided urine was collected every 2 weeks from all dogs, and weekly from some dogs, for analyses including protein:creatinine ratio (UPC) determination. Aliquots of urine were stored at Ϫ80ЊC and subsequently assayed for albumin concentration using a canine-specific competitive binding ELISA method. Urine albumin (UAlb) concentrations (mg/dl) were expressed as values normalized to 1.010 urine specific gravity, as well as in a ratio to urine creatinine (UCr) concentration. A total of 421 urine samples were analyzed.

Among 92 samples from 10 normal dogs (3-12; median, 10 samples/dog), only 4 samples from 3 dogs had UAlb values Ն 1 mg/dl, and no dog had 2 consecutive samples containing excessive albumin. In affected dogs, onset of persistent proteinuria (UPC Ն 2 then and thereafter) occurred at 14 to 30 (median, 20) weeks of age. In these dogs, onset of persistent albuminuria (UAlb Ն 1 mg/dl then and thereafter) occurred at 8 to 23 (median, 15) weeks of age. Equivalent results were obtained when albuminuria was alternatively defined as UAlb/UCr Ն 30 mg/gm. Thus, persistent albuminuria was observed for 0 to 16 (median, 4) weeks before onset of overt proteinuria. Only one dog failed to demonstrate albuminuria on at least one occasion before onset of proteinuria, but his urine was not tested weekly.

We conclude that persistent albuminuria is a reliable indicator of incipient nephropathy in dogs with XLHN, a rapidly progressive glomerular disease. These results support the concept that screening for albuminuria could be an effective method for early detection of other progressive renal diseases involving glomerular damage in dogs. In cats with chronic renal failure (CRF) decreased plasma potassium concentration appears to be a risk factor for hypertension. Since aldosterone plays a major role in both potassium homeostasis and control of blood pressure, the aim of the present study was to determine if plasma aldosterone concentrations (PAC) are altered in cats with hypertension and/or CRF and to determine if these changes occur due to adrenal disease, or secondary to stimulation of the renin-angiotensin system.

Aldosterone was extracted from plasma with dichloromethane and measured by radioimmunoassay (RIA). Plasma renin activity (PRA) was measured by RIA of generated angiotensin I. All cats were evaluated by history, physical examination, plasma biochemistry and systolic blood pressure (SBP) measurement (Doppler method). Cats were considered hypertensive if SBP was Ͼ175 mmHg repeatedly or was accompanied by ocular lesions. Hyperthyroid cats were excluded from the study. PAC and PRA were compared in normal aged cats, cats with CRF and normal SBP (CRF-NT), hypertensive cats with CRF (CRF-HT) and hypertensive cats with normal creatinine (i-HT) by a Kruskal-Wallis test. In addition, data were obtained from NT-CRF cats after feeding a renal care diet (WALTHAM, Low Phosphorus Low Protein diet), and from hypertensive cats (CRF-HT and i-HT) after treatment with amlodipine (0. In this study hyperaldosteronism was not commonly detected in hypertensive cats. When evaluating aldosterone in cats, diet must be considered. Although treating hypertension with amlodipine increases PRA there is no resulting increase in PAC. The mechanism for this suppression of aldosterone secretion remains to be determined. Preliminary work in our laboratory showed that higher dietary NaCl intake increases urine output without increasing urine CaOx saturation. As dietary NaCl might negatively affect blood pressure, the purpose of this next study was to compare blood pressure in healthy cats fed a normal or moderately increased NaCl acidifying diet.

Ten healthy cats (mean age 2.5 years) were used in the study. Prior to the start of the trial, cats were familiarized to metabolic cages and blood pressure measurements. Cats were then randomly divided in 2 groups and fed according a cross over design either a normal (0.46% Na, 1.33% Cl on a DM basis) or moderatly elevated (1.02% Na, 2.02% Cl on a dm basis) NaCl dry expanded diet for 2 weeks. Between the 2 treatment periods cats were fed a maintenance diet for 1 week. During the experimental period food and water intake, urine pH and volume were measured daily, body weight weekly. Blood pressure was measured 5 times, twice daily, by Doppler ultrasonic sphygmanometry. Results are expressed as a mean Ϯ SD. The F-test for cross over designs was used for the statistic analysis.

Bodyweights remained stable over the study (4.5Ϯ 0.89 kg). Food intake was similar between the 2 diets (70 Ϯ 8 g vs 70 ϩ 8 g). Water intake was increased (pϽ0.01) on the higher NaCl diet (34 Ϯ 3 vs 29 Ϯ 3 ml/kg bw/d) as was urine volume (70 Ϯ 22 ml/cat/d vs. 47 Ϯ 14 ml/cat/d). Urine pH was not different between the 2 diets. Mean systolic blood pressure on the higher NaCl diet was 134 Ϯ13 mm Hg versus 133Ϯ14 mm Hg on the control diet.

Feeding a moderately increased NaCl diet enhanced water intake and diuresis but did not affect systolic blood pressure compared to the level of NaCl commonly found in acidifying diets. Moreover, systolic blood pressures on those 2 diets were within normal limits (Ͻ 160 mm Hg).This study suggests that a moderate increase in dietary NaCl has no deleterious effect on blood pressure of healthy adult cats. This study was performed to describe clinical and laboratory findings, potential causes, outcome, and prognostic indicators in dogs with acute renal failure (ARF). Medical records were reviewed to identify dogs with ARF from 1986 to 2001 and pertinent information was retrieved. Statistical analysis was done to test for association between prognostic indicators and outcome and detect differences in continuous measures between groups based on category (cause) of ARF and outcome. Data are presented as mean Ϯ SE.

Of 643 cases reviewed, 80 had ARF and were included in the study. Clinical signs included depression/lethargy (58 dogs), vomiting (51 dogs), neurologic signs (32 dogs), inappetence (21 dogs), hypothermia (16 dogs), and abdominal pain (14 dogs). Mean duration of signs prior to admission was 3.88 Ϯ 0.37 days. There was no association between duration of signs and outcome; however, duration was lowest in dogs with ethylene glycol toxicosis (EGT) (2.34 Ϯ 0.40 days) (P ϭ 0.001). EGT and leptospirosis were diagnosed in 35 and 6 dogs, respectively. Twenty-four dogs had more than one concurrent disorder (eg, DIC, sepsis, EGT) associated with ARF. Single miscellaneous disorders (eg, hypercalcemia, hepatic disease) were present in 8 dogs. A potential cause was not found in 7 dogs. Sixteen dogs (20%) survived, 55 dogs were euthanized (69%) and 9 dogs (11%) died. Mortality rates were 100% for EGT, 75% for multiple disorders, 63% for miscellaneous disorders, and 33% for leptospirosis. There was an association between category of ARF and outcome (P ϭ 0.0003). Mean duration of hospitalization was greater in dogs that survived (10.15 Ϯ 1.25 days) compared with dogs that did not (2.19 Ϯ 0.27 days) (P Ͻ 0.0001). Initial anion gap was significantly higher in dogs that did not survive (31.42 Ϯ 1.93) compared with dogs that lived (19.16 Ϯ 1.45) (P ϭ 0.006). While there was no association between presence of severe initial or maximal azotemia (creatinine Ͼ 10 mg/dl) and outcome, there was an association between presence of severe maximal azotemia and category of ARF (P ϭ 0.038). Urinalysis revealed proteinuria in 70% (35/51), cylindruria in 11% (6/53), glucosuria in 31% (17/54), and calcium oxalate crystalluria in 25% (14/56) of dogs. Urine volume was lowest in EGT (1.38 Ϯ 0.40 ml/kg/hr) and highest in dogs with multiple disorders (4.75 Ϯ 1.74 ml/kg/hr) (P ϭ 0.04). There was no association between urine volume and outcome (P ϭ 0.25).

In conclusion, the most common cause of ARF in this study was EGT; however, many dogs had multiple disorders. While there were many diagnostic and prognostic indicators in these dogs with ARF, it is important to consider that multiple factors interact to affect diagnostic findings and determine outcome. Clinically, mild to moderate chronic renal insufficiency is a frequent problem in cats. While serum creatinine is most often used to assess renal function in cats, each laboratory traditionally establishes its own normal ranges and the utility of these normal ranges to diagnose early chronic renal insufficiency has been challenged.

To assess the utility of measurement of serum creatinine concentration in the diagnosis of renal insufficiency in cats, serum samples were taken from 15 cats with either reduced renal mass (Group RF; n ϭ 10) or intact renal mass (Group N; n ϭ 5). Renal function was assessed as GFR by measurement of urinary clearance of inulin to confirm normal GFR of all cats in Group N and a reduced GFR for all cats in Group RF. Simultaneously obtained serum samples from each cat were then sent to 12 clinical pathology laboratories in the US following a standard protocol. The utilized laboratories were located at academic institutions (n ϭ 8) or commercial facilities (n ϭ 4).

Although intra-laboratory coefficient of variation was generally less than 6%, there was substantial variation among laboratories for serum samples obtained simultaneously from the same cat, supporting the need for separate reference ranges for each laboratory. However, when reported values were compared to established reference or ''normal'' ranges at each laboratory, there was considerable variation in the utility of these ''normal'' ranges. Ten cats in this study had a reduced GFR but one laboratory reported all of their serum creatinine values as ''normal''. This is consistent with previous suggestions about the lack of sensitivity of this test as currently employed. However, using the ''normal'' ranges and accompanying results from several laboratories resulted in ''elevated'' serum creatinine concentrations being reported for some, or all, of the 5 cats with a normal GFR from Group N.

Our results indicate that a reference range individualized to each laboratory is needed and that results from different laboratories cannot be directly compared. Critically, veterinarians relying upon currently published laboratory ''normal'' ranges for serum creatinine concentration for identifying the presence of chronic renal insufficiency in cats will experience a high probability of both false negative AND false positive results. The purpose of this study was to develop a technique to measure equine plasma endothelin-1 (ET-1) in order to evaluate the association between plasma levels of ET-1, laminitis and season. The pathophysiology of equine acute laminitis remains poorly understood, but ischaemia and reperfusion of the sensitive laminae may be important. Specific factor(s) mediating this ischaemia have yet to be elucidated but the potent vasoconstrictive peptide ET-1 may be one candidate.

Twenty-one horses and ponies prone to developing laminitis and 23 control animals from the same farm were studied. Whole blood was collected every 3 months by jugular venipuncture into EDTA (Ethylene-diamine-tricarboxylicacid) with aprotinin (444.7 TIU/1 ml plasma). The sampling periods represented the seasons of the year (summer, fall, winter, spring). The plasma was harvested and stored at Ϫ80ЊC for batch assay. A human Endothelin-1 Enzyme Immunometric assay kit, with a limit of detection of 0.16 pg/ml, was validated for the measurement of ET-1 in equine plasma. The repeatability and precision of the assay was measured by the intra-assay and interassay variance, and the recovery ascertained by measuring the amount of ET-1 in 2 equine samples spiked with 8 pg/ml and 50 pg/ml of human ET-1 respectively. Specificity was evaluated by diluting 4 equine plasma samples 2-fold and correcting the measured ET-1 by the dilution factor. A 2-way repeated measures analysis of variance was used to compare the ET-1 levels between the 2 groups (laminitic and control) and within the 2 groups (season). Significance was set at P Յ 0.05.

The mean intra-assay and interassay CV was 12.3% and 17.3%. The mean ET-1 recovered from the samples spiked with 8 pg/ml was 67.9% while 87.1% was recovered from the samples spiked with 50 pg/ml. Parallel dilution was evident with ϳ100% of the ET-1 recovered following correction. The control group's mean plasma ET-1 level was 0.75 Ϯ 0.09, 0.68 Ϯ 0.09, 0.74 Ϯ 0.10 and 0.76 Ϯ 0.12 pg/ml for fall, winter, spring and summer while the mean level for the laminitic group was 0.67 Ϯ 0.18, 0.58 Ϯ 0.10, 0.63 Ϯ 0.11 and 0.69 Ϯ 0.14 pg/ml. The ET-1 level of the 2 groups was not significantly different at any of the time points, nor were there any differences between the ET-1 levels during the different times within each group.

These results indicate that a human ET-1 Immunometric assay kit can be used to measure equine plasma ET-1. No association was found between plasma ET-1 levels, time of year or predisposition to laminitis. These results, however, do not rule-out a role for ET-1 in the development of laminitis as a locally acting vasoconstrictor mediator. We investigated the potency and cyclooxygenase (COX) isoform selectivity of ML-1,785,713, a novel, experimental COX-2-selective inhibitor in horse blood and evaluated its efficacy in an experimental model of acute lameness. In vitro whole blood COX-2 inhibitory potency and selectivity over COX-1 was determined for ML-1,785,713 and phenylbutazone (PBZ) as described by Brideau, et al (Am J Vet Res 2001 , 62:1755 . In vivo efficacy was evaluated in a pressure-induced lameness model. Lameness was induced in adult horses using a modified horseshoe, with an adjustable screw to apply pressure to the sole of the hoof. Pressure was applied to the hoof until animals experienced a 15-20% reduction in weight bearing (ϭ baseline lameness), measured by force plate analysis. In a cross-over design, horses (n ϭ 9) were dosed orally with vehicle, PBZ (4.4 mg/kg) or ML-1,785,713 (0.0625 mg/kg to 0.25 mg/kg) 1 to 1.5 hours after induction of baseline lameness. Weight bearing on the lame limb was then measured 2, 4 and 6 hours after compound or vehicle treatment and was compared among treatments at each time after compound administration using paired t-tests. Statistical significance was set at p Յ 0.05.

ML-1,785,713 was 265-fold selective for COX-2 vs. COX-1. The IC50 (mean Ϯ sem) for COX-1 was 23.9 Ϯ 8.1 M and for COX-2 was 0.09 Ϯ 0.01 M. PBZ was essentially non-selective for COX-2 vs COX-1 (IC50 for COX-1 ϭ 6.2 Ϯ 1.8 M and for COX-2 ϭ 3.79 Ϯ 0.72 M). In the lameness model, horses in all treatment groups had an average baseline lameness of approximately 17 percent reduction in weight bearing before drug treatment. Horses in the vehicle group became progressively more lame during the trial, and had a reduction in weight bearing (mean Ϯ sem) of 30 Ϯ 9 percent at the 6 hour time point. In contrast, horses receiving ML-1,785,713 became significantly less lame during the trial, with a reduction in weight bearing of 9 Ϯ 2 percent (0.125 mg/kg) and 10 Ϯ 1 percent (0.25 mg/kg) at the 6 hour time point. Efficacy of ML-1,785,713 at these dosages was similar to that observed with PBZ.

These results suggest that selective inhibition of the COX-2 enzyme by ML-1,785,713 provides comparable anti-inflammatory efficacy to PBZ. Given the improved therapeutic index of COX-2 inhibitors over non-selective non-steroidal antiinflammatory drugs (NSAIDS) in humans, we anticipate that a potent and selective COX-2 inhibitor such as ML-1,785,713 will provide a greater therapeutic index than non-selective NSAIDS in the horse. It has been hypothesised that acute laminitis may be triggered by the release of factor(s) from the gut resulting in digital vasoconstriction. This phenomenon commonly follows the fermentation of carbohydrate in the cecum and colon. Although the factors causing this condition have not yet been elucidated, we have recently shown that a number of amines are formed in the equine cecum and the rate of formation increases with fermentation. Many monoamines, such as tryptamine and phenylethylamine, have been shown to be potent vasoconstrictors of isloated digital blood vessels in vitro, due to their structural similarities with 5-hydroxytryptamine (5-HT; serotonin) and catecholamines. The aim of the present study was to to examine the effects of infusions of low doses of tryptamine and phenylethylamine on digital blood flow in the normal horse.

Four adult thoroughbred horses, two females and two males, aged 3-7 years, were used in this study. Tryptamine or phenylethylamine were infused at concentrations of 64 and 48 mcg/kg, respectively, in 500 ml sterile saline, and given over a period of 30 minutes, with saline alone being used as a control. Heart rate and systemic blood pressure, measured by an indirect oscillometric method, were monitored throughout the experimental period. Blood flow in the lateral digital artery was measured at 0, 15, 30 min and 15 min post infusion by the Doppler ultrasound method, using a 10 MHz linear transducer, and in addition, coronary band temperature was measured using an infared thermometer. Heparinised plasma samples were also taken from the contralateral jugular vein for measurement of platelet-free plasma 5-HT and amine concentrations.

Both amines caused a decrease in blood flow in the lateral digital artery compared to controls, with tryptamine causing a 21.8 Ϯ 9.0% decrease in blood flow at 15 min (pϽ0.05) and a decrease in coronary band temperature from 26.5 Ϯ 2.0 to 25.8 Ϯ 2.1 ЊC. This amine also caused a 1.5-2 fold increase in plasma 5-HT concentration. There were no detectable effects of the amines observed on heart rate or systemic blood pressure at the doses infused.

These data suggest that if monoamines formed in the equine cecum are released as a result of lactic acid production in carbohydrate overload, they are capable of causing selective vasoconstriction in the equine digit. Therefore they should be considered as potential trigger factors in the pathogenesis of acute laminitis. Furosemide is the most widely used diuretic in horses. It is routinely administered as intravenous (IV) bolus doses of 0.25-1 mg/kg q8-24 h. This intermittent administration (IA) regimen causes fluctuations in electrolyte concentrations, plasma volume and activity of the renin-angiotensin-aldosterone system (RAAS). Studies in man show that continuous rate infusion (CRI) of furosemide increases efficacy, but reduces fluctuations in fluid and electrolyte balance. We hypothesized that CRI of furosemide would have the same positive effects in horses with less RAAS activation.

Six mares were used in a randomized, crossover study. During a 24-hour period each horse received a total of 3 mg/kg furosemide by either CRI (0.12 mg/kg/h preceded by a loading dose of 0.12 mg/kg IV) or IA (1mg/kg q8h IV). The mares were catheterized with indwelling 28F Foley catheters to allow constant collection of urine. Urine volume and concentrations of electrolytes (Ca 2ϩ , Na ϩ , K ϩ , Cl Ϫ , HCO 3 Ϫ , Mg 2ϩ ) and aldosterone in urine were recorded. Serial blood samples were obtained and analyzed for packed cell volume, total protein, electrolytes and BUN. Indirect blood pressure and ECGs were monitored. Urine output and clinical values were initially examined graphically for outliers and normality. Urine output and clinical values of horses after they received IA and CRI were compared independently at specific times post-treatment with the Wilcoxin's signed rank test. A P value of Ͻ 0.05 was considered statistically significant.

Only 4 of 6 horses produced more urine in 24 hours with CRI compared to IA. However, there was significantly greater urine output after CRI in the first 8 hours. Significantly more K ϩ and Ca 2ϩ were excreted after CRI compared to IA after 24 hours. After 8 hours, horses on CRI excreted more K ϩ and Cl Ϫ . Na ϩ excretion increased after CRI, but not significantly. HCO 3

Ϫ and Mg 2ϩ were lost to the same extent with both methods. Serum K ϩ dropped significantly after each bolus and the CRI loading dose. While K ϩ recovered between doses for the IA horses, it stayed low throughout the study with CRI. Serum Ca 2ϩ decreased with both methods, but significantly only for CRI after 8 hours, with 3 CRI horses becoming mildly hypocalcemic after 24 hours. Serum Cl Ϫ concentrations decreased significantly for both methods. Serum Na ϩ decreased, but HCO 3 Ϫ and Mg 2ϩ increased, though not significantly, throughout the study. Five of 6 horses excreted less aldosterone after CRI, indicating less RAAS activation.

We conclude that furosemide CRI provides more vigorous diuresis than IA during the first 8 hours and produces less RAAS activation. Serum levels of K ϩ and Ca 2ϩ should be monitored and supplementation provided as needed. Fentanyl, a synthetic opioid, can be delivered by a transdermal therapeutic system (TTS) for continuous analgesia. This study investigated the pharmacokinetics of fentanyl following intravenous (IV) and TTS administration in healthy adult horses.

In the first portion of the study, horses were given fentanyl by an IV bolus of 2 mg (N ϭ 6) and by application of two 10 mg TTS patches (Duragesic; Janssen Pharmaceutica, Titusville, NJ) to shaved skin on the antebrachium or gaskin for 48 h (N ϭ 3) or 72 h (N ϭ 3), on two study days, with one week between treatments. In the second portion of the study, two 10 mg TTS fentanyl patches were applied every 48 h for 8 days (N ϭ 3) or every 72 h for 9 days (N ϭ 3). Serum was collected before and for up to 96 h after all treatments. A radioimmunoassay was used to determine serum fentanyl concentrations. A three-compartment model described the disposition of IV fentanyl, whereas the Loo-Reigelman method was used to determine the rate and extent of absorption of the first dose of TTS fentanyl.

Complete blood counts, biochemistry panels, and physical examination findings remained within normal limits during and after TTS fentanyl administration. Following IV fentanyl administration, the mean (Ϯ SD) extrapolated peak serum concentration at time zero was 44.0 Ϯ 15.3 ng/ml, the volume of distribution (Vd) of the central compartment was 0.10 Ϯ 0.04 L/kg, Vd at steady state was 0.66 Ϯ 0.18 L/kg, total body clearance was 5.92 Ϯ1.37 ml/kg·min, and the elimination half-life was 2.2 Ϯ 0.6 h. The bioavailability of the first dose of TTS fentanyl was 99% Ϯ 10%, with a peak serum concentration of 2.62 Ϯ 0.61 ng/mL at 8.7 Ϯ 2.3 h. After multiple doses of TTS fentanyl, the average serum concentration at steady state was 1.2 Ϯ 0.3 in horses in which the patches were changed every 72 h, whereas it was 1.91 Ϯ 0.51 ng/mL in horses in which patches were replaced every 48 h.

Following administration of TTS fentanyl patches in horses, serum fentanyl rapidly reached and maintained concentrations (1-2 ng/mL) that are analgesic in other species. Due to the rapid absorption of fentanyl, TTS patches may need to be replaced every 48 h, rather than at 72 h intervals as is recommended by the manufacturer, in order to provide uninterrupted pain relief in horses. Meropenem (Merrem) is a member of a new class of beta-lactam antibiotics, carbapenems. These drugs are advantageous due to their bactericidal and post antibiotic effects as well as being resistant to most beta-lactamases. Their broad spectrum of activity includes drug-resistant bacteria responsible for some small animal infections. To determine a dosage regimen for dogs, pharmacokinetic parameters were determined from a single intravenous (IV) and subcutaneous (SC) administration at 20 mg/kg to 6 healthy dogs. To characterize the distribution of meropenem in dogs and demonstrate a unique tissue concentration sampling method, an in-vivo ultrafiltration device was used to collect extracellular fluid (ECF). To measure meropenem's potential for treating urinary infections, urine concentrations were measured. Plasma, urine and tissue fluid samples were analyzed via HPLC with UV detection. Plasma data was analyzed by compartmental and non-compartmental pharmacokinetic methods. An ultrafiltration device determined plasma protein binding of meropenem. Mixed effects models with different correlation structures were used to analyze the data statistically. Plasma pharmacokinetic characteristics of meropenem include: short half-life, high clearance (CL), excellent absorption from SC, and a small volume of distribution (VD). Half-life, VD, and CL after IV meropenem for plasma were 0.67 Ϯ 0.07hr, 371.51 Ϯ 52.77 ml/kg, and 391.99 Ϯ 90.72 ml/hr/kg, respectively. The half-life for meropenem in tissue fluid was 0.84 Ϯ0.5 hr. The half-life of meropenem after SC administration was 1.0 Ϯ 0.21 hr and 1.07 Ϯ 0.54 hr for plasma and tissue fluid, respectively. Protein binding was 11.87% and the bioavailability after the SC dose was 84%. Analysis of our data show that tissue fluid concentrations are almost identical to protein unbound plasma concentrations. Urine concentrations were at least 50 ϫ plasma concentrations during the collection periods. Owing to the kinetic similarity of meropenem in the extravascular and vascular space, tissue fluid concentrations can be predicted from plasma. We concluded that a dosage of 30 mg/kg SC every 12 hrs, or 13 mg/ kg every 8 hours would achieve adequate tissue fluid and urine levels for sensitive bacteria with a MIC up to 1 mcg/mL. We investigated the potency and cyclooxygenase (COX) isoform selectivity of ML-1,785,713, a novel, experimental COX-2-selective inhibitor in dog and cat blood and evaluated its efficacy in experimental models of inflammation. In vitro COX-2 inhibitory potency and selectivity vs. COX-1 was determined for ML-1,785,713 in whole blood assays from each species as described by Brideau, et al (Am J Vet Res 2001 , 62:1755 . In vivo efficacy was evaluated in a canine inflammatory synovitis model and in a feline pyrexia model. Dogs (n ϭ 4-6) were treated PO with vehicle or ML-1,785,713 (2 or 4 mg/kg) 2 hours before receiving an injection of monosodium urate crystals in one femoropatellar joint. Weight bearing on the injected limb was evaluated by force plate analysis 0 (baseline), 2, 4, 6 and 8 hours after joint injection. Treatment efficacy was evaluated by comparing the area under the percent of baseline weight bearing vs. time curve (AUC) between groups. ANOVA was performed on Tukey's normal scores of the ranks of the AUC. Cats (n ϭ 3-4) were treated PO with vehicle or ML-1,785,713 (1 or 3 mg/kg) 1 hour before IV challenge with LPS (E. coli O55:B5, 0.5 g/kg). Body temperature 0-14 hours after LPS challenge was monitored continuously by radiotelemetry; ANOVA was used to compare the average temperature after LPS challenge between groups. For both studies statistical significance was set at p Ͻ 0.05.

In dog blood ML-1,785,713 was 430-fold selective for COX-2 vs. COX-1. The IC50 (mean Ϯ sem) for COX-1 was 129.1 Ϯ 58.6 M and for COX-2 was 0.30 Ϯ 0.07 M. In cat blood ML-1,785,713 was 58-fold selective for COX-2. The IC50 for COX-1 was 7.5 Ϯ 2.0 M and for COX-2 was 0.13 Ϯ 0.03 M. In the synovitis model, ML-1,785,713 significantly improved lameness in a dosagedependent manner. The AUC (mean Ϯ sem, percent weight bearing-hour) for each treatment was: vehicle-170.6 Ϯ 28.6; ML-1,785,713 (2 and 4 mg/kg)-569 Ϯ 57.4 and 771.0 Ϯ 8.7, respectively. In the feline pyrexia model ML-1,785,713 significantly prevented LPS-induced fever in a dosage dependent fashion. The average temperature (mean Ϯ sem, ЊC) for each treatment was: vehicle-39.4 Ϯ 0.1; ML-1,785,713 (1 and 3 mg/kg)-38.7 Ϯ 0.1 and 38.3 Ϯ 0.1, respectively.

We conclude that a selective COX-2 inhibitor is efficacious in the dog and cat in acute models of inflammation and pyrexia, respectively. Given the improved therapeutic index of selective COX-2 inhibitors in humans, we anticipate that selective COX-2 inhibitors in the dog and cat will provide veterinarians with a new class of antiinflammatory agents with improved therapeutic indices for these species. The immunomodulator Cyclosporine A (CsA) is characterized by great interpatient variability in disposition, including absorption. Bile is necessary for the absorption of the original CsA oral formulation (Sandimmune) but apparently not the newer microemulsion formulation (Neoral). Higher blood concentrations are achieved with the microemulsion formulation in human beings. The purpose of this study was to determine the pharmacokinetics of CsA microemulsion capsules in dogs.

Six apparently healthy adult dogs were administered a single 5mg/kg oral dose of the microemulsion formulation of CsA. Dogs were fasted 12 hours before oral CsA administration and food was withheld for an additional 12 hours following oral CsA administration. Twenty-three whole blood samples were collected in EDTA containing vacutainer tubes from each dog through an indwelling jugular catheter and analyzed for CsA concentrations by using HPLC method and radioimmunoassay (RIA). Both assays were validated for canine whole blood at concentrations from 33.5 to 1404.9 ng/mL. Coefficient of variation for each assay was less than 15% for all except the lowest concentration (20%). Using WinNonLin, noncompartmental analysis (NCA) was performed for both HPLC and RIA and compartmental analysis (CA) for HPLC. Comparisons were made between the RIA and HPLC values using a paired t-test.

Mean results did not differ for any parameter resulting from NCA between the two assays. Mean results for CA of HPLC data is presented in the accompanying table. Maximum concentration (Cmax) was 1024 Ϯ 488 ng/ml. The disappearance half-life (t1/2d), based on that portion of the curve that represented the greatest portion of the area under the curve (AUC) and therapeutic concentrations was 47 minutes. Cmax and AUC of CsA when administered as microemulsion were approximately 2 fold or more higher than the same parameters (when adjusted for dose differences) previously reported for the original formulation. The results of this study suggest that the bioavialability of the microemulsion product is improved in dogs compared to the original product. Enrofloxacin (ENR) is a fluoroquinolone (FQ) antimicrobial approved for use in veterinary medicine. As with other FQs, it can accumulate in phagocytic white blood cells (WBC). Previous studies in our laboratory have demonstrated up to 84 fold accumulation of ENR in circulating canine WBC compared to plasma. Presumably, accumulation in circulating WBC may result in higher concentrations of ENR at the site of inflammation such as occurs with infection. The purpose of this study was to investigate the effect of accumulation of ENR in circulating WBC on the concentration of ENR at an inflammatory site. Dogs (n ϭ 5) were surgically implanted with bilateral subcutaneous tissue chambers. Localized inflammation was induced in one chamber with a sterile irritant (carrageenan 1%); the other chamber served as a control. Peripheral leukocytes were collected from each dog, exposed in vitro to radiolabeled enrofloxacin to allow accumulation, and injected back into the donor 10 hours after the inflammatory insult. Mean dose of ENR administered in WBC was 1085 Ϯ 178 ng/kg. Samples were collected from blood (circulating WBC and serum) and each the inflamed and non-inflamed chambers prior to and at 2, 4, 6, 12 and 24 hours after administration of WBC loaded with ENR. Chamber fluid was separated into WBC and extracellular fluid (ECF) and, along with plasma and WBC isolated from peripheral blood, were subjected to ENR analysis using liquid scintillation counting. The lower and upper limits of the assay were, respectively, 0.22 and 6.6 ng/ml; controls that spanned these concentrations predicted within 15%. ENR was not detected in any plasma, circulating leukocytes or in non-inflamed chambers. ENR was detected in both ECF and WBC collected from the inflamed tissue chamber of each dog. Maximum ENR concentration 2.00 Ϯ 0.50 ng/ml in the ECF occurred at 6.4 Ϯ 3.2 hours after administration and in chamber WBC 0.55 Ϯ 0.27 ng/ml at 4.8 Ϯ 4.1 hours. Disappearance half-life of ENR in ECF and WBC collected from the inflamed chamber were 17.8 Ϯ 7.8 hours and 23.1 Ϯ 9.2 hours respectively. This study demonstrated the ability of circulating WBC containing ENR to increase the concentration of ENR at the site of inflammation. The North American Llama and Alpaca population is approximately 275,000 animals respectively. In spite of this, few studies have been done to determine pharmacokinetic behavior of drugs in SAC. Consequently, most health decisions regarding antimicrobial use are based upon extrapolation from other species. TMPS combinations are widely used to treat bacterial infections in Llamas and Alpacas. Information regarding appropriate drug doses and dosage intervals specific to Llamas and Alpacas are not available. The purpose of this study was to determine the fate of orally administered TMPS combinations in Llamas and Alpacas. Six healthy male-castrate alpacas were used in this study. The animals were housed in stalls and fed hay and water ad libitum. After intravenous administration of 15mg/kg IV (2.5mg/kg trimethoprim, 12.5 mg/kg sulfamethoxazole) concentration-vs.-time profile suggest that drug disposition is modeled most appropriately using a 2 compartment equation. Maximum concentration (Cmax) of trimethoprim (T) and sulfamethoxazole (S) in plasma was 6.5Ϯ2.1g/ ml and 111Ϯ96g/ml respectively. Elimination half-life was 1Ϯ.7 hr (T) and 2.2Ϯ0.6 h (S) respectively. The mean residence times were T ϭ 1.0Ϯ0.5 and S ϭ 2.8Ϯ.6 hours. The areas under the respective concentration versus time curves (AUC) were 17Ϯ1.3 g*hr/ml (T) and 124Ϯ60 g*hr/ml (S). After intragastric administration of 15 mg/kg, 30mg/kg and 60 mg/kg the respective Cmax and Tmax for sulfamethoxazole were 1.9Ϯ0.8, 2.6Ϯ0.4 and 2.8Ϯ0.7 at 1.2, 1.5 and 2 h respectively. The AUC for S at oral doses 15mg/kg, 30mg/kg and 60mg/kg was 9.1Ϯ5 g*hr/ml, 25.9Ϯ3.3 g*hr/ml and 39.1Ϯ4.1 g*hr/ml respectively. Based upon these AUC values and correcting for dose, the bioavailability of sulfamethoxazole was F ϭ 7.7% at 15mg/kg, F ϭ 10.5% at 30mg/kg and 7.94% at 60mg/kg. T was not detected in plasma after intragastric administration (15 mg/kg) except in one animal where T was detected at 4 hours (Cmax ϭ 0.49g/ ml). T was not detectable after oral doses of 30mg/kg or 60mg/kg. These data demonstrate that therapeutic concentrations of trimethoprim-sulfamethoxazole are not achieved after oral administration to Alpacas. Hydrated sodium calcium aluminosilicate (HSCAS) is a harmless phyllosilicate clay commonly used as an anticaking agent in animal feeds. Preliminary research has shown that HSCAS is capable of tightly and selectively adsorbing aflatoxin in vitro and in vivo (Phillips, 1995; Mayura et al., 1998; Grant and Phillips, 1998) . When included in the diet at levels between 0.5-2.0% (w/w), HSCAS clay significantly protected animals including chickens, turkeys, and swine from levels of aflatoxin as high as 7,500 ppb. As recently as 1998, about 250 dogs died in Texas from exposure to aflatoxin in a variety of dog food products. Dogs ingested aflatoxin amounts in the range of 150 ppb over a threemonth period. The corn used in the specific pet foods fed was found to be contaminated with aflatoxin. Studies were conducted to investigate the efficacy of HSCAS dietary clay in offering protection for dogs consuming aflatoxin in the diet.

Five dogs were studied and randomly fed a commercially available complete and balanced diet containing either HSCAS clay (0.5%) or no clay (control) for a period of 7 days in a crossover design. After the 7-day feeding period, the dogs were given a predetermined low level non-toxic single dose of aflatoxin B1. Urine samples were collected from the dogs at specified intervals during the next 48 hours. The urine was analyzed for aflatoxin metabolites M1, P1 and Q1 using HPLC. Following a 5-day washout period, the dogs were switched to the other diet and the process repeated. Normal aflatoxin clearance curves for clayunprotected dogs were determined. Results showed that the urinary metabolite aflatoxin M1 regularly clears the dog's system in 48 hours. On average, 71.5% of this metabolite is cleared in the first 6 hours after dosing, increasing to 90.4% after 12 hours. These findings concurred with work by Emafo (1976) indicating that dogs do not demethylate aflatoxin B1 to the urinary metabolite aflatoxin Q1. In this study, no aflatoxin Q1 was found in the urine in contrast to large amounts of M1. Urinary metabolites were measured and compared between the control diet fed period and clay-supplemented diet fed period to determine the effects of HSCAS clay in the diet. The dietary HSCAS clay diet provided an average decrease in urinary aflatoxin metabolites of 58.1% as compared to the control diet. It is concluded that the HSCAS dietary clay is efficacious in protecting dogs fed pet foods even at minimal aflatoxin contamination. The significance of this finding is that despite regular and careful screening of ingredients for aflatoxin, low levels may reach the final product undetected. HSCAS clay may therefore provide the pet food industry further assurance of canine diet safety for pet foods. The signalment, medical history, laboratory findings, histopathology, and hepatic copper concentrations were characterized in 10 cases of Dalmatian dogs with hepatic copper toxicosis. All hepatic copper concentrations were determined at Colorado State University Diagnostic Laboratory using atomic absorption analysis and concentrations were expressed as mg/g dry weight liver (dwl). In 9 of 10 cases, hepatic tissue was available for histopathologic evaluation by a single pathologist (DJM).

Nine of the 10 Dalmatians in this study had a history of anorexia, vomiting, or diarrhea. In all cases, the initial chemistry panel revealed marked increases in liver enzyme activity. The relative increase in ALT activity was much greater (average 10.8 times above normal) than the relative increase in ALP activity (average 5.5 times above normal). The mean hepatic copper concentration for nine Dalmatians was 3197 g/g dwl (normal Ͻ 450 g/g). In five of the nine dogs, hepatic copper concentration exceeded 2000 g/g dwl. In 6 cases copper granules appeared to be predominantly located in zone 1, in zone 3 for 3 cases, and in the liver sample with the highest copper content it was densely distributed throughout the lobule. The mean hepatic copper concentration of those dogs found to have a predominantly Zone 1 distribution was 2458 g/g liver dry weight, and in those with a predominantly Zone 3 distribution was 2698 g/g liver dry weight. Necroinflammatory alterations intimately attendant to copper-laden parenchymal cells was the notable histopathologic finding. The inflammatory infiltrate was either primarily lymphocytic or neutrophilic. Morphologic features of cholestasis were generally not prominent except in those with severe pathology. The relative magnitude of the ALT increase also was consistent with a primary hepatocellular toxicity rather than cholestatic liver disease with secondary copper accumulation.

A hepatic copper toxicosis was identified and characterized in ten Dalmatian dogs. Early recognition is essential for instituting potentially effective treatment. Future studies should focus on identifying the probable defect in copper metabolism in this breed.

Microbubble contrast agents have been used in a variety of diagnostic studies in human medicine. These agents were first used in cardiovascular studies, and have expanded into vascular studies involving several other organ systems. Our goal was to investigate the usefulness of contrast harmonic ultrasound (CHU) for acquired and congenital vascular diseases. Dogs and cats with suspected vascular diseases were imaged with a General Electric Logic 700 ultrasound machine system with power doppler (PD), and single pulse (HH) and angiographic (HA) harmonic settings. Ultrasound contrast (Definity (TM), Bristol-Myers Squibb Medical Imaging) was injected intravenously (0.1-0.2 ml) for each harmonic sequence. Imaging sequences were directly transferred to a digital video camera.

Dogs with acquired (n ϭ 2) and congenital (n ϭ 2) portosystemic shunts (PSS) and a cat with aortic and renal thromboembolic (TE) disease were imaged. In both dogs with congenital PSS the shunting status was easily detected by determining the peak time to perfusion of the liver. Compared to published norms (mean ϭ 23 seconds), both dogs with congenital PSS had shorter peak perfusion times (5 & 7 seconds). In one dog with acquired PSS due to cirrhosis, the CHU was useful in detecting hepatofugal flow and patency of the portal vein, both of which were not detected by PD.

Both PD and CHU accurately detected the location of the aortic obstruction in a cat. The CHU was more useful in detecting low velocity flow in the peripheral arteries of the thrombosed leg and confirming poor perfusion in the musculature of the affected compared to the contralateral limb. Both CHU and PD identified renal vessels to the level of the arcuate arteries in effected and unaffected kidneys. The HH characterized the perfusion pattern of normal and thrombosed kidney better than either PD or HA.

We conclude that CHU is useful in clinical cases of suspected vascular disease. In cases of suspected PSS the time to peak measurement may be a quick and simple test to identify a large hepatic artery component to liver blood flow and, therefore, portosystemic shunting. Further testing is necessary to validate these preliminary data. CHU is clinically useful to detect regions of poor perfusion due to TE diseases and is more accurate than conventional ultrasound methodology. This study was designed to test the treatment effects of S-adenosylmethionine (SAMe), a hepatoprotective antioxidant, in a feline model of acetaminopheninduced oxidative damage.

Cats received either SAMe-only for fourteen days, a single dose of acetaminophen and no further treatments, or a single dose of acetaminophen followed by fourteen days of SAMe treatment. Measures of oxidative damage included methemoglobin concentration, Heinz body formation, packed cell volume (PCV), thiobarbituric reacting substances (TBARS), and glutathione concentration in both the blood and the liver. A repeated measures ANOVA was used to determine the significance of time, treatment, and the time-treatment interaction.

All cats that received acetaminophen had a significant (pϽ0.0001) increase in methemoglobin and Heinz body production. A significant interaction of time and treatment (pϽ0.0001) was found for Heinz body formation among groups; those cats treated with SAMe had numerically lower percent Heinz bodies than acetaminophen-only cats. A significant effect for the interaction of time and treatment was also found for PCV (p ϭ 0.008), with SAMe-treated cats demonstrating a smaller decline than untreated, acetaminophen-only cats. Reduced glutathione (GSH) concentration increased significantly (pϽ0.0001) in the blood of cats that received SAMe, and in those that received acetaminophen. No significant changes were observed in blood or hepatic TBARS concentrations. Changes in the hepatic GSH:GSSG ratio were not significantly different between groups, although cats receiving only acetaminophen tended to increase their liver oxidized glutathione (GSSG) concentrations, while those cats receiving only SAMe tended to decrease their liver GSSG concentrations.

S-adenosylmethionine therapy, instituted one hour following drug administration, showed evidence of protecting against acetaminophen-induced changes in the formation of Heinz bodies and the decline in PCV over time. Growth-hormone (GH) release by the pituitary gland is controlled by two antagonistic hypothalamic peptides: somatostatin inhibits GH release and GH releasing hormone (GHRH) stimulates GH release. GH release can also be stimulated by synthetic GH-secretagogues (GHSs), such as growth hormone releasing peptide-6 (GHRP-6). These GHSs elicit their effect on GH release via a specific GHS receptor that is not activated by GHRH or somatostatin. Recently, the endogenous ligand for the GHS receptor has been identified, it is called ghrelin. In rats, ghrelin is a selective GH releaser. Cushing's syndrome, a state of chronic hypercortisolism, is associated with impaired GH secretion. In humans both, GHRH and GHRP-6 are unable to induce significant GH release, neither in patients with pituitary-dependent hyperadrenocorticism (PDH) nor in patients with hyperadrenocorticism due to an adrenal tumour.

In 7 dogs with PDH we studied the effects of intravenous administration of GHRP-6 and ghrelin, both in a dose of 2 mcg. per kg body weight. The results were compared with those obtained in 8 healthy dogs, that also were tested with GHRH (2 mcg./kg) and control solution (NaCl 0.9%).

In healthy dogs ghrelin, GHRH and GHRP-6 caused significantly higher GH responses than the control solution. Ghrelin induced significantly higher GH release than GHRP-6. Ghrelin elicited a higher ACTH and cortisol response than the control solution. GH release after administration of ghrelin in healthy dogs (20.1 Ϯ 4 ng/ml) was significantly higher (PϽ0.001) than in dogs with PDH (1.0 Ϯ 0.5 ng/ml). GH release after administration of GHRP-6 did not differ significantly between healthy dogs (2.0 Ϯ 0.9 ng/ml) and dogs with PDH (0.3 Ϯ 0.1 ng/ml). ACTH, cortisol and TSH responses after administration of both ghrelin and GHRP-6 in healthy dogs were not significantly different from those in dogs with PDH.

In conclusion, in the dog ghrelin appeared to be the most potent GHS. In healthy dogs, ghrelin is not a selective GH releaser, as it also elicited ACTH and cortisol secretion. Dogs with PDH have an impaired GH release after administration of ghrelin. The pharmacokinetic and pharmacodynamic effects of glargine, a new synthetic human insulin analogue, protamine zinc beef-pork insulin (PZI), and Caninsulin, a purified pork lente insulin (lente) were evaluated in 9 healthy, neutered adult cats (5 male, 4 female). A 3-way triple crossover study was performed in which the serial plasma concentrations of insulin (insulin) and plasma concentrations of glucose (glucose) were determined over a 24hr period after SC administration of the three insulins (0.5U/kg of bodyweight) at 3-day intervals. Following administration of all three insulins, glucose decreased significantly (pϽ0.05). There was no significant difference in time to onset of action between insulins. Time to reach nadir glucose concentration was longer for glargine (16ϩ1.9hrs) and statistically different (pϽ0.05) from both PZI (6ϩ2.3hrs) and lente (4ϩ0.5hrs). There was no statistical difference between nadir glucose concentration for all three insulins. Time for glucose to return to baseline was significantly shorter for lente, than for glargine or PZI. PZI and glargine were not statistically different. Mean daily glucose concentration following insulin administration was similar for glargine (3.2ϩ0.1mmol/L) and PZI (3.1ϩ0.1mmol/L) and were both significantly lower than lente (4.0ϩ0.1mmol/L). The area under the 24hr glucose curve was also similar for glargine (70ϩ5mmol/L.hr) and PZI (71ϩ4mmol/L.hr) which were both significantly lower than lente (96ϩ7mmol/ L.hr).

The radioimmunoassay used to measure insulin was less sensitive to glargine than to PZI or porcine lente. After administration of all 3 insulins, mean insulin increased significantly above baseline. There was no statistical difference between the three insulins for time to reach peak insulin concentration. The area under the 24hr insulin curve was similar for PZI and lente, which were both significantly higher than glargine.

The results of this study indicate that insulin glargine produced a glucose nadir later than PZI or lente, and had greater duration of action than lente. In humans, glargine is marketed as a 'peakless' insulin, but this trial in healthy cats has shown there are definite peaks in insulin concentration and glucose lowering effects. A study in diabetic cats is required to compare these insulins and to fully evaluate the potential of insulin glargine for treatment of diabetes mellitus in cats. The antithyroid drug methimazole is widely used for the medical management of feline hyperthyroidism. Recently, custom veterinary pharmacies have offered methimazole in a transdermal gel containing pluronic and lecithin (PLO), with anecdotal evidence of efficacy. The purpose of this study was to determine the bioavailability, relative to intravenous and oral routes of administration, of transdermal methimazole in a PLO gel in cats.

Six healthy adult cats were assigned to receive 5 mg of methimazole by the IV, oral, or transdermal routes, in a randomized triple crossover protocol with one week washout between doses. Blood samples were taken for HPLC determination of serum methimazole, at 0, 5, 15, 30, 60 minutes, and 2, 4, 6, 12 and 24 hours after dosing.

Methimazole absorption following single dose transdermal administration was poor but variable, with only 2 out of 6 cats achieving detectable serum methimazole concentrations at any time point following transdermal administration. AUC, Cmax, and absolute bioavailability were all markedly lower for the transdermal route (0.39 Ϯ 0.63 mcg h/ml, 0.05 Ϯ 0.09 mcg/ml, and 11.4 Ϯ 18.7%, respectively) than for either the oral (2.94 Ϯ 1.24 mcg h/ml, 0.51 Ϯ 0.15 mcg/ ml, 40.4 Ϯ 8.1%) or IV routes (7.96 Ϯ 4.38 mcg h/ml, 3.34 Ϯ 2.00 mcg/ml, 100%). The results of this study indicate generally low to undetectable bioavailability of methimazole in a lecithin/pluronic gel given as a single transdermal dose to healthy cats, although one individual cat did achieve nearly 100% transdermal bioavailability relative to the oral route. These results are in discordance with the apparent efficacy of transdermal methimazole with chronic dosing in hyperthyroid cats. We hypothesize that chronic dosing may enhance absorption due to changes in the stratum corneum with chronic PLO administration. Controlled studies are underway to evaluate the efficacy of transdermal methimazole compared to oral methimazole, with chronic dosing, in the treatment of hyperthyroidism in cats. Reverse iontophoresis is a process by which small amounts of interstitial fluid are extracted transdermally without tissue disruption or pain. Extracted fluid can be analyzed for chemical content, including glucose. Because interstitial glucose correlates well with blood glucose, it has been successfully applied as a method of monitoring glucose levels in human diabetic patients. This study aimed to evaluate the viability of reverse iontophoresis as a method of glucose monitoring in dogs and cats.

Non-diabetic laboratory animals (7 dogs, 7 cats) were used. A central and peripheral catheter were placed in each for the purpose of blood sampling and medication administration, respectively. Between 1 and 3 non-invasive monitors (GlucoWatch, Cygnus, Inc.) were placed in various locations around the neck or trunk of each subject after clipping and scrubbing of the skin. Periods of normoglycemia, hypoglycemia and hyperglycemia were produced in random order through no treatment or intravenous administration of regular insulin or dextrose, respectively. To coincide with GlucoWatch readings, blood samples were collected every 20 minutes during a 12-hour monitoring period for glucose determination utilizing a portable glucose monitor (HemoCue, HemoCue, Inc.). Results of blood glucose measurements and signal measured by each monitor were compared utilizing standard statistical methods. Blood samples were also analyzed hourly in a laboratory for verification of HemoCue accuracy.

Measured blood glucose concentrations ranged between 23 and 529 mg/dl. Portable blood glucose monitor and clinical chemistry determinations of glucose concentrations were well correlated (R ϭ 0.97 and 0.98 for dogs and cats, respectively; pϽ0.001). Average R value from sensor-A electrical signal data collected by the non-invasive monitors was 0.78 for both dogs and cats (pϽ0.001). Corrections for individual patient factors resulted in R values of 0.85 and 0.87 for dogs and cats, respectively (pϽ0.001).

As a result of experimental design and the use of non-diabetic animals, rapid, non-physiologic alterations in blood glucose exceeding 300 mg/dl were observed to occur between 20 minute sampling periods resulting in reduction in the overall correlation between electrical signal generated through non-invasive monitoring and blood glucose determination. However, a significant correlation was observed suggesting that the use of reverse iontophoresis as a method of noninvasive glucose monitoring may be a viable option for performing glucose curves in diabetic dogs and cats. Further evaluation utilizing diabetic patients is needed to determine the true correlation between methods under more typical conditions occurring during performance of a glucose curve. Hypertension has been found to occur in over 50% of cases of canine hyperadrenocorticism (HAC). Increased vascular responsiveness to endogenous cathecolamines is one of the possible causes for glucocorticoid-induced hypertension. The objective of this study was to evaluate pressor sensitivity to cathecolamines in dogs with iatrogenic HAC (I-HAC) by serial arterial blood pressure measurements during infusions of increasing dose rates of norepinephrine.

Eight dogs with I-HAC induced by administration of oral hydrocortisone at a mean dose of 3.3 mg/kg PO TID for 42-49 days and 8 control dogs which received placebo gelatin capsules at the same interval and length of treatment were used in the study. Direct systolic, diastolic, and mean arterial blood pressure and heart rate were recorded at baseline and 10 minutes after administration of increasing dose rates of norepinephrine (0.1, 0.125, 0.2, 0.3, 0.4, 0.6 and 0.8 mcg/kg/min). Infusions were stopped when the systolic blood pressure reached 240 mmHg or the mean blood pressure reached 120 mmHg to avoid potential complications associated with severe hypertension. The mean change in direct systolic, diastolic, and mean blood pressure and heart rate in response to norepinephrine administration was compared between groups by a two-way analysis of variance.

Infusions could not be completed in 7 of the dogs in the I-HAC group due to severe hypertension. In these dogs, infusions were stopped at 0.15 mcg/kg/min in 1 dog, 0.2 mcg/kg/min in 2 dogs, 0.3 mcg/kg/min in 1 dog, 0.6 mcg/kg/min in 1 dog and at 0.8 mcg/kg/min in 2 dogs. Three of the dogs in the control group had the infusions stopped at the highest dose rate, while the remaining 5 received all dose rates. The mean change in systolic blood pressure was consistently higher in the I-HAC group versus the control group and was significant (pϽ0.05) at 0.2 mcg/kg/min dose rate. The mean change in heart rate was consistently lower in the I-HAC group and was significant (pϽ0.05) at 0.2 mcg/kg/ min dose rate. No significant differences were found in diastolic or mean blood pressure.

Increased pressor responsiveness to norepinephrine infusion is present in dogs with I-HAC. Obesity is often associated with alterations in the serum lipid profile. In man, an increase in cholesterol, triglycerides, and LDL is most often seen, while HDL is decreased. These changes predispose the individual to diabetes, cardiovascular disease and hypertension. In obese dogs, an increase in the LDL fraction of triglycerides (TG) has been observed. This study was designed to compare lipid profiles in carefully monitored lean and obese cats.

In the present study, 10 neutered obese (O) cats were compared to 10 lean (L) cats (5 females and 5 males each). The cats were fed a commercially available dry ration (Purina Pro Plan, St. Louis, MO, USA) twice daily. The body weights (BW; kg) were: 5.9 Ϯ 0.6 (O) and 3.5 Ϯ 0.6 (L; PϽ0.0001), and body mass indices (BMI; %) were 55.2 Ϯ 5.6 (O) and 39.7 Ϯ 3.6 (L; pϽ 0.0001). Body Fat (BF) compositions (DEXA scan; %) were 44.3 Ϯ 11.9 (O) and 16.7 Ϯ 4.5 (L), and girths (cm) were 46.6 Ϯ 3.5 (O) and 25.7 Ϯ 4.6 (L). The obese cats had significantly higher serum concentrations of TG (mg/dl; 44.3 Ϯ 11.9 vs. 25.7 Ϯ 4.6; p ϭ 0.0013), and cholesterol (mg/dl; 140 Ϯ 22.5 vs 117 Ϯ 25.2; p Ͻ 0.05), but no significant difference was observed between groups for nonesterified fatty acids or phospholipids. Measurement of the different lipoprotein fractions (VLDL, LDL, HDL2, HDL3 and VHDL) by density ultracentrifugation revealed that there were no significant differences in the overall density between lean and obese cats; however, the TG concentration of the VLDL fraction of obese cats was significantly higher than that of lean cats (p ϭ 0.012). Serum TG concentrations correlated well with girth measurements (R ϭ 0.81), BW (R ϭ 0.80), BMI (R ϭ 0.78) and BF (R ϭ 0.77). Cholesterol concentrations correlated best with girth and BF (R ϭ 0.39 each) and least well with BMI (R ϭ 0.30).

In conclusion: Obesity in the cat leads to marked changes in serum triglyceride and cholesterol concentrations; however, unlike in man, there were no differences in the lipoprotein fractions. Further studies are needed to determine if there is pathologic significance to the observed lipid alterations in the cat. Recent evidence suggests that diets high in protein and low in carbohydrate may be beneficial in prevention and management of diabetes mellitus by minimising postprandial glucose and insulin concentrations.

The objective of this study was to compare the effect that diets high in one of three macronutrients (protein, fat and carbohydrate) had on postprandial glucose and insulin concentrations and glucose tolerance.

Twenty-four (12F, 12M) clinically healthy, neutered cats with mean weight 4.97 kg were used. Cats were blocked into 3 groups based on gender, bodyweight and plasma concentrations of glucose and insulin.

Prior to trial, cats were fed a commercial diet (caloric distribution protein 28%, fat 46%, carbohydrate 26%) for 4 weeks. Groups of cats were then randomly assigned to 1 of 3 test diets. Caloric distribution of test diets were: high protein diet (protein 46%, fat 26%, carbohydrate 27%), high fat (protein 26%, fat 47%, carbohydrate 26%), high carbohydrate (protein 25%, fat 26%, carbohydrate 47%).

Ad libitum (AL, 12 hr) and meal response tests (MR, 24 hr) were conducted on consecutive days after 4 weeks of feeding the test diets. Ad libitum and MR tests were performed to compare the effect that different feeding patterns had on glucose and insulin concentrations. During AL testing, cats were allowed free access to food prior to and over the 12-hr test. During the MR test, fasted cats ate Ͼ90% of the 12-hr AL intake in 0.5 hr, immediately after time 0.

Mean glucose and AUC glucose for both the AL and MR tests were significantly higher in the high carbohydrate diet compared to the high protein and high fat diets (pϽ0.001; 0.008). The maximum increase in glucose was significantly higher (pϽ0.012) for the high carbohydrate diet compared to the high fat and protein diets in the MR test. Mean glucose and AUC glucose tended to be lower for the high protein diet compared to the high fat diet in both feeding tests, but the difference was not statistically significant. AUC insulin tended to be higher during the MR test for the high carbohydrate diet compared to the high protein and high fat diets, although the difference was not significant.

We conclude that a diet with 46% of the calories from protein is associated with lower postprandial glucose concentrations compared to diets with 47% of the calories from fat or carbohydrates. Therefore, diets with increased protein content and modest quantities of fat and carbohydrates may be advantageous in the prevention and management of impaired glucose tolerance or diabetes in the cat. The 65kDa isoform of glutamic acid decarboxylase (GAD65) and insulinoma antigen 2 (IA-2) are important pancreatic autoantigens in the pathogenesis of Type 1 diabetes mellitus in humans and in the non-obese diabetic mouse. Anti-GAD and anti-IA-2 antibodies are associated with and often precede the development of clinical signs of disease. The aim of this project was to clone canine GAD65 and IA-2 for future immunological studies in canine diabetic patients.

The entire canine GAD65 gene (1773bp) was amplified from canine insulinoma cDNA by polymerase chain reaction (PCR), using primers based on conserved regions of human and porcine GAD65. The gene encoding the C-terminus of the intra-cytoplasmic domain of canine IA-2 (IA-2/3Ј-648bp), the most antigenic region in humans, was amplified from canine insulinoma cDNA by PCR, using primers based on the human IA-2 sequence. Canine GAD65 and IA-2/3Ј were each ligated into the pCRII vector and recombinant E.coli clones for each gene were sequenced.

Canine GAD65 was found to have 92% nucleotide identity and 96% amino acid identity with the human gene. Canine IA-2/3Ј had 95% nucleotide identity and 99% amino acid identity with the human gene. It is anticipated that expression of these canine antigens will allow the development of a screening test to identify autoantibodies to GAD65 and IA-2 in canine diabetic patients. The objective of this study was to determine whether dogs with atherosclerosis are more likely to have concurrent hypothyroidism, diabetes mellitus, or hyperadrenocorticism than dogs that do not have atherosclerosis.

A retrospective mortality case-control study was performed. The study included 30 dogs with results of a complete necropsy and histopathologic evidence of atherosclerosis that were examined between January 1986 and April 2000, and 142 control dogs with results of a complete necropsy and no histopathologic evidence of atherosclerosis that were examined during the same time period. Proportionate changes in the prevalence of hypothyroidism, diabetes mellitus, and hyperadrenocorticism were calculated and presented using exact prevalence odds ratios (POR), 95% confidence intervals (95% CI), and p-values. Multivariate analysis, adjusted for age and year of necropsy was performed.

Eighteen of 30 dogs (60%) with atherosclerosis had hypothyroidism, six (20%) of dogs with atherosclerosis had diabetes mellitus, and three (10%) of dogs with atherosclerosis had hyperadrenocorticism. In the age-matched control group of dogs with no histopathologic evidence of atherosclerosis, five of 142 dogs (3.5%) had hypothyroidism, five dogs (3.5%) had hyperadrenocorticism, and one dog (1%) had diabetes mellitus. Dogs with atherosclerosis were over 58 times more likely to have concurrent hypothyroidism than dogs with no histopathologic evidence of atherosclerosis (POR ϭ 58.6, 95% CI ϭ 11.2-179.7, p Ͻ0.001), and over 44 times more likely to have concurrent diabetes mellitus than dogs with no histopathologic evidence of atherosclerosis (POR ϭ 44.8, 95% CI ϭ 4.5-765.2, p ϭ 0.002). Dogs with atherosclerosis were not found to be more likely to have concurrent hyperadrenocorticism than dogs that did not have atherosclerosis (POR ϭ 1.8, 95% CI ϭ 0.2-17.6, p ϭ 0.59).

We conclude that hypothyroidism and diabetes mellitus, but not hyperadrenocorticism, are more prevalent in dogs with atherosclerosis compared to dogs verified atherosclerosis-free on post-mortem examination. Radiofrequency heat ablation of small tumors has been efficacious in humans. An insulated needle is guided into target tissue; a radiofrequency pulse applied, and heat formation at the needle tip results in coagulation necrosis. The purpose of this study was to evaluate efficacy of this treatment(Rx) modality for cats with naturally occurring hyperthyroidism.

For inclusion in this study, each cat must have had: clinical signs of hyperthyroidism (polyphagia, weight loss, etc.), persistent increase in serum thyroxine concentration (sTT 4 ) and no other condition that would contraindicate Rx. Nine cats met these criteria, and sTT 4 ranged from 5.2-20.1 mcg/dl, (mean, 12.1 mcg/ dl; reference range 1.1-3.9 mcg/dl). A solitary thyroid mass was identified via pertechnetate scintigraphy and cervical ultrasound in 4 cats (UNI). Two nodules were found in each of the other 5 cats (BI). Each cat was anesthetized and positioned in dorsal recumbency. With ultrasound guidance, an insulated 20gauge catheter stylet was inserted into the thyroid mass and radiofrequency pulses applied for 30-60 seconds using an RF 2000 unit, Radiotherapeutics Redwood City, CA. In cats with bilateral disease, the larger mass was ablated. During ablation, treated tissue became hyperechoic secondary to nitrogen sublimation.

Heat ablation was performed once on each of 3 thyroid nodules and twice on 1 nodule in the 4 UNI cats. Three of the 4 cats were euthyroid after the procedure, and hyperthyroidism recurred in all 3 cats: 4, 7 & 19 months post-Rx, respectively. One of these cats had a second ablation performed and remained euthyroid for 9 months. One of the 4 cats had reduction in sTT 4 but did not become euthyroid. In the BI cats, heat ablation was performed once in 2 cats, twice in 2 cats, and three times in 1 cat. Each of these cats had heat ablation performed only on the larger nodule. Each of 2 cats became euthyroid after 1 heat ablation, for periods of 7 & 9 months, respectively. In each of the 2 cats that had 2 procedures performed, neither cat was euthyroid after the first procedure. One was euthyroid for 2 months after the second procedure, and 1 is euthyroid at the time this abstract was written (9 months post-Rx). The cat that had 3 procedures performed became euthyroid after the second and third procedure, for 5 months each time. Horner's syndrome developed in 2/9 cats, and resolved within 2 months in both cats.

Unipolar radiofrequency heat ablation is feasible and effective as a short-term treatment for feline hyperthyroidism, but is not effective as a permanent treatment. Thyroglobulin antibody (TgAA) has been reported to be a useful serum marker for canine immune mediated thyroiditis. This questionnaire survey study attempted to identify clinical associations with TgAA status in euthyroid dogs that could direct future research of canine thyroiditis and hypothyroidism.

Responses were received from 233 serum TgAA positive (TgAAϩve) and 203 clinic-matched TgAA negative (TgAA-ve) euthyroid dogs. Thyroid functional status was determined by a combination of serum thyroxine and thyrotropin measurements. There were no significant associations between serum TgAA and gender, vaccination frequency, veterinarian-observed lesions/clinical signs, owner-reported seizures, type of food (homemade, dry, canned) or the proportion of healthy animals being screened for thyroid disease (ϳ50%) versus those with clinical signs suggestive for hypothyroidism. However, TgAA ϩve dogs were younger (median age 3.5yrs) than TgAA-ve (5.1yrs; p Ͻ 0.001) and there were positive associations between serum TgAA and owner-dissatisfaction with their dog's weight (Odds ratio (OR) 1.53; 95%CI 1.01-2.32; p Ͻ0.05), skin/hair (OR 1.87; 1.24-2.81; pϽ0.002), personality/behavior (OR 3.38; 1.56-7.53; pϽ0.001) and within females, there were more intact animals in the TgAAϩve group (OR 1.92; 1.07-3.60; p Ͻ0.02). Within detailed vaccination history, there were positive associations between reported exposure to parvovirus (OR 5.40; ; p Ͻ 0.0001) and corona virus vaccines (OR 2.92; ; p Ͻ 0.0005). There was no association with exposure to rabies vaccination but there was with reported 3-year frequency of rabies vaccine administration (OR 4.66; p Ͻ 0.005) . No exposure or frequency associations were found for other reported vaccine components (leptospirosis, hepatitis, distemper, bordetella and parainfluenza). There was a negative association between serum TgAA and reported concurrent immune mediated disease (OR 0.29; 0.07-0.97; p Ͻ 0.05).

The results highlight possible causal contributors to canine thyroiditis that warrant further investigation to confirm, refute or better understand these associations. Tricyclic antidepressants have been shown to alter thyroid function in man and laboratory animals, but have not been evaluated in the dog. The effect of administration of clomipramine on canine thyroid function was studied in a prospective protocol in which 14 mature, healthy dogs were administered clomipramine (3 mg/kg PO q12h) for 112 days. Thyroid-stimulating hormone (TSH), total thyroxine (T4), total 3,5,3Ј triiodothyronine (T3), free thyroxine (fT4), and 3,3Ј,5Ј triiodothyronine (reverse T3; rT3) concentrations were measured on days 0, 7, 28, 42, 56, and 112. Thyrotropin-releasing hormone (TRH) response tests were performed concurrently by measuring serum TSH before and 30 min after IV administration of 0.2 mg TRH. Repeated measures analysis of variance was applied to test for effects of day of treatment. When a significant effect was noted, it was further investigated using orthogonal polynomial trends (p-value Ͻ 0.05 considered significant).

Significant decreases were found in the serum T4, fT4, and rT3 concentrations at days 28-112 of clomipramine treatment. Serum T4 concentration declined from a mean Ϯ SE pretreatment value of 26 Ϯ 1.2 nmol/L to 17 Ϯ 0.5 nmol/ L at day 112 (p Ͻ 0.001). Mean serum fT4 concentration decreased from 29 Ϯ 2.4 pmol/L to 19 Ϯ 1.3 pmol/L on day 112 (p Ͻ 0.0002). Mean serum rT3 concentration decreased from 1.2 Ϯ 0.1 nmol/L to 0.83 Ϯ 0.08 nmol/L on day 112 (p Ͻ 0.0001). The effect of time on serum T3 concentration was significant (p Ͻ 0.0001) as well, but the deviation in T3 from pre-treatment concentration was variable, and no consistent trend in T3 concentrations could be identified. No significant effect of time was noted in either basal or post-TRH TSH concentrations.

The results of this study indicate that significant and substantial decreases in T4, fT4 and rT3 can occur during clomipramine administration. Long-term administration of clomipramine may result in a misdiagnosis of hypothyroidism if a dog is tested while taking this medication and, since decreased serum fT4 occurs, hypothyroidism may result. The parathyroid gland is designed to respond to changes in extracellular ionized calcium (Ca 2ϩ ) concentrations by controlling parathyroid hormone (PTH) secretion. Abnormalities in Ca 2ϩ homeostasis are reported in horses with several pathological conditions; however, there is little information on the regulation of calcium in horses. The objectives of the present study were to determine the Ca 2ϩ set-point (the calcium concentration corresponding to 50% of maximal PTH secretion) in healthy horses, to determine whether the Ca 2ϩ /PTH response curves during hypocalcemia and hypercalcemia were characterized by hysteresis (different PTH concentrations for the same serum Ca 2ϩ concentration depending on the direction of changes in Ca 2ϩ concentrations), and to determine if the order of experimentally-induced hypocalcemia or hypercalcemia had an effect on PTH secretion (area under the time concentration curve, AUC). The Ca 2ϩ set-point and hysteresis were determined in 12 healthy horses in a cross-over design by infusing EDTA (30-90 mg/kg/h) and calcium gluconate (4-16 mg/kg/h). Four healthy horses were infused with 0.9% NaCl to serve as controls.

The Ca 2ϩ set-point was 1.37 Ϯ 0.05 mmol/L, which is higher than values reported for humans and dogs (1.0-1.2 mmol/L). In horses in which hypercalcemia was induced first the Ca 2ϩ set point was lower (1.30 Ϯ 0.05 mmol/L). We demonstrated that the phenomenon of hysteresis was present during experimentally-induced hypocalcemia and hypercalcemia. At the same serum Ca 2ϩ concentration, the serum PTH concentration was higher when the serum Ca 2ϩ concentration was falling (induction of hypocalcemia and recovery from hypercalcemia) than when the serum Ca 2ϩ was rising (induction of hypercalcemia and recovery from hypocalcemia). Horses in which hypocalcemia was followed by hypercalcemia secreted more (PϽ0.05) PTH (7440 Ϯ 740 pmol min/L) than horses in which hypercalcemia was followed by hypocalcemia (5990 Ϯ 570 pmol min/L).

In conclusion, this study has demonstrated that the Ca 2ϩ set-point in the horse is higher than in other domestic animals and man. We have shown that the Ca 2ϩ / PTH relationship in horses is sigmoidal and displays hysteresis during both hypocalcemia and hypercalcemia, and that extracellular Ca 2ϩ concentrations may affect the response of the parathyroid gland to hypocalcemia. Defining the Ca 2ϩ set-point in horses may help to explain the pathophysiology of hypocalcemia and hypercalcemia in this species. Parathyroid hormone (PTH) is secreted by the chief cells of the parathyroid gland in response to changes in extracellular ionized calcium (Ca 2ϩ ) concentrations. In vitro studies on parathyroid cells from different species have improved our understanding of the physiology and pathophysiology of the parathyroid gland. Several conditions in the horses are associated with abnormal parathyroid gland function. In a previous study we found that hypocalcemia and abnormal parathyroid gland function were present in some horses with sepsis. In the present study we evaluated equine parathyroid cell PTH secretion, and PTH mRNA and calcium-sensing receptor (CaR) mRNA expression. We also evaluated the effects of interleukin (IL)-1beta, IL-6, and tumor necrosis factor (TNF)-alpha, cytokines known to be increased during sepsis in humans and animals, on PTH secretion, and PTH and CaR mRNA expression.

There was an inverse sigmoidal relationship between PTH secretion and Ca 2ϩ . PTH secretion in a low Ca 2ϩ (0.8 mM) medium decreased from 100% (day 0) to 13% (day 30). The inhibitory effect of high Ca 2ϩ concentrations on PTH secretion also declined. By day 5 of culture, higher Ca 2ϩ concentrations were required to inhibit PTH secretion, and there was a rightward shift (increase) of the Ca 2ϩ set-point. After day 10 of culture, there was no significant difference in PTH secretion between low (0.8 mM) and high (2.0 mM) Ca 2ϩ concentrations. Parathyroid cells exposed to high Ca 2ϩ concentrations had lower PTH mRNA expression (PϽ0.05) than cells exposed to low Ca 2ϩ concentrations. PTH mRNA expression declined from 100% (day 0) to 43% (day 5), and to 25% (day 30). CaR mRNA also decreased from 100% (day 0) to 31% (day 5), and to 16% (day 30). Low and high Ca 2ϩ concentrations had no effect on CaR mRNA expression. IL-1beta (2000 pg/ml) decreased both PTH secretion (75%) and PTH mRNA expression (73%) (PϽ0.05). IL-1beta concentrations greater than 200 pg/ ml resulted in significant (PϽ0.05) overexpression of CaR mRNA (up to 142%). The effects of IL-1beta were blocked by an IL-1 receptor antagonist. IL-6 (5-10 ng/ml) decreased PTH secretion, but had no effect on PTH and CaR mRNA expression. TNF-alpha (1-10 ng/ml) had no effect on PTH secretion, and PTH and CaR mRNA expression.

We conclude that the decreased responsiveness of parathyroid cells to Ca 2ϩ from 0 to 30 days could be explained, in part, by the reduced CaR expression. We also believe that IL-1, IL-6, but not TNF-alpha may be important in regulating PTH secretion in the horse. Obesity is the most common nutrition-related problem in pet dogs, yet it often goes unrecognized. The objective of this study was to determine how accurately dog owners could determine their own pet's body condition, and to evaluate means of enhancing that estimation.

Owners of 201 healthy adult dogs were recruited to participate in this study. Owners were asked to complete a questionnaire that included approximately 30 questions about their dog (breed, age, weight, fitness and exercise levels, and eating habits) and to determine their dog's body condition score (BCS) using a 9-point BCS system (Laflamme 1997). Independently, a pet professional trained in BCS evaluation also evaluated each dog's BCS (ExpertBCS) and weighed each dog. Eighty percent of the data were used to build regression models for predicting BCS, while 20% was reserved to validate the selected model(s).

Body weight of dogs evaluated ranged from 2.0 to 69.9 kg, and ExpertBCS ranged from 3 to 9. The mean (Ϯ s.d.) of ExpertBCS was 6.3 Ϯ 1.0, while the owner BCS score averaged 5.3 Ϯ 0.8. The experts considered 79% of dogs to be above ideal body weight, while only 28% of owners estimated their dog to have a BCS above 5. Of the various questions, those related to rate of food consumption (Q1); frequency of food consumption (Q2) and frequency of treats being given (Q3)-all rated on a 5-point scale-best correlated with ExpertBCS. Logistic regression resulted in an equation that predicts if a dog is obese (BCS ϭ or Ͼ7): Logit(p) ϭ Ϫ10.286 ϩ 1.77*ownerBCS ϩ 0.64*Q1 Ϫ 0.74*Q2 ϩ 0.34*Q3. Testing of this model with the validation data set resulted in a sensitivity (ability to predict an overweight dog as overweight) of 78% and a specificity (ability to predict a not-overweight dog as not overweight) of 71%, compared to a sensitivity of 17% by owners using the BCS chart alone.

In conclusion, despite access to an illustrated BCS chart, most pet owners do not recognize their overweight dogs as overweight without professional guidance. It is possible to correct for owner under-prediction of BCS by statistical modeling using input from the owner. We sought to characterize clinician proficiency at detecting the presence of dehydration in intensive care unit (ICU) patients using history and physical examination. We used the percent change in ICU patient body weight after the first 24-48 hours of treatment as our gold standard for detecting dehydration. Our objective was to determine whether clinician estimation of hydration status at the time of admission into the ICU corresponded to actual hydration status based on weight change subsequent to administration of intravenous fluids and oral water. We also sought to determine whether baseline PCV or TS, or the percent change in these measures following therapy, were related to clinician estimates of hydration or changes in weight after treatment.

Dogs and cats with various medical disorders that had not had surgery within 48 hours of admission into the ICU were consecutively admitted into the study. Animals with gastrointestinal fluid loss, edema or diseases predisposing to edema, oliguria, diuretic therapy, and body fluid drainage or hemorrhage during the observation period were excluded. Physiologic data were collected at the time of admission and 24-48 hours later. Each admitting clinician used clinical judgement to estimate the hydration status of his or her patient. Measurements included baseline packed cell volume (PCV) and total plasma solids (TS), amount and type of fluids or blood products administered, access to water, percent change in PCV and TS, and percent change in body weight. A change in body weight Ͼ Ϯ 5% was considered clinically significant.

Of 143 dogs, 22 gained, 7 lost, and 114 had no change in body weight. Of 40 cats, 4 gained, 6 lost, and 30 had no change in weight. Neither clinician estimates of dehydration nor baseline PCV/TS corresponded to changes in body weight following treatment, and there was no relationship between changes in PCV/TS and weight change or between change in PCV and volume of fluid administered. There was a slight correlation between volume of fluid administered and change in TS in dogs only (Pearson correlation coefficient ϭ Ϫ.22, p ϭ 0.047).

We concluded that a clinical diagnosis of dehydration based on history and physical exam findings, at least as practiced in our hospital, does not predict weight gain in critically ill patients following treatment, and measurement of baseline PCV/TS does not improve upon this. Furthermore, changes in PCV/TS did not correlate with changes in body weight, and there was no relationship between total fluids administered and changes in PCV. Future studies are needed to determine whether refinement and better standardization of physical examination skills can improve our ability to diagnose dehydration. The role of leukotrienes (LTs) in the pathogenesis of feline asthma is unknown, and the use of drugs to inhibit LTs such as zafirlukast (Accolate) in cats has not been scientifically studied. The purpose of this study was to evaluate changes in LTs in urine and bronchoalveolar lavage fluid (BALF) prior to and after inducing asthma in 11 cats. Healthy cats with negative skin tests to bermuda grass or house dust mite allergen and normal BALF (mean eosinophil % ϭ 5.4) were sensitized. At baseline and post-sensitization, urine (n ϭ 11) and BALF (n ϭ 5) were added to a LT inhibitor, nordihydroguaiaretic acid. Pulmonary function testing with methacholine was used to determine the ED200RL (200% increase in the airways resistance compared with post-saline challenge value). A competitive ELISA kit with monoclonal antisera recognizing LTC4, D4 and E4 was used per manufacturer's instructions. The concentration of LTs in samples (ng/ ml) were calculated by using a standard curve. Urinary LT concentrations were referenced to urinary creatinine concentrations (mg/dl) measured using a modified kinetic Jaffe method.

Results showed that post-sensitization, all cats had positive skin tests, an increase in mean BALF eosinophil % to 39.5, and a 7 fold decrease in the ED200RL compared with baseline values. No significant differences were found in urine LT:creatinine ratios post-sensitization compared with baseline, and BALF LTs were undetectable. Based on this experimental model, the role of LTs in the pathogenesis of feline asthma is questionable as an increase in BALF eosinophil % and a decrease in the ED200RL could be documented without an increase of LTs in urine or BALF. Further studies evaluating changes in LTs at multiple time points after allergen challenge are warranted, as a transient increases in LTs may occur but were not detected in this study. Inhalant therapeutics are commonly used in the management of human respiratory disorders, most notably asthma. The use of inhaled medications has been recommended in cats with respiratory disease however it has not been demonstrated whether cats will tolerate this method of drug delivery. The purpose of this paper was to determine whether awake, untrained cats would inhale a nebulized radiopharmaceutical from a facemask and if the material would reach the lower airways.

Twenty adult, privately owned cats were included in the study. The cats were determined to be free of major disease on the basis of history, physical exam, CBC, biochemical profile and thoracic radiographs. Technetium-diaminetriaminopentaacetic acid ( 99m Tc-DTPA) was nebulized and administered to the cats using a closely-fitting facemask. The facemasks were held on the cats for approximately 10 breaths. Images were then captured using a gamma camera within a few minutes of administration of the compound and analyzed for deposition of 99m Tc-DTPA. The cats were not sedated or anesthetized for any of these procedures.

All 20 cats had evidence of the radiopharmaceutical reaching the lower airways. Uniform distribution of the technetium was noted in all cases.

Based on our findings, inhalation is a potential route for administering medications to cats. Specific therapeutic agents will need to be examined individually. Cats will tolerate having a facemask applied to them. The aims of this study were to quantitate and compare the effects of rapid infusion of either normal (0.9%) saline or lactated Ringer's solution (LRS) on the acid-base status of anesthetized, hypovolemic dogs, and to determine whether the use of either solution creates a clinically relevant metabolic acidosis (arterial pH Ͻ 7.2).

Twelve anesthetized dogs that were about to undergo terminal surgical procedures at Mississippi State University were used. Dogs were pre-screened with a complete blood count, serum chemistry, and physical exam. After induction of anesthesia, a metatarsal arterial catheter, cephalic venous catheter, and jugular catheter were placed. Dogs were bled 20% of their blood volume over 60 minutes (18 ml/kg total; 6 ml/kg at Times 0, 30, and 60) from their jugular catheter in order to mimic an ongoing moderate surgical bleed. The dogs were volume resuscitated with either LRS or 0.9% saline (6 dogs in each group) at a volume of 54 ml/kg over 90 minutes starting at Time 0 via their cephalic catheter. The crystalloid dose was calculated to be three times the volume of blood lost. The following blood parameters were measured at Times 0, 30, 60, and 90: sodium, chloride, potassium, total carbon dioxide, anion gap, lactate, packed cell volume (PCV) and total protein (jugular venous samples); and PaO 2 , PaCO 2 , bicarbonate, base excess, and pH (arterial samples). Data was analyzed using a two-way repeated-measures analysis of variance to compare intergroup and intragroup differences at Times 0, 30, 60, and 90 of crystalloid infusion. A 0.05 level of significance was used.

While there were no significant pH differences between or within the two groups, the arterial pH was significantly below the normal reference range in both groups at all times: the 0.9% saline group had a pH of 7.30 Ϯ 0.03 (mean Ϯ SD) at Time 0 and a pH of 7.30 Ϯ 0.05 at Time 90, and the LRS group had a pH of 7.29 Ϯ 0.03 at Time 0 and a pH of 7.32 Ϯ 0.01 at Time 90. With the exception of total protein and PCV, all other blood parameters remained within normal reference range. The 0.9% saline group had a drop in total protein from 6.3 Ϯ 0.2 g/dL at Time 0 to 4.5 Ϯ 0.2 g/dL at Time 90, and the LRS group had a similar drop from 6.1 Ϯ 0.2 g/dL to 4.2 Ϯ 0.37 g/dL. The 0.9% saline group had a drop in PCV from 40.9 Ϯ 6.2% at Time 0 to 26.0 Ϯ 8.4% at Time 90, and the LRS group had a similar drop from 39.6 Ϯ 6.5% to 28.9 Ϯ 5.1%.

In conclusion, healthy anesthetized dogs with moderate amounts of blood loss can be volume replaced with either LRS or 0.9% saline without significantly altering acid-base status or electrolyte levels. Ease of blood collection from donor cats contributes to the success or failure of blood donor programs. Blood collections are perceived as difficult due to loss of vein quality with repeated venipunctures, cancelled collection secondary to handling difficulties, and increased personnel needs for handling and monitoring. The authors hypothesized that blood donation via subcutaneous totally implantable vascular access ports [VAPs] allows circumvention of these problems and facilitates blood collection.

Eight healthy adult male cats were divided into two groups for blood collection. Group A (n ϭ 4): conventional jugular phlebotomy under anesthesia; group B (n ϭ 4): subcutaneous totally implantable VAPs using manual restraint. Blood donors were evaluated for a 6-month period during monthly blood donations. The parameters: anesthesia and surgery duration for port implantation, port acceptance by the donor (post-surgical pain, discomfort during port manipulations, self mutilation), and post-surgical complications were evaluated for group B. Time for blood collection, number of personnel, changes in RBC morphology, and donor behavior were compared between groups. Ease of phlebotomy (group A) and port associated complications during and after blood collection (group B) were evaluated. Units of blood were cultured for anaerobic and aerobic bacteria following collection.

Mean anesthesia duration was 110 min, (range 86-150 min) and mean surgery duration was 71 min, (range 54-102 min), respectively. Post-surgical discomfort to the port was only noted in one cat with port infection. Post-surgical VAPrelated complications occurred in one other cat and included disconnection of the port from the catheter and seroma formation. Blood was successfully collected 24/24 and 20/24 times, for groups A and B respectively. Mean total blood collection times were significantly (pϽ0.02) longer for group A (6.8 min.), when compared to group B (4.2 min.). Group A cats had larger variability in collection time (max 14.3 min.) when compared to Group B cats (max 6.1 min.) and required three and two persons, respectively. Group A cats had inferior behavioral scores (less tolerant to handling) during initial venipuncture when compared to port manipulation in group B cats (pϽ0.01). Bacterial cultures of blood were negative 22/24 and 16/20 times, for groups A and B respectively. No differences were seen in RBC morphology between groups.

Positive results for collections via VAPs were increased donor acceptance, decreased number of personnel and reduced collection time. Potential drawbacks included contamination of blood products and port-related complications. Whilst a multitude of laboratory techniques have been developed to quantify various parameters of the inflammatory response in man, the applicability of such techniques to the dog is less well established. Leucocytes in blood that participate in the inflammatory response can be divided into two sub-populations, namely neutrophil polymorphs (PMNs) and peripheral blood mononuclear cells (PBMCs). Blood samples were obtained from 38 healthy dogs into sodium heparin tubes and plain tubes with no anticoagulant. The PMN and PBMC fractions were isolated from the sodium heparin samples obtained. Reactive oxidant production in both fractions was measured by chemiluminescence, whilst LTB4 production was quantified by HPLC following incubation with A23187. The intracellular and extra-cellular concentrations of three acid hydrolase enzymes were measured using spectroflourometry, whilst the percentage conversion of C3 to C3b in serum was assayed by immunoelectrophoresis. Mean values for reactive oxidant production were similar for both cell populations (17,853 Ϯ 9,695 vs 19,138 Ϯ 14,569 for the PMBC and PMN fractions, respectively). However, the PMN fraction exhibited a biphasic response incorporating a slower increase to maximal reactive oxidant release (10.7 vs 5.1 min for PMNs and PMBCs, respectively). The intracellular concentrations of beta-glucuronidase, beta-galactosidase and N-acetyl-beta-glucosaminidase were higher in the PBMC fraction. Both cell fractions released less than 20% of their intracellular concentrations following stimulation with opsinized zymosan. LTB4 production in the PMBC fraction was higher than that for the PMN fraction (4.45 and 0.96 ng/106 cells, respectively). A mean value of 57.3% of complement C3 was found to have converted to C3b. The results generated provide a reference range of values for the above parameters in healthy dogs, which can subsequently be used for comparative purposes with canine patients suffering from a variety of inflammatory disorders. Sulfonamide hypersensitivity in humans is characterized by fever, rash, blood dyscrasias, or hepatotoxicity, observed 5 or more days after sulfonamide initiation. A similar syndrome is observed in dogs, and may also include arthropathy, uveitis, or keratoconjunctivitis sicca. In humans, anti-drug antibodies and antibodies to liver microsomal proteins have been documented in patients with sulfonamide hypersensitivity, suggesting haptenization of liver proteins by a reactive drug metabolite generated in the liver. The purpose of this study was to summarize the clinical findings in dogs with sulfonamide-associated hypersensitivity identified at the authors' institutions from 1993-2001, and to determine whether IgG or IgM anti-drug antibodies, or antibodies to liver microsomal proteins, were detectable in the serum of these patients.

Forty-five dogs with suspected hypersensitivity to a sulfonamide antimicrobial were evaluated. Nineteen dogs had been exposed to trimethoprim-sulfamethoxazole (SMX, human generic), 11 dogs to ormetoprim-sulfadimethoxime (Primor), and 3 dogs to trimethoprim-sulfadiazine; sulfonamide type was not available for 12 dogs. Clinical signs included thrombocytopenia (17/45; 38%), hepatotoxicity (12/45; 27%), KCS (8/45; 18%), hemolytic anemia (6/45; 13%), neutropenia (5/ 45; 11%), arthropathy (5/45; 11%), skin eruptions (2/45; 2%), and uveitis (2/45; 2%). The mean dose of potentiated sulfonamide was 48.4 Ϯ 15.0 mg/kg/day (range, 29.3-81.4) , and the mean time from drug initiation to the onset of adverse signs was 11.7 Ϯ 4.5 days (range, 5-25) .

Twenty dogs, all sampled within one month of the adverse reaction and with confirmed type of sulfonamide exposure, were selected for initial screening for anti-drug antibodies. Serum was tested by ELISA for IgG or IgM antibodies to either SMX, sulfadimethoxime, or sulfadiazine, as appropriate. Rabbit anti-SMX serum was used as a positive control. Negative controls included preimmune rabbit serum, and serum from 10 healthy dogs, 3 sulfonamide-tolerant dogs, and 4 systemically ill dogs without sulfonamide exposure. ELISA results indicated that no dog was positive for either IgG or IgM antibodies to specific sulfonamides. This indicates that sulfonamide hypersensitivity in dogs is not associated with a humoral response to the drug itself, and may instead be due to nonhumoral mechanisms (cell-mediated or direct cytotoxicity), or a humoral response to drug-modified proteins. Western blotting studies are underway to determine whether dogs with sulfonamide-associated hypersensitivity have antibodies to native or metabolite-modified tissue proteins. Recently, Giardia vaccines for use with dogs and cats have been shown to lessen clinical signs of giardiasis and greatly lessen cyst shedding on challenge when used as a preventative. Treatment of giardiasis with currently available drugs may be ineffective and some drugs have significant side-effects. Use of Giardia vaccination as an immunotherapy led to cessation of diarrhea and cyst shedding in seven naturally-infected dogs. This study was undertaken in order to assess the efficacy of Giardia vaccination as a treatment of giardiasis in experimentally-infected cats.

Sixteen young adult, FeLV and FIV naïve, mixed sex cats were used. The cats were shown to be microscopically negative for Giardia cysts in feces after zinc sulfate centrifugation and a commercially available immunofluorescent antibody assay (IFA; Meridian Diagnostics, Cincinnati, OH) three times prior to infection. The cats were housed individually in SPF facilities to avoid crossinfection. On day 0, an equal number of Giardia cysts of a strain initially isolated from a human were administered to all cats via an oral-gastric tube under brief sedation. On weeks four, six, and 10, eight randomly selected cats were administered Giardia vaccine (Fel-O-Vax Giardia, Fort Dodge Animal Health, Overland Park, KS) subcutaneously. For each of 28 weeks post-infection (PI), three fecal samples per cat were examined for Giardia cysts. By use of IFA, cyst numbers were counted and a score from 0 to 4 assigned. Results from the vaccinated and unvaccinated cats were compared by logistic regression.

All cats were shedding cysts by the end of week 2. Diarrhea was rarely detected and when occurred was mild and transient. By week 28, five of eight vaccinated cats and seven of eight control cats maintained patent Giardia infections. Magnitude of infection based on the number of positive samples and number of cysts per sample decreased progressively in both groups over time. There was no statistically significant difference between groups over time (p value ϭ 0.3173).

Administration of three Giardia vaccinations was ineffective for elimination of this strain of Giardia from cats. Since clinical signs were minimal in both groups of cats, it could not be determined whether vaccination lessened clinical disease. Whether Giardia vaccination is an effective treatment for giardiasis in naturally-infected cats remains to be determined. Information concerning treatment of Giardia spp. infections of cats is for the most part lacking. Fenbendazole has been used successfully for the treatment of giardiasis in dogs, but has not been assessed for this purpose in cats. The objective of this study was to determine whether administration of fenbendazole was effective for eliminating Giardia from eight chronically infected cats with concurrent Cryptosporidium parvum infection.

Eight, one to two year old, mixed sex, clinically normal cats with concurrent Giardia spp. (three months duration) and C. parvum infections (Ͼ six months duration) were studied. Fenbendazole was administered at 50 mg/kg, PO, q24 hr for 5 days. Feces from each cat were collected and processed (as available) each day for 23 days after treatment. By use of a commercially available immunofluorescent assay for detection of Giardia lamblia cysts and C. parvum oocysts (Meridian Laboratories, Cinncinati, OH), cyst numbers were counted and a score from 0 to 4 assigned. Fecal results from treated cats were compared by logistic regression to those of eight untreated, Giardia spp. infected, C. parvum naïve cats that were collected over a similar time period after Giardia infection but not treated.

All 16 cats were positive for Giardia cysts in feces on all samples assayed the two weeks prior to use in this study and all fecal samples from untreated cats were positive throughout the study period. After treatment with fenbendazole, four of eight cats were Giardia negative for the remainder of the study. These four cats were persistently positive for C. parvum oocysts prior to fenbendazole treatment and persistently negative after. One treated cat was positive for Giardia cysts on all samples and three treated cats had one to three negative fecal samples and then resumed shedding large numbers of cysts (two, six, and nine days after treatment); each of these cats was persistently positive for C. parvum oocysts in feces. When compared to untreated cats, the fenbendazole treated group shed less Giardia cysts during the first week but not the second week post-treatment.

Administration of fenbendazole decreased Giardia cyst shedding to less than detectable numbers in 50% of the cats for the duration of the study period. However, since the cats were only followed for 23 days and were not immunosuppressed, it is unknown whether infection was eliminated. Further study will be required to determine whether persistent C. parvum infection was associated with treatment failure in the four cats that continued detectable Giardia cyst shedding after fenbendazole treatment or whether fenbendazole lessened C. parvum oocyst shedding in the four cats that became persistently negative. Mycoplasma haemofelis and Mycoplasma haemominutum are the causes of feline haemobartonellosis and commonly induce hemolytic anemia. Clinical and laboratory abnormalities usually resolve after administration of tetracyclines or enrofloxacin. However, based on PCR results, treatments utilized to date do not always clear the infections and so recurrent disease is possible.

Five feline leukemia virus and feline immunodeficiency virus seronegative cats were treated by their veterinarians for recurrent haemobartonellosis. Of the cats, two were infected by M. haemofelis, two were infected by M. haemominutum, and one was diagnosed by use of cytology only. Anemia was recurrent after treatment with glucocorticoids and doxycyline (2 cats); glucocorticoids, doxycycline, and enrofloxacin (1 cat); glucocorticoids, doxycycline, tetracycline, and enrofloxacin (1 cat); and glucocorticoids, doxycycline, enrofloxacin, and chloramphenicol (1 cat). Imidocarb diproprionate was administered at 5 mg/kg, IM as the sole agent or added to a protocol that was currently failing.

For one cat, clinical abnormalities resolved after one dose of imidocarb and the cat remained normal for a 14 month followup period. For one cat, clinical abnormalities resolved after two doses of imidocarb (q14 days) and the cat remained normal for a 9 month followup period. For two cats, clinical and laboratory abnormalities resolved after 4 injections of imidocarb (q14 days) and the cats remained normal for followup periods of 10 weeks or 7 months. One cat requires imidocarb injections monthly to maintain a normal PCV (13 injections to date). Recheck PCR test results were available for 2 cats; one M. haemofelis infected cat was PCR negative 10 weeks after the 4 th imidocarb injection and one M. haemominutum infected cat was PCR positive 3 weeks after the 4 th imidocarb injection.

These results suggest that imidocarb diproprionate is useful for the treatment of resistant haemobartonellosis in cats. Leptospirosis has increased in dogs in the United States and parts of Canada (Ontario, Québec) in the last few years. There was a marked increase in the number of cases in Ontario in the fall of 2000. This study presents the epidemiological features (clinical, clinicopathological, histopathological) of thirty-one dogs diagnosed with leptospirosis at the Ontario Veterinary College (OVC).

Clinical, clinicopathological, and histopathologic data was obtained for 31 dogs and diagnosed with leptospirosis at OVC between 1998 and 2000. Criteria for diagnosis of leptospirosis were a microscopic agglutination test (MAT) titre of Ն 320 to one or more serovars autumnalis, bratislava, canicola, grippotyphosa, icterohaemorrhagiae, pomona; seroconversion; or a positive leptospira polymerase chain reaction (PCR) on urine.

The dogs consisted of 17 males and 14 females of which the most common breed was ''mixed breed'' (6) followed by Shih Tzu (3), Labrador or Golden Retriever (2), Dalmation (2), German Shepherd dog (2), Siberian Husky (2) and Yorkshire Terrier (2). Dogs presented with nonspecific signs of lethargy (28), inappetence (25), dehydration (16), weight loss (9), vomiting (25), abdominal or lumbar pain (20), polyuria/polydipsia (13), tachypnea (11), diarrhea (11), stiff gait (11), icterus (9), lymphadenopathy (6), pyrexia (4), renomegaly (3), petechiation (1), and ascites (1). Ninety percent of dogs demonstrated biochemical evidence of injury to several organs, notably combination of kidney (23), muscle (22), pancreas (16), and liver (9). Observations included elevated urea or creatinine (29), elevated bilirubin (21), elevated serum alkaline phosphatase (18), electrolyte and mineral imbalances (27); acid-base disturbances (9); and altered total protein, albumin or globulins (22). The significant microscopic lesions were restricted to the kidney and liver. A striking feature of the histologic changes was the contrast between the subtlety of histologic lesions and the severity of the clinical signs.

This study demonstrated the resurgence of canine leptospirosis as an important disease of dogs in the last few years. Clinical signs are nonspecific and specific diagnosis requires MAT or PCR. However, further work is required to isolate and definitively identify the serovars involved and the common sources of infections for dogs. Idiopathic lower urinary tract disease (I-LUTD) is the most common cause of frequent and painful urination, hematuria, and inappropriate urination in cats. Feline caliciviruses (FCVs) have been implicated in the etiopathogenesis of I-LUTD. However, earlier studies investigating the causative role of FCVs in I-LUTD were hindered by lack of a sensitive means of detecting FCV urinary tract infections. The purpose of this study was to create a sensitive and specific FCV reverse-transcriptase polymerase chain reaction (RT-PCR) assay for rapid detection of FCV infections.

Multiple FCV cDNA sequences corresponding to the helicase gene were obtained from GenBank. Oligonucleotide primers delineating a conserved 123 bp region of the helicase gene were synthesized at a commercial DNA synthesis laboratory. The reverse primer was degenerate in 5 nucleotide positions. Total RNA was extracted from FCV-infected tissue-culture cell preparations. Viral RNA was then reverse transcribed into cDNA and amplified using a one-step RT-PCR and FCV helicase gene-specific primers. The 123 bp RT-PCR products were identified by agarose and acrylamide gel electrophoresis. The RT-PCR assay was optimized for primer concentrations and cycling conditions. Assay specificity was evaluated by direct sequencing of RT-PCR products and by amplification of viral RNA from ten wild FCV strains. Assay sensitivity was determined by extracting and amplifying viral RNA from serial 10-fold dilutions of FCV strain 2351774 (starting titer 9.2 ϫ 10 8 TCID 50 ).

Optimal RT-PCR amplification occurred when the concentration of the reverse degenerate primer was four times that of the forward primer. Specific FCV amplification product bands were observed for all 10 wild FCV strains. The nucleotide sequence of the RT-PCR product was 95% identical with published FCV sequences. Amplification products from the 9.2 ϫ 10 Ϫ1 TCID50 per ml dilution of FCV were detected visually.

In conclusion, our one-step RT-PCR assay appears to be a highly sensitive and specific means of detecting FCV. Tick-borne diseases appear to be more common in tropical regions of the world. In our hospital, 97% of dog ticks are Rhipicephalus sanguineus, whereas the remaining 3% are Amblyomma spp. Thus, we attempted to determine serologically the epidemiology of two diseases transmited by R. sanguineus to dogs, canine babesiosis and ehrlichiosis. Three-hundred and eighty-three dogs were randomly selected to represent the dog population presented to the Londrina State University Veterinary Teaching Hospital in south Brazil. Indirect immunofluorescence was used to confirm previous exposure to Babesia canis, whereas a commercially available dot-elisa test was used to detect antibodies against Ehrlichia canis.

Babesiosis had a prevalence rate of 35.8% (137/383), whereas 23% (88/383) of the dogs were positive to E. canis. Fifty-four dogs (14.1%) seroreacted to both agents. Dogs older than 1 year (OR ϭ 3.39), that lived in suburban areas (OR ϭ 2.76) or were presented with bleeding (OR ϭ 19,29) were more likely to have been exposed to B. canis. Dogs older than 1 year (OR ϭ 5.85), previously exposed to ticks (OR ϭ 5.84), that lived in suburban areas (OR ϭ 2.76) with other dogs (OR ϭ 1.99), or had neurological signs (OR ϭ 2.94) were more likely to have been exposed to E. canis. Dogs seropositive to B. canis were also more likely to be seropositive to E. canis (OR ϭ 4.33). Bleeding (OR ϭ 19.29) and lameness (OR ϭ 5.18) were more prevalent in dogs that seroreacted to both diseases than in the general hospital population. Owners of dogs with ticks were more likely to have been exposed to ticks themselves (OR ϭ 3.18). Only 10% of the owners that interacted occasionally with their dogs have had ticks, whereas more than 25% of the owners that interacted frequently or very frequently had ticks in the past. Canine babesiosis and ehrlichiosis appear to be highly prevalent at our hospital population. Ehrlichiosis is a chronic rickettsial infection of dogs. Antibody titers reflect the prevalence of exposure which varies by country and region. The goals of this study were 1) to identify the prevalence of exposure in stray adult dogs in East Tennessee 2) to compare the results of three commercially available serologic tests for identifying E. canis antibodies.

Blood from 90 adult stray dogs admitted to a Humane Society shelter in East Tennessee had complete blood counts. Serum antibodies to E. canis were measured by IFA and 2 different ELISA kits. IFA testing for E. canis was done using antigen slides and FITC anti-dog IgG (VMRD, Inc., Pullman, WA). Positive samples were diluted to their endpoint. The original ELISA test kit identified Dirofilaria immitis antigen and E. canis antibodies. The updated ELISA kit detected D. immitis antigen, E. canis antibody, and Borrelia burgdorferi antibody (IDEXX, Westbrook, Maine).

Prevalence of heartworm antigen was 12% with the original kit and 13% (95% CI 6, 21) with the updated kit. Seropositivity on IFA for E. canis antibody was defined as Ͼ1:20. Thirteen of 90 dogs (14%) were positive on IFA (95% CI 7,22). Only one dog was positive on the original ELISA kit. This dog was not positive by IFA or the updated test kit. One dog was positive on the updated ELISA kit. This dog had an IFA titer of 1:10, 240. Only one dog was positive for both heartworm antigen and E. canis antibody on IFA. There were no relationships between sex, breed, body weight, WBC count, platelet count, PCV, or heartworm status and E. canis antibodies identified by IFA.

We conclude that commercially available ELISA test kits have lower sensitivity for E. canis antibody than IFA. The prevalence of E. canis exposure in East Tennessee as determined by IFA is approximately 14% in stray adult dogs, which is similar to the prevalence of heartworm antigen in the same population. Purpose: To develop a sensitive and specific PCR assay for detection of T. foetus in feline fecal samples that are unsuitable for protozoal culture.

Methods: Total DNA was extracted from 200 ul aliquots of serially diluted cultured T. foetus (10,000-0.01 organisms/200 ul) and from 180 mg samples of feline feces spiked with 20 ul aliquots of similarly diluted organisms using a commercial kit modified for optimal elimination of PCR inhibitors (QIAamp DNA Stool Mini Kit). Two primer pairs amplifying T. foetus ITS DNA were evaluated alone and in combination; TFR3/TFR4 as described by Felleisen (J Clin Micro 1998;36) and TFITS-F (5Ј-CTGCCGTTGGATCA GTTTCG-3Ј)/ TFITS-R (5Ј-GCAATGTGCATTCAAAGATCG-3Ј) designed by the authors. Reaction conditions were optimized for annealing temperature (65-74ЊC), MgCl 2 (1-5 mM), TFR3/TFR4 (1.25-5 pmol) and template DNA (1-10 ul) concentration. Successful amplification of T. foetus genomic DNA after its addition to negative samples was used to rule out PCR inhibitors. Controls (Ϯ) were included in all reaction processing steps and DNA extraction. Sequencing was done by a commercial laboratory.

Results and Conclusions: The absolute detection limit of each primer pair alone was 1 organism/200 ul. The detection limit of T. foetus in feces was 10 organisms/200 mg using single-tube nested PCR (Table 1) . Optimal conditions for nested PCR were 1.25U AmpliTaq Gold (Perkin-Elmer), 1.25 pmol of each primer TFR3/TFR4, 25 pmol of each primer TFITS-F/TFITS-R, 200 M of each dNTP, 5 mM MgCl 2 , 1X PCR Buffer II (Perkin-Elmer), 10 g BSA and 5 ul of DNA template/100l reaction. Cycling conditions were; 95ЊC 5-min once, 95ЊC 30-sec/72ЊC 1-min for 30 cycles, 57ЊC 30-sec/72ЊC 30-sec for 50 cycles, and 72ЊC 5-min once. Sequence analyses confirmed amplification of T. foetus ITS DNA from cultured organisms, organisms added to feces, and feces obtained from cats with T. foetus diarrhea. Primers did not amplify feline DNA, feline Giardia lamblia (ATCC50163), Cryptosporidium parvum (Pleasant Hill Farms) or Pentatrichomonas hominis (ATCC30098) DNA.

FOR FELINE SPUMAVIRUS VECTORS. A.C. German, C. Helps, D. Harbour. University of Bristol, Langford, UK.

Traditional vaccination strategies, stimulating humoral immunity, are not protective against feline infectious peritonitis (FIP) and may even enhance disease severity. A protective cell-mediated response could be generated by using a virus vector from a different family to express selected coronavirus proteins. We have been developing replication-incompetent feline spumavirus (FeSV) transducing vectors to use as potential vaccines against FIP. Such vectors require a helper cell line to supply the viral transactivator (Tas), which enables production of infectious virus particles. Therefore, the aim of the current study was to develop a helper cell line for these recombinant retroviral vectors.

A cationic lipid reagent (Lipofectamine, Life Technologies) was used to transfect Crandell feline kidney (CRFK) cells with pHook3 plasmids (Invitrogen) containing either the FeSV Orf 1 gene (transactivator) or beta-gal gene (reporter plasmid). Transfection was optimised by standard methods and cells selected by magnetic beads or zeocin resistance. Cells underwent single cell cloning in 96well plates and successful clones were collected for analysis. Assessment of cell function included transfection with the beta-gal gene (under control of the FeSV long terminal repeat [pLTR/FeFV/beta-gal]), PCR, immunofluorescence (IF) and western blotting (WB) using FeSV reactive cat sera and recombinant Orf 1 antiserum.

The beta-gal assay demonstrated that, when high purity DNA was used, transfection efficiencies of at least 50% were achieved. CRFK cells showed poor recovery on magnetic selection due to poor expression from the pHook3 plasmid RSV promoter, confirmed by immunofluorescence. Selection by zeocin was more effective. Of thirty-two single cell clones, only six were functional, shown by transfection with pLTR/FeFV/beta-gal. IF, WB and PCR studies confirmed stable transactivator transfection, and differentiated levels of Tas expression between clones.

We have developed a helper cell line using CRFK cells for the expression of FeSV transactivator, to allow the production of replication-incompetent FeSV transducing vectors. Assessment of generation of infectious virions using a recombinant spumavirus vector containing GFP is currently underway. Feline heartworm (HW) infection is a growing concern for veterinarians in HW endemic areas. The purpose of this study was to compare the performance of several currently available commercial laboratory and in-house serologic tests against the current gold standard of necropsy, in naturally HW infected cats in a canine heartworm endemic area.

This prospective study examined 380 cats (36% male, 64% female) euthanized at an animal shelter in Alachua County Florida, between June 1998 and August 2001. Blood was collected from the vena cava within 90 minutes of death. Thoracic cavity, heart, and lungs were dissected and examined for the presence of adult HW. Serum was assayed for HW antibody (Ab) and antigen (Ag).

Overall, Ag tests detected 79.3-86.2% of HW and were highly specific. Most of the cats with false negative Ag tests had only a single male worm on necropsy. Ab tests detected 62.1-72.4% of HW infections, and had a wider range of false positives (1.4-19.1%) than Ag tests (0.3-2.0%) in the cats studied.

Results of the Ab tests studied found the Antech Laboratories Feline Ab test to demonstrate a 72.4% specificity and an 80.9% sensitivity. The Synbiotics Witness Ab test resulted in a 62.1% sensitivity and a 98.6% specificity. The Ag assays used resulted in the IDEXX Snap Feline Ag test giving a 79.3% sensitivity and a 98.0% specificity. The SA Scientific CHAT Ag test had a 79.3% sensitivity and a 99.7% specificity. Lastly, the Synbiotics DiroCHEK Ag test demonstrated an 86.2% sensitivity and a 99.1% specificity.

Feline HW infection in north-central Florida is prevalent in the shelter cat population, making it an important parasitic disease. Serologic tests continue to vary in sensitivity, specificity, and overall diagnostic performance. A negative Ab test does not preclude adult HW infection. The Ag tests used in this study were more sensitive for adult HW infection in cats compared to Ab assays. The overall measure of test performance (AUC:ROC) was higher for the Ag tests compared to the Ab tests. Combining results from serum Ab and Ag tests achieved higher sensitivities than using serum Ab and Ag test results alone. There are many infectious agents of cats associated with fever and so infectious disease tests are commonly applied to samples from cats with fever in an attempt to determine causation. New diagnostic tests have been developed for ehrlichial infections, Bartonella spp. infections, and both Haemobartenella felis strains (Mycoplasma haemofelis [Hflg] and M. haemominutum [Hfsm] ). The purpose of this study was to apply a battery of infectious disease tests to blood and serum samples from cats with fever (n ϭ 121) and geographically matched cats without fever (n ϭ 99). Assays performed included: FeLV p27 antigen ELISA; Overall, serologic evidence of exposure or PCR evidence of infection by one or more of the infectious agents was detected in 54.5% and 50.1% of the cats with and without fever, respectively. All 3 Ehrlichia spp. PCR positive results were identified as E. equi (Anaplasma phagocytophila) by DNA sequence evaluation. With the exception of Ehrlichia PCR, positive test results were detected in cats with and without fever and there were no statistically significant differences between groups. These results emphasize that since each of the assays used can be positive in cats with and without fever, causation of fever cannot be proven based on the test result alone. Multidrug resistance is a major obstacle to successful chemotherapy in canine lymphoma and leukemia. Drug efflux by P-glycoprotein (P-gp) is considered to be a common mechanism of multidrug resistance. In dogs, several reports have shown the expression of P-gp in drug-resistant lymphoma cells, however, there are discrepancies observed between the P-gp expression and the drug resistance. In this study, we carried out a study on the functional detection of multidrug resistance in 20 canine lymphoma and leukemia cases as well as 2 canine lymphoid tumor cell lines.

The canine cases examined in this study included 17 lymphoma cases and 3 acute lymphoblastic leukemia (ALL) cases. Tumor cells were collected from neoplastic lymph nodes or peripheral blood mononuclear cells. Response to the multidrug chemotherapy using vincristine, cyclophosphamide, doxorubicin, Lasparaginase and predonisolone was evaluated based on the size of lymphoma or the count of leukemia cells. Two canine lymphoid tumor cell lines included a canine ALL cell line (GL-1) and its vincristine-resistant subline (GL-1/VCR). Functional detection of multidrug resistance was carried out by a flowcytometric assay using a fluorescent dye, rhodamine123 (Rh123) and cyclosporine A (CsA) as a substrate/modulator of the P-gp pump. Drug efflux activity by P-gp was evaluated as the ratio of mean fluorescence intensity of Rh123 in the tumor cells treated with Rh123 alone to that in the tumor cells treated with Rh123 and CsA.

The 20 cases with lymphoma and ALL were divided into two groups: a good response group (n ϭ 11) showing complete response or partial response and a poor response group (n ϭ 9) showing no change or progressive disease. The ratio of Rh123 accumulation in the absence of CsA to that in the presence of CsA was significantly higher in the poor response group than the good response group. The ratio was less than 1.50 in all 11 cases in the good response group, whereas it was more than 1.50 in 5 of 9 cases in the poor response group. The ratio of Rh123 accumulation was 11.25 in drug-resistant GL-1/VCR cells, which was 10 times larger than that in drug-sensitive GL-1 cells. We also carried out an immunohistochemical analysis on P-gp with a monoclonal antibody directed to P-gp (C219) and obtained the results almost identical to those by the functional assay of the multidrug resistance.

The present study indicates that the functional assay of the multidrug resistance will provide an useful information on the drug resistance of canine lymphoid malignancies. This multi-site study was designed to evaluate the clinical utility of a veterinary bladder tumor antigen test (V-BTA) manufactured by Alidex, Inc. as a screening test for transitional cell carcinoma (TCC) of the canine urinary tract.

Urine samples were collected at four participating institutions between June 1999 and November 2001. Samples were shipped to Alidex Inc. to facilitate testing within 48 hours of collection. Individuals performing the test were blinded with regard to patient diagnosis. Sample categories included: 1) TCC, 2) healthy controls, 3) controls with non-TCC urologic disease, and 4) controls with non-urologic disease. A diagnosis of TCC was categorized as suspected when clinical signs and imaging suggested a mass, and as confirmed when cytological or histopathological evaluation was consistent with a diagnosis of TCC. For the purpose of analysis, prostatic carcinoma and prostatic TCC cases were categorized together with other TCC cases. Test sensitivity and specificity were calculated using standard epidemiologic methods. Logistic models were developed to assess the effect of disease status, test conditions, urine composition and signalment on the performance of the V-BTA test.

A total of 229 specimens were analyzed including 48 with suspected (n ϭ 3) or confirmed (n ϭ 45) TCC. There were 31 dogs with TCC confined to the bladder, 13 with TCC of non-bladder sites (kidney, urethra, or prostate) and 4 with both bladder and another urinary site involved. Of the 45 confirmed TCC cases, confirmation was by cytology (n ϭ 32), histopathology (n ϭ 9), or both (n ϭ 4). The non-TCC control groups consisted of 82 healthy dogs, 71 dogs with non-TCC urologic disease, and 28 dogs with non-urologic disease. Calculated sensitivities were 88, 87, and 85% for all TCC cases, confirmed TCC cases, and confirmed bladder TCC cases, respectively. Calculated specificities were 84, 41, and 86% using samples from healthy dogs, dogs with urinary tract disease (non-TCC), and dogs with non-urinary disease, respectively. The test performed better on spun samples than on unspun samples, as per the label directions.

We conclude that the V-BTA is a relatively sensitive test for the detection of canine urinary tract TCC. Specificity was reasonable with the exception of dogs with non-TCC urinary tract disease. Test limitations and indications will be discussed in detail. Microbubble contrast agents have been used in a variety of diagnostic studies in human medicine. These agents were first used in cardiovascular studies, and have expanded into vascular studies involving several other organ systems. Doppler ultrasound is currently the standard for ultrasound imaging of tumor vascularization. Our goal was to characterize tumor vascular and perfusion patterns using contrast harmonic ultrasound (CHU).

Dogs and cats with spontaneous neoplastic masses and control dogs with normal lymph nodes were imaged with a General Electric Logic 700 ultrasound machine system with power doppler (PD), and single pulse (HH) and angiographic (HA) harmonic settings. Ultrasound contrast (Definity (TM), Bristol-Myers Squibb Medical Imaging) was injected intravenously (0.1-0.2 ml) for each harmonic sequence. Imaging sequences were directly transferred to a digital video camera.

In all cases the HA settings allowed better characterization of the vascular pattern compared to PD. Significantly increased number of vessels and more accurate imaging of the vessels' size was seen in the HA sequences (pϽ0.05). Images with PD led to exaggerated vessel diameter measurements and underestimated the tumor vasculature. Malignant lymph nodes in dogs with lymphoma had significantly increased number of aberrant vessels compared to normal dogs (pϽ0.05).

All malignant lymph nodes of dogs with lymphoma had uniform patterns of perfusion as determined by HH. The peak enhancement was seen approximately 15 seconds after injection. Masses had variable perfusion patterns. Perfusion patterns were similar in cases that had both HH computed tomographic (CT) imaging. There was a tendency for CT to underestimate the perfused regions of masses compared to HH.

We conclude that CHU is a valuable clinical tool for characterizing the vascular and perfusion patterns of masses, including malignant lymph nodes. These patterns may be valuable in determination of malignant versus reactive lymph nodes, response to chemo-or radiation therapy and as a measure of malignant neovascularization. In selecting sites for biopsy, characterization of the vascular pattern may assist with avoiding areas with high numbers of vessels and the perfusion pattern to avoid areas of necrosis. The purpose of the study was to evaluate the radiotherapeutic effects of combined carboplatin (radio-sensitizing/chemotherapeutic) and the radiopharmaceutical Sm-153-EDTMP (Samarium-153-ethylenediaminetetramethylene phosphate) on naturally occurring canine osteosarcoma (OSA) of the distal radiusulna. Six histopathologically confirmed cases were included in the trial. The sizes of the primary tumors were recorded from radiographs and staged using the TNM method for bone tumors.

The dogs were sedated (metomidine HCL 0.1ml/kg iv.) and an intravenous continuous infusion (200ml normal saline) for one hour was started with the required dose of carboplatin (150 mg/m2) added. At thirty minutes, 37 MBq/kg of Sm-153-EDTMP was given intravenously over 1 minute. At 24 hours, scintigraphy was done (energy peak 140 keV, window 15%). For comparative purposes, a reference region of interest (ROI) was identified over the tumor site and opposite limb. Counts per pixel were recorded for the tumor (T) and contralateral limb (NTC) and calculated as a ratio (T/NTC). All cases returned for weekly CBC (4 weeks) and follow-up radiographs (1, 2, 3, 4 months). Five cases received an additional 300mg/m2 carboplatin i.v. at three weekly intervals for five treatments post radiotherapy. The results were compared to another group of distal radius (osteoblastic) OSA.

All dogs had histologically confirmed osteoblastic osteosarcomas, however radiological appearance differed with four being osteoblastic and the balance osteolytic in appearance. Pain control was evident in all dogs (based on limb use) within two weeks of therapy. This differs substantially from Sm-153-EDTMP as a single agent. The mean uptake ratio (T/NTC) (9.7 SDϮ4.9) in comparison to other OSA cases (2.7 SDϮ0.3) showed an increased uptake indicating that the carboplatin infusion improved uptake of Sm-153-EDTMP. Leukocyte count and platelet count reached a nadir at 2 weeks with mean lows of 5.2 Ϯ 1.4 ϫ 10(9)/l and 53.7 ϫ 10(3)/l respectively. These results do not differ significantly from hematological changes when Sm-153-EDTMP is used alone. All OSA underwent tumour progression as confirmed clinically and radiologically. These results serve as a pilot trial for a larger study to evaluate survival and time to metastasis. Frameless stereotactic radiosurgery (FSR) delivers a single high conformal dose of radiation to the target, with steep dose gradients resulting in minimal dose to the surrounding normal tissues. Frameless methods were developed in people and have proven easy to use in animals. We report on the use of FSR for the treatment of canine and feline nasal tumors. A custom bite plate is fitted to the patient's dentition at the time of CT scan. A reference array composed of six infrared light emitting diodes (IRLED) is rigidly fixed to the bite plate for both imaging and treatment. Image sets are fused to facilitate computerized treatment planning. Treatment is performed under GA. A ceiling mounted infra-red detecting camera localizes each of the IRLED's in space, defining the bite plate position in reference to the isocenter of the linear accelerator to which the camera is calibrated. The linear accelerator is mounted on a rotating gantry; radiation is delivered to the target (defined as the contrast-enhancing region) in a series of non-coplanar arcs. All patients received 1250 to 1750 cGy as a single dose.

Seventeen nasal tumors have been treated by the technique of which 12 were canine and four feline. Tumors included chondrosarcoma (CSA)(n ϭ 1), adenocarcinoma (ACA)(n ϭ 6), lymphoma (LSA)(n ϭ 2), osteosarcoma (OSA)(n ϭ 3), squamous cell carcinoma (SCC)(n ϭ 2), myxosarcoma, soft tissue and undifferentiated sarcomas. Acute complications have been minimal, and consist of mild erythema of the skin, decreased lacrimation and mild keratoconjunctivitis. Late effect (cataract) in a single case has occurred. Three cases were bilateral, 12 cases were unilateral, divided equally. Two cases involved the frontal sinuses. Both LSAs occurred in cats; one had CR and the second cat showed progression and was euthanized (3m). The longest surviving case is a CSA (8m), with no evidence of metastases. ACA (n ϭ 6) cases (accruing), are all alive with static disease (range 1-2m). Of the three OSA, two have static disease; one underwent progression with metastases and was euthanized (4m). A single SCC underwent progression (5m) with metastases. In most cases, clinical improvement was apparent within 1-week post therapy. Cases are still currently accruing to the study.

The current followup period is too short to draw conclusions as to survival but FSR may offer an alternative to conventional fractionated radiation therapy for certain nasal tumors. Benefits to the patient and owner would include a single treatment session requiring less hospitalization time, and minimal complications (normal tissue sparing). The Hereditary Ataxia in the Jack Russell Terrier is a gait disturbance with symmetric generalized ataxia, hypermetric and spastic movements as seen in cervical lesions. Histopathology reveals a systemic disease of the central nervous system, predominantly a central axonopathy.

33 phenotypic affected dogs were examined. Gait abnormalities started at 2 to 9 months of age. In 13 dogs seizures occurred in addition to the ataxia, 7 dogs developed respiratory distress. Further investigations, such as routine haematological and biochemical analyses, examination of the cerebrospinal fluid, myelography and computed tomography (thecography to show cerebellar structures), revealed no abnormalities. However, brainstem auditory evoked potentials were abnormal in 4 of 7 examined dogs. Only wave I and II could be detected. These findings could support the clinical diagnosis of Hereditary Ataxia in Jack Russell Terriers. Histopathology was performed on 9 of the examined dogs and confirmed the clinical diagnosis in all cases.

There was no significant influence of sex, hair coat, hair color, withers height, number of puppies born and inbreeding coefficient on the occurrence of Hereditary Ataxia in the examined dog population. Investigations regarding the mode of inheritance were performed using complex segregation analysis on 3 pedigrees with a total of 115 Jack Russell Terriers (27 clinically affected and 88 unaffected litter mates and ancestors). Four different modes of inheritance were tested: monogenic, mixed (major gene with polygenic variation), polygenic and environmental effects (without any genetic effects). The polygenic model explained the pedigree data best (''maximum-likelihood''). Because of the small amount of examined dogs, a segregation of a major gene could not be excluded.

Our clinical examinations, histopathology and the segregation analysis revealed differences between the Hereditary Ataxia in the Jack Russell Terrier and the Hereditary Ataxia in the Fox Terrier. Further investigations, also based on molecular genetic studies, are necessary to determine the pathogenesis and whether a major gene effect is responsible for the axonopathy. Medical record, seizure survey, and phone interview information was obtained for 29 Vizslas with idiopathic epilepsy (IE), 72 unaffected siblings, and 41 parents from 13 families to determine the common clinical characteristics, feasibility of a genome scan, and most likely mode of inheritance of IE in this breed. IE was diagnosed based on age of onset of seizures, seizure history, blood work results, and neurological examinations. CT or MRI scan with CSF analysis was required for inclusion of individuals with an age of onset of seizures of less than 6 months or greater than 5 years. Simple segregation analysis was performed with the Davie ascertainment correction and chi-square analysis.

Seventy eight percent of affected Vizslas exhibited partial onset seizures. Partial onset signs included leg tremors, staring, pupil dilation and/or salivation without loss of consciousness in over 50% of the dogs with partial signs. The median age of onset of seizures was 3.0 years of age. The chi-square analysis with the ascertainment correction and the ascertainment corrected estimated segregation frequency of p ϭ 0.28 (95% confidence interval p ϭ 0.11 to 0.45) of families with unaffected parents were both consistent with autosomal recessive inheritance (expected p ϭ 0.25). Polygenic inheritance remains a possibility, however, and these findings should be considered preliminary information until more Vizsla families with IE are analyzed. Simulated linkage with FASTSLINK on all 13 families using an age dependent penetrance model indicated that the average maximum Log of Odds (LOD) score would be 3.23 with a maximum LOD score of 6.56 and a 55.5% chance of exceeding a statistically significant LOD score of 3.0.

This analysis indicated that IE in Vizslas may be an autosomal recessive trait, and that the families ascertained are sufficient for a whole genome scan to potentially find the chromosomal segment containing the epilepsy gene.

NIDAZOLE TOXICOSIS IN THE DOG. Jason Evans, Donald Levesque, Scott Plummer, Randy Longshore, Kim Knowles. Las Vegas, Nevada/Phoenix, Arizona.

Central nervous system toxicity secondary to metronidazole administration has been reported in dogs and cats. The currently recommended treatment is drug discontinuation and supportive therapy with reported recovery rates of 1 to 2 weeks. This investigation was instituted to determine if the administration of diazepam had an effect on the recovery of dogs with metronidazole toxicosis. The records of 20 dogs with metronidazole toxicosis were retrospectively analyzed.

The dose and duration of metronidazole therapy and the response and recovery times of twelve dogs treated with diazepam were compared to eight dogs receiving only the traditional supportive care. Response time is defined as the time to resolution of the debilitating clinical signs. Recovery time is the time to resolution of all residual clinical signs. The average dose of metronidazole and duration of administration for the diazepam-treated and control group was 60.3 mg/kg/day for 44.9 days and 65.1 mg/kg/day for 37.25 days, respectively. There was no significant difference between the groups with respect to metronidazole dose or duration of therapy. The protocol for diazepam administration consisted of an initial intravenous bolus followed by oral dosing every 8 hours for 3 days. The average dose of both the intravenous and oral diazepam was 0.43 mg/kg. The average response time of the diazepam treated dogs was 13.4 hours compared to approximately 4.25 days for the control group. Recovery time was also significantly improved for the diazepam-treated dogs (38.8 hours) compared to the control group (14 days).

The results of this investigation show that dogs with metronidazole toxicosis recover faster when treated with diazepam. The exact mechanism by which diazepam exerts its favorable effect in reversing signs of metronidazole toxicity is not known but is likely related to GABA receptor dysfunction within the cerebellar and vestibular systems. We postulate that diazepam competitively displaces metronidazole from the benzodiazepine sites on the GABA receptors. Speculation for metronidazole affinity for the GABA receptor site is based on the similarity of both structure and clinical signs of toxicity of metronidazole and the benzodiazepine antagonist, flumazenil. Atlantoaxial instability (AAI) of dogs usually results from either congenital malformation of, or trauma to, the atlantoaxial (AA) joint and/or its associated ligaments. The primary goals of this study were to describe the variable clinical presentation of this disease, a modified ventral surgical stabilization technique, and long term outcome in dogs with AAI.

The medical records of 17 dogs diagnosed and treated surgically for AAI were retrospectively reviewed at the UC Davis VMTH from September 1995 to February 2001. All dogs were evaluated by physical and neurological examinations. Complete vertebral column radiographs were done under general anesthesia in all dogs; myelography in 13 dogs; computed tomography (CT) following myelography in two dogs; and magnetic resonance imaging (MRI) in 1 dog. Ventral surgical stabilization was done in all dogs using a modified version of the technique described by Schulz, et al (Vet Surg 26:317-25, 1997) . Two transarticular Steinmann pins, 2 pins or cortical bone screws in the body of C1, and 2 pins or bone screws in the body of C2 were placed. The implants were incorporated in polymethylmethacrylate.

Clinical signs ranged in duration from 1 day to 3 years. Five dogs had a history of trauma. Seven dogs presented with non-ambulatory tetraparesis (NAT), 4 dogs with cervical pain only, and the remainder with varying degrees of ataxia and paresis Ϯ cervical pain. Two dogs with NAT presented with clinical signs of brainstem dysfunction. Radiographic abnormalities of the dens were seen in 13 dogs. Myelography, CT, and MRI delineated compression and/or deviation of the spinal cord.

Outcome was assessed to be good, fair or poor at the time of follow-up examinations completed in 16 dogs surviving the immediate post-operative period (6-69 months). 12 of the 17 dogs had a good outcome (either absence of detectable neurological abnormalities or mild gait abnormalities on discharge or recheck examination). Two dogs had a fair outcome (persistent ataxia), and 3 dogs had a poor outcome (1 dog died post-operatively, and 2 dogs had persistent paresis and ataxia). 2 of the 3 dogs with a poor outcome initially presented with signs of brainstem dysfunction.

It is concluded that ventral surgical stabilization of AAI may result in good long term function in a majority of dogs. AAI should be considered in any dog presenting with signs of cervical pain and/or myelopathy or caudal brainstem dysfunction. The modified ventral surgical technique presented here provided excellent long term stabilization in the AA joint in all cases. Long term neurological recovery was adversely affected in dogs with greater severity and chronicity of presenting neurological signs, especially when brainstem dysfunction was present.

FIRST CHOICE ANTICONVULSANT THERAPY IN THE CA-NINE EPILEPTIC. Boothe DM, Dewey C and Slater M. Texas A&M University, College Station, TX 77845.

Efficacy and safety of bromide (BR) were compared to phenobarbital (PB) in 46 dogs with spontaneous epilepsy using a parallel, randomized double blinded study design. Acceptance was based on seizure history, physical and neurologic examinations and clinical pathology. Dogs were loaded over a 7 day period to achieve the minimum end of the therapeutic range of the assigned drug. PB (3.5 mg/kg) or BR (15 mg/kg) was administered every 12 hours. Data (clinical pathology and drug concentrations) were measured at baseline and at 30 days intervals for 6 months. Drug doses were modified based on drug concentrations and response to therapy. Comparison of 6 month data to baseline data within groups was based on, for continuous data tests (p p Յ 0.025 to compensate for multiple testing) and for categorical, Wilcoxin signed rank (within group) and using Wilcoxin ranked sum (between groups) (pϽ0.05). Proportions were compared using chi-square (pϽ0.05).

All but 3 patients completed the study. Seizures initially worsened in 3 dogs on BR but not in any PB patient. Mean seizure number, frequency and severity were reduced at 6 months compared to baseline for both drugs; seizure duration was shorter for PB but not BR. Seizure activity was eradicated in a greater percent of PB (85%) compared to BR (65%) patients, but successful control (at least 50% reduction in seizure number) did not differ between drugs at 6 months. Mean bid dose and drug concentrations were dose 4.1 Ϯ 1.1 mg/kg and 27 Ϯ 6 mg/ml, respectively for PB and 31 Ϯ 11 mg/kg and 1.9 Ϯ 0.6 mg/ml for BR. Both drugs caused abnormal behaviors. Weight increased by 10% in both groups. Changes in clinical pathology were limited to increased (but within normal) serum alkaline phosphatase and decreased (but within normal) serum albumin at 6 months for PB compared to baseline and compared to BR at 6 months. Side effects at one and six months, respectively for each drug were: ataxia (PB: 55 vs 5%; BR: 22 vs 9%), obtundation (PB: 50 vs 5%; BR: 35 vs 13%), polydypsia (PB: 40 vs 0%; BR: 39 vs 4%), polyuria (PB: 35 vs 0%; BR: 13 vs 0%), hyperactivity (PB: 35 vs 10%; BR 43 vs 4% [one failure]), polyphagia (PB: 30 vs 0%; BR: 43 vs 4%) and vomiting (PB: 20 vs 0%; BR: 57 vs 21% [one failure]). One PB dog failed due to neutropenia. The incidence of obtundation and vomiting were greater in BR compared to PB at 6 months. This study suggests that both PB and BR are reasonable first choices for control of epilepsy in dogs, although PB may provide better control. Side effects can be expected to be greater in BR following chronic dosing. Magnetic resonance imaging and cerebrospinal fluid analysis are useful for diagnosing animals with suspected intracranial disease. However, even with the advanced neurodiagnostic tools currently available, it is challenging to differentiate between the different types of encephalitis and round cell tumors without obtaining a surgical biopsy. While establishing a definitive diagnosis is essential in treating the patient and counseling clients, it is often difficult to persuade clients to pursue aggressive diagnostic efforts such as CT-guided brain biopsy. The purpose of this study was to determine if it is possible to differentiate between granulomatous meningoencephalitis (GME), necrotizing encephalitis (NE), fungal encephalitis, lymphosarcoma (LSA), and other encephalitides based on MRI alone or MRI and CSF analysis together in the absence of a biopsy.

Medical records of 46 dogs and cats with a confirmed necropsy diagnosis as listed above were reviewed. Only animals in which an MRI, CSF analysis, or both were performed were included. Parameters evaluated on MRI included axial origin, lesion location and shape, distribution, signal intensity, enhancement patterns, and characterization of edema. Results of cerebrospinal fluid analysis included total protein in mg/dl (TP), nucleated cell count per microliter (NCC), and differential analysis of the nucleated cells.

The most useful parameters on MRI for differentiating among the inflammatory diseases or round cell tumors were signal intensity on T1-weighted images and lesion location. In general, lesions that were isointense on T1-weighted images were only found in GME or LSA, whereas lesions that were hypointense were only associated with NE, cryptococcus, or other, non-specific encephalitides. For lesions that were hypointense on T1, those that were located both infraand supra-tentorially were only found with fungal disease or non-specific encephalitis and not with NE. For LSA or GME, the differentials could be narrowed down further since many of the LSA cases also had an extra-axial mass. While MRI was helpful in eliminating some of the differentials, there were exceptions that made definitive diagnosis impossible. However, CSF results were a valuable adjunct in further eliminating many of the differentials and complementing the MRI findings. Significant differences in TP and NCC were found among GME, NE, LSA, and fungal encephalitis. The combination of MRI and CSF findings in animals with intra-cranial disease can be used to help distinguish the various inflammatory diseases and round cell tumors and contribute to establishing an accurate diagnosis in the absence of a surgical biopsy. Spinal epidural empyema (SEE) is a rarely reported suppurative, septic process in the epidural space that develops secondary to hematogenous spread or direct extension of local infection. In previous reports, 3/5 dogs with SEE were euthanized due to the severity of their disease. The purpose of this report is to characterize the clinical signs, diagnostic and surgical findings, and longterm outcome in 7 dogs with SEE treated with laminectomy and antibiotics, and to review the literature on this subject.

Dogs with spinal epidural empyema were identified retrospectively from the medical records at University of California-Davis, Veterinary Medical Teaching Hospital between 1992 to 2001. Selection criteria required that all dogs had a neurological examination, a minimum database, spinal radiographs, myelography, culture, and a definitive diagnosis of spinal epidural empyema confirmed by surgery and histopathology or necropsy. Seven dogs fit the selection criteria.

Common presenting signs were lethargy, fever, anorexia, apparent spinal pain and paraparesis/plegia. Common laboratory abnormalities were a peripheral neutrophilia (mean ϭ 24,247/uL reference range, 6,000 to 17,000/uL) and a neutrophilic pleocytosis in CSF (mean total nucleated count 89.9/uL, reference range Ͻ5/uL), mean percent neutrophils ϭ 58%, (reference range ϭ 0). Three dogs had concurrent discospondylitis. One dog with concurrent discospondylitis had a spinal luxation. On myelography, extradural spinal cord compression was noted in all dogs. Two dogs had focal compression, three dogs had multifocal lesions, and two dogs had diffuse compression. Bacteria were cultured from the blood, surgical site, pleural fluid or urine in 6 of the 7 dogs. Bacteria were not cultured from CSF in any dog. All dogs were treated with antibiotics and surgical decompression with hemilaminectomy Ϯ dorsal laminectomy. Two dogs had hemilaminectomies at multiple sites. One dog had a hemilaminectomy and dorsal laminectomy at separate sites. Two dogs with discospondylitis were surgically stabilized with pins and methylmethacrylate. One dog that did not have discospondylitis initially, four months later developed vertebral osteomyelitis and multifocal bone infarcts in the ilium and long bones, but was neurologically normal. Five of seven dogs improved neurologically and were assessed to have a good long term outcome. Two dogs were euthanized: One dog was euthanized with worsening of neurological signs and pneumonia, and another due to herniation of a cervical intervertebral disc and multiple orthopedic problems.

Results of this study suggest that treatment with surgical decompression and long term antibiotics may result in a good long term outcome in dogs with SEE. SEE should be included as a differential diagnosis in dogs with clinical signs of fever, apparent spinal pain and myelopathy.

BROSPINAL FLUID D-DIMER CONCENTRATIONS. Rossmeisl JH 1 , Troy GC 1 , Inzana KD 1 , Jortner B 2 , Boon GD 2 . Virginia-Maryland Regional College of Veterinary Medicine, Blacksburg, VA. 1 Department of Small Animal Clinical Sciences, 2 Department of Biomedical Sciences and Pathobiology. D-dimers are plasminolytic cleavage products formed by the cross-linkage of fibrin by Factor XIIIa. Peripheral blood D-dimer analysis has been demonstrated to be beneficial in the diagnosis of several systemic hemostatic and thrombotic disorders in people and dogs. In humans, CSF D-dimer testing is a useful adjunct for the differentiation of pathologic subarachnoid from iatrogenic CNS hemorrhage, head trauma prognostication, and identification of cerebral ischemia and infarction. The objectives of this study were to determine if D-dimers could be quantitated in canine CSF, establish a reference range for D-dimers in canine CSF, and to establish if D-dimers were elevated in CSF experimentally contaminated with peripheral blood.

Cerebellomedullary CSF samples and citrated blood samples were collected from 16 mixed breed dogs with no evidence of neurologic disease on clinical or necropsy examination, and divided into 50 l aliquots. CSF samples were grossly, biochemically, and cytologically normal in all dogs. Following initial CSF analysis, 1 l of variably diluted peripheral blood (mean 9,500 Ϯ 2,000 RBC/ l, range 0-15,000 RBC/l) was added to an additional aliquot of CSF from each dog. Both qualitative latex agglutination (LA) and quantitative enzymatic immunoassay (EIA) D-dimer human monoclonal antibody detection systems were then performed on all experimental samples (peripheral blood, normal CSF, blood contaminated CSF).

The mean peripheral blood EIA D-dimer concentration was 54.6 Ϯ 19.8 ng/ ml (range of 0-190 ng/ml), and the mean CSF EIA D-dimer concentration was 16.2 Ϯ 4.3 ng/ml (range 0-54 ng/ml). The mean EIA D-dimer concentration (19.4 Ϯ 2.1 ng/ml) of the blood contaminated CSF was not different (p ϭ 0.78) from unadulterated CSF. Peripheral blood and CSF LA assays were negative in all samples. Results of this study indicate that D-dimers can be measured in canine blood and CSF, and that acute blood contamination of CSF, as may occur with a traumatic spinal tap, does not result in an abnormal elevation of the CSF D-dimer concentration in vitro.

DOGS WITH SUSPECTED BRACHIAL PLEXUS AVULSION. D. Faissler 1 , S. Cizinauskas 2 , A. Jaggy 3 . 1 Tufts University, Department of Clinical Science, North Grafton, MA, 2 University of Helsinki, Finnland, 3 University of Berne, Department of Clinical Neurology, Switzerland.

Acute unilateral neural forelimb dysfunction in dogs is commonly due to brachial plexus root avulsion. Nerve lesions of the brachial plexus are reported to be rare. Prognosis for complete functional recovery is often unfavorable. In polytrauma, large breed and workings dogs, prognostic indicators for functional recovery in the early course of the disease would be beneficial. The goal of this retrospective study was to search for these prognostic factors. Clinical signs and electrodiagnostic findings were related to recovery. Only dogs with a known outcome were included in the study.

Thirty seven patients were presented to the University of Berne with acute motor dysfunction due to suspected brachial plexus nerve root avulsion. Most of the dogs were hit by car and the mean age was 3.4 Ϯ2.6 years. The first consultation was done on average 34 days after the trauma. Clinical signs were non weight bearing monoparaparesis (40.5%) or monoplegia (59.5%), abnormal pain perception (81%), abnormal proprioception in the ipsilateral hind limb (24%), Horner's syndrome (54%) and abnormal panniculus reflex (46%). Follow up period was 1 day to 2 years. Five (13.5%) dogs regained normal front limb function. Fifteen (40.55%) dogs improved or showed stationary dysfunction, but none of these 15 patients were able to bear weight. The diseased limb was amputated in 5 (13.5%) patients and 12 (32.5%) dogs were euthanized. The reasons for amputation or euthanasia were selfmutilation, infections or the inability to be a working dog.

At first presentation, the best predictor of complete recovery was pain perception. Out of 7 dogs with normal sensation, 5 recovered completely. In 9 dogs with abnormal sensation in the autonomous zone of the radial nerve, only one dog was euthanized and 8 showed improvement. Out of 21 dogs with abnormal pain perception below the elbow, only 6 improved or remained stationary, 5 needed limb amputation and 10 were euthanized. In dogs with partial motor function, the outcome was slightly better than in dogs with monoplegia, but motor function was less predictive of functional recovery than pain perception. The additional presence of Horner's syndrome, ipsilateral proprioceptive deficit or abnormal panniculus reflex was not indicative of negative outcome.

Electromyography and motor nerve conduction of the radial or ulnar nerve were done in 14 dogs. Abnormal motor nerve conduction was correlated with a negative outcome. Seven of 14 dogs were euthanized, 2 needed limb amputation, 2 had no further improvement, one dog improved and one dog recovered completely. The clinical signs and pathology of ACP are thought to be due to immunemediated damage to spinal nerve roots and peripheral nerves. Axonal damage and demyelination are most pronounced at the level of the ventral nerve roots, leading to breakdown of the blood-CSF barrier. ACP is considered an animal model for the acute axonal form of Guillain-Barré syndrome in humans. In humans with GBS, lumbar cerebrospinal fluid (CSF) analysis reveals protein elevation with normal cellularity. In this study, cisternal and lumbar CSF analyses were performed in dogs previously diagnosed with ACP.

Serum and CSF (lumbar and cisternal) was evaluated in 15 dogs with ACP and 10 healthy control dogs. Total protein quantitation and cytology were performed on all CSF samples. Polyacrylamide gel electrophoresis, membrane transfer, and densitometry were performed on all CSF and serum samples to calculate albumin and immunoglobulin (IgG) concentrations. These values were used to determine albumin quotient (AQ), an estimate of blood-CSF barrier integrity, and IgG index (IGI), a measure of intrathecal immunoglobulin production.

Mean lumbar CSF protein concentration was significantly higher in ACP dogs (83.6 mg/dl) compared to controls (22.4 mg/dl). Fourteen dogs (93.3%) had abnormally elevated lumbar CSF protein concentration. The mean albumin quotient was significantly higher in lumbar CSF of ACP dogs (1.63) compared to control dogs (0.4). There was no significant increase in protein concentration or AQ in cisternal CSF samples. No significant difference in IgG index was found between ACP and control groups for either lumbar or cisternal CSF. Cell count and cytology were normal in all CSF samples in both groups.

The results of this study show a significant increase in lumbar, but not cisternal, CSF protein concentration in dogs with ACP, consisting predominantly of albumin. The increased AQ seen in these samples is consistent with transudation of protein rather than intrathecal immunoglobulin production. Lumbar CSF analysis is a clinically useful diagnostic test in dogs suspected of having ACP. The systemic immune system is thought to play an important role in the pathogenesis of Acute Canine Polyradiculoneuritis (ACP). Although many dogs that develop this polyradiculoneuropathy have a history of exposure to a raccoon 7-14 days prior to the onset of clinical signs, a significant number of dogs have no history of exposure and develop identical clinical signs and pathology. ACP is considered to be an animal model of the acute axonal form of Guillain-Barré syndrome (GBS) in humans. Various antecedent events have been shown to occur 7-14 days prior to onset of neurological signs in GBS, and are thought to trigger an immune response resulting in peripheral nerve pathology via a bystander effect. These factors include bacterial and viral infections, vaccination, and surgical procedures. In this study, the relationship between ACP and prior infection or exposure to various infectious agents was examined.

Serum from forty-four dogs with ACP was assayed for antibodies against select infectious agents. Seroprevalences for Ehrlichia canis (9.1%), Borrelia burgdorferi (11.9%), Toxoplasma gondii IgG (55.8%), T. gondii IgM (4.7%), Neospora caninum (6.8%), Campylobacter jejuni (4.5%), and canine distemper virus (100%; CDV) infections were determined. Only the T. gondii IgG seroprevalence rate was greater than that reported in previously published seroprevalence studies. Serum CDV and T. gondii IgG antibody titers were then measured in 45 age and region-matched dogs without ACP. There was no significant difference in CDV mean reciprocal titers or seroprevalences between dogs with ACP and control dogs. Dogs with ACP were more commonly positive for T. gondii IgG than control dogs (PϽ0.001). Presence of T. gondii IgG in serum generally denotes current tissue infection. Based on the results of this study, a positive association exists between dogs with ACP and T. gondii infection. A causal relationship, however, has not been determined. Blood contamination of samples may adversely affect results of cytologic, serologic, biochemical, and molecular tests performed on CSF. A clear consensus of these effects does not exist in the veterinary literature. It was hypothesized that iatrogenic blood contamination of CSF would result in significant increases in both the CSF total protein (TP) and nucleated cell count (WBC), when analyzed in vitro.

Cerebellomedullary cisternal CSF samples were collected from 16 dogs considered to be normal based on clinical and histologic examination of the CNS, and divided into 50 l aliquots. Initially, an aliquot of CSF was assayed using standard, previously described techniques. CSF samples were grossly, biochemically, and cytologically normal in all dogs. Following initial analysis, CSF aliquots from each dog were contaminated in vitro with serial dilutions of peripheral blood (ie 1 l of blood with a final concentration ranging from 500 to 32,000 RBC/l added to 50 l CSF sample), and each aliquot assayed in manner identical to initial samples.

In vitro CSF blood contamination resulted in statistically significant linear increases in both the TP (p Ͻ 0.001) and WBC (p Ͻ 0.01) with increasing RBC concentrations. The relationships between the CSF TP, CSF WBC, and CSF RBC concentrations can be defined by the following equations, respectively: Although this study demonstrates that the CSF TP and WBC are significantly increased statistically by blood contamination, the clinical impact of such contamination on commonly performed clinicopathologic tests is negligible until gross blood contamination occurs (Ͼ10,000 RBC/l).

IN TWO DOGS. Y. Fujii 1 , Y. Wakao 1 , T. Awazu 1 , T. Yamane 1 , N. Machida 2 , and H. Ninomiya 1 . 1 Azabu University, Kanagawa, and 2 Tokyo University of Agriculture and Technology, Tokyo, Japan.

In most cases, when a continuous murmur is heard at the heart base, a diagnosis of patent ductus arteriosus (PDA) would be made. However, the cases reported here were diagnosed as an anomalous shunt between the bronchoesophageal artery and pulmonary artery, mimicking PDA. The purposes of this study were to describe the clinical feature of an anomalous shunt and to determine the difference between an anomalous shunt and PDA.

Case 1: A 1-year-old 3.5 kg female Miniature Dachshund was presented for evaluation of a continuous murmur (grade 3/6) which was loudest at the heart base. Thoracic radiography revealed mild cardiomegaly (VHS ϭ 10.6). On electrocardiograph (ECG), P pulmonale was observed. Echocardiography demonstrated shunting blood flow presumably from the arterial duct at the pulmonary artery carina. Selective contrast angiography revealed that the bronchoesophageal artery and other blood vessels with complicated branches were shunting into the pulmonary artery.

Case 2: A 1-year-old male Beagle dog was evaluated for a continuous murmur (grade 2/6) loudest at the heart base. Findings of ECG and thoracic radiography were normal. Echocardiography demonstrated turbulent flow in the proximal portion of the left pulmonary artery. Selective contrast angiography revealed a dilated bronchoesophageal artery with complicated branches shunting into the left pulmonary artery. This was an experimental dog that was euthanized for another research purpose. At the necropsy, latex resin was poured into bronchoesophageal artery and pulmonary artery to clearly identify their relationship. A large, well-developed bronchoesophageal artery with many branches coiled around the great vessels.

No cyanoses, lung disease, pulmonic stenosis, pulmonary hypertension, or heartworm infection were found in our two cases. Therefore, congenital anomaly of vessels was the most likely diagnosis. The following three findings were observed in both cases: 1) relatively small continuous murmur loudest at the heart base, mimicking PDA; 2) shunt flow signals in a pulmonary artery on echocardiography; and 3) no PDA on selective angiography, but evidence of anomalous shunting vessels. Relatively small intensity of murmur and the color flow Doppler findings would differentiate this anomaly from PDA, but selective angiography was essential for definitive diagnosis. Medical records of cats diagnosed with congestive heart failure (CHF) seen at the University of Minnesota from January 1992 through October 2001 were reviewed. Of 271 cats with CHF, 29 (11%) had a history of corticosteroid administration (CSA) within the preceding 90 days. Six cats received more than one type of corticosteroid. Injectable corticosteroids administered were methylprednisolone acetate (MPA, 16), unspecified ''cortisone injection'' (2), triamcinolone acetonide (TCA, 6), and dexamethasone sodium phosphate (5). Oral corticosteroids administered were prednisone (7), and methylprednisolone (1). The reasons recorded for CSA were dermatologic (8), neurologic (3), gastrointestinal (4), allergic airway disease (8), recurrent sneezing (2), and non-specific illness (4). When compared to a case control population of sick cats without CHF, administration of injectable long-acting glucocorticoids was highly associated with CHF with an odds ratio of 28.8 (p ϭ 0.005).

In 18 cats treated with only a long-acting injectable glucocorticoid, mean Ϯ SD time from injection to onset of clinical signs of CHF was 4.4 Ϯ 2.7 days. Five cats presented in CHF within 1 day of receiving MPA or TCA. Five cats treated with only oral prednisone developed signs of CHF after 8 to 85 days (34.4 Ϯ 30.7). Of 28 cats where echocardiography was performed, 19 (68%) had evidence of left ventricular (LV) posterior wall and/or interventricular septal concentric hypertrophy.

Six cats that received MPA did not survive to discharge after admission for CHF. The acute mortality rate was similar to that for cats with CHF not associated with CSA. However, median survival time was 439 days for cats with a history of CSA, compared to 65 days for cats with CHF without a history of CSA (p ϭ 0.0006). (See figure) Of 10 cats that died later, only 5 deaths were attributable to signs associated with cardiac disease. Follow-up echo-cardiography was performed in 11 cats with previous hypertrophy. In 9 of these cats the LV hypertrophy resolved and medication was discontinued without recurrence of CHF. Corticosteroid-associated LV hypertrophy and CHF appears to be reversible in many cases, and has a better prognosis than CHF not associated with corticosteroids. Medical records of cats diagnosed with arterial thromboembolism (ATE) at the University of Minnesota from January 1992 through October 2001 were reviewed. Cats (n ϭ 44) that survived an acute episode of ATE were included. Long-term anticoagulant used was either low-dose aspirin (LDA ϭ 5 mg PO q3d, n ϭ 24), high-dose aspirin (HDA ϭ 40-162 mg PO q 2-4d, n ϭ 18), or none (n ϭ 2).

Median survival time (MST) for the 44 cats was 117 days. Eleven (25%) cats had 17 ATE recurrences, 9 of which were fatal. Two cats with left atrial thrombi (identified with the initial ATE) had long-term survival (275, 590 days) without any embolic episodes. Five (11%) cats developed limb problems [necrosis requiring amputation (2), necrosis requiring wound management (2), limb contracture (1)].

Cats with congestive heart failure during the initial episode had significantly shorter long-term survival than cats without congestive heart failure (CHF MST: 77 days, no-CHF MST:223 days, p ϭ 0.016, see figure) .

There was no significant difference in survival or recurrence between cats receiving HDA and cats receiving LDA. Adverse effects of aspirin were uncommon at both doses, but less frequent and milder for LDA. Comparison of survival curves and recurrence rates in cats receiving aspirin vs. published data from cats receiving warfarin (Moore et al 2000*) indicated no difference in survival or rate of recurrence with warfarin vs. aspirin. LDA is an inexpensive and safe option for thromboprophylaxis that appears to be at least as effective as HDA and warfarin. A high prevalence of low-grade, left-basilar ejection murmurs is recognized in boxer dogs. These murmurs are believed to be related to increased aortic ejection velocity (AV) consequent to mild aortic stenosis or undefined flow disturbances in the left ventricular outflow tract (LVOT).

Most reports suggest that a peak AV exceeding 2.0 to 2.25 m/s is abnormal. However, the relationship between increased AV and the soft ejection murmurs often heard in boxers has not been established conclusively. The objective of this retrospective study was to report AV and auscultatory findings in a large group of healthy boxers without overt structural evidence of LVOT obstruction.

There were 228 mature boxers evaluated by history, auscultation, 2D and Doppler echocardiography. Two independent examiners auscultated each dog, and a consensus was reached regarding the presence or grade of any cardiac murmur. A right parasternal, long-axis 2D echocardiogram was used to examine the LVOT for overt structural lesions. The maximal instantaneous AV was obtained from the subcostal position by continuous wave Doppler studies in unsedated dogs. Dogs were excluded from subsequent analysis for any of the following reasons: previously diagnosed systemic disorder; therapy with beta adrenergic antagonists or L-thyroxine; a loud (Ͼ3/6) murmur; or a structural lesion of the LVOT identified by 2D echocardiography.

A total of 201 boxers met the inclusion criteria. Of these, 88 dogs (44%) had no cardiac murmur (Group I); whereas, 113 (56%) manifested a soft (grade 1-3/6), left-basilar ejection murmur (Group II). The median AV for all dogs was 1.91 m/s (range: 1.31-4.02 m/s). The median AV in Group I dogs was 1.72 m/ s (1.31-2.38 m/s). The median AV in Group II dogs was 2.11 m/s (1.57-4.02 m/s), which was higher than in Group I dogs (pϽ0.001, Mann-Whitney). The AV exceeded 2 m/s in 11 (13%) of the Group I dogs, compared to 73 (65%) of the Group II dogs. Auscultation of a murmur was 86.9% sensitive, but only 65.8% specific, for the identification of AV Ͼ2 m/s. There was a non-significant trend for higher AV in male dogs compared to females (median of 1.96 vs. 1.87 m/s, p ϭ 0.1; Mann-Whitney). The AV for all dogs was correlated weakly to body weight (r ϭ 0.2; p ϭ 0.03), but not to age (p ϭ 0.06).

These data demonstrate that AV is more likely to be higher in dogs with soft ejection murmurs compared to dogs without obvious murmurs. Furthermore, in this population auscultation was a sensitive screening test for the identification of dogs with AV Ͼ2 m/s. Limitations of this study include single-plane echocardiographic evaluation of the LVOT, the potential for varying autonomic tone impacting results, and exclusion of pulmonary artery flow from data analysis. Computerized ECG recordings have been used lately for ECG acquisition, and the values are compared to published reference values obtained in strip recordings. To evaluate possible differences in measurement accuracy among different recording displays, ECG tracings were obtained from 20 healthy cats in right lateral recumbency using a computerized (method I) and a conventional, onechannel strip ECG machine (method II), without changing electrode positioning. Measurements were made on lead II by three observers and included P wave and QRS duration and amplitude, PR and QT interval duration, T wave amplitude, heart rate/ R-R interval and the main electrical axis was determined. For method I measurements were obtained on screen and on ECG printed on paper (at 50mm/s) and for method II, on the actual ECG strip. A caliper was used for measurements made on paper and the values considered to the fourth of a millimeter. Comparisons were made among methods and observers. Differences in measurements among method displays were significant for PR and QT intervals, P and QRS duration. There was also inter-observer difference for on-screen measurements. These differences demonstrate that standards for computerized electrocardiograms are needed. Arrhythmogenic cardiomyopathy in the boxer dog is typically characterized by ventricular tachyarrhythmias, syncope and sudden death. Ambulatory electrocardiography (AECG) is frequently used as a method for screening at risk dogs, but risk factors for the development of syncope have not been well defined. The objective of this study was to evaluate AECGs from boxers with ventricular tachyarrhythmias and syncope to identify important prognostic criteria.

Dogs were prospectively evaluated for a history of the presence or absence of syncope, and with a physical examination, echocardiogram, and 24-hour AECG. Analysis of AECG included the total number of VPCs/ 24 hours, maximum (MAX) and minimum (MIN) sinus heart rate (HR), and grading of the arrhythmia: 1 ϭ single VPCs, 2 ϭ bigeminy and/or trigeminy, 3 ϭ ventricular couplets and/or triplets, 4 ϭ R on T, 5 ϭ ventricular tachycardia. Dogs were selected for participation in the study if they had an echocardiogram demonstrating normal cardiac chamber sizes and function, absence of other concurrent cardiac or systemic disease, and an AECG with at least 50 VPCs/24 hours. For purposes of analysis, dogs were grouped based on history of syncope. A t-test was used to compare age, heart rate and VPCs/24 hours. A Mann Whitney was used to compare grade. A pϽ 0.05 was considered as significant.

A total of 139 boxers were evaluated. Thirty-four (16 female (F), 18 male (M)) dogs were syncopal. The median age of syncopal dogs was 8 years (range 1-13). and 45 (36-74) beats per minute (bpm), respectively. Median number of VPCs/24 hours was 1423 (57-17,858) with a median grade of 4 (1-5).

Of the 105 dogs (63 F, 42 M) without syncope, the median age of dogs was 7 years (range 1-12). bpm. Median number of VPCs/24 hours was 373 (54-62,622) with a grade of 3 (1-5).

Syncopal dogs had a significantly greater number of VPCs/24 hours (p ϭ 0.008) and grade of arrhythmia (p ϭ 0.024), but not MAX or MIN HR. Age and gender were not significantly different between the two groups.

We conclude that the presence of syncope in the boxer dog with arrhythmogenic cardiomyopathy and normal cardiac function is related to number of VPCs/ 24 hours and grade of arrhythmia. However, significant variability exists with some asymptomatic boxers having thousands of VPCs/24 hours and a high grade of complexity. It is likely that other factors may also be involved in the development of syncope. Further evaluation of cardiac electrical activity during the syncopal episode is needed. Diltiazem is given to cats with HCM to improve diastolic performance. Cardizem is administered TID and Cardizem CD once daily. Dilacor XR (60mg) is often given once daily, but serum concentrations have not been reported. The purpose of this study was to determine serum diltiazem concentrations over a 24-hr period when cats were administered 60 and 30mg of Dilacor XR once daily.

Ten clinically healthy cats and 3 cats with HCM (1-13 yrs; 3.6-7.3kg) were administered Dilacor XR at 60mg orally for 7 days. Five of these cats later received 30mg once daily for 7 days. Blood samples were drawn 6, 12, 18 and 24 hrs post-pill administration on the 7th day. Serum diltiazem concentrations were determined by gas chromatography. The reported therapeutic window is 50-200 ng/ml.

All 13 cats receiving 60mg had therapeutic or supratherapeutic serum concentrations at 6, 12, and 18 hrs. 3 (23%) had subtherapeutic levels at 24-hrs. Four of 5 cats receiving 30mg had subtherapeutic levels at 24-hrs, while 1 had subtherapeutic levels at 18 hrs. The 60mg dosage produced mean (ϮSD, range) serum concentrations at 6, 12, 18 and 24 hrs of 787 (Ϯ488,71-1500), 583 (Ϯ394,100-1640), 286 (Ϯ180,35-650) and 196 (Ϯ232,5-920) ng/ml, respectively. The 30mg dosage produced mean (ϮSD, range) serum concentrations of 448 (Ϯ370,20-800), 445 (Ϯ157,84-680), 117 (Ϯ65,47-200) and 43 (Ϯ24,20-88) ng/ml, respectively. There was no correlation between dosages (mg/kg) and serum concentrations.

Both intercat and intracat serum concentrations produced by either 30mg or 60mg dosages were erratic. Dilacor XR at 60mg once daily usually produced therapeutic blood levels for 24hrs, but produced very high serum concentrations at 12 (11/13), 18 (7/13) and 24 (3/13) hrs. The 30mg dose produced supratherapeutic serum levels at 6 or 12hrs in 4/5 cats and subtherapeutic levels at 24hrs in 4/5 cats. Although we saw adverse effects (decreased appetite) in only 1 cat, lethargy, anorexia and GI disturbances are reported and may be due to high serum concentrations. Supratherapeutic levels may be a problem at either dosage, while subtherapeutic levels suggest that once daily administration of 30mg is inadequate. Also, breaking/cutting the tablets is difficult.

Israël, AT French, EM Welsh. Edinburgh University, Scotland.

The objective of the study was to evaluate if older dogs with persistent flow across the ductus arteriosus (PDA) should still undergo closure and to hypothesize why some animals still continue to deteriorate despite closure.

The case records of 21 dogs with left to right PDA that had reached 24 months of age before diagnosis were reviewed and those animals that had survived were requested to participate in a long-term follow-up study (n ϭ 11). Animals represented to the Cardiology clinic were fully assessed (including clinical examination, electrocardiography, thoracic radiography and Colour flow Doppler echocardiography).

The age range at the time of diagnosis was 24-108 months (mean 51, median 48). Fourteen dogs underwent surgical ligation of their PDA and two dogs had coil embolisation. Haemorrhage occurred in 3/14 animals, which were all less or equal to 3 years of age, and was fatal in only 1. After successful closure of the ductus (n ϭ 15), the clinical signs disappeared in all but one animal. On follow-up of these animals (n ϭ 9), there was echocardiographic evidence of left ventricular systolic (n ϭ 6) and diastolic dysfunction (n ϭ 7) in many. Development of mitral valve endocardiosis was a common feature (n ϭ 6). Time of survival after surgery ranged from 10-112 months (median 36 months) and age of death ranged from 46 months to 168 months (median 96 months). Three of the 5 non-treated animals had died. Two of these 3 dogs lived respectively 5 months and 60 months after surgery and died secondary to heart failure at the age of 101 and 114 months. The two alive non-treated dogs are currently 84 and 72 months old. From the 4 dogs that were in heart failure at initial presentation, 2 dogs had died and two were still alive.

In conclusion, older animals with PDA follow an individual course, independent of pre-existing heart failure. Irreversible left ventricular dysfunction is common, however it does not seem to affect the clinical course. Late closure does not appear to carry a higher risk in the hands of experienced surgeons and it relieves clinical signs. Outcome without intervention was less favorable. The accuracy of Doppler-derived blood flow velocity depends on the angle of incidence between the ultrasound beam and the direction of flow. Failure to align the ultrasound beam with blood flow results in under estimation of velocity. In dogs with subaortic stenosis, it is known that the subcostal transducer site provides higher left ventricular ejection velocities (LVEV) than does the apical site. LVEV obtained from the subcostal site in healthy dogs have not been reported; accordingly, we examined 42 healthy, random-source dogs to test the hypothesis that aortic velocities obtained from the subcostal (S) and apical transducer (A) sites differ.

In 38 unsedated dogs in which apical and subcostal studies were completed, high-pulsed repetition frequency (HPRF) and continuous-wave (CW) Doppler studies of the proximal aorta (Ao) and left ventricular outflow tract (LVOT) were performed using two-dimensional (2D) echocardiographic guidance. The principal sample volume of the HPRF beam was placed at the tips of the open aortic valve leaflets. The peak velocities (V p ) reported represent the average of five, not necessarily consecutive, cardiac cycles. Heart rate (HR) was calculated from the average of three RR intervals measured from a simultaneously recorded electrocardiogram. The effects of Doppler modality, transducer site, HR and body weight (BW) on peak aortic velocity were tested through the use of AN-OVA. P values less than 0.05 were considered to indicate significance.

The dogs weighed between 14.4 and 28.4 kg; 14 of them were male, they were of various breeds and their ages, unknown. As part of unrelated investigations, the dogs had been subject to variable numbers of training sessions intended to acclimate them to lateral restraint. 2D and color-Doppler echocardiographic studies were normal in all dogs. There was no statistical effect of Doppler modality, HR or BW on V p . The mean peak aortic velocities measured in m/ s (Ϯ standard error) were: S-Ao HPRF ϭ 1.45 (Ϯ0.03), A-Ao HPRF ϭ 1.39 (Ϯ0.03), S-LVOT CW ϭ 1.48 (Ϯ0.03), A-LVOT CW ϭ 1.39 (Ϯ0.03). Aortic velocities obtained using the subcostal site were significantly greater than those obtained from the cardiac apex (p ϭ 0.0001). Lower and upper limits of 95% reference intervals for aortic velocities by method of measurement and transducer location were as follows: S-Ao HPRF ϭ 1.07-1.83 m/s, S-LVOT CW ϭ 1.10-1.86 m/s. In a population of healthy dogs without cardiac murmurs, peak aortic velocities obtained from the subcostal site exceeded those obtained from the cardiac apex but did so only to a marginal degree.

Annika Linde, Nuala Summerfield, Evelyn Ivey, Matthew Johnston, Angela Keffer. University of Pennsylvania, Philadelphia, PA.

Heart disease has been anecdotally described in the chinchilla and, with increasing popularity as a pet, the demand for diagnostic work-up and treatment has increased. The goal of this study was to establish normal echocardiographic parameters for the chinchilla. Twenty normal adult chinchillas were included in the study with an equal distribution of males and females. All animals were anesthetized with isoflurane by mask. Standard echocardiographic views were used. Data for both genders are combined in the table below. No statistical significant difference was found between means of each echocardiographic parameter with the exemption of left atrial diameter. This study establishes normal reference values for echocardiography in the chinchilla.

RD Marshall, JS Rand. University of Queensland, Australia.

Insulin preparations currently available for the treatment of feline diabetes mellitus usually result in resolution of clinical signs, but provide only moderate to poor glycemic control, largely because of inadequate duration of insulin action, and in some cats, poor absorption. Recently the new synthetic analogue, insulin glargine, was approved for the treatment of type 1 and type 2 diabetes in humans. This report compared the kinetics and dynamics of once daily (SID) versus twice daily (BID) administration of glargine in normal cats. A double crossover study, with 2 days between tests, was performed in 6 healthy, neutered adult cats (3 male, 3 female). Serial plasma insulin (insulin) and plasma glucose concentrations (glucose) were measured following SC administration of glargine SID (0.5U/kg bodyweight) or BID (0.25U/kg bodyweight, q 12hrs).

Glucose concentration decreased significantly (pϽ0.05) below baseline following administration of glargine SID and BID. There was no significant difference in time to onset of action for SID and BID administration, and no significant difference in nadir glucose or time to reach nadir glucose for SID and BID. Time for glucose to return to baseline was significantly longer for BID than SID. There was no significant difference between the area under the 24hr glucose curve for SID and BID administration.

Peak insulin concentration and time to reach peak insulin were not statistically different for administration of glargine SID and BID. The time for insulin to return to baseline was significantly longer for BID than for SID. There was no significant difference between the area under the 24hr insulin curve for SID and BID.

In summary, while once daily administration of insulin glargine at 0.5U/kg provided a significant blood glucose lowering effect, a longer effect was achieved by administering glargine at 0.25U/kg BID. In 1/3 of cats treated BID, mean blood glucose remained significantly suppressed at 24hrs, indicating a carryover effect of glargine to the next day. A study using diabetic cats is required to compare these dosing regimes, and to fully evaluate the potential of insulin glargine for treatment of diabetes mellitus in cats.

A DIABETIC CAT FOLLOWING TREATMENT WITH INSULIN GLARGINE. RD Marshall 1,2 , JS Rand 2 , MN Gunew 1 . 1 Creek Road Cat Clinic, 2 University of Queensland (Australia).

A 9 year old neutered female 3.5kg Burmese cat was diagnosed with diabetes mellitus and treated with Caninsulin (porcine lente). Initial clinical response was good, and body weight increased to 4.5 kg. After 12 months, insulin was changed to Monotard (human lente) because of inadequate glycemic control, presumed from inadequate duration of insulin action. This was characterized by PU/PD, weight loss, marked hyperglycemia, and hypoglycemic episodes when the insulin dose was increased. There were no signs of hyperadrenocorticism or acromegaly, and insulin dose averaged 0.5-0.75U/kg bid.

Twenty-two months after initial diagnosis with diabetes, and 2 weeks after treatment with enrofloxacin for a bacterial urinary tract infection, the cat was diagnosed with urinary candidiasis (Canidida glabrata). Fluconazole (50mg/cat PO bid for 10wks) was begun and insulin changed to Ultratard (human ultralente). Urine was culture negative after 3 weeks of fluconazole, but was positive at 4 weeks and remained culture positive until the end of 10 weeks of fluconazole therapy. Diabetic control remained poor on ultralente with weight loss (to 4.1 kg), PU/PD (SG 1.024), and marked glycosuria and hyperglycemia. Amphotericin B (2mg) was instilled directly into the bladder under general anaesthesia and fluconazole therapy reinstituted. Urine was negative on sediment exam for 4 days after treatment, but on day 5 there was recurrence of candidiasis. Fluconazole was withdrawn and insulin changed to glargine (2 U/cat bid), a new recombinant human synthetic analog with long duration of action. Good glycemic and clinical control was achieved with resolution of polyuria (SG 1.060), polydipsia and weight gain (to 4.7 kg). Urine was culture negative after 2 weeks of glargine therapy and has remained culture negative for 5 months.

This is the first reported use of insulin glargine in a diabetic cat and the first report of successful resolution of urinary candidiasis due to C. glabrata in a diabetic cat. Good glycemic control and minimization of glycosuria appear to be important factors in the successful management of urinary tract candidiasis. Glargine resulted in superior glycemic control compared to lente and ultralente, resulting in reduced glycosuria and increased urine specific gravity. Nonsteroidal anti-inflammatory drugs alter thyroid function tests in dogs and man. Etodolac is a selective cyclo-oxygenase-II inhibitor. Nineteen healthy, mixed breed dogs with normal thyroid function based on normal serum thyroxine (T4) response to TSH stimulation were treated with etodolac at 15 mg/kg twice daily. Serum concentrations of T4, triiodothyronine (T3), free T4 (fT4) measured by radioimmunoassay following equilibrium dialysis, and canine thyrotropin (cTSH) measured by radioimmunometric assay, and plasma concentrations of albumin, and globulin were measured on day Ϫ7 and day 0 prior to initiation of etodolac administration, and on days 14 and 28 after initiating treatment. The mean of the two pretreatment basal hormone concentrations was used to compare to subsequent measurements. Statistical analysis was accomplished by repeated measures analysis of variance with post-hoc testing using Dunnett's test. All data are expressed as mean Ϯ SE. A P-value of Ͻ 0.05 was considered significant.

No side effects of etodolac administration were noted. The mean serum concentrations of T4 were 28 Ϯ 1.3, 27 Ϯ 1.6, and 27 Ϯ 2.0 nmol/L, serum T3 were 1.39 Ϯ 0.08, 1.20 Ϯ 0.12, and 1.49 Ϯ 0.15 nmol/L, serum fT4 were 24 Ϯ 1.5, 23 Ϯ 1.6, and 26 Ϯ 2.4 nmol/L, and serum cTSH were 0.24 Ϯ 0.03, 0.20 Ϯ 0.03, and 0.23 Ϯ 0.03 ng/ml before and on days 14 and 28 after beginning etodolac treatment, respectively. There was no significant effect of etodolac treatment on any of the thyroid function tests. The plasma albumin concentration was significantly decreased on both day 14 (3.04 Ϯ 0.09 g/dl) and day 28 (3.07 Ϯ 0.10 g/dl) of treatment compared to pretreatment (3.49 Ϯ 0.07). Plasma globulin concentration was also significantly decreased on day 14 (2.69 Ϯ 0.17 g/ dl) and day 28 (2.63 Ϯ 0.17 g/dl) compared with pretreatment (3.02 Ϯ 0.13 g/ dl). Etodolac administration to dogs for up to 28 days had no effect on thyroid function tests. This study was undertaken to assess the impact of dietary carbohydrate source on glucose and insulin concentrations following a test meal or a glucose challenge, and on food intake and body weight in overweight cats (Ͼ30% body fat as determined by DEXA), which were previously shown to have reduced glucose tolerance and insulin resistance.

Sixteen overweight cats were divided into 2 groups and randomly allocated one of 2 extruded diets formulated to contain similar starch content (33%) from different cereal sources (sorghum and corn versus rice). Meal response and glucose tolerance tests were performed before and after a 6-week weight-maintenance phase and after a further 8-week free-access feeding phase. Body weight was determined weekly and food intake was measured daily for the duration of the study.

During the weight maintenance phase, all cats maintained a constant body weight yet cats fed the sorghum/corn diet consumed (P Ͻ0.001) more energy than cats fed the rice diet. During the free-access phase, cats fed the rice diet had higher (P ϭ 0.001) food intake and gained significantly (P ϭ 0.001) more body weight than cats fed the sorghum/corn diet. After the weight maintenance phase, cats fed the rice diet had higher (P ϭ 0.01) incremental peak glucose concentrations after a test meal and tended to have larger area under the glucose curves and average glucose concentrations compared to the sorghum/corn diet. The modal time to peak glucose concentration after a test meal for rice occurred 8 hours before the sorghum/corn diet. After the free-access phase, total (P ϭ 0.004) and incremental (P ϭ 0.004) areas under the glucose curves and average glucose concentrations (P ϭ 0.006) following a glucose challenge were significantly increased for cats fed the rice diet but remained unchanged for cats on the sorghum/corn diet. Most insulin concentrations and derived indices tended to be higher for the rice diet following a test meal or a glucose challenge for both phases, although results generally did not reach statistical significance. During the free access phase, average insulin concentrations following a test meal were significantly (P ϭ 0.02) higher for the rice diet /compared to sorghum/corn diet.

We conclude that a sorghum and corn blend is a superior carbohydrate source than rice for overweight cats with glucose intolerance and reduced insulin sensitivity and may help minimize overeating and further weight gain. Many dogs with chronic illness have serum biochemistry abnormalities consistent with hyperadrenocorticism (HAC). Lymphoma (LSA) is a common chronic disease of dogs. The purpose of this project was to study adrenocortical screening test results in dogs with LSA to evaluate their specificity. Criteria for inclusion into the study included 1) a diagnosis of LSA, 2) expected survival time of 4-14 months, 3) no glucocorticoid treatment for longer than 4 weeks after the start of their LSA chemotherapy protocol, 4) no evidence of HAC, and 5) owner consent. Post-ACTH stimulation plasma cortisol concentrations (ACTHs) (reference range: 6-17 mcg/dl, ''borderline'' 17-22 mcg/dl, ''abnormal'' 22 mcg/ dl) , urine cortisol:creatinine (UC:CR) (reference range: Ͻ1ϫ10 Ϫ5 , ''borderline'' 1-1.35ϫ10 Ϫ5 , ''abnormal'' Ͼ1.35ϫ10 Ϫ5 ; urine was collected at home on day of evaluation), and maximal left adrenal width measurements (reference range: Յ0.74 cm, ''borderline'' 0.75-0.9 cm) were performed at the time of diagnosis of LSA prior to initiation of chemotherapy and at 4, 6, 8, 10, and 12 months after diagnosis or until loss of LSA remission or development of another disease.

Ten dogs met the criteria for inclusion (LSA stage IIIA-1, stage IVA-3, stage VA-2, stage IVB-1, stage VB-3). Forty-two ACTHs were performed; 1 was abnormal and 2 were borderline (range, 6.2-22.7 mcg/dl; median ϭ 11.4 mcg/ dl; mean ϭ 11.8 mcg/dl). The abnormal or borderline values were detected at various times prior to (1 dog) or during therapy (1 dog, 2 values), and all returned to normal on a subsequent test. Thirty-five maximal left adrenal width measurements were performed; 5 borderline measurements were detected prior to (2 dogs) and during (2 dogs, 3 values) therapy (range, 0.37-0.87 cm; median 0.65 cm, mean 0.66 cm). Thirty-six UC:CR were evaluated; 26 abnormal and 4 borderline values were detected prior to and throughout the treatment period with 9/10 dogs having at least one abnormal UC:CR (range, 0.5-21.7ϫ10 Ϫ5 ; median 4ϫ10 Ϫ5 , mean 5.4ϫ10 Ϫ5 ). Five dogs were removed from the study when their LSA was out of remission (range: 3-12 months post-diagnosis, mean ϭ 6.8 months, median ϭ 6 months); 1 dog was withdrawn at 9 months when an oral tumor was detected; 4 dogs were followed for 12 months.

These data suggest that in dogs with chronic illness, UC:CR is frequently abnormal and may not be a specific test for HAC. Alternatively, the UC:CR may be the most sensitive test for increases in cortisol secretion as a result of chronic illness. Conversely, post-ACTH stimulation cortisol and maximal left adrenal width measurements were almost always normal and may be more specific for HAC or less sensitive for demonstrating chronic increases in cortisol secretion. Hypothyroidism in dogs and hyperthyroidism in cats are two of the most frequently appreciated endocrinopathies in these species. Although various methods testing for thyroidal disease have been devised, measurement of total serum thyroxine (T4) levels are commonly used for the initial screening of thyroid status as well as for monitoring therapeutic responses to medical therapy of thyroid disorders. The radioimmunoassay (RIA) is considered to be the gold standard for performance of total T4 measurements. Performance of this test, however, typically requires samples be delivered to a commercial laboratory, increasing the time and costs associated with thyroid testing. The availability of an in-house method for determination of serum total T4 levels may be beneficial by providing immediately available, cost-effective results. The purpose of this study was to evaluate the accuracy of an enzyme-linked immunosorbent assay (ELISA) developed for in-clinic use.

Total serum T4 concentrations obtained using a commercially available ELISA (Snap T4 Test kit, IDEXX Laboratories, Inc.) were compared to values obtained by a validated RIA. Serum samples from 50 dogs and 50 cats submitted to a reference laboratory for total T4 measurement by RIA were used. Samples represented a wide range of total T4 concentrations based on RIA results and varied from Ͻ6 to 124 nmol/L. Results of ELISA and RIA testing were compared. Correlation coefficients, bias, precision, and clinical agreement were determined for canine and feline results.

Correlation coefficients between RIA and ELISA results for canine and feline samples were 0.84 and 0.59, respectively. A non-linear relationship was identified in both. Bias plots identified poor agreement between testing methods. Although an overall overestimation of ELISA T4 measurement was identified for both dogs and cats, both significant over-and under-estimations were observed for individual data points. Clinical discordancy between ELISA and RIA results occurred in 48-58% of canine results and 36-56% of feline results, depending on classification guidelines used. Clinically inappropriate decisions were suggested by ELISA results in 62% of canine and 50% of feline samples. Evaluation of test precision resulted in coefficient of variations of 18 and 28% in dog and cat samples, respectively.

Significant discrepancies between ELISA and RIA results were observed in this study. The in-house ELISA test kit evaluated cannot be recommended as an accurate method for determining total T4 levels in dogs and cats. Erythropoiesis is often compromised in patients with naturally-occurring chronic renal failure. We postulated that cats with surgically reduced renal mass would respond similarly and have a reduction in hematocrit as a consequence. A previous study of cats (unpublished data) demonstrated that hematocrit decreased from 38.05 Ϯ 0.98 (mean Ϯ SE) before surgery to 22.3 Ϯ 0.95 at 23 weeks after surgical reduction of renal mass. We examined postoperative hematocrit values within this interval to determine the pattern of change.

Twenty-five young adult male and female cats had renal mass reduced by approximately 11/12 with 2-stage surgical procedures. During stage 1 (day 0), selective ligation of branches of the renal artery was performed to infarct approximately 5/6 of the kidney. Stage 2 (right uninephrectomy) was performed on day 14. Following stage 2, cats were treated with amlodipine (0.625 mg/cat/ day) until day 42 to ameliorate systemic hypertension. Serum creatinine concentration was measured on days 28 and 56 to assess degree of renal dysfunction. Between days 28 and 112, a 1 mL sample of blood was obtained at 14 day intervals for a complete blood count.

Renal function was markedly altered by the surgical procedures. At day 28, serum creatinine was 5.2 Ϯ 1.67 and on day 56 values were 3.7 Ϯ 1. We conclude that this model of renal failure in the cat causes a significant reduction in hematocrit which reaches a plateau approximately 2 months after renal mass is reduced. For the next 2 months, there is relative stability in hematocrit values, with a slight increase. We speculate that these changes in hematocrit may parallel the quantity of renal tissue available for production of erythropoietin (EPO). Immediately postoperatively EPO production may be markedly diminished, followed by a relative increase with hypertrophy of residual nephrons. Detection of the antibodies on red blood cells (RBC) and platelets (PLT) is important for the diagnosis of immune-mediated hemolytic anemia (IMHA) and thrombocytopenia (IMT). Recently, flow cytometry (FCM) has been applied to identify immunoglobulin (Ig) and complement C3 bound to RBC and PLT. In this study, we examined the presence of Ig and C3 on RBC and PLT in dogs with IMHA, IMT and other immune-mediated disorders including pure red cell aplasia (PRCA), systemic lupus erythematosus (SLE) and polyarthritis.

Tests were performed on RBC and PLT from eight dogs with IMHA, two dogs with IMHA and IMT, one dog with PRCA, one dog with SLE and one dog with polyarthritis. Blood samples from ten healthy dogs were use as controls. RBC and PLT were isolated from sodium citrate anticoagulated blood and immediately treated with 0.1 mcg of prostaglandin E 1 per ml. Washed RBC and PLT were incubated with fluorescein isothiocyanate (FITC)-conjugated goat antibodies to canine IgG, IgM, IgA and C3. Each sample was analyzed by FCM equipped with a 488-nm argon laser and bandwidth filter of 530 Ϯ 30 nm to measure the green light emission of FITC.

Of eight dogs with IMHA, three dogs had IgG, three had IgG and IgM, and one had IgG, IgM and IgA on RBC. Two dogs with IMHA and IMT had IgG and IgM on both RBC and PLT. One dog with PRCA had no detectable Ig and C3 on RBC but the case had IgG on PLT, and the dog subsequently showed severe thrombocytopenia. One dog with SLE had no detectable Ig or C3 on RBC and PLT. One dog with polyarthritis had IgG on RBC, and IgG, IgM and IgA on PLT, and the dog subsequently showed mild thrombocytopenia.

In conclusion, it was shown that Ig attaching on the surface of RBC and PLT were demonstrated by FCM not only in IMHA and IMT but also in other immune-mediated disorders including PRCA and polyarthritis. Because of the high sensitivity of FCM on the detection of antibodies, FCM can be useful to detect an early stage of anemia and thrombocytopenia occurring in various immunemediated disorders. In 1998, a bovine hemoglobin based oxygen carrier, Oxyglobin, received FDA approval for use in anemic dogs. Since then it has also been administered to cats at the Veterinary Hospital of the University of Pennsylvania (VHUP) under offlabel use with owner consent when compatible blood was not available. The purpose of this retrospective study was to describe the initial experience with use of Oxyglobin in cats at VHUP, including patient condition, underlying disease, reason for administration of Oxyglobin, volume and rate of infusion, concurrent treatments, adverse events, and outcome. Medical records of 72 cats which had received Oxyglobin from June 1998-December 2000 were reviewed.

The most common presenting clinical signs and physical examination findings prior to Oxyglobin infusion were those associated with anemia, but vomiting (15), neurologic signs (2), and respiratory abnormalities (14) were also noted. Oxyglobin was given as a supportive measure in the treatment of anemia in 70 cats, most often due to lack of available, blood type and crossmatch compatible blood. Two cats with ischemic extremities received Oxyglobin in an attempt to improve tissue oxygenation without success. There were 80 separate Oxyglobin infusion events, and dose and rate varied markedly (mean dose per infusion was 14.6ml/kg and mean rate of administration was 4.8ml/kg/hr). Improvement was noted in 86% of cats evaluated and included rise in body temperature, whole blood [Hb] , and blood pressure, as well as increased appetite and activity. Adverse events noted in 44 cats after the infusion of Oxyglobin included pulmonary edema (n ϭ 8), pleural effusion (21), vomiting (4), and neurologic abnormalities (4); however, some of these problems could also be attributed to pre-existing conditions. The dose and rate of Oxyglobin infusion may have contributed to the development of pulmonary edema or pleural effusion. Mucous membrane discoloration and pigmenturia were expected transient events related to Oxyglobin.

Only 23 of 72 (32%) cats were discharged from the hospital. All of the 13 (18%) cats that died and 36 (50%) that were euthanized had severe underlying disease prior to receiving Oxyglobin. The general reason for euthanasia of 35 cats was based upon a poor prognosis associated with the cat's severe underlying disease and financial constraints. There was no evidence of direct hemoglobin toxicity on histopathologic review of 23 necropsies.

In conclusion, although administration of Oxyglobin may provide temporary support to anemic cats as much as dogs when compatible red blood cells are not available, controlled prospective studies need to be performed prior to recommending its use in cats. The purpose of this study is to investigate D-dimer concentrations in clinically healthy dogs, clinically ill dogs without thromboembolic disease, and dogs with known thromboembolic disease. Secondary objectives are to determine the sensitivity and specificity of D-dimer in detection of pathologic thromboembolic disease in the dog.

Group 1 consists of 30 clinically healthy dogs presenting for routine care. Group 2 represents a population of 60 clinically ill dogs without thromboembolic disease. This group is sub-divided into the categories: post-op orthopedic surgical procedures, congestive heart failure, renal failure, hepatic disease, and neoplastic disease. Group 3 represents dogs diagnosed with thromboembolic disease by supportive laboratory findings, blood gas determination, ultrasonography, MRI, contrast angiography, histopathology, necropsy or combinations of these tests. All dogs have the following tests as a minimum part of their coagulation profile: CBC, PT, APTT, FDPs, and plasma D-dimer titer. D-dimer results are as follows:

The sensitivity of D-dimer concentrations Ͼ500ng/dl for predicting thromboembolic disease was 100%; however, the specificity of D-dimer for thromboemboli at that concentration was 46%. The specificity of D-dimer concentrations Ͼ1000ng/dl to predict thromboembolic disease was 79% and Ͼ2000ng/dl was 100%. FDPs may be an insensitive indicator of thromboembolic disease. TEST IN DOGS WITH DISSEMINATED INTRAVASCULAR  COAGULATION, THROMBOEMBOLISM, D-dimer is formed as a result of plasmin degradation of cross-linked fibrin but not fibrinogen. High plasma D-dimer concentrations have been reported in humans and dogs with disseminated intravascular coagulation (DIC) and thromboembolic disease (TED). The purpose of this study was to evaluate a point-of-care immunochromatography test (AGEN, Brisbane, Australia) used for qualitative detection of D-dimer in canine whole blood or plasma in the following groups of dogs: healthy, DIC, other thromboembolic disease, and bleeding unrelated to DIC. Unlike D-dimer tests previously described in the veterinary literature, this test utilizes a canine specific monoclonal antibody, provides rapid results (positive or negative), and is simple enough to perform in a primary care facility. Results of the pointof-care test were also compared to a quantitative laboratory test (electroimmunoassay [EIA], AGEN) using the same canine monoclonal antibody.

Citrated blood samples were collected from 4 groups of dogs: healthy (n ϭ 7), DIC (8), TED (14), and internal bleeding due to trauma or coagulopathy (5). Inclusion criteria for the DIC group were clinical bleeding, presence of a primary disease process associated with DIC and 2 of the following: PT or PTT Ͼ25% prolonged compared to control, elevated fibrin degradation products, schistocytes, thrombocytopenia, and decreased antithrombin III activity. Inclusion in the TED group required the diagnosis of an underlying disorder associated with thromboembolism, and consistent clinical signs, laboratory test results, imaging studies or necropsy findings.

All healthy dogs (7/7) Renal insufficiency, behavior problems, or endocrine diseases can all cause polyuria and polydipsia (PU/PD) in horses. Diabetes insipidus (DI) is an endocrine cause of PU/PD that is termed central or neurogenic when due to a lack of vasopressin (antidiuretic hormone) production and secretion or nephrogenic when due to a lack of response of collecting ducts to vasopressin. Water deprivation and exogenous vasopressin administration are two diagnostic tests that can be used to differentiate neurogenic from nephrogenic DI. Recently, desmopressin acetate (DDAVPÒ ), a synthetic vasopressin analog, has been used for diagnosis and treatment of neurogenic DI in human and animal patients. The purpose of this study was to assess the antidiuretic effect of desmopressin acetate in six normal horses in both the euhydrated (Experiment 1) and hyperhydrated state (Experiment 2). Experiment 1 was performed to assess potential hemodynamic effects of desmopressin acetate (20 ug IV) and no changes in heart rate or indirect blood pressure or other adverse effects were observed. In experiment 2, hyperhydration was induced by nasogastric administration of 40 ml/kg of water q 12 hours for 3 days. One hour after the last intubation of water, either desmopressin acetate (20 ug, equivalent in antidiuretic activity to 80 IU of vasopressin) or saline (control) was administered IV and urine and blood samples were collected for 8 hours. Total urine production over the 8 hour study period was 14.2 Ϯ 1.7 L with the control treatment compared to 2.8 Ϯ 0.5 L after administration of desmopressin acetate (pϽ0.01). Further, both urine specific gravity and urine osmolality increased after administration of desmopressin acetate and values were greater (pϽ0.01) than for the control treatment. In conclusion, these data demonstrate that intravenous administration of 20 ug of desmopressin acetate is both a safe and useful diagnostic tool for evaluation of horses with DI.

ADRENOCEPTOR-MEDIATED SIGNAL TRANSDUCTION PATHWAYS. Westropp, JL, Buffington, CAT.

We previously have reported increased plasma concentrations of norepinephrine (NE) in cats with feline interstitial cystitis (FIC) compared to healthy (H) cats. Alpha-2 adrenoceptors ( 2-AR) can function as autoreceptors to inhibit the release of NE. Activated 2-AR, through inhibitory G-proteins (G i ), inhibit adenylyl cyclase (AC) and production of cAMP, an important second messenger in cell signal transduction pathways. The pontine locus coeruleus (LC) contains a dense population of NE-containing neurons bearing 2-AR; we have previously reported increased tyrosine hydroxylase immunoreactivity in this region in cats with FIC. LC cell membrane preparations were prepared from 2 FIC and 2 H cats to investigate the effects of FIC on 2-AR-mediated signal transduction by measuring [ 32 ]P-cAMP formation from [ 32 P]ATP. After measuring basal cAMP production, GTP, forskolin (a direct stimulator of AC), mastoparan (a direct G iprotein stimulator), and medetomidine (a selective 2-AR agonist) were used to test G i and AC activity. Membrane samples were assayed in triplicate, and results expressed as pmol of [ 32 P]cAMP formed per mg protein per 15 minutes. Results were compared using unpaired 2-tailed t-tests (using Welch's correction) and are reported as meanϮsem; pϽ0.05 was considered significant.

Basal The reduced basal and effector-stimulated [ 32 P]cAMP production indicates decreased AC activity in FIC membranes. Moreover, the lack of response to mastoparan suggests the presence of hypofunctional Gi proteins. The absence of an effect of medetomidine could have resulted from decreased binding to the receptor (due to decreased receptor numbers or affinity), or to reduced coupling to the G i protein complex.

Overall, these results suggest multiple abnormalities in the 2-AR-mediated signal transduction pathway in cats with FIC. Abnormalities of this pathway have been identified in many diseases of humans, including fibromyalgia, panic disorder, and migraine headaches, all of which are comorbid conditions in human patients with interstitial cystitis. Feline nephroliths comprise a small number of the uroliths seen in veterinary medicine; however, the complications seen with nephroliths can make clinical management difficult. Traditional modalities of nephrolith management have consisted of surgical and/or medical management. However, surgical intervention and medical management are not without complications and/or poor success rate. Extracorporeal shock-wave lithotripsy (ESWL) has been widely and safely used in human medicine for the treatment of nephrolithiasis. Studies in human and dogs report intrarenal hemorrhage resulting from ESWL, but minimal decrease in glomerular filtration rate (GFR). The purpose of this pilot study was to evaluate the safety of extracorporeal shock-wave lithotripsy in cats.

A group of 4 adult cats was entered into the study. The health status of each cat was screened with a physical exam, CBC, serum chemistry profile, FELV/ FIV test, urinalysis, and abdominal ultrasound. Glomerular filtration rate was evaluated before ESWL using nuclear scintigraphy with 99m Tc-DTPA. ESWL was performed using a Dornier MFL-5000 Lithotripter. General anesthesia was used during ESWL, and each cat received a dose of 2000 shock waves to the left kidney generated at a voltage of 24 kV. Post-ESWL evaluation consisted of an abdominal ultrasound 24 hours and fourteen days post treatment to look for any evidence of renal hematoma formation or change in gross renal morphology. In addition, 14 days post treatment a urinalysis, and nuclear scintigraphy to evaluate GFR, were performed.

The group of cats had normal pre-screening labwork with no evidence of renal insufficiency. The pre-treatment and post-treatment GFR for each cat were within the normal range. The mean pre-treatment GFR was 2.46 Ϯ 0.82 ml/min/kg, and the mean post treatment GFR was 2.52 Ϯ 0.75 ml/min/kg. A paired t-test was used to evaluate for a significant difference between pre and post GFR results. No significant difference was found between the pre and post GFR (p ϭ 0.92). Ultrasound evaluation of the kidneys and abdomen at 24 hours and 14 days post ESWL treatment did not reveal any gross changes or abnormalities.

The results of this study show that ESWL can be safely performed in cats with no significant change in kidney function or morphology. Future studies of ESWL in clinical cats with nephroliths are needed to investigate ESWL as an effective treatment modality for nephroliths. The risk factor analysis technique was first developed by Robertson et al in 1978 to distinguish between human calcium oxalate (CaOx) and calcium phosphate stone-formers (SF) and normal individuals (N), and involved the use of overlapping frequency distributions of urinary factors found to be significantly different between the two groups. From these a set of risk curves were developed, and combined using Bayes theorem, to give an overall measure of the relative risk of CaOx formation in a given individual.

A similar risk factor analysis technique was applied to a group of 17 CaOx SF dogs and age-, breed-and sex-matched controls (N), where urinary calcium and oxalate were significantly higher in the SF group, and uric acid concentrations tended to be higher when compared to the N group. The relative risk factors of these three analytes were combined into an overall measure of the probability of being a CaOx stone-former (P SF ), where P SF lay between 0 (low risk) and 1 (high risk). P SF was compared with urinary relative supersaturation (RSS) as a technique for discriminating between the two groups (table).

Both RSS and P SF were significantly different between the N and SF groups, with good correlation between the two techniques (r 2 ϭ 61.6%, PϽ0.0001). However, RSS more clearly discriminated between the two groups than P SF , suggesting that RSS was the better tool for assessing the risk of CaOx stone formation in dogs. This study compared the urine composition of calcium oxalate (CaOx) SF dogs with N dogs on a 'free-choice' (FC) or standardised diet (S)*, to establish the influence of diet on the risk of CaOx stone-formation.

17 dogs with confirmed CaOx urolithiasis were age-, breed-and sex-matched against N dogs, living in a UK home environment. Urine was collected over 48 hours and frozen immediately after voiding from all dogs on a FC diet, and after 1 month (SF & N), and 1 year (SF only) on Diet S. Urinary CaOx RSS was measured by previously described methods. Results were compared using nonparametric methods, where PϽ0.05 was significant (table). A different superscript letter in a row indicates a significant difference within a group; different superscript numbers indicates a significant difference between groups.

Increased urinary calcium and oxalate concentrations of the SF group on a FC diet, resulted in a significantly higher CaOx RSS (greater than the estimated formation product; RSSϳ12 Determination of urolith mineral composition is critical for effective management of urolithiasis. Prediction of a urolith's mineral composition can be made using several variables; however, none of these variables allows accurate in vivo determination of urolith mineral composition. Uroliths are distinguishable from surrounding tissues using noncontrast computed tomography. Tissue density can be quantified using peak beam attenuation measurements (Hounsfield units; HU). HUs have been used to predict composition of uroliths retrieved from people in vitro, and to differentiate calcium oxalate and urate uroliths in vivo in people. This study was designed to establish in vitro reference ranges for three types of uroliths retrieved from dogs. 66 uroliths (22 urate, U; 21 calcium oxalate, O; 14 struvite, S; 9 mixed, M) retrieved from dogs were placed in a phantom array. Uroliths were scanned at 120 kVp/400 mAs and 80 kVp/400 mAs. The region-of-interest (ROI) for HU calculation was determined using two techniques. In ROI 1, an area was hand-drawn just within the urolith border, using the largest cross-sectional image obtained. In ROI 2, an oval encompassing 2/3 to 3/4 of the urolith area was manually centered over the largest urolith image. Reference ranges for each urolith type were calculated by log-transforming HUs, calculating 99.99% confidence intervals using the t-distribution, and back-transforming the results. Intervals were adjusted to prevent overlap of ranges. Two blinded investigators used these ranges to predict the mineral composition of each urolith using both ROI techniques at either kVp; positive and negative predictive values (PPV, NPV) were determined.

HUs for urolith types of pure composition were statistically different (Kruskal-Wallis, pϽ0.0083) using both ROI techniques at either kVp. S were not statistically different from M (pϾ0.0083); the M group was excluded from future calculations. Calculated reference ranges were: PPV and NPV were higher at 80 kVp using either ROI technique. At 80 kVp, PPV and NPV for O and U were greater than or equal to 89% using either ROI technique. PPV for S was greater than 70% with either ROI technique at 80 kVp; NPV was greater than 90%.

HUs can be used to differentiate urolith mineral composition in vitro. Further studies are needed to determine the predictive value of HUs in vivo. Vitamin D metabolism is altered in hypercalcemia. Serum 1,25-dihydroxycholecalciferol (1,25-[OH]2-D3) is expected to be low or normal with functional hypercalcemia. In primary hyperparathyroidism (PHPTH) it might be high due to increased 1 a-hydroxylation of 25-hydroxycholecalciferol (25-OH-D3). 25-OH-D3 might be influenced by decreased nutritional intake or increased urinary loss of vitamin D. We recently reported 1,25-[OH]2-D3 and 25-OH-D3 in 50 dogs with renal failure to be low compared to control dogs. The aim of this study was to evaluate the two vitamin D metabolites in dogs with hypercalcemia and to compare their concentrations with those of healthy control dogs.

Causes responsible for hypercalcemia (total serum calcium Ͼ 3.01 mmol/l) were, lymphoma (LSA) in 9 dogs, PHPTH in 5 dogs and chronic renal failure (CRF) in 9 dogs. The diagnosis of LSA was based on histology or cytology of affected tissue. The diagnosis of PHPTH was based on hypercalcemia with normal to low phosphorus together with a histological diagnosis of parathyroid neoplasia or a parathyroid mass detected by ultrasound of the neck. The diagnosis of CRF was based on a history of renal azotemia, on ultrasound results suggesting chronic renal disease or on histopathology of the kidney. The 24 control dogs had normal complete blood counts and serum chemistry panels. Serum 1,25-[OH]2-D3 was measured by a radioimmunoassay test and serum 25-OH-D3 was measured by a protein binding assay.

1,25-[OH]2-D3 ranged from 26 to 309 pmol/l (median 62.0) in LSA dogs from 61 to 398 pmol/l (median 248.0) in PHPTH dogs, from 28 to 310 pmol/l (median 84.0) in CRF dogs and from 60 to 239 pmol/l (median 157.5) in control dogs. 1,25-[OH]2-D3 was significantly higher in PHPTH dogs and significantly lower in LSA and CNI dogs compared to control dogs. Although not significant 1,25-[OH]2-D3 appeared higher in PHPTH dogs compared to LSA and CNI dogs. 25-OH-D3 ranged from 64 to 291 nmol/l (median 107.0) in LSA dogs from 66 to 298 nmol/ l (median 91.0) in PHPTH dogs from 35 to 184 nmol/l (median 70.0) in CRF dogs and from 48 to 350 nmol/l (median 306.5) in control dogs. 25-OH-D3 was significantly lower in LSA, PHPTH and CRF dogs compared to control dogs. There was no significant difference between LSA, PHPTH and CRF dogs.

As expected 1,25-[OH]2-D3 is higher than normal in PHPTH and lower than normal in LSA and CRF. However as the difference between PHPTH dogs and the LSA and CRF dogs is not significant it might only be of minor value for the differentiation of reasons for hypercalcemia. Different causes of hypercalcemia affect 25-OH-D3 in the same way. Decreased intake or urinary loss in CRF might be the only reasons for low 25-OH-D3 in hypercalcemic dogs.

Urinary tract inflammation is thought to be a cause of proteinuria in dogs. In one study of urinary tract inflammation experimentally induced by either cystotomy or infection, all dogs developed proteinuria with urine protein:creatinine ratios Ͼ1.5. However, when proteinuria was evaluated in dogs presenting to a veterinary teaching hospital that had abnormal urine microscopic examinations, 66% had qualitative proteinuria (sulfosalicylic acid) but only 35% had urine protein:creatinine ratios Ͼ1.0. Methods used in these studies measure not only urine albumin (UAlb) but also urine globulin. A method to quantify UAlb in dogs has recently become available. The purpose of this study was to determine the effect of urinary tract inflammation on UAlb concentrations.

Urine samples were obtained from dogs presenting for routine health screening, elective procedures and evaluation of health problems at the Veterinary Teaching Hospitals at Colorado State University and North Carolina State University and from dogs in a colony of soft coated wheaten terriers (SCWT) and SCWT ϫ beagle crosses (SCWTx). SCWT and SCWTx with known protein-losing nephropathy were excluded. UAlb concentrations were measured in samples with pyuria (Ͼ5 WBC/hpf) using a canine albumin specific competitive ELISA.

Of 70 samples with pyuria, 40 (57%) had negligible UAlb concentrations (Ͻ1 mg/dl); whereas microalbuminuria (1-30 mg/dl) and macroalbuminuria (Ͼ30 mg/dl) were present in 25 (36%) and 5 (7%) of the samples respectively. Bacteriuria (trace to 4ϩ) and hematuria (Ͼ5 RBC/hpf) were concurrent microscopic abnormalities in 26 and 17 dogs, respectively. UAlb concentrations were significantly higher in urine samples with bacteriuria ( Some studies in human beings have suggested that serum cystatin C (SCys) concentration may be a better indicator of declining glomerular filtration rate (GFR) than serum creatinine concentration (SCr). Others have validated a commercially available immunoturbidimetric method for SCys measurement in dogs. During a previous study, we performed serial GFR estimations in dogs with progressive renal disease due to X-linked hereditary nephropathy (XLHN). The goal of this study was to examine the correlation of SCys with GFR in these dogs and to compare it with the correlation of SCr with GFR in the same dogs.

Twenty-three male dogs were studied; 18 had XLHN and 5 were unaffected littermates. Starting at 2 months of age, GFR was estimated monthly from renal scintigrams obtained after IV injection of 99mTc-diethylenetriaminepentaacetic acid (99mTc-DTPA). Serum obtained in weeks when GFR was estimated was stored at Ϫ80ЊC until assayed for SCys by the previously validated method. Contemporary SCr values, which had been determined weekly, were also available for comparisons.

All dogs had normal GFRs until 6 months of age, but GFR declined thereafter in dogs with XLHN. Dogs were 8 to 15 (median, 10.5) months old at their study end-point, which was when SCr Ն 5.0 mg/dl occurred. Thus, to focus on the portion of the dogs' lives when GFR was declining, only results obtained for the 152 occasions (4-11; median, 6/dog) when GFR was estimated in dogs Ն 6 months old were used for data analysis. Serum was available for SCys assay from 97 of these occasions (1-9; median, 4/dog).

Linear regression was used to assess the correlation between GFR and the reciprocals of SCr and SCys (i.e., 1/SCr and 1/SCys, respectively). Significant (p Ͻ 0.0001) correlation was found between 1/SCr and GFR, both overall (r ϭ 0.911) and for the occasions when SCys was also determined (r ϭ 0.903). Significant (p Ͻ 0.0001) correlation was also found between 1/SCys and GFR, but the strength of correlation (r ϭ 0.655) was much less than the strength of correlation found between 1/SCr and GFR on the same occasions.

We conclude that SCr was superior to SCys as an index of GFR under the conditions of this study, which involved serial evaluation of declining renal function in individual dogs throughout the course of their disease. In this setting, SCr was a useful index of GFR. The lesser correlation of SCys with GFR was unexpected. Use of SCys as an index of GFR in dogs will require further study before it can be recommended. Hypertension and chronic renal insufficiency are frequently observed in cats. Therapy is complicated by potential effects of both hypertension and antihypertensive agents on renal function. Other target organs, such as the eye and brain, may be adversely impacted by systemic hypertension. We hypothesized that the calcium channel antagonist, amlodipine besylate, would lower systemic arterial blood pressure (BP) and reduce the prevalence of complications in a feline model of hypertensive renal insufficiency without adversely impacting renal function. We studied 20 cats using the remnant kidney model. Cats were treated orally with amlodipine besylate (n ϭ 10) or placebo (n ϭ 10) once daily. Systolic blood pressure (BP) was measured by implanted radiotelemetric devices continuously for 36 days.

Compared with values for normal cats, systolic BP, diastolic BP, and mean BP were markedly elevated (PϽ0.05) in placebo treated cats. In contrast, cats treated with amlodipine exhibited significant (PϽ0.05) reductions in BP compared with those receiving placebo. Albuminuria, but not urine/protein-to-creatinine ratio, was directly related to BP in hypertensive cats (PϽ0.05). The prevalence of ocular lesions attributable to systemic hypertension in cats receiving placebo (70%) was greater (PϽ0.05) than that observed in cats receiving amlodipine (20%). Renal function, as assessed by serum creatinine and BUN determinations, was not adversely affected by treatment with amlodipine.

Our results demonstrate that the calcium channel antagonist, amlodipine besylate, normalized BP in cats with coexistent systemic hypertension and renal insufficiency and prophylactically reduced the frequency of ocular lesions. We previously have reported that resting plasma catecholamine concentrations ([CCE] ) are higher in cats with interstitial cystitis (IC) than in healthy cats, whereas responses to corticotrophin releasing factor appear to be similar. To investigate this observation further, we compared the effects of stress on [CCE] and urine cortisol excretion (cortisol:creatinine-C:Cr) in normal cats and cats with IC. Twelve normal and twelve IC cats were food deprived overnight, and confined and transported to our laboratory for determination of [CCE] by HPLC (Day 1). The procedure was repeated on all cats on day 3, and on 6 cats per group on days 8 to assess habituation to the stressor. Data were assessed by ANOVA with post-hoc and t-tests as appropriate; based on our previous studies a 1-tailed PՅ0.05 was considered significant (sig).

All plasma [CCE] (dihydroxyphenylalanine (DOPA), dopamine (DA), dihydroxyphenylacetic acid (DOPAC), norepinephrine (NE), epinephrine (E), and dihydroxyphenylglocol (DHPG)) were higher in IC than in normal cats. DOPA was sig. increased in IC cats at all times, DHPG at all but day 3. All [CCE] but E (0.06) were sig. increased on Day 8; P values were DOPA Ϫ 0.008, DA Ϫ 0.02, DOPAC Ϫ 0.01, NE Ϫ 0.03, DHPG Ϫ 0.04.

In contrast, no differences between groups were identified in C:Cr at any time. The [CCE] were increased in both groups by the stressor. Moreover, major differences in [CCE] were found at day 8, at a time when most other stress-related parameters had declined toward baseline values. The marked difference in DOPA suggests a stress-induced increase in activity of tyrosine hydroxylase, the ratelimiting step in CCE synthesis, and is in accord with previously published findings of increased tyrosine hydroxylase immunoreacivity in the locus coeruleus of cats with IC. In contrast, no effects on C:Cr were identified, suggesting an uncoupling of these two parameters of the stress response. Similar findings have been reported in human patients with chronic pelvic pain, panic disorder, and post-traumatic stress disorder. Bacterial urinary tract infections occur commonly in dogs. Diagnosis is made by aerobic culture of urine collected by cystocentesis. Despite appropriate technique, many practitioners have difficulty acquiring reproducible results. The purpose of this study was to determine if reliable aerobic bacterial culture results could be obtained by using incandescent lighting for incubation of samples.

Twenty-nine canine and three feline urine samples were included in this study; animals had not received antimicrobials. Specimens were obtained via cystocentesis and were cultured within six hours. Of the samples, 22/32 were positive for bacteriuria based on microscopic examination of sediment and 10/32 were negative. All samples were submitted to microbiology for routine aerobic culture and inoculated onto four blood agar plates as follows: plate one was incubated at 35C, plate two was incubated under incandescent lighting at a distance of 11cm to maintain a temperature of 35C at the plate surface, and plates three and four were incubated at ambient temperature for 24 hours and up to 72 hours, respectively. A fifth plate was inoculated as a control by a microbiology technician. If growth was present at 24 hours, plates were packaged on ice and transported for 24 hours. Plate four was monitored up to 72 hours for microbial colonies. All 32 samples were submitted for microbial identification. Colony growth of blood agar plates incubated under incandescent lighting was similar to those plates incubated routinely, however, growth on plates maintained at room temperature required at least 48 hours for microbial colonies to be adequate enough to properly quantify.

All samples were evaluated with urine test strips containing the leukocyte esterase and nitrite pads. Of those with confirmed positive urine cultures, 6/15 were positive for leukocyte esterase and 2/15 were positive for nitrite. Of those with confirmed negative urine cultures, 2/15 were positive for leukocyte esterase and none were positive for nitrite.

Sensitivity and specificity results were 100%. Positive and negative predictive values were 100%. We conclude that utilization of incandescent lighting to incubate blood agar plates may be reliable for obtaining aerobic bacterial urine cultures in private practice.

OF CATS EXPOSED TO A NOVEL CAGE ENVIRONMENT. Hague, DW, Buffington, CAT.

We have observed that cats with feline interstitial cystitis (FIC) do not appear to respond to their environment as healthy (H) cats do. To investigate this observation further, we compared the behavioral responses of FIC and healthy cats to a novel environment. A person unfamiliar to the cats moved 11 FIC and 9 healthy cats (one FIC and one H at a time) into individual stainless steel cages in a room that was different from their home room for videotape recording of their behavior. Each cage contained food and water bowls, a litter box, a covered bed, a ball, and two shelves. The cat's behavior was videotaped for four hours. Weights of each cat's food and litter pan were recorded prior to and after the videotape period, and episodes of urination and defecation were noted for each cat. After each videotaping, the cages and included items were thoroughly washed before the next pair of cats was moved into them.

All videotapes were viewed in 15-second intervals. In cases where more than one activity occurred during a 15-second period, only the activity that occupied the majority of the time was recorded. Recorded behaviors included sitting (in box, on floor, or on shelf), resting recumbent with eyes closed (in box, on floor, or on shelf), lying with eyes open (in box, on floor, or on shelf), activity, hiding in the covered bed, hiding with head out, genital grooming, non-genital grooming, use of box, interest in box, eating, drinking, and time out of camera view. Minutes engaged in each behavior were compared between groups using a 2tailed t-test.

During the period of videotaping, FIC cats engaged in more eating and drinking behaviors (p ϭ 0.05) and grooming behaviors (p ϭ 0.02) than did H cats. The mean food intake was 16 grams for FIC vs. 3.8 grams for H cats (p ϭ 0.06). Additionally, three FIC, but no H, cats urinated during the four-hour period.

These results show that FIC cats tend to engage in more displacement activities, such as eating, drinking, and self-grooming, than do H cats upon exposure to a novel environment. These behaviors may result from exaggerated arousal associated with placement into the environment. Sodium bicarbonate (NaHCO 3 ) is often administered to Standardbred or Thoroughbred race horses before racing. Detection of administration of NaHCO 3 is problematic because it is based on measurement of serum concentrations of total carbon dioxide or bicarbonate and these substances are normally present in variable concentrations in horse serum. The purpose of this study was to determine the effect of administration of commercial NaHCO 3 on 13 C enrichment of bicarbonate in horse serum.

NaHCO 3 (Baking soda, Arm & Hammer, Princeton, NJ) was administered to 7 fit Thoroughbred geldings (2-7 years, 426 to 541 kg). Horses had been fed the same source diet and had access to the same source water for 3 months before the study. NaHCO 3 was administered (450 g per horse) in 3 l of tap water by nasogastric intubation. Horses had feed withheld for 14 hours before and for 6 hours after NaHCO 3 administration. Horses had free access to water at all times. Blood samples were collected immediately before and 2, 4, 6 and 24 hours after NaHCO 3 administration. Blood was allowed to clot and then refrigerated before collection of serum. Serum total carbon dioxide concentration (tCO 2 ) was measured using a commercial analyzer (Beckman EL-ISE). 13 C enrichment of serum bicarbonate was estimated by measurement of 13 C enrichment of CO 2 released by acidification of serum. Data were analyzed by one way repeated measures analysis of variance and Dunnett's post-test.

Administered NaHCO 3 had a 13 C enrichment significantly different (P Ͻ 0.05) from that of serum of untreated horses (Ϫ18.53 Ϯ 0.26 and Ϫ21.00 Ϯ 0.05 delta 13 C, mean Ϯ SE, respectively). NaHCO 3 administration increased (P Ͻ 0.05) serum tCO 2 concentration from pre-treatment values of 28.0 Ϯ 0.8 mmol/ l to 34.8 Ϯ 0.8, 36.8 Ϯ 0.3 and 36.6 Ϯ 0.5 at 2, 4 and 6 hours. Serum tCO 2 concentration at 24 hours (29.6 Ϯ 0.7) was not significantly different from values before treatment. 13 C enrichment of serum-derived CO 2 was only transiently and minimally affected by NaHCO 3 . 13 C values were Ϫ21.00 Ϯ 0.05 (delta 13C), Ϫ20.97 Ϯ 0.14*, Ϫ21.15 Ϯ 0.36, Ϫ21.45 Ϯ 0.07, and Ϫ21.50 Ϯ 0.07* (* ϭ P Ͻ 0.05 vs pretreatment) at pretreatment and 2, 4, 6 and 24 hours after treatment. There was no correlation (r ϭ Ϫ0.0133, p ϭ 0.946) between tCO 2 and 13 C enrichment. Changes in 13 C enrichment of individual horses were variable. We conclude that, under the conditions of this study, administration of NaHCO 3 cannot be detected by measurement of 13 C enrichment of serum-derived CO 2 . Studies in human medicine have shown a high incidence of hypomagnesemia in patients admitted to intensive care units, and hypomagnesemia has been associated with increased death rates. Supplementation with magnesium (Mg) is used routinely, and in some disorders has been shown to improve survival. The purpose of this study was to determine the prevalence of hypomagnesemia in horses admitted to the large animal clinic at NCSU, CVM; to determine whether or not there is an association between hypomagnesemia and organ system affected; to determine if hypomagnesemia affects clinical outcome; and to determine which other serum chemistries correlate with Mg.

Data were obtained for all horses admitted to the large animal clinic at NCSU, CVM between Jan 1999 and May 2001 by a database search. Only horses that had a blood chemistry panel analyzed within 24 hours after arrival were included in the study. If more than one panel was analyzed in a horse, only the first one was considered. Data were recorded for the serum chemistry panel results, age, breed, gender, number of hospitalization days, discharge status, organ system affected and diagnosis. A series of independent and multivariable logistic regression models were used to assess the potential association of demographic variables and the organ system affected with low magnesium values. Odds ratios and corresponding 95% confidence intervals calculated from the beta coefficient were evaluated to add variables to improve model validity or delete variables to improve model precision.

Of all horses admitted to the clinic 48.8% had Mg values below the normal reference range. Horses with low magnesium were more likely to be thoroughbreds (OR ϭ 1.62, 95% CI ϭ 1.06-2.48) and greater than 1 month of age (OR ϭ 1.43, 95% CI ϭ 1.25-1.63). Low magnesium was also more likely to be observed in horses that were admitted for gastrointestinal problems (OR ϭ 2.86, 95% CI ϭ 1.47-5.55) than horses admitted for other disorders. This study provides pharmacokinetic data on the distribution of ketoprofen in elephants. It also yielded oral absorption and bioavailability data following a single administration allowing for rational dose design in a clinical setting. Finally, the pharmacokinetic parameters obtained from these elephants were compared to the published data in domestic species to assist with establishing relationships for the proper extrapolation and treatment of other therapeutic agents in the elephant.

Five adult Asian elephants (Elephas maximus), housed at the Center for Elephant Conservation, Ringling Bros. and Barnum & Bailey Circus in Orlando, Florida were used in this study. The elephants were randomly assigned to 1 of 2 groups. The 1st group received a single intravenous (IV) dose of ketoprofen (2.2 mg/kg). The second group received ketoprofen orally (PO) at a dose of 2.2 mg/kg. Blood samples were collected prior to drug administration and at 0.17, 0.33, 0.67, 1, 1.5, 2, 3, 4, 5, 6, 8, 12, & 24 h post-administration. After a period of at least one week, the groups were switched in a cross-over design so that all 5 animals, at the completion of the study, had received both IV and PO ketoprofen. Plasma samples were assayed for ketoprofen enantiomers using liquid chromatography/mass spectrometry via chiral separation. Pharmacokinetic parameters were determined using non-compartmental analysis. Parameters calculated following the IV dose were: area under the plasma concentrations vs. time curve (AUC), area under the first moment curve (AUMC), mean residence time (MRT), volume of distribution (V d ), plasma clearance (Cl p ), elimination rate constant (k el ), and apparent terminal half-life (t ½ ). Following PO administration, the following parameters were determined: AUC, AUMC, MRT, mean absorption time (MAT), k el , t ½ , and F.

In all animals, both the IV and PO administrations were well tolerated. No adverse effects were noted following either route of ketoprofen administration. Azithromycin (AZ) is the first of a class of antibiotics classified as azalides. Six ball pythons (Python regius) were given a single dose of AZ at 10 mg/kg orally (PO) and intravenously (IV) in a cross-over design. Serial blood samples were collected for unchanged AZ quantitation and to determine, if possible, the structure and number of AZ metabolites circulating in plasma.

After a 4 month wash-out period, the snakes were given AZ PO as a single dose of 10 mg/kg for the study of tissue distribution and metabolism. Bile, liver, lung, kidney, and skin samples were analyzed for the metabolites identified in the plasma from the first experiment.

Unchanged AZ accounted for 80%, 68%, and 60% of the total material at 12, 24, and 48 h post administration in plasma. At 24 and 72 h post administration, AZ accounted for 70% of total AZ associated material in bile. Thirteen metabolites were identified in the plasma of all snakes at 12 h, independent of route of administration. Two of these metabolites, 3Ј-desamine-3-ene-azithromycin and 3Ј-N-nitroso-9a-N-desmethylazithromycin are unique to this species. Only 4 of the 13 metabolites were identifiable at 24 or 72 h in python bile. Compared to the dog and cat, a larger number of metabolites were identified in the ball python plasma. The percentage of unchanged AZ in bile is not different between the three species. Piroxicam (PIRO) is a non-steroidal anti-inflammatory drug that has been shown to have clinical value in the treatment of canine squamous cell, transitional cell, and mammary carcinomas. In addition, laboratory studies have proven PIRO's value as an anti-tumor and/or chemopreventative agent against a variety of epithelial tumors. Although PIRO is being used in cats and is frequently administered in combination with an H 2 receptor antagonist or other gastro-protective agent, neither the multiple-dose pharmacokinetics nor the safety of PIRO or cimetidine (CIM) has been evaluated in cats. The purpose of this study was to evaluate the multiple-dose pharmacokinetics and safety of PIRO alone and in combination with CIM in cats.

Seven healthy cats weighing Ͼ3.5kg each were included in this randomized crossover study. Cats were assigned to groups designated to receive CIM alone, PIRO alone, and PIRO combined with CIM. The cats were dosed orally with 15mg/kg of CIM every twelve hours and 0.3mg/kg of PIRO once daily for 10 days followed by at least a two-week washout period after which the groups were rotated in a randomized block design. Blood samples were collected utilizing vascular access ports on day 0 prior to the first dose and at 0.25, 0.5, 1, 2, 4, 6, 8, 12 hours post-administration on days 1 and at 0.25, 0.5, 1, 2, 4, 6, 8, 12, 24, and 48 hours post-administration on day 10. Additional one-hour postadministration samples were collected on days 3, 5, and 7. Blood samples were analyzed utilizing a liquid chromatography/mass spectroscopy assay. Pharmacokinetic parameters were determined using compartmental analysis. Endoscopic evaluation of the gastric mucosa was performed on days 5 and 10 following each treatment. Serum urea nitrogen (SUN), creatinine, alkaline phosphatase (ALP), and alanine transaminase (ALT) activities were evaluated on day 11.

In all samples evaluated, the harmonic mean t 1/2 of PIRO administered alone and combined with CIM was 10.8 and 9.58 hours (h), respectively. The mean C max for PIRO administration alone was 724ng/mL, and the mean T max was 2.1 h. For PIRO combined with CIM, the values were 571ng/mL and 2.1 h, respectively. The mean plasma clearances of PIRO alone and combined with CIM were 0.024 L/h/kg and 0.033L/h/kg, respectively. One cat receiving CIM alone had mild gastric erosions present on day 10. One cat receiving both drugs had mild erosions present on day 5. These lesions resolved by day 10. Two cats receiving both PIRO and CIM had mild erosions on day 10. Four cats receiving PIRO alone had moderate to severe erosions on day 10. The SUN, creatinine, ALP, and ALT activities on day 11 were within reference ranges for all cats.

ASSAY METHOD TO MEASURE CANINE WHOLE BLOOD CY-CLOSPORINE CONCENTRATIONS. PB Peterson, DM Boothe, G Inan, & GE Lees. Texas A&M Univ., College Station, TX.

Cyclosporine A (CsA) absorption and metabolism are highly variable, so therapeutic drug monitoring is often employed. High-pressure liquid chromatography (HPLC) is the gold standard method for assaying CsA, but HPLC is expensive and requires specialized training and facilities. The purpose of this study was to compare CsA concentrations in canine blood using HPLC and a simpler monoclonal radioimmunoassay (RIA) method.

Six healthy adult dogs and 6 adult heterozygous female dogs with X-linked hereditary nephritis were given a single 5mg/kg oral dose of the microemulsion formulation of CsA. Twenty-three serial whole blood samples were then collected in EDTA from each dog through an indwelling jugular catheter and subsequently analyzed for CsA concentration in duplicate using RIA (Cyclo-Trac SP) and HPLC.

Calibration curves for both assays were prepared by adding CsA to canine whole blood at 11 different concentrations from 33.5 to 1404.9 ng/mL. Precision and accuracy of both RIA and HPLC were determined by analyzing canine whole blood samples containing CsA at 33.5, 50.2, 100.4, 501.8, 1087.1, and 1505 .3 ng/mL.

Results were pooled and analyzed for correlation with a linear regression plot. Correlation between HPLC and RIA assay results using a first-order linear regression model was high (r 2 ϭ 0.970, PϽ0.001). Correlation was described by the linear equation [RIA value ϭ (1.074Ϯ0.015) HPLC value ϩ (38.8Ϯ7.58)].

Correlation between HPLC and RIA was good at clinically relevant drug concentrations, but RIA yielded greater CsA concentrations than those obtained by HPLC even after a single oral dose. The clinical significance of the RIA overestimating actual CsA concentration depends on therapeutic goals, but might be minimal because target drug concentrations that are commonly recommended for CsA therapy in dogs have a wide range. This is a multi-center trial to establish the clinical efficacy and safety of Buscopan (n-butylscopolammonium Br, 0.3 mg/kg body weight, single dose) for the alleviation of abdominal pain (colic) in horses. There were 22 investigators, with 225 qualified cases and 12 disqualifications. Treatments were assigned randomly (115 pla-cebo, 110 Buscopan). Investigators were masked with respect to treatment. Horses underwent a pretreatment physical examination that focused on parameters considered important in the evaluation of colic. Examinations were completed and a total colic score, based on clinical signs of colic, calculated at 5, 15 and 30 min following treatment administration. Following the 30-min examination, the investigators provided their clinical impression of the effectiveness of the test article. Continuous data were analyzed by repeated measures ANOVA using the PROC MIXED in SAS. Discrete data were subjected to repeated measures analysis using the Genmix model of SAS.

Of the horses used in the study, Quarter Horses, Arabians, and Thoroughbreds accounted for 40%, 14%, and 12% of the cases, respectively. Among the eligible cases, the majority of the horses treated were geldings (53%), followed by mares (39%) and stallions (8%). The range of body weights for animals from eligible cases was 300 to 1700 lb (136 to 771 kg), with a mean of 968 lb (439 kg). Ages ranged from 4 months to 35 years, with an average of 10.6 years. Together, spasmodic, flatulent, and mild impaction colic accounted for 82% of the tentative diagnoses provided by the investigators.

Total colic scores decreased throughout the 30-min post-treatment observation period for Buscopan and placebo. Scores were lower for the Buscopan-treated horses. Buscopan management was rated as a ''success'' by 88% of the investigators while 43% rated placebo a ''success.'' The beneficial effect of Buscopan was further buttressed by the horses' improved behavioral attitudes. A greater percentage of horses treated with Buscopan were rated as alert/calm at 15 and 30 min post-treatment, and a greater proportion of placebo-treated horses were nervous/restless. The parasympatholytic effect of Buscopan was evident in elevated heart rate at 5 and 15 min and its effects on smooth muscle evident in reduced intensity and frequency of borborygmi in the Buscopan group and not seen with placebo treatment. By 30 min the heart rate for the Buscopan group did not differ from that observed prior to treatment.

Buscopan clearly provided relief from the abdominal pain associated with the diagnosis of spasmodic, flatulent, and mild impaction colic. No adverse reactions were reported during this study. Therefore, the drug was both safe and effective. Soft-coated wheaten terriers (SCWT) commonly develop a syndrome of protein-losing enteropathy (PLE) and/or protein-losing nephropathy (PLN). PLE is associated with inflammatory bowel disease and is usually diagnosed at an earlier age (mean 4.5 yrs) than is PLN (mean 6 yrs). Fecal ␣ 1 -proteinase inhibitor (␣ 1 -PI) is a marker of enteric protein loss that has been used to screen SCWT for PLE. The purpose of this study was to determine the prevalence of increased fecal ␣ 1 -PI concentrations in a colony of SCWT with PLE/PLN that have been studied longitudinally as well as to determine the prevalence of increased fecal ␣ 1 -PI in SCWT in the general population.

The SCWT colony consists of 18 dogs (10 purebred SCWT, 8 SCWTxbeagles), 7 of which have overt PLE defined as increased fecal ␣ 1 -PI (Ͼ5.67 g/ g) with panhypoproteinemia (serum albumin and globulin Ͻ2.5 g/dl). Fecal ␣ 1 -PI concentrations of 3 consecutive defecations were measured in these dogs every 3 months, while dogs ranged from 0.08 to 7 yr of age (total # submissions, 356; median, 25%, 75% per dog, 19, 18, 20.8). All dogs had at least 1 sample in which either the maximum or mean of 3 samples was increased. However, those dogs developing overt PLE had significantly greater number of fecal ␣ 1 -PI concentrations in which either the mean or maximum of the 3 samples was Ͼ15 g/g (p Ͻ0.05, Wilcoxon test). Both the mean and maximum fecal ␣ 1 -PI concentrations of the 3 consecutive defecations were significantly greater at 1 yr of age when compared to values obtained at 3 yr of age (pϽ0.05, paired T test).

A total of 193 fecal ␣ 1 -PI submissions were available from 157 SCWT (median, 25%, 75% per dog, 1, 1, 1) in the general population (49 M, 94 F, 14 unspecified; 0.33 to 12 yr of age). Of the 49 dogs for which clinicopathologic data was available, PLE, PLN or PLE/PLN was diagnosed in 2, 5 and 5 dogs, respectively. Fecal ␣ 1 -PI was increased in 82 (52%) of the dogs, 32 of which had concentrations Ͼ15 g/g. Of the 7 dogs with PLE, 6 had increased fecal ␣ 1 -PI concentrations. One of these 6 dogs had concentrations Ͼ15 ug/g.

Increased fecal ␣ 1 -PI is common in SCWT. Concentrations may decrease with advancing age and fecal ␣ 1 -PI Ͼ15ug/g may be predictive for panhypoproteinemia later in life. More studies are needed to determine if these observations are representative of the population at large. An enzyme-linked immunosorbent assay (ELISA) for measurement of canine pancreatic lipase immunoreactivity (cPLI) has been developed and validated in our laboratory. Serum cPLI has been shown to be highly specific for exocrine pancreatic function and has also been shown to be highly sensitive for the diagnosis of canine pancreatitis. A reference range of 2.2 to 102.1 g/L was established for cPLI in serum. The goal of this study was to determine the stability of cPLI concentrations in serum samples stored at varying temperatures over a period of 21 days and evaluate differences in cPLI concentrations between serum and plasma samples from the same dog.

Eight serum samples submitted to the Gastrointestinal Laboratory at Texas A&M University were randomly selected. Each serum sample was divided into twenty 30 l aliquots. Five aliquots of each sample were stored at different temperatures: room temperature (approximately 24ЊC), 4ЊC, Ϫ20ЊC, and Ϫ80ЊC. Aliquots from each temperature group were analyzed for cPLI on days 0, 3, 7, 14, and 21. Statistical analysis was performed by repeated measures ANOVA. Additionally, cPLI concentrations were analyzed in paired serum and plasma samples from 30 randomly selected dogs submitted to the Clinical Pathology Laboratory at Texas A&M University College of Veterinary Medicine. Serum and plasma cPLI concentrations were compared using a paired t-test.

Mean cPLI concentrations from samples kept at either room temperature, refrigerated, or frozen did not vary significantly from day 0 to day 21 (pϾ0.05).

Mean (ϮSD) serum cPLI concentration was 64.1 (Ϯ70.9) g/L with a range of 0.0 to 257.0 g/L. Mean plasma cPLI concentration was 62.1 (Ϯ69.8) g/L with a range of 0.0 to 255.6 g/L. There was no significant difference between serum cPLI and plasma cPLI concentrations (p ϭ 0.3803).

In conclusion, serum cPLI concentrations measured in serum samples kept at room temperature, refrigerated, or frozen at Ϫ20ЊC or Ϫ80ЊC are stable for at least 21 days. Serum and plasma cPLI concentrations do not differ significantly. Small intestinal bacterial overgrowth (SIBO) in dogs is challenging to diagnose and treat. This disorder is usually chronic and persistent, requiring longterm or repeated therapy for optimum therapeutic response. Most clinicians rely on antibiotics for management of SIBO. Dietary modification using fructo-oligosaccharides (FOS) and subsequent alteration of the small intestinal flora has been proposed by some investigators. The aim of this study was to compare the clinical efficacy of a broad-spectrum antibiotic medication and a diet containing high concentrations of FOS in the management of dogs with clinical histories and laboratory findings suggestive of SIBO.

Client-owned pet dogs (n ϭ 17) were recruited from accessions to the GI Laboratory. Criteria for entry into the trial included clinical signs and history consistent with SIBO, an abnormally high serum folate concentration (Ͼ14 g/ L) or abnormally low serum cobalamin concentration (Ͻ230 pg/L), and normal serum TLI concentration. Dogs were randomized to receive either antibiotic therapy for 30 days (tylosin, 15 mg/kg PO BID, n ϭ 10) or a diet containing 1% FOS (Eukanuba Low-Residue Adult Canine, Dry) for 60 days (n ϭ 7). Dogs were weighed weekly and clinical signs recorded daily. Sera were obtained at days 0, 30, and 60. Serum concentrations of cobalamin, folate, unconjugated bile acids (SUBA), and unconjugated cholic acid (SUCA) were measured. Changes in concentrations of these analytes and body weight were analyzed by repeated measures ANOVA followed by Dunnett's multiple comparison test.

Dogs receiving antibiotic therapy had a significant decrease in SUBA (11,759Ϯ8,157 vs 994Ϯ672 nmol/L (meanϮSE)) and SUCA (5,518Ϯ3,650 vs 604Ϯ371 nmol/L) at day 30 (both pϽ0.05). Body weights of dogs receiving antibiotics significantly increased at day 60 (32.98Ϯ4.5 vs 30.8Ϯ4.7 kg, pϽ0.01). Dogs receiving FOS had significantly increased cobalamin at day 60 (542Ϯ125 pg/L) when compared to day 0 (397Ϯ79 pg/L, pϽ0.01), no other changes achieved statistical significance in this group.

We conclude that the use of tylosin is associated with a decline in SUBA and SUCA at 30 days, and increased body weight at 60 days, in dogs suspected of having SIBO. Dietary FOS supplementation was associated with increasing serum cobalamin concentration, suggesting a potentially beneficial alteration in small intestinal flora. Short chain fatty acids (SCFA) influence several aspects of gastrointestinal function: fluid and electrolyte absorption, energy metabolism, and enterocyte differentiation and proliferation. We have previously shown that SCFA stimulate contraction of canine colonic longitudinal smooth muscle. In this study, we examined the effect of SCFA on colonic smooth muscle contraction in the cat, a species at increased risk for colonic motility disorders. We studied the effects of SCFA on longitudinal and circular smooth muscle, from ascending, transverse, and descending colon of kittens and adult cats.

Feline longitudinal and circular colonic smooth muscle strips were suspended in physiologic (HEPES) buffer solution, attached to isometric force transducers, and set to optimal muscle length (L o ) with acetylcholine (ACh; 10 Ϫ4 M) or potassium chloride (KCl; 80 mM). Muscles were then incubated with individual short chain fatty acids, i.e., sodium acetate, butyrate, or propionate (1-100 mM). SCFA responses were compared to those obtained with ACh. The role of extracellular calcium in the SCFA responses was investigated by incubating muscle strips with nifedipine (1 micromolar) or verapamil (1 micromolar) prior to treatment with SCFA.

Acetate, butyrate, and propionate elicited isometric stress responses (0.25-1.98 ϫ 10 4 N/m 2 ) in longitudinal, but not circular, smooth muscle from the proximal and distal colon of adult cats. Maximal responses were attained at 50-75 mM butyrate, propionate, and acetate concentrations. Maximal butyrate and propionate responses were 29 and 19% of the maximal ACh response (10 Ϫ4 M), respectively. Sodium acetate was least effective (Ͻ5%) in stimulating contractile responses. Nifedipine and verapamil abolished the acetate, butyrate, and propionate responses. Responses in kittens were similar to those observed in adult cats but smaller (25-48%) in amplitude.

We conclude that SCFA stimulate longitudinal colonic smooth muscle contractions in kittens and adult cats. SCFA-induced contractions involve activation of calcium influx. These in vitro findings may account for some of the effects of fiber on feline colonic motility. A balanced interaction between the commensal intestinal flora and the immune system is central to maintaining gastrointestinal health. Dysregulation of the intestinal mucosal response may be a crucial early event in the etiopathogenesis of inflammatory bowel disease (IBD). Differences between the systemic antibody responses in normal and IBD cats could reflect the state of tolerance and the potential bacterial strains responsible for perpetuation of the inflammatory environment. As an initial part of this investigation, a study was designed to compare IgG responses to commensal GIT bacteria in normal cats and cats with IBD.

A library of aerobic and anaerobic bacteria has been established from feline duodenal aspirates. Bacteria were purity plated, cultured in FAB broth, assessed for growth and used whole and disrupted to coat 96 well plates. Plates were incubated with serum from normal cats and cats with a diagnosis of idiopathic IBD in a standard phosphatase ELISA. Plates were read at dual wavelength (405 & 495nm). The diagnosis of idiopathic IBD was based on compatible clinical signs (weight loss, vomiting, diarrhea), a full work-up to exclude underlying causes, and histopathological evidence of inflammation on intestinal biopsy.

All cats showed a broad response to a variety of bacteria. In general, responses to Gram-negative aerobes were greater than to Gram-positive aerobes that were greater than anaerobes in normal cats. Cats with IBD tended to show greater responses to a variety of bacteria particularly anaerobes when compared to normal cats. Total IgG levels were similar in both groups.

Cats were not equally reactive to all components of their intestinal flora and a difference in the pattern and magnitude of reactivity was noted in the sera of cats with IBD. If certain bacterial strains are more likely to induce an inflammatory response then modification of the bacterial flora may reduce the chance of IBD developing and provide therapeutic options in its management.

PANCREATIC INSUFFICIENCY (1997 INSUFFICIENCY ( -2000 . AL Williams, JM Steiner, CG Ruaux, and DA Williams. Gastrointestinal Laboratory, Texas A&M University, College Station, TX.

Exocrine pancreatic insufficiency (EPI) is a syndrome that occurs due to insufficient synthesis and secretion of digestive enzymes by pancreatic acinar tissue. EPI was previously thought to be a rare disorder in cats, due in part to the difficulty of definitive diagnosis. Availability of the serum feline trypsin-like immunoreactivity (fTLI) test has simplified the definitive diagnosis of this condition. In cats with EPI malabsorption and maldigestion may occur, leading to certain clinical signs. The goal of this project was to define clinical manifestations and outcome of EPI in cats.

Clinical manifestations and outcome of EPI in cats were assessed by a postal survey. Diagnosis was based on the measurement of serum fTLI, with concentrations Յ 8 g/L being considered diagnostic for EPI. Surveys were sent to veterinarians who had submitted feline serum samples to the Gastrointestinal Laboratory at Texas A&M University from 1997 through 2000 with serum fTLI concentrations Յ 8 g/L. Information was requested about the observed clinical signs during the disease history of these cats. Data were compared with previously published data on dogs with EPI using contingency table analysis.

A total of 52 replies were received. Cats diagnosed with EPI showed one or more of the following clinical signs: weight loss (84.6%), diarrhea (65.4%), voluminous stools (44.2%), vomiting (42.3%), polyphagia (42.3%), greasy soiling of the haircoat (13.5%), flatulence (7.7%), borborygmus (3.9%), and bleeding tendencies (3.9%). In three (5.8%) of the cats weight loss was the only documented clinical sign.

It was found that voluminous stools (44% vs. 95%, pϽ0.001) and flatulence (8% vs. 88%, pϽ0.001) were reported significantly less frequently in cats than in dogs. There was no significant difference in the observance of diarrhea (34% vs 22%, p ϭ 0.158) and vomiting between cats and dogs. There was also no significant difference in mortality rate between cats and dogs with EPI (31% vs 26%, p ϭ 0.215).

We conclude that cats with EPI show a variety of clinical signs including weight loss, diarrhea, voluminous stools, vomiting, and polyphagia. Some cats with EPI exhibit greasy soiling of the haircoat, flatulence, borborygmus, and bleeding tendencies. Cats with EPI exhibit voluminous stools and flatulence less frequently than dogs. Cats with EPI may show no clinical signs other than weight loss at the time of diagnosis. Serum feline trypsin-like immunoreactivity (fTLI) has been shown to be clinically useful for the diagnosis of exocrine pancreatic insufficiency and pancreatitis in cats. Currently, an in-house ELISA is used for the measurement of serum fTLI concentrations. This assay detects trypsinogen, the main form of trypsin in serum of healthy individuals, trypsin, and also a portion of trypsin bound to trypsin inhibitors. Feline trypsinogen and trypsin are small cationic proteins with a molecular mass of approximately 24 kDa. Therefore, both proteins are filtered in the kidney and renal function may affect renal excretion of these proteins and in turn may influence serum fTLI concentrations. The goal of this project was to examine the influence of experimentally induced chronic renal failure (CRF) on serum fTLI concentrations.

Serum fTLI concentrations were measured in serum samples from 20 cats with CRF induced by subtotal nephrectomy. Serum fTLI concentrations in cats with CRF were compared with those from 63 clinically healthy cats using a twosided t-test (GraphPad Prism software package).

Mean ϮSD serum fTLI concentration in 20 cats with CRF was 117.8 Ϯ63.6 g/L with a range of 44.0 to 273.0 g/L and was significantly different from serum fTLI concentrations in clinically healthy cats (mean ϮSD; 46.9 Ϯ17.5 g/L; range: 15.7 to 123.6 g/L; pϽ0.0001). Serum fTLI concentrations in cats with CRF were above the upper limit of the reference range (82 g/L) in 13/20 cats (65%) and above the currently recommended cut-off value for a diagnosis of feline pancreatitis of 100 g/L in 11/20 cats (55%).

We conclude that decreased renal function has a significant effect on serum fTLI concentrations and cats with renal failure may have increased serum fTLI concentrations. Therefore, evaluation of serum fTLI concentrations in azotemic cats may lead to a false positive diagnosis of feline pancreatitis. Medical records for 16 horses with histological evidence of myodegeneration of the masseter muscles were examined. These cases were unique because the muscle damage was most extensive in the muscles of mastication, and most of them were adult horses. In half of these horses only the masseter muscle was involved, or the masticatory muscles were more severely involved. The horses ranged in age from 6 months to 22 years, and 10 of the 16 were over 10 years old.

The most frequently found presenting signs in the horses examined while they were still alive included: difficulty eating or opening their mouth, weight loss, difficulty moving, or noticeable masseter atrophy. Serum concentrations of muscle enzymes were elevated in all of the horses. Whole blood and/or liver selenium levels were less than the reference range in some of the horses, and vitamin E levels were also decreased in some but not all of the horses. Unfortunately some of the cases had been given a selenium or Vitamin E injection prior to presentation.

Lesions varied depending on the stage of the disease and consisted of swelling and discoloration, or muscle atrophy and fibrosis. Histologically, the muscle alterations ranged from acute degenerative changes, to subacute with regenerative changes concurrent with ongoing degeneration, to chronic with fibrotic replacement of muscle tissue. All animals had masticatory muscle changes but, in some the lesions were widespread, and a few had myocardial lesions as well.

It was concluded that adult horses as well as foals could have myodegenerative diseases, and the muscles of mastication seemed more likely to be affected. Selenium or vitamin E could be involved. Obesity is a relatively common occurrence in pets. Although obesity has been associated with increased risk for some health problems in dogs, reduction of these risks subsequent to obesity therapy has yet to be demonstrated. One reason for this circumstance is the difficulty commonly encountered in sustaining weight loss after it has been achieved. To begin to investigate alternative approaches to weight loss maintenance, we compared the effects of a conventional (C) approach to weight loss used at The Ohio State University Veterinary Hospital with a weight management program (I) that incorporated interventions intended to effect behavioral changes in clients with obese dogs. The study objectives were to: 1) determine if weight loss during a 6-month period would be different in dogs enrolled in the I compared to the C program; and 2) determine if participation in the I program would result in better maintenance of the lost weight during the subsequent 18 months.

Fifty-two healthy obese male and female dogs of a variety of breeds from 1 to 10 years of age with a body condition score (BCS) of 8/9 or 9/9 were enrolled; 27 in C and 25 in I. After a thorough diet history to estimate daily energy intake, the energy intake reduction to induce a 1% weight loss per week was calculated using the following equation: Kcal reduction ϭ BWLB X 0.01X 3500 kcal/LB/ 7days/week. All dogs were fed Ralston Purina Veterinary Diets OM Canine Formula Dry food; 10% of the estimated daily energy intake was permitted to consist of treats.

As of January 2002, 26 (14 C, 12 I) of the 52 dogs enrolled have completed the entire 24 months of the study. In addition, 20 dogs (11 C, 9 I) have left the study for various reasons, and 6 dogs will complete the study in March, 2002. Of the 26 dogs that have completed the study, dogs in both treatment groups achieved similar weight losses at 6 months. The average weight loss was 15.0Ϯ4.6% for C group and 14.6Ϯ7.1% for I group. At 24 months, the overall average weight loss for C dogs was 18.2Ϯ6.8%, compared with 14.2Ϯ12.3% in the I group. The BCS changed from 8.3Ϯ0.5 to 6.1Ϯ1.0 in the C and 8.2Ϯ0.3 to 5.9Ϯ0.9 in the I group. If sustained in the final analysis, these results suggests that a weight loss of ϳ15% (BCS of ϳ6/9) can be sustained in dogs, and the more intensive behavioral intervention approach is not necessary to achieve these results. Asthma in humans as well as in cats is characterized by reversible bronchoconstriction and chronic airway inflammation. Reactive oxygen species (ROS) may contribute to the pathogenesis and perpetuation of the disease. In the present case control study the antioxidant/oxidant balance in plasma from cats with spontaneous feline asthma was determined. In 7 cats diagnosis of asthma was established by history, clinical signs, X-rays, a dominance of eosinophils in bronchoalveolar lavage, and detection of bronchoconstriction during an asthmatic attack by use of conscious unrestrained barometric whole body plethysmography. 10 healthy cats served as controls. Plasma lipid-soluble and water-soluble antioxidants were determined independently by photochemiluminescence. Plasma ceruloplasmin, an acute phase protein and indicator of inflammation was evaluated by electron spin resonance. To elucidate the role of ROS, malondialdehyde, a determinant of lipid peroxidation due to oxidative stress was measured in plasma as thiobarbituric acid reactive substances (TBARS). Plasma lipid-soluble antioxidants were significantly lower in asthmatics compared to controls (7.76Ϯ5.05 vs 13.52 Ϯ 2.3 arbitrary units (AU), mean ϮSD; p ϭ 0.012), whereas watersoluble antioxidant content was slightly, but not significantly higher in the asthmatic group (64.82Ϯ36 vs 37.89Ϯ18 AU, p ϭ 0.055). Patients had a significantly higher plasma ceruloplasmin (10304.25Ϯ4002.49 vs 4538.77Ϯ1813.49 AU, p ϭ 0.0002). TBARS (697.42 Ϯ156 vs 527.50Ϯ192.53 fluorescence units; p ϭ 0.001) were significantly increased in asthma cats vs controls. The more severely affected patients showed a tendency for higher TBARS and ceruloplasmin levels. Our data suggest that similar to humans, cats with asthma have a significant plasma oxidant/antioxidant imbalance reflecting increased oxidative stress. Low lipid-soluble antioxidant levels may be explained by an increased utilization, whereas elevations in acute phase proteins and water-soluble antioxidants may be the consequence of upregulation in response to continuing oxidative stress. Longitudinal studies evaluating the influence of antiinflammatory therapy on the oxidant-antioxidant balance in feline asthma are underway. In the future, measurement of the antioxidant-oxidant status may contribute to objective monitoring of therapeutic effects beyond resolution of clinical signs. It might serve as a surrogate marker for the control of the inflammatory process in the airways. The purpose of this study was to summarize characteristics of metaldehyde toxicoses in dogs as well as to discuss demographic aspects of poisoning in North America.

Retrospective data was compiled from reports to the ASPCA Animal Poison Control Center between January 1999 to May 2001.

Ninety-two cases of suspected metaldehyde toxicoses in dogs (ranging in age from 2 months to 14 years) were evaluated. Eighty-eight dogs (95.7%) exhibited clinical signs relating to the CNS. The incidence of clinical signs reported was tremors (60 reports-65%), seizures (30 reports-33%), hyperthermia (19 reports-21%), tachypnea (15 reports-16%) salivation (13 reports-14%), and ataxia (13 reports-14%.) Other signs reported were vomiting, tachycardia, hyperesthesia, and depression. In 53% of the cases, clinical signs developed within 30 minutes to 3 hours post ingestion. Twenty eight percent of cases involved dogs in California. The next most common states/provinces were New York (8.70%), Ontario (8.7%), and Texas (5.7%). Fifty-nine percent of cases were received between May and July. Treatment of metaldehyde toxicoses in dogs includes standard decontamination procedures, such as emesis and activated charcoal, seizure control, and supportive care. Intravenous methocarbamol have been recommended to treat metaldehyde-induced tremors and seizures in dogs with consistent success in APCC case records. Monitoring serum concentration of acute phase proteins (APPs) provides valuable information on the host's innate immune response. Haptoglobin (Hp) is a moderate APP present in low concentrations in normal canine serum which reaches peak concentrations five days after surgery or trauma. Raised Hp concentrations are a result of increased production and secretion from hepatocytes following stimulation by pro-inflammatory cytokines such as interleukin-6. Previous studies have demonstrated low concentrations of Hp in advanced liver pathology and increased concentrations following prednisolone administration.

Hp was measured by a biochemical assay (Eckersall et al, 1999, Comp Haem Inter, 9:117-124) in canine serum samples submitted to the University of Glasgow Veterinary School Diagnostic Services (n ϭ 146, median ϭ 4.38, range ϭ 0-31.8). Results were grouped as follows: Group 1, Hyperadrenocorticism n ϭ 33, Group 2, dogs receiving steroids n ϭ 24, Group 3, diabetes mellitus n ϭ 16, Group 4, immune-mediated disease n ϭ 36, Group 5, neoplasia n ϭ 24. Results were compared statistically to the Hp concentration in samples from healthy animals (Hp ϭ 0-3g/l). HpϾ10g/l was considered to be evidence of major infection or inflammatory disease incident.

In group 1, (n ϭ 33, median ϭ 4.9g/l) 7 samples were within the reference range and none had HpϾ10g/l. In group 2, (n ϭ 24, median ϭ 8.52) only 2 samples were within reference range, while 9 cases had Hp values Ͼ 10g/l. In group 3, (n ϭ 16, median ϭ 4.14g/l) highest concentrations of Hp were found in cases with diabetes ketoacidosis, while only 2 cases were within reference range. In group 4, (n ϭ 36, median ϭ 4.98g/l) only 6 cases were within reference range. In group 5, (n ϭ 24, median ϭ 5.6g/l) 14 cases had HpϾ3g/l and of these 7 had HpϾ10g/l. Preliminary data suggests that monitoring Hp concentrations will prove to be a valuable aid in identifying and quantifying the inflammatory response in canine patients. Further studies are warranted to evaluate the use of Hp in monitoring progression of disease and response to therapy. The purpose of the study reported here was to characterize the immunologic responses of elk to brucellosis and tuberculosis vaccines. Previous studies have suggested that, unlike bison, cattle, and swine, vaccination of elk with Brucella abortus strain RB51 (SRB51) does not induce protection against challenge with virulent strains of B. abortus during gestation. Elk (n ϭ 7/trt) were subcutaneously vaccinated at 7 to 8 months of age with single inoculations of saline, 10 10 CFU of SRB51, or B. abortus strain 19 (S19), or twice with 5 ϫ 10 6 CFU of bacillus of Calmette and Guerin (BCG). Elk vaccinated with S19, SRB51, or BCG had greater (PϽ 0.05) antibody responses to Brucella or M. bovis antigens, respectively, when compared to responses of nonvaccinated elk. Antibody responses of elk to SRB51 were greater than responses of bison and cattle in comparable studies. Proliferative responses of PBMC from S19-or SRB51-vaccinated elk to Brucella antigens were greater at 22 and 26 weeks after vaccination when compared to nonvaccinated elk. Proliferative responses of S19-and SRB51-vaccinated elk to Brucella antigens were delayed when compared to responses of SRB51-vaccinated cattle and bison. Flow cytometry data suggest that proliferative responses to Brucella antigens were primarily confined to B cell subsets. At 2 weeks after the first BCG vaccination, PBMC from vaccinated elk had greater proliferative responses to M. bovis antigens as compared to nonvaccinated elk. Proliferating cells at this time included B cells, CD4 and T cells. However, at all other sampling times, which included samples out to 22 weeks after the second BCG vaccination, no significant differences in proliferative responses were detected between vaccinated and nonvaccinated elk. As compared to cattle, elk immune responses to BCG are less robust and of much shorter duration, with greater proliferative responses to M. bovis antigens within B cells subsets, and reduced proliferative responses within CD4 and T cell subsets. All BCG-vaccinated elk had positive responses to skin testing to M. bovis, but not M. avium antigens. Only 1 of 7 elk were skin test positive to brucellin. The results of these studies suggest that elk primarily respond to brucellosis and tuberculosis vaccines with a humoral response that is unlikely to be associated with long-term protection. Our data also suggests that the immunologic responses of elk may substantially differ from other ruminants. As in most other species, kittens acquire nearly all of their passive immunity from colostrum during the first day of life. Immunoglobulin supplementation protocols exist for correction of failure of passive transfer in kittens, but it is not known if these treatments result in improvement of innate immune function. The purpose of this study was to determine if correction of IgG deficiency in neonatal kittens results in improvement of opsonizing capacity and neutrophil function.

Specific pathogen free kittens were separated from their mothers at birth and randomly placed into 1 of 4 treatment groups: colostrum-fed (CF) (n ϭ 6), colostrum-deprived (CD) (n ϭ 6), colostrum-deprived supplemented with equine IgG (EQ) (n ϭ 5), or colostrum-deprived supplemented with feline IgG (FE) (n ϭ 5). CF kittens were returned to the queens to nurse after initial blood sampling. All other kittens were formula-fed for 48 hours, then returned to the queens for the duration of the study. Kittens in the FE group were supplemented with 400 mg feline IgG SC. Kittens in the EQ group were supplemented with 400 mg equine IgG SC. Blood samples were collected from each kitten at 0 (birth), 2, 7, 14, 28, 42, and 56 days of age. Plasma IgG was measured by radial immunodiffusion. Kitten neutrophil function was assessed at each time point by simultaneous measurement of bacteria-induced phagocytosis and oxidative burst using flow cytometry. The opsonizing capacity of kitten plasma was determined by incubating the plasma (untreated or heat-treated) with bacteria prior to adding neutrophils from adult donor cats. The phagocytic response of the donor neutrophils and subsequent oxidative burst were measured by flow cytometry.

IgG was undetectable in all kittens at birth. Plasma IgG concentrations were significantly higher in CF, EQ, and FE kittens compared to CD kittens at 2-28 days. There were no significant differences in phagocytosis or oxidative burst by kitten neutrophils or of opsonizing capacity of kitten plasma among the treatment groups. Heat-treated plasma had significantly less opsonizing capacity than untreated plasma in all treatment groups.

Opsonizing capacity of kitten plasma and function of kitten neutrophils were not correlated with IgG concentration. However, a measurable decrease in both phagocytosis and oxidative burst of donor neutrophils was seen when plasma from all groups was subjected to heat. This suggests that a kitten's first line of defense against bacterial infection may be more dependent upon a heat-labile plasma component, such as complement, than on IgG acquired via passive transfer. Infectious diseases are a common cause of morbidity and mortality in kittens. Kitten susceptibility to infections may be dependent upon the intrinsic ability of their neutrophils for phagocytosis of bacteria, or to their plasma opsonic capacity for promoting phagocytosis of bacteria. The purpose of this study was comparison of neutrophil function and plasma opsonic capacity in kittens to that in adult cats.

The neutrophil function assay was performed on heparinized whole blood collected from adult specific pathogen free (SPF) cats and SPF kittens from birth to 8 weeks of age. Neutrophil function was assessed by simultaneous measurement of bacterial-induced phagocytosis and oxidative burst by flow cytometry of whole blood samples. The plasma opsonic capacity was assessed by adding bacteria preincubated with the plasma fraction to donor adult neutrophils followed by flow cytometry measurement of phagocytosis. The contribution of complement to plasma opsonic capacity was evaluated by comparison of the opsonic capacity before and after complement inactivation by heating of the plasma at 56ЊC for 30 minutes. The contribution of plasma IgG to opsonic capacity was evaluated by correlating the plasma IgG concentrations with the opsonic capacity of plasma before and after complement inactivation.

Compared to adult cats, neutrophil phagocytosis and the subsequent oxidative burst response were significantly decreased in kittens from birth to 4 weeks of age, but reached adult values by 6 to 8 weeks of age. The neutrophil function assay cannot differentiate between intrinsic phagocytic capacity of the neutrophils and opsonic capacity of the plasma when whole blood is used. The small size of neonatal kittens precludes the measurement of phagocytic capacity in kitten neutrophils independently of their plasma due to the substantial amount of blood required for neutrophil separation. However, the plasma opsonic capacity of kittens was lower than adults when measured independently of their neutrophil function. Heat inactivation of complement significantly decreased the opsonic capacity of kitten plasma. Plasma IgG did not correlate with kitten plasma opsonic capacity either prior to or after complement inactivation.

The results of this study suggest that the greater susceptibility of kittens to bacterial infections compared to adults may be more related to the activity of phagocytosis-promoting factors in the plasma such as complement, rather than to IgG concentrations or neutrophil function per se. Polymorphonuclear cells (PMN) and their ability to phagocytize and kill invaded microorganism are important parts of the host defence system. Defects in these functions can lead to recurrent bacterial infections. Flow cytometry is a fast and easy method for investigating some functional disorders of neutrophils. Tremendous intra-and interindividual variations have been observed in healthy dogs. The purpose of this study was to elucidate the reasons for such variations and to standardize the evaluation of the phagocytic capacity and of the generation of reactive oxygen species (ROS) of canine PMN using flow cytometry.

Phagocytosis assay: PMN purified from blood of healthy dogs by density gradient centrifugation were incubated with FITC-labelled, heat-inactivated Staphylococcus aureus bacteria with and without complement opsonization. Opsonization improved the phagocytosis by canine PMN. Different concentrations of phorbol-myristate-acetate (PMA) were used to modulate PMN phagocytosis. However, no positive effect was found. Experiments with a known killed control cell population have shown a high loss of evaluated cells, most probably by adherence at tubes or wells. It was possible to avoid this loss by layering all ingredients on cushions of Histopaque. However, the cushion had a negative influence on the phagocytic activity of canine PMN. After incubation phagocytosis was stopped by storing the cells on ice. Quenching of extracellular fluorescence was achieved by adding crystal violet, thus allowing the distinction between adherent and ingested microorganisms.

ROS assay: Purified canine peripheral blood PMN were stimulated with different concentrations of PMA and subsequently labelled with dihydrorhodamine-123 (DHR-123). The non-fluorescent DHR-123 was converted into the fluorescent rhodamine-123 (R-123) by reactive oxygenic species. PMA was able to increase the ROS formation dose dependently. Cushions of Histopaque were used as described above. No influence of this cushion was observed on ROS formation.

Despite these different methods, the interassay-variation remained high. Monitoring the PMN function of patient and healthy control-dog on the same day and under strictly identical conditions is still recommended. Neutrophils are the primary professional phagocytic cells that provide the host with a first line of defense against infections through their innate ability to ingest and kill invading microorganisms. The objective of this study was the development and validation of a flow cytometric assay for simultaneous measurement of neutrophil phagocytosis and oxidative burst responses in whole blood samples from cats.

The neutrophil function assay was performed on heparinized whole blood collected from adult spefici pathogen free (SPF) cats. The targets for phagocytosis were heat-killed Staph aureus bacteria labeled with propidium iodide. The oxidative burst response was measured by conversion of a nonfluorescent probe, dihydrorhodamine (DHR), to fluorescent rhodamine. The basic assay consisted of incubation of 100 microliters of DHR-loaded whole blood with the labeled bacteria. The samples were prepared for flow cytometry using the Immunoprep reagent kit and Q-Prep machine (Coulter Diagnostics, Hialeah, FL), and acquired on a flow cytometer using an argon laser with 488 nm excitation. Neutrophils were gated for analysis by the light scatter density plot. The validity and specificity of the assay was evaluated by 2 approaches: comparison of phagocytosis measured by flow cytometry to light microscopy, and pretreatment of the samples with selective inhibitors of phagocytosis and/or oxidative burst.

The maximal oxidative burst response occurred with a 5 micromolar loading dose of DHR. Phagocytosis and oxidative burst reached a plateau at a target: neutrophil ratio of 30:1 and an incubation time of 30 minutes. There was a significant correlation between percent phagocytosis measured by flow cytometry and light microscopy. EDTA, a calcium chelator, inhibited phagocytosis and its subsequent oxidative burst. Paralysis of the actin microfilament network with cytochalasin D also abolished phagocytosis and oxidative burst. Staurosporine, a tyrosine kinase inhibitor, did not affect the phagocytic activity of the neutrophils, but did prevent their oxidative burst response. Using the standardized assay in normal SPF adult cats, the mean number of neutrophils that phagocytosed bacteria was 78% (range ϭ 48-96%), and the mean number responding with a subsequent oxidative burst was 55% (range ϭ 36-79%).

This assay provides a clinically useful tool for evaluation of neutrophil function in a small volume of whole blood compared to previous assays requiring neutrophils purified from large blood samples. This feature is particularly relevant in cats, a species in which blood sampling volume is frequently limited.

Of 26 cats with CRF, 8 cats (31%) exhibited secondary hyperparathyroidism (PTH Ͼ 44 pg/mL). PTH levels in cats with HPTH-CRF ranged from 45 to 1162 pg/mL (mean 234 pg/mL), as compared to 1 to 36 pg/mL (mean 13 pg/ mL) in NPTH-CRF cats. PTH in normal cats ranged from 1 to 13 pg/mL (mean 3 pg/mL). Concentration of iCa in HPTH-CRF cats ranged from 5.01 to 5.45 mg/dL (mean 5.23 mg/dL), as compared to 5.18 to 6.13 mg/dL (mean 5.52 mg/ dL) in NPTH-CRF cats. Concentration of iCa in normal cats ranged from 5.03 to 5.37 mg/dL (mean 5.23 mg/dL)

Creatinine in normal cats ranged from 1.2 to 2.1 mg/dL (mean 1.66 mg/dL). Serum P in cats with HPTH-CRF ranged from 5.8 to 17.6 mg/dL (mean 10.25 mg/dL)

56 mg/dL) in NPTH-CRF. Phosphorus in normal cats ranged from 3.7 to 5.6 mg/dL (mean 4.63 mg/dL)

Serum iCa concentrations were significantly increased in NPTH-CRF cats, and P levels were significantly increased in both HPTH-CRF and NPTH-CRF cats. Serum iMg levels in both HPTH-CRF and NPTH-CRF cats were decreased as compared to normal cats. Serum Cr levels were similar in both HPTH-CRF and NPTH-CRF cats, but significantly increased as compared to normal cats. Correlation of PTH concentration to iCa was 0.46, PTH to iMg was 0.29, PTH to P was 0.60, and PTH to Cr was 0.57. In conclusion, serum iCa, iMg, Cr, or P concentrations cannot be used to predict PTH levels and/or the presence of secondary hyperparathyroidism

Intestinal permeability and mucosal function testing is a sensitive method for evaluating intestinal integrity and function. Testing consists of oral administration of a sugar solution and determination of the percent urinary recovery of the sugars administered. Because of the need for complete urine collection over a specified time, a urethral catheter and a closed urinary collection system is useful for this procedure. Because most cats require sedation or anesthesia for catheter placement, anesthesia for gastroduodenoscopy and biopsy (GB) is an opportunity for placement. The goal of this project was to determine if GB would affect 24-hour urinary recoveries of 5-orally administered sugars in healthy cats. Six clinically healthy male experimental cats were evaluated. All cats were anesthetized and GB was performed. During anesthesia an indwelling urethral catheter was placed in a sterile fashion in each cat

Urine samples were collected aseptically from a closed urinary collection system prior to solution administration, every 2 hours for 12 hours, and again 24 hours after administration. Total urine volume was recorded for each time period. The same 6 cats underwent the same procedure 2 months later without prior GB. Urinary recoveries of the sugars were determined by high-pressure anion exchange liquid chromatography with pulsed amperometric detection. Mean cumulative urinary recovery for each sugar was compared between the 2 study periods

At the time of initial physical examination 17 cats (49%) were dehydrated, 9 (26%) had pale mucous membranes, and 7 (20%) were icteric. All cats had a systemic illness concurrent with chronic pancreatitis, fourteen (40%) of which had hepatic disease and eight (23%) had renal disease. Clinicopathologic findings included mild nonregenerative anemia, azotemia, hyperbilirubinemia and elevated ALT and AST. Isosthenuria was reported in 14/ 23 (60%) cats. TLI was measured in five cats, 2 (40%) of which were elevated. In twenty-three cats that had thoracic radiography, ten (44%) were normal, eight (35%) had pleural effusion and five (23%) had cardiomegaly. In twelve cats that had abdominal radiography, eight (67%) had decreased abdominal detail, three (25%) had gas filled intestinal loops and two (17%) had hepatomegaly. Abdominal ultrasound was performed in all cats. Twenty (57%) were deemed to have a normal pancreas, three (8%) had a hypoechoic pancreas with surrounding hyperechoic mesentery, ten (29%) did not have the pancreas identified and 2 (6%) had a pancreatic mass

2 , and DA Williams 1 . 1 Gastrointestinal Laboratory

In human beings measurement of serum pepsinogen concentrations is used as a marker for gastric disorders. The aim of this study was to evaluate the concentrations of serum cPG A in dogs with gastric lesions in order to evaluate serum cPG A as a possible marker for canine gastritis. Sera were obtained from 72 dogs: 19 dogs with histopathologically confirmed spontaneous gastritis and 53 dogs with endoscopically visible gastric lesions after completion of an endurance sled race. Endoscopic appearance was qualitatively scored as no lesions, mild lesions, or severe lesions (bleeding ulcers). Concentrations of serum cPG A were measured using an in-house ELISA. Data were analyzed with a statistical software package (GraphPad Prism 3.0). The mean serum cPG A concentrations for dogs belonging to different severity score groups were compared to the mean serum cPG A concentration of clinically healthy dogs (n ϭ 64) using one-way ANOVA. Relationships between high serum cPG A and the severity of gastric lesions were analyzed using a Chisquared test

Increased serum cPG A concentrations were associated with an increased probability of endoscopically visible gastric lesions. The mean serum cPG A concentration was significantly higher in dogs with gastric lesions than in clinically healthy dogs but did not exceed the upper limit of the reference range. We conclude that serum cPG A concentrations are elevated in dogs with gastric lesions. However, at least in this group of dogs

420, and 480 minutes after feeding. Concentrations of serum cPG A were measured by an in-house ELISA. Data were analyzed using a statistical software package (GraphPad Prism 3.0). Variations in concentrations of cPG A were analyzed by time-point relative to baseline using one-way, repeated measures ANOVA followed by Dunnett's multiple comparison test, comparing each time-point to the baseline sample. Statistical significance was assigned for values of pϽ0.05. All eight dogs showed similar postprandial alterations in cPG A concentraany time-point. We conclude that feeding has a significant influence on serum pepsinogen A concentrations in dogs. It is crucial that food is withheld for at least 8 and preferably 12 hours, when determining baseline pepsinogen A concentrations in serum as a diagnostic tool for canine gastric disorders

Canine pepsinogen A (cPG A) was purified from gastric mucosa by anionexchange chromatography, hydroxyapatite chromatography, and size-exclusion chromatography. Antibodies against cPG A were raised in sheep and purified by affinity chromatography. A sandwich ELISA was developed. Capture and reporter anti-cPG A antibodies were produced in two different sheep. The reporter antibody was biotinylated and a streptavidin horseradish peroxidase preparation and a horseradish peroxidase substrate were used for development. The assay was validated by determination of sensitivity, working range, linearity, accuracy, precision, and reproducibility. A reference range for serum cPG A was determined in 64 clinically healthy dogs using the 95 th percentile. Assay sensitivity was 18 g/L, and the maximum detectable concentration was 1,080 g/L. Observed to expected (O/E) ratios for three serial dilutions of 3 serum of variation for intra-and interassay variability of 3 serum samples ranged from 7.6 to 11.9% and from 10.1 to 13.1%, respectively. The mean serum concentration for cPG A in clinically healthy dogs was 63.8Ϯ31.0 g/L (meanϮSD) and the reference range was 18 to 129 g/L. Results suggest that the ELISA for measuring canine pepsinogen A in canine serum is sufficiently sensitive

However, it is accepted that serum lipase activity is not specific for exocrine pancreatic function, and that many cell types other than pancreatic acinar cells synthesize lipases. Recently an assay for measurement of canine pancreatic lipase immunoreactivity has been developed and validated. This assay has been shown to be specific for exocrine pancreatic function and sensitive for the diagnosis of pancreatitis in the dog

Tracer was produced by iodination ( 125 I) of feline pancreatic lipase using the chloramine T method. A radioimmunoassay was established and validated by determination of sensitivity, dilutional parallelism, spiking recovery, intraassay variability, and inter-assay variability. A control range for fPLI in cat serum was established from 30 clinically healthy cats using the 95 th percentile. The sensitivity of the fPLI assay was 1.2 g/L. The observed to expected dilutions of 1 in 2, 1 in 4, and 1 in 8. Observed to expected ratios for spiking recovery ranged from 76.0 to 156.5% for 4 different serum samples and 6 different spiking concentrations. Coefficients of variation for intra-assay variability for 4 different serum samples

Animal Health Diagnostic Laboratory

Hepatic lipase (HL) may be involved in the clearance of chylomicron remnants. Altered LPL/HL activity may be a possible cause of lipid abnormalities in primary hyperlipoproteinemia (PHLP). Objectives were to compare lipid metabolic parameters in dogs with PHLP, to those of normal

English Cocker Spaniels, and 1 Poodle/Cocker mix) were identified with PHLP, after all causes of secondary hyperlipoproteinemia had been ruled out. Fasting serum and postheparin plasma were collected from each dog with PHLP, and all control dogs (n ϭ 8). Serum cholesterol concentrations were significantly (PϽ0.05) higher in PHLP dogs (mean Ϯ SD; 532 Ϯ 256 mg/dL)

On lipoprotein (LP) electrophoresis distribution, percentages of chylomicrons and alpha-2 migrating LP (HDL1) were similar in both groups. Percentage of beta-migrating LP (VLDL and LDL) in PHLP dogs was significantly higher (mean 45.2 Ϯ 12%) as compared to normal dogs (23.5 Ϯ 5%). Percentage of alpha-1 migrating LP (HDL2) of PHLP include significantly elevated cholesterol and TG levels, increased percentage of beta-migrating LP, decreased percentage of HDL2, decreased LPL activity, and increased HL activity. Decreased LPL activity may contribute to accumulation of beta-migrating LP

In this way it is possible to eliminate interference by high-molecularweight proteins on the migration of the low-molecular-weight proteins. Distinction of the individual protein bands makes it possible to identify proteins of tubular origin (MW Ͻ70,000) and proteins of glomerular origin (MW Ͼ70,000). The scope of the study was the qualitative determination of physiological proteinuria in the dog. Fourteen dogs aged 24 to 48 months (6 intact males, 1 castrated male

for immunofluorescence examination. Conventional light microscopic examination was done on samples stained with hematoxylin-eosin, periodic acid-Schiff, trichrome according to Goldner, methenamine and Congo red. Immunofluorescence examination was performed by testing the samples with IgG, IgA, C 3 complement fraction and fibrinogen. At the time of the biopsy, a pre-established volume (10 ml) of urine was collected by centesis and kept (after the addition of 1% sodium azide equal to 1 l/ml of urine) at 4-8ЊC. The following analyses were performed on the urine samples: physico-chemical, sediment, determination of proteinuria by a quantitative (staining with pyrogallol) and a qualitative method (SDS-PAGE HYDRAGEL PRO-TEINURIA), urine protein/creatinine ratio. All the dogs with inert sediment and with normal conventional light microscopic

SDS-PAGE can be considered a useful method, specific and sensitive, for the qualitative determination of physiological proteinuria in the dog

There appears to be a high incidence of pancreatitis in the Miniature Schnauzer. In human beings, two major mutations (R117H and N21I) of the cationic trypsinogen gene have been linked with hereditary pancreatitis. These mutations are thought to cause structural changes in the protein. The major mutation, R117H, leads to a decreased ability of prematurely activated trypsin to digest itself, thus allowing for activation of other zymogens and leading to pancreatic autodigestion. The purpose of this study was to evaluate the cationic trypsinogen gene in Miniature Schnauzers for possible mutations.Four Miniature Schnauzers were selected on the basis of their clinical record including either: 1) presence of a pancreatic biopsy showing pancreatic inflammation, or 2) the presence of clinical signs compatible with pancreatitis (such as vomiting or abdominal pain) and a 5-fold elevation of serum lipase activity, or 3) the presence of clinical signs compatible with pancreatitis and ultrasonographic evidence of pancreatitis (pancreatic edema, ascites in the area of the pancreas, and changes in echogenicity of the pancreas). In addition, a clinically healthy Miniature Schnauzer and a mixed breed dog were selected for comparison. Whole blood samples were obtained from all dogs and DNA was extracted from white blood cells using a commercially available kit (Gentra Systems, PURE-GENE). Primers were designed and optimized to amplify the entire canine cationic trypsinogen cDNA sequence. PCR was performed and each product was purified. (Promega Wizard PCR Preps DNA Purification System). These PCR products were then sequenced on an automated sequencer. The sequence of the cationic trypsinogen gene of the healthy and the diseased Miniature Schnauzers was compared with that of the healthy control dog and the published sequence for canine cationic trypsinogen.The clinically healthy control dog showed an identical sequence of the cationic trypsinogen gene to the published sequence. The clinically healthy Miniature Schnauzer as well as all 4 Miniature Schnauzers with pancreatitis also showed the same sequence of the cationic trypsinogen gene.We conclude that, in contrast to human beings with hereditary pancreatitis, pancreatitis in the Miniature Schnauzer is not due to mutations of the cationic trypsinogen gene. Recently, serum canine pancreatic lipase immunoreactivity (cPLI) concentration has been shown to be highly specific for exocrine pancreatic function and also highly sensitive for the diagnosis of pancreatitis in dogs. Given these encouraging results in the dog and the current limitations in diagnosing feline pancreatitis we established a project in order to develop an immunoassay for measurement of feline PLI in serum. The objective of this project was to purify fPL from pancreatic tissue and partially characterize this protein as a prelude to the development of an assay for measurement of feline PLI in serum.Pancreata were collected from cats, sacrificed for unrelated research projects. After removal of grossly visible fat, pancreatic tissue was crushed in acetone and filtered through a Buchner funnel. In order to completely remove all fat this procedure was repeated a total of 8 times, using acetone and a chloroform butanol mixture. The final extract was resuspended in ethyl ether, and filtered in the Buchner funnel to dryness. Two grams of the delipidated pancreatic extract were further purified by extracting the enzymes in a tris-buffer containing two different protease inhibitors, benzamidine and phenylmethylsulfonyl fluoride, followed by anion-exchange, size-exclusion, and cation-exchange chromatography. The purified protein was partially characterized by molecular mass determination, estimation of specific absorbance, and N-terminal amino acid sequencing of the first 25 amino acid residues.Feline pancreatic lipase was successfully purified from feline pancreatic tissue. The purified product showed a single band on sodium dodecyl sulfate polyacrylamide gel electrophoresis with a molecular mass of approximately 52.5 kDa. Exact molecular mass was determined by mass spectrometry to be 52.4 kDa. Approximate specific absorbance at 280 nm of fPL was 1.18 for a 1 mg/ml solution. N-terminal amino acid sequence of the first 25 amino acid residues showed the sequence Lys-Glu-Ile-X-Phe-Pro-Arg-Leu-Gly-X-Phe-Ser-Asp-Asp-Ala-Pro-Trp-Ala-Gly-Ile-Ala-Gln-Arg-Pro-Leu. Of 23 identified amino acid residues 19, 16, and 20 matched corresponding residues of canine, human, and porcine classical pancreatic lipase, respectively.We conclude that fPL was successfully purified from feline pancreatic tissue. We recently developed and validated an enzyme-linked immunosorbent assay (ELISA) for measurement of canine pancreatic lipase immunoreactivity (cPLI). This assay has been shown to be highly specific for exocrine pancreatic function and highly sensitive for a diagnosis of pancreatitis. Based on previous findings that feeding leads to significant increases of serum activities of several digestive enzymes, we currently recommend that serum samples are collected after withholding food for 12 hours. The goal of this study was to evaluate the influence of feeding on serum cPLI concentrations.

Eight clinically healthy adult Beagle dogs belonging to a research colony were enrolled in this study. None of the dogs showed any abnormalities on physical examination, complete blood count, or serum chemistry profile. An indwelling jugular catheter was placed, and food was withheld from the dogs for at least 17 hours. A baseline serum sample (0 min) was collected and dogs were fed a commercial dry canine maintenance diet (Canine Maintenance; Hill's Pet Nutrition). Further serum samples were collected at 15, 30, 45, 60, 75, 90, 105, 120, 150, 180, 210, 240, 300, 360, 420, and 480 minutes after feeding. Serum cPLI was measured using an in-house sandwich ELISA. Serum cPLI concentrations at different times after feeding were compared to baseline concentrations using a one-way, repeated measures ANOVA, followed by a Dunnett's multiple comparison test (GraphPad Prism 3.0). Statistical significance was assigned for values of pϽ0.05.Serum cPLI concentrations at baseline ranged from 1.8 to 73.7 g/L (mean ϮSD; 28.9 Ϯ28.6 g/L). There was no significant difference in serum cPLI concentrations at any time after feeding when compared to baseline concentrations (p ϭ 0.131). However, there were fluctuations in serum cPLI concentrations over time. Three of the dogs had baseline serum cPLI concentrations in the lower range of the reference range. Serum cPLI concentrations decreased below the lower limit of the reference range for several of the time points for these 3 dogs. None of the serum cPLI concentrations measured at any time point was increased over the upper limit of the reference range for serum cPLI.We conclude that, in clinically healthy dogs, feeding has no significant influence on serum cPLI concentrations.

We have derived predictive equations to estimate plasma triglyceride (TG) and phospholipid (PL) and neutrophil PL enrichment from dietary polyunsaturated fatty acid (PUFA) composition data. Using the TG predictive equation and our recently determined Michaelis-Menton kinetic constants (Km) for delta-6 desaturase, concentrations of the dietary competitors (linoleic (LA) and alpha-linolenic (ALA) acids) needed to assure essential fatty acid status and adequate n-3 fatty acid tissue enrichment were calculated. In our earlier kinetic study, concentrations of total LA and ALA of canine liver microsomes were determined (373 uM and 11 uM respectively). These dogs had been fed a commercial extruded-type diet (32 energy % (en%) total fat with 0.94 en% ALA and 8.75 en% LA). From this diet information, plasma TG-fatty acid concentrations were calculated as 420 uM and 20 uM for LA and ALA respectively which are similar to those seen in the canine microsomes. Thus plasma TG LA and ALA concentrations estimate microsomal total LA and ALA concentrations. These latter fatty acids serve as a substrates for hepatic delta-6 desaturase. A diet with known LA and ALA concentration was then formulated so that the resultant hepatic microsomal LA and ALA content would exceed the delta-6 desaturase Km values for these acids. A 34.6 en% as fat diet was formulated blending beef tallow/safflower oil/linseed oil as fat sources and contained poultry meal and rice. This theoretical diet contained 6.5 en% ALA and 7.4 en% LA. Substituting these values into the derived 2nd order polynomials resulted in estimated concentrations for microsomal ALA (121 M) and LA (377 M). These values both exceed the Km values of 12 and 42 M respectively by a similar factor. Calculated amounts of n-6 LCPUFA and n-3 PUFA using the PL equations for plasma and neutrophils were also consistent with anticipated functional modifications due to PUFA composition alterations from our earlier feeding studies. Diets such as these will need to be produced and tested for safety and measures of efficacy. However, they serve as a point of embarkation to define the amount of more stable sources of n-3 fatty acids needed to achieve desired LCPUFA effects and the utility of our predictive equations in canine nutrition. Exercise is associated with an increased rate of production of reactive oxygen species and a presumed increased requirement for antioxidants. We hypothesized that plasma disappearance of deuterated vitamin E would be more rapid in dogs performing 3 days of strenuous exercise than in sedentary dogs.

Twelve moderately trained Alaskan sled dogs (5 females, 7 males, 2-7 years, 16-25 kg) were randomly divided into 2 groups (exercise, 48 km in a team pulling a sled over snow, 16 km/h for 3 d, or no exercise). Dogs in both groups were administered orally 200: L of D6-RRR-alpha tocopherol (97.8%) with 10 grams of fat 20 hours before start of exercise. Blood samples for measurement of plasma total ''-tocopherol ([vit E]) and D6-'' tocopherol (D6) concentrations were collected before administration of D6, and at regular intervals thereafter (see figure) . Plasma [vit. E] concentrations were affected by exercise (2 way, repeated measures ANOVA, group x time interaction p ϭ 0.002, see figure) with the exercise group having higher concentrations. Serum [D6] were not different between groups. Results were not different when tocopherol concentrations were adjusted for plasma cholesterol concentration. These results demonstrate that prolonged, repetitive exercise by dogs does not alter the kinetics of plasma concentrations of D6-'' tocopherol after oral administration suggesting that exercise either does not alter the elimination or utilization of vitamin E or affect its apparent volume of distribution. Desaturation of essential fatty acids by delta-6 desaturase is considered the rate-limiting step in their conversion to long chain polyunsaturated fatty acids. This enzyme can utilize either linoleic (LA) and alpha-linolenic acids (ALA). When both acids are present, competition between them for conversion exists. Consequently, methods to characterize enzyme activity against ALA in unpurified or partially purified preparations are fraught with error due to the presence of high endogenous amounts of the competitive substrate, LA, in such preparations. A radioisotopic technique was thus developed to enable the kinetic characterization of the enzyme using radiolabeled ALA. It employed a graphical correction for the presence of varying and competitive amounts of LA in canine hepatic microsomal enzyme preparations. Microsomes were prepared using fresh canine liver which were incubated with 14 C labeled ALA under defined conditions. Lipids were extracted, saponified and phenacylated. Fatty acid phenacyl esters were separated by HPLC and counted. Radioactive product formation (i.e. 18:4n-3) was converted to activity (pmol/min/mg protein). Before correction, delta-6 desaturase activity (V max ) was 50.9 pmol/min mg protein with an apparent K m of 20.8 uM. The V max was lower than that reported in dogs and other species probably the result of the presence of high amounts of the competitive endogenous LA inherent in the microsomal preparations. To correct for this competition, the amounts of LA in each enzyme preparation was first determined by extraction and gas chromatography with internal standardization. Next, all individual microsomal activities measured using 14 C-ALA at known LA concentration were plotted against their respective LA concentration. This technique yielded a series of 12 linear graphs in which LA concentration varied from approximately 15 to 75 uM and which spanned an ALA concentration of from 4 to 200 uM. Extrapolation of each line to its respective y-intercept yielded activity values at LA concentation of zero (i.e. no competitive substrate present). The y-intercept activty values were then plotted against ALA substrate concentration and kinetic analyses performed. The resultant V max increased to 62.4 uM and Km decreased to 12.4 uM. This graphical technique provides a more accurate characterization of canine delta-6 desaturase activity with ALA substrates when unpurified enzymes preparations are studied.