key: cord-023353-2yoz1t6a
authors: nan
title: ABSTRACTS FROM THE 2010 EVDI ANNUAL MEETING
date: 2010-12-28
journal: Vet Radiol Ultrasound
DOI: 10.1111/j.1740-8261.2010.01774.x
sha: 
doc_id: 23353
cord_uid: 2yoz1t6a

nan

Oral malignant melanoma is the most common canine oral tumour being characterized by local invasion and a high rate of regional and distant metastases. The overall prognosis is guarded, mainly because of the poor response to adjuvant systemic treatment like chemotherapy. Melanoma is sensitive to radiation, which may reduce/eliminate tumour volume in macroscopic tumours or prolong local control in microscopic settings. Neither surgery nor RT can control the high risk of distant metastases though. Interesting approaches for immunogenic tumours such as melanoma are the use of tumour vaccines. The objective of this study was to evaluate toxicity associated with concurrent administration of an intradermally administered tumour vaccine and radiation. Materials and Methods: Nine dogs with histologically confirmed oral malignant melanoma were treated with RT in combination with a human recombinant tyrosinase DNA vaccine. Patients were staged according to the WHO TNM-staging-scheme for oral tumours1. In 5 dogs a CT scan was performed to assess local disease, 4 dogs were evaluated clinically for local disease extension. For assessment of metastatic disease, regional lymph node aspiration, thoracic radiographs and abdominal ultrasound were performed. All dogs received 4 fractions of 8 Gray each at 7-day intervals to a total dose of 32 Gray (6MV linear accelerator) to the primary tumour and the regional lymph nodes (mandibular and medial retropharyngeal). The vaccine was given 4 times at 14-day intervals with the first vaccine being administered at the first fraction of RT. The acute radiation toxicity was evaluated according to the VRTOG scoring scheme2 at the last RT and 2 to 4 weeks after the last RT.

The anatomic tumour locations were as follows: mandible (n ¼ 4), maxilla (n ¼ 3) and buccal mucosa (n ¼ 2). Metastases were initially identified in 8 of 9 dogs: ipsilateral mandibular lymph node (n ¼ 5), ipsi-and contralateral mandibular lymph node (n ¼ 1), ipsilateral mandibular lymph node and lungs (n ¼ 1), liver (n ¼ 1). Local acute RT toxicity (mucositis, alopecia/dermatitis, conjunctivitis) never exceeded grade 2 toxicity and did not have an appreciable negative impact on quality of life. The melanoma vaccine was well tolerated by all dogs, minimal side effects were observed in few dogs: fever (n ¼ 2), lethargy (n ¼ 2) and local swelling (n ¼ 2). Locally and loco-regionally all tumours could be controlled with a mean follow-up of 195 days. Three dogs developed lung metastases after 41 to 68 days. Three dogs died of tumour unrelated reasons. Three dogs are still alive (for 109 to 351 days). Discussion/Conclusion: We could not observe an increase in severity of acute local radiation toxicity administering RT and vaccine concurrently, as it has been suspected initially. This is an important fact, since early timing of the vaccination might be essential for its activity. More patients will be collected to evaluate these preliminary results.

Owen LN: TNM classification of tumors in domestic animals, World Health Organisation Geneva, 1980 LaDue T, Klein MK: toxicity criteria of the veterinary radiation therapy oncology group. VRUS 42(5) [475] [476] 2001 Introduction/Purpose: Pulmonary angiography with multi-detector computed tomography (MDCT) is the standard of care in humans for non-invasive diagnosis of acute and chronic pulmonary embolism, pulmonary hypertension and other diseases or anatomical variants of the pulmonary arteries. However, there is paucity of data in the animal literature despite the expected diagnostic performance of MDCT for pulmonary angiography. Therefore, aim of this feasibility study was to evaluate the anatomy and visualization of the normal canine pulmonary vasculature as the basis for further application of MDCT in diagnosing canine pulmonary arterial disease. Materials and Methods: Experiments were previously approved by the responsible animal care and use committee (V1430). In four one-year old beagles, pulmonary MDCT angiography (MDCTA) was timed based on an individualized timing bolus and calculated delay after bolus injection, scanning the entire thorax in caudal to cranial direction using helical scan mode at 0.625 mm collimation, 0.625 helical reconstruction interval utilizing all 64 detectors. Images were retrospectively reconstructed at 1.25 mm and 2.5 mm collimation for each dog. The pulmonary arteries were evaluated for anatomy (relationship to major vessels, cardiac and bronchial structures visible), visibility and visible generations of branching using all three reconstruction thicknesses. Results: All lobar pulmonary arteries and adjacent anatomical structures were visualized at all reconstruction thicknesses. Subjective blurring of small sublobar arteries occurs using 1.5 mm and 2.5 mm reconstruction thickness, particular with multiplanar reconstructions. Fewer generations of sublobar arterial branches were detected on 1.5 mm and 2.5 mm as compared to 0.625 mm reconstructions. Computed tomography of the brain in small animals can be limited by contrast resolution. The ability of the observer to differentiate grey and white matter is an important measure of image quality, but this can be reduced on images with low contrast. Various factors affect the contrast resolution, including exposure, slice thickness and patient size. While slice thickness can be adjusted on retro-reconstruction, the exposure settings are fixed for the study. The purpose of this study was to determine the effect of kVp and mA on objective and subjective measures of image quality, and to identify the technique that results in the best image quality. Materials and Methods: Six fresh canine cadavers of medium size were imaged using a 16-slice CT set in axial mode with 2.5 mm slice thickness and soft tissue algorithm. Twelve different protocols were used to image the brain in each dog: kVp was set at 140, 120 or 100 and for each kVp the mA was set at 360, 280, 200 and 120. All other parameters were kept constant. A slice at the level of the internal capsule was selected for analysis. Using image analysis software identical regions of interest were placed in the white matter and grey matter, and the mean Hounsfield units and standard deviation calculated for each slice. Image quality was evaluated objectively by measuring a contrast to noise ratio (CNR), calculated as CNR ¼ (mean GM ROIÀmean WM ROI)/ f(SD GM ROI) 2 þ (SD WM ROI) 2 g 1/2 . Images were evaluated subjectively by a radiologist blinded to the technique who graded the image quality on a 5-point scale (subjective score; SS). A two-factor ANOVA with repeated measures was used to determine the effect of mA and kVp on CNR and SS. The spearman test was used to correlate CNR and SS. Significance was set at Po0.05.

The exposure factors had a significant effect on objective and subjective measures of image quality. Increasing mA significantly increased CNR (Po0.001) and SS (Po0.001). Increasing kVp also significantly increased CNR (P ¼ 0.0176) and SS (P ¼ 0.0055). A technique of 360 mA 140 kVp resulted in the greatest CNR, whereas 360 mA 120 kVp resulted in the highest subjective score. A technique of 360 mA 140 kVp resulted in a higher CNR, but a lower SS than 280 mA 140 kVp. SS was positively correlated with CNR (r ¼ 0.61; Po0.0001). Discussion/Conclusion: CT acquisition with a high mA and kVp technique is recommended for canine brain imaging. It results in improved contrast resolution, primarily due to a reduction in image noise. While this was expected for mA, the reduction in noise with increasing kVp also offset the corresponding reduction in contrast. When tube heating limits the use of a 360 mA technique, acquisition at 280 mA provides subjectively good contrast resolution. Introduction/Purpose: Classroom performance systems (CPS) [audience response systems (ARS); ''clickers''] provide a means of encouraging student interaction in a traditional didactic lecture setting. CPS TM is the leading response system in education, being used by over 5 million students worldwide. CPS TM response pads (''clickers'') are handheld devices that allow students to respond to questions asked during a lecture enabling students and instructors to instantly assess their comprehension of lecture material. Materials and Methods: Student performance in and student evaluations of the 3rd year (5th semester) didactic radiology course were evaluated from [2005] [2006] [2007] [2008] [2009] . Additionally, student interactions were evaluated subjectively by instructors who used CPSTM. Results: From 2005 -2008 , all lectures and laboratory sessions were held in conventional fashion. In 2009, CPSTM was utilized by 3/5 instructors. There was no significant difference in overall student performance and student performance on individual instructor's exam questions. On a scale from 1 (worst) to 5 (best) the radiology course was evaluated higher by students when CPSTM was used than not, however, this difference was not significant (P ¼ 0.075). Subjectively, students appeared more interested and involved when CPSTM was used. Discussion/Conclusion: CPS may be beneficial in radiology education. Benefits include interactive involvement of students in a lecture setting and immediate feedback for students and instructors. Disadvantages include time constraints, technical problems and dependence on willing student participation. Introduction/Purpose: Non-invasive imaging techniques have been developed for phenotyping genetically engineered mice created to model human cardiovascular diseases like atherosclerosis. B-mode and Doppler ultrasound may be useful because it is a fast and real-time imaging technique which gives structural, functional and hemodynamic information and it makes possible to perform temporal longitudinal studies. Atherosclerotic lesions do not occur at random sites. The spatial heterogeneity of their distribution is in part dependent on hemodynamic factors. In murine models of atherogenesis lesions tend to develop at the ascending aorta, near the ostia of the three major branches, and on the lesser curvature of the aortic arch. The purpose of this study is to describe normal hemodynamics in mouse aorta and its major branches in the thorax by ultrasonography. Materials and Methods: 27 inbred C57BL/6 adult mice were studied (10 males and 17 females; 17-25 g body weight). Anesthetized mice were scanned in supine position with an ultrasound biomicroscope (Vevo 770, Visualsonics, Toronto, ON, Canada) provided with a 30 MHz transducer. Aorta was studied at five different regions (ascending aorta, near the ostia of the three major branches in the arch, and in the proximal portion of the descending aorta). Brachiocephalic trunk (BT), left common carotid artery (LCCA), and left subclavian artery (LSCA) were imaged near its junction with the aortic arch. Systolic area was measured in B-mode, and peak systolic velocity (PSV), minimum diastolic velocity (MDV), and time average maximum velocity (TAMX) were measured with pulsed Doppler ultrasound. Pulsatility (PI) and resistive indices (RI), and mean blood flow (FV) were calculated. Mean values and standard deviations were determined averaging a total of three Doppler waveforms in every mouse. Flow velocity profile and flow pattern were recorded in each region. The statistical analysis was performed with the Statistical Analysis System (SAS Institute, Cary, USA) by applying the GLM and LS MEANS procedures. A P-value less than 0,05 was considered statistically significant. All anatomical terms are in accordance with the Nomina Anatomica Veterinaria. Results: Aorta in the thorax has a plug flow profile and a typical high resistance flow pattern. BT, LCCA and LSCA have an intermediate flow velocity profile and a high resistance flow pattern. There were not differences between Doppler data obtained in males and females. The ostium of the brachiocephalic trunk, because of its topographical disposition and flow direction changing, was the region were Doppler has been most difficult to record.

Statistical differences were found between systolic area, PSV, MDV, TAMX, and FV of the aorta at the five different regions described above. No significant differences were found between systolic area, PSV, and FV of the aorta major branches, although differences were observed between the MDV and TAMX in these branches. Values of PI and RI don't differ along aorta and its major branches in the thorax.

This study provides normal Doppler wave pattern and vascular values of the aorta and its major branches in the thorax of adult C57BL/6 mice by ultrasonography. These values may be used as normal reference to compare with mutant mice with cardiovascular alterations.

Pascaline Pey 1 , Massimo Vignoli 2 , Hendrik Haers 1 , Luc Duchateau 3 , Federica Rossi 2 , Jimmy Saunders 1 . 1 Department of Medical Imaging of Domestic Animals and Orthopedy of Small Animals, Ghent University, Belgium. 2 Clinica Veterinaria dell'Orologio, via Gramsci 1/4, 40037 Sasso Marconi (BO), Italy. 3 Department of Physiology and Biometry, Ghent University, Belgium Introduction/Purpose: The microbubble-based contrast agents for ultrasonography have become more used in recent years in dogs1, and have been reported to be useful in differentiating adenomas from nonadenomatous lesions in human patients with adrenal masses2. The purposes of this study were to evaluate the feasibility and to define the perfusion parameters of contrast-enhanced ultrasonography of the adrenal gland in the normal beagle dog. Materials and Methods: Six healthy female beagles were injected with an intravenous bolus of SonoVue s (0.3 mL/10 kg, mechanical index ¼ 0.10). Subjective and objective analyses were evaluated. Objective measurements were performed using an off-line image analysis system (ImageJr). Results: Subjective evaluation: the aorta was first rapidly opacified after injection of the contrast medium, followed by the renal artery and the medulla. Subsequently, a homogeneous centrifugal enhancement from the adrenal medulla to the adrenal cortex was observed. At this point, a homogeneous decrease in opacification of both cortex and medulla started, associated with an enhancement of the phrenicoabdominal vessels. Then a persistent enhancement in the splenic parenchyma, in the renal vein, the caudal vena cava and phrenicoabdominal veins was noticed. At the end of the parenchymal phase a second contrast enhancement was observed, corresponding to the refilling time. Discussion/Conclusion: The shape of the time-intensity curve reflecting the adrenal perfusion was similar in all dogs. However, the absolute values of the measurements and of the parameters of perfusion displayed considerable variations. Ratios of the peak values of the cortex and the medulla to the peak values of the renal artery showed significant difference allowing differentiation between the cortex and the medulla for both adrenal glands. Contrast-enhanced ultrasonography of the adrenal glands is feasible in dogs and the optimal time for adrenal imaging is between 5 to 90 seconds after injection. Introduction/Purpose: Testicular anomalies are common in dogs. The aim of this study was to describe contrastenhanced ultrasonographic features of focal and diffuse testicular lesions, compared with normal tissue. Materials and Methods: Seventeen dogs with scrotal anomalies were considered. All dogs underwent B-mode ultrasound, anesthesia, contrast-enhanced ultrasound (CEUS), bilateral orchiectomy and histological examination. CEUS was performed with CnTI mode (Esaote s , Italy) and a secondgeneration contrast medium (SonoVue s -Bracco, Italy) injected at a dosage of 0.03 mL/Kg, IV. Parenchymal perfusion was recorded for 2 minutes. Time-intensity curves were reconstructed by a commercial software (Qontrast s -Bracco, Italy), using the gamma-variate bolus-corrected model. The peak intensity (P), time to peak (TTP), regional blood flow (RBF) and mean transit time (MTT) were calculated. Data relative to neoplasms were compared to those of normal/non neoplastic tissue by Fisher's exact test. A Po0.05 was considered significant. Results: Nine normal testes, 18 tumors (11 interstitiomas, 3 sertoliomas and 4 seminomas), 9 degenerated testes and a chronic orchitis were histologically diagnosed. Eight dogs had bilateral involvement, while 9 monolateral. Two seminomas (intratubular type) were not visualized. Perfusion parameters of each group are summarized in the following table.

Peak intensity Time to Introduction/Purpose: Studies with contrast enhanced ultrasonography (CEUS) on focal splenic lesions have shown discrepancies in accuracy for differentiating benign from malignant lesions.1-3 A speculative explanation for false positives may be the absence of a dual blood supply to the spleen compared to the liver. We therefore hypothesized that the early wash in/early wash out (EWEW), and the hypoechogenicity during all phases are unreliable in differentiating malignant from benign splenic lesions. Materials and Methods: A retrospective study reviewing CEUS studies of 17 splenic lesions (7 malignant, 10 benign), with concurrent cytology, core biopsy and/or excision biopsy was conducted. Sonovue& at a dose of 0.03 mL/kg was used in all dogs.

Tortuosity of feeding vessels, presence of EWEW, and hypoechogenicity during all phases were assessed visually and used to characterize the lesions.

There were 4/10 benign, and 4/7 malignant lesions with EWEW. Therefore, sensitivity, specificity, and accuracy for EWEW in differentiating malignant from benign lesions are 57%, 60%, and 59% respectively. There were 3/10 benign, and 3/7 malignant lesions with persistent hypoechogenicity throughout all phases. Therefore, sensitivity, specificity, and accuracy are 43%, 70%, and 59%. There were 0/10 benign, and 5/7 malignant lesions with tortuous and persistently visible feeding vessels. Sensitivity, specificity, and accuracy are 71%, 100%, and 88%. Finally, there were 7/10 benign and 7/7 malignant lesions with a combination of at least one of the above parameters. Sensitivity, specificity, and accuracy are 100%, 30%, and 59%. Discussion/Conclusion: These preliminary results support our hypothesis that the interpretation of splenic lesions cannot be performed accurately on the basis of EWEW, or persistent hypoechogenicity throughout all phases. Instead, we recommend assessing the presence of tortuous vessels in the periphery of the lesion during the early phase, and the persistence of these hyperechoic tortuous vessels in the late phase. Further studies with larger numbers are necessary to better assess these findings. Introduction/Purpose: CEUS has been used to detect focal perfusional changes in the spleen in humans and dogs. No earlier studies exist of contrast ultrasound of the spleen in cats. The purpose of this study was to evaluate perfusion of the cat spleen, and to compare the perfusion patterns in awake and anesthetized cats.

The spleen of 18 healthy cats was imaged with CEUS (Acuson Sequoia 512, Siemens or Esaote MyLab70, Biosound Esaote Inc.). Two groups of cats were imaged: Group I comprised of ten young, anesthetized cats and group II of eight young to middle aged cats that were initially imaged awake (IIa) and later anesthetized (IIb). Mechanical index (MI) was set at 0.32 (group I) or 0.04 (group II) in the focal zone area. Parameters were standardized including depth gain, power, number, and location of focal zones in relation to the spleen. All cats received multiple bolus injections (0.1 ml) of contrast medium (Definity s o) per imaging. A 1-to 2-min digital clip was recorded for each contrast injection and analyzed off-line. The timing and grade of heterogeneity of the spleen were estimated in a standardized manner, blinded and in a randomized order. The grade of heterogeneity (1) (2) (3) (4) was estimated when the spleen was the most heterogeneous. The grade of homogeneity (I-III) was estimated when the spleen was the most homogeneous. Student's t-test was used to compare the time to first appearance of contrast in the spleen (AT), and the time to the spleen becoming heterogeneous or homogeneous. Frequencies of each heterogeneity and homogeneity grade within separate groups were calculated with statistical software (Statistix version 9, Analytical Software). Statistical significance for analysis was set at Po0.05. Results: Enhancement of the spleen was bi-phasic in all cat groups. AT was significantly faster in awake than anesthetized cats. A marked heterogeneous perfusion pattern (grade 4) was significantly more prevalent in anesthetized (50%) than awake (12.5%) cats.

These results indicate focal perfusion defects, more pronounced with general anesthesia, of the spleen with CEUS should be evaluated with caution and only after disappearance of the initial heterogeneity. Introduction/Purpose: Diabetic nephropathy in small animals is a controversial issue. It is known that some spontaneously diabetic dogs develop significant proteinuria and hypertension. In cats ''diabetic nephropathy'' is a term describing fatal, chronic renal insufficiency with low urine specific gravity and proteinuria often occurring in feline diabetes. Measurement of intrarenal resistivity index (RI) values is described in human medicine as an useful method in early diabetic nephropathy diagnostics. Up to the present time the results of such tests in dogs (Novellas and others)three diabetic ones and four suffering from diabetes caused by hyperadrenocorticism-did not reveal any RI value deviations in animals suffering from diabetes not caused by hyperadrenocorticism. The aim of this study was to compare the RI values in diabetic dogs and cats with the healthy ones and evaluate the possible relationship between RI, age and duration of diabetes.

One of the study groups consisted of 10 dogs of both sexes and different breeds. The mean age was 10,9 AE 0,3, the duration of diabetes was from 1 month to 6 years. Another study group were cats -five European shorthairs of both sexes (mean age 7,8 AE 3,56, the duration of diabetes from 1month to 2,5 years). The data confirming diabetes, using blood and urine analysis were strongly documented. Patients with indication for ultrasound examination, but without any signs of serious or systemic diseases (eg. the animals with diseases of musculoskeletal system, routinely monitored geriatric patients) were assigned to control groups. Control groups consisted of 15 dogs and 15 cats of both sexes and different breeds and ages (mean age in canine group 6, 9 AE 4, 5, 1 AE 3, 07 years) . Animals with the suspicion of urinary tract infection or with history of previous kidney conditions were excluded from this study.

Renal RI values were higher in diabetic dogs and cats than in healthy animals. In the group of diabetic dogs in four animals RI exceed the accepted limit 0,71. Mean RI for the group of the dogs with diabetes was 0,66 AE 0,07, in control group 0,62 AE 0,05 (P ¼ 0,08). As many as four diabetic cats had significantly elevated RI and in two of them the diabetic nephropathy was suspected. Mean RI for the group of diabetic cats was 0,71 AE 0,07 and for the healthy cats 0,59 AE 0,04 (Po0,001). In all animals the positive correlation between the RI and age was observed (healthy dogs r ¼ 0,38, healthy cats r ¼ 0,51, diabetic dogs r ¼ 0,39, diabetic cats r ¼ 0,71). Correlation between RI and duration of diabetes was more relevant in cats (r ¼ 0,56) than in dogs (r ¼ 0,06). Discussion/Conclusion: Our study proves that renal vascular resistance value is elevated in some of small animals with diabetes. It is not possible to conclude much about usefulness of Doppler ultrasound in monitoring the kidney condition in animals with diabetes, from study performed on such a small group of patients. Our results may serve as an introduction to the research on bigger group of animals, which would be enriched in detailed laboratory and histopathological data providing new information about diabetic nephropathy in dogs and cats. Brkljacic B., Renal vascular resistance in diabetic nephropathy: Duplex Doppler US evaluation. Radiology 1994; 192: 549-554 Introduction/Purpose: Ultrasound (US)-guided percutaneous renal biopsy has become invaluable as diagnostic method in kidney diseases that is associated with minor or major complications, reportedly being more extensive using large-bore 14-gauge needles. 1 Contrast harmonic US (CHUS) increases the intensity of the blood pool echo signals in parenchymal organs and enhances existing lesions. This study investigated the value of CHUS as a screening method of post-biopsy renal complications, and evaluated the traumatic importance of 14-gauge biopsy needles.

Renal biopsy was performed in 11 healthy Beagle dogs at 3 occasions (0, 4 and 6 months) for a total of 33 biopsies using 14-gauge needles (Vet-Core TM , Surgivet, USA). A CHUS examination was performed 30 minutes after biopsy using a MyLab30Vet US machine (Esaote, Genoa, Italy) with a multifrequency linear transducer of 5-7.5 Mhz. Approximately 0.3 mL/10 kg of contrast medium (Sonovue, Bracco), was injected twice IV. The CHUS study was repeated each week, until complete dissolving of the lesions. The images were recorded and reviewed using subjective and semi-quantitative methods for description of the lesions including their number, shape, size, sharpness, echogenicity, and evolution over time.

Results: CHUS performed 30 minutes after biopsy revealed hypoechoic lesions in all kidneys, visible as 2 tracts (n ¼ 17), 1 tract (n ¼ 13), triangular or nodular area (n ¼ 2), or area þ tract (n ¼ 1). One Beagle had perirenal bleeding. Ten lesions were small to mild and 23 lesions were moderate to large. Ten tracts had a thickened nodular base, subcapsular at the renal cortex. At week 2, all lesions were still present but slightly (n ¼ 8) or significantly (n ¼ 25) decreased in size and sharpness, with increased echogenicity. The perirenal bleeding disappeared. One Beagle showed ureter and renal pelvic dilation and anechoic round lesions were present compatible with hematoma. At week 3, 27 kidneys were normal, 5 showed vague lesions and one was still abnormal showing the same features as at week 2. At week 4, all kidneys were normal. Discussion/Conclusion: CHUS enables assessment of the evolution of post-biopsy renal lesions and can be an adequate screening method. Moreover, 14-gauge needles were not found to be quite as traumatic as generally believed. Introduction/Purpose: Variable ultrasonographic measurements of normal and abnormal canine adrenal glands have been reported.1-3 Moderate correlation was present between ultrasonographic and gross measurements of thickness (dorsoventral measure) for both glands, but no correlation was found for width (mediolateral measure) or length.1 However, the repeatability of such measurements has not been determined. The purpose of the study was to assess the intra-and interobserver variability of sonographic measurements of adrenal glands. Materials and Methods: Six healthy Beagle dogs were used. Three observers, two ECVDI diplomates (J.H.S. and V.B.) and one ECVDI resident (P.P.) scanned both adrenal glands of each dog three separate times. The observers were asked to measure the maximal length (craniocaudal direction), height (dorsoventral direction) and width (mediolateral direction) of each gland. Length and height of both cranial and caudal poles were measured in longitudinal plane. Height and width of both cranial and caudal poles were measured on transverse images. A random effects model was fit to the data with dog, observer and measurement within observer as random effects.

Measurements of the length of both glands had the highest intra-and interobserver variability. Height of the caudal pole measured on longitudinal images had the lowest intra-and interobserver variability. Measurements that also presented low intra-and interobserver variability were: height and width of the caudal pole on transverse images and height of the cranial pole but on longitudinal images only. Discussion/Conclusion: Length measurements appeared unreliable due to difficulties in obtaining the complete length of the gland on one image and in defining clearly the cranial border of the right gland. Measurements of height and width of the cranial pole of both glands on transverse images were unsatisfactory due to its inconsistent shape and position. Hence, height of the caudal pole measured on longitudinal images should be preferentially used, as it is the less variable and the most accurate sonographic estimation of gross adrenal size.

Grooters AM, Biller DS, Merryman J. Ultrasonographic parameters of normal canine adrenal glands: comparison to necropsy findings. Vet Radiol Ultrasound 1995; 36 (2) Introduction/Purpose: Dogs and cats presented with signs of upper airway obstruction frequently have to undergo bronchoscopic examinations for a final diagnosis. Although usually followed by surgery to relieve the obstruction the procedure carries a high risk due to the need for general anesthesia. The goal of this study was to assess computed tomography and virtual bronchoscopy in awake dogs and cats with signs of upper airway obstruction. Materials and Methods: Five dogs and 2 cats presented with signs of upper airway obstruction were included. Final diagnosis was made by bronchoscopy, physical examination of upper airway under general anesthesia, fluoroscopy or histopathology. The patients underwent head and neck CT examination with a 16 detector-row helical CT unit without sedation or general anesthesia. All patients received oxygen during the entire CT examination.

Five patients presented with respiratory distress (dogs 1, 2 and 4 and all cats) and one with chronic cough (dogs 3 and 5). Other clinical signs included cough (dog 2) and stertor (cat 1). The imaging findings and final diagnosis were: Dog 1: diagnosed on CT as focal severe tracheal collapse and final diagnosis of tracheal collapse on bronchoscopy. Dog 2: diagnosed on CT as tracheal stenosis; final diagnosis on necropsy of severe submucosal granulation tissue and stricture due to scar tissue. Dog 3: diagnosed on CT as everted sacules and elongated soft-palate; final diagnosis on bronchoscopy of everted sacules, elongated softpalate and grade 2-3 laryngeal collapse. Dog 4: diagnosed on CT as laryngeal paralysis and final diagnosis of laryngeal paralysis on physical examination. Dog 5: diagnosed on CT as tracheal collapse and final diagnosis of tracheal collapse on fluoroscopy. Cat 1: diagnosed on CT as mass, probable neoplasia; final diagnosis on histopathology of pyogranulomatous laryngitis. Cat 2: diagnosed on CT as regional wall thckening (inflammation or neoplasia); final diagnosis of severe suppurative laryngitis. Discussion/Conclusion: CT of awake patients using the VetMousetrap was successful in accurately demonstrating upper airway disease. CT VetMousetrap imaging was a fast and non-invasive procedure that allowed simultaneous oxygen administration with minimal stress to respiratory compromised patients. Introduction/Purpose: The purpose of the study was to design, construct and test a device for clinical computed tomographic imaging of awake dyspneic cats. This abstract reports on device design and clinical testing on normal eupneic cats. Materials and Methods: Ten normal cats were placed inside the device for measurement of carbon dioxide (CO2), fraction of inspired oxygen (FiO2), temperature at rest. Twenty two awake normal cats were imaged with two different kVps (80 and 120) and 2 different pitches (0.562 and 1.75). The signal intensity of the pulmonary parenchyma (SI-lung), signal intensity of background (SI-backgr.), contrast, noise, signal-to-noise ratio (SNR) and contrast-to-noise ratio (CNR) were calculated. Images were evaluated for sharpness of liver margins, motion artifact, aliasing and windmill artifacts. Significance was set at P ¼ 0.001.

The final device design was a transparent acrylic tube with a wall thickness of 5 mm, outer diameter of 21 cm, 40 cm in length, with oxygen and catheter line access, and no metallic components. No artifacts related to the device were detected and the cats could be visually monitored throughout the imaging and quickly removed. Oxygen and catheter line access was uncomplicated and effective. Carbon dioxide levels were mildly elevated by 5 minutes. The mean FiO2 at 25 minutes was 68%. Temperature levels inside the device increased over time with the highest mean of 26.81C at 30 minutes. Overall, 75 out of 80 (94%) examinations were judged to have absent or minimal motion artifact. Higher kVp protocols had higher SNR and CNR. Higher pitch protocols had lower noise, higher SNR and CNR, and less motion artifact. Low pitch protocols were deemed clinically better because of significantly lower windmill and aliasing helical artifacts. Discussion/Conclusion: The VetMousetrap was an effective device for imaging awake cats with minimal elevation in carbon dioxide and significant elevation in ambient oxygen. The only clinically relevant artifacts noted were windmill and aliasing in the large pitch (1.75) protocols. Motion was not a clinically significant issue. Section of Diagnostic Imaging, Department of Companion Animal Clinical Studies, 3 Section of Epidemiology, Department of Production Animal Studies, Faculty of Veterinary Science, University of Pretoria, Onderstepoort 0110, South Africa Introduction/Purpose: Canine chylothorax is a problematic condition to manage due to poor response rates with both medical and surgical management. Surgical ligation of the thoracic duct (TD) has been the most frequently reported treatment; however success rates with TD ligation alone have been reported to be between 53% and 59%. Failure to ligate all of the TD branches has been described as the most common cause of surgical failure. Lymphography has conventionally been conducted by catheterization of a mesenteric lymphatic, but has recently been described using the less invasive technique of percutaneous injection into a popliteal lymph node. Computed tomography (CT) has been shown to be superior to radiographs in detecting TD branches and providing TD topographical information. This study assessed the CT comparability of the two lymphography techniques and determined the equivalency of helical and sequential CT modalities for TD lymphography. Materials and Methods: Seven beagles underwent both mesenteric lymphography (ML) and ultrasound guided percutaneous popliteal lymphography (PPL). Iohexol 300 mg/ml was administered at 1 ml/kg for both techniques. For the PPL, contrast medium was administered at 100 ml/hr where as for ML it was given as a bolus over a one minute period. Helical CT was initiated immediately after completion of contrast medium injection, followed by sequential CT of the same area. For both administration techniques and CT modalities, images were taken at the mid-vertebral bodies of T9-L1. Images were assessed for the total number of TD branches and their position relative to the aorta, as well as the cross-sectional area and mean Hounsfield units (HU) for the largest TD branch. The effects of administration technique, CT modality, vertebral site and animal on the observed TD number were assessed using zero-truncated Poisson regression, while their effects on cross-sectional area and mean HU, were analyzed using multiple linear regression.

The number of TD branches detected did not differ significantly between the two contrast administration techniques (P ¼ 0.256) or CT modalities (P ¼ 0.417), although the count ratio (CR) indicated a trend for the PPL administration technique (CR ¼ 0.830) and the helical CT modality (CR ¼ 0.876) to detect slightly fewer TD branches. ML and helical CT both resulted in significantly greater largest TD branch cross-sectional area and mean HU (Po0.001).

This study suggests that PPL is an acceptable alternative to ML for identifying TD branches when using either helical or sequential CT. Introduction/Purpose: Thoracic computed tomography (CT) is superior to conventional radiography by eliminating superimposition and enhancing resolution, thus improving detection of pathology. Acquisition timing during computed tomography angiography (CTA) is critical to optimize contrast medium enhancement. Due to the large differences in veterinary patients optimal timing of contrast enhancement cannot be standardized. This study serves to determine if there is any statistically significant difference in the two CTA techniques, namely bolus tracking (BT) and time delay (TD) utilizing a test bolus in terms of enhancement, scan time and CT dose index (CTDI). Materials and Methods: 6 healthy beagles (36-48 months), underwent dual slice thoracic CTA using the BT and TD techniques. The time interval between studies was 6 weeks and the order was randomized. Images were interpreted at the level of the cranial aspect of the 8th thoracic vertebra (T8), mid and caudal T8, where the aorta (Ao), pulmonary artery (PA) and caudal vena cava (CVC) were visualized simultaneously. Images were viewed in a mediastinal window. In this within-subject study design the BT and TD techniques were compared with respect to the Hounsfield units in the Ao, PA and CVC during the arterial and venous phases, respectively. Findings were compared using the students paired t-test and in each case the result was confirmed by the non-parametric Wilcoxon's method pairs signal-ranks test. The Shapiro-Wilkes test for normality was done. Additionally, the 95% confidence interval for the difference of the means between the BT and TD techniques was rewritten to a 95% confidence interval for the BT as a ratio of the TD studies, where the latter was considered the gold standard for an individual. (GSD) . Although various studies dealing with radiographic appearance of the lumbosacral region in GSD exist, radiographic anomaly of the lumbosacral transition has not been confirmed by an adequate population size so far. The aim of this study was to analyse the anatomical variance and frequency of early onset pathologies of the lumbosacral transition between GSD and other large breed dogs. Furthermore, a genetic analysis of radiographic signs related to lumbosacral stenosis was performed for the GSD.

A total number of 1.267 lateral lumbosacral radiographs of clinically sound large breed dogs -733 of which GSD (mean age 15 months) and 534 dogs of other breeds (mean age 31 months) -were studied using established methods (1, 2) . Nine specific findings and 17 anatomic dimensions were assessed with emphasis on malalignment and vertebral canal height at the lumbosacral level as well as lumbosacral osteochondrosis and transitional vertebrae. The results of radiographic examinations of 572 GSD (95% born in [2000] [2001] [2002] [2003] [2004] [2005] [2006] [2007] with complete pedigree data were included into heritability estimation.

There were significant differences between the GSD and the other breeds in most of the assessed morphologic parameters. The prevalence of lumbosacral osteochondrosis and 4 other specific findings was significantly higher in the group of the GSD. The ratio of ventrodorsal lumbosacral subluxation was 2:1 between GSD and other breeds and was significantly higher for the GSD. Dorsoventral dimension of the lumbosacral vertebral canal was significantly lower for the GSD in any location, whereas the ratio of caudal lumbar vertebral canal height to cranial sacral vertebral canal height was significantly higher in this group. Genetic analysis of the collected parameters in the group of the GSD resulted in moderate to high heritability estimates for specific findings (h 2 0,22-0,64) and predominantly high heritability estimates (h 2 0,31-0,73) for anatomic dimensions. Discussion/Conclusion: Compared with other breeds there is a primary aberrant anatomic conformation of the lumbosacral transition and a higher frequency of promoting factors such as sacral osteochondrosis, malalignment and inherently narrow vertebral canal, that predispose GSD's to lumbosacral stenosis. Genetic determination of these predisposing factors suggests that a selective breeding scheme based on radiographic parameters could be promising. Introduction/Purpose: Previous studies reported the German Shepherd Dog (GSD) being overrepresented in the occurrence of cauda equina compression syndrome (CES) as a consequence of degenerative lumbosacral stenosis (DLSS). A congenital predisposition for the early degeneration of the lumbosacral intervertebral disc has been suspected. Male dogs show a higher incidence of CES and multiple anatomic variations such as lumbosacral transitional vertebra or osteochondrosis of the sacrum are reported as predisposing factors. The purpose of the present prospective study was to evaluate degeneration of the lumbosacral intervertebral disc in the GSD compared to other breeds.

The lumbosacral spines of 126 GSDs and 54 other large breed dogs were examined using Magnetic resonance imaging (MRI). The signal intensities of the entire disc and the nucleus pulposus were determined. The dimensions of the disc and the lumbosacral angle were measured, and step formation as well as conformation of the lumbosacral junction was recorded. In addition the degree of degeneration of each intervertebral disc was graded using three different classification systems. Results of GSDs without clinical signs of CES were compared to healthy dogs of other breeds and to dogs showing signs of CES. Statistical regression analysis was performed to identify possible associations between the morphology of the intervertebral disc and the training level of the dogs. Results: A significant difference in the severity of degeneration of the lumbosacral intervertebral disc in GSDs in comparison to other breeds was found. Degeneration of the lumbosacral intervertebral disc was not associated with the degree of degeneration of other lumbar discs. Morphologic differences concerning gender and breed could be deminstrated. Discussion/Conclusion: We concluded that male dogs and GSD dogs have a higher risk to develop degenerative changes of the lumbosacral intervertebral disc. Introduction/Purpose: When interpreting MR images in veterinary diagnostic imaging a thorough understanding of brain anatomy and species characteristics are of particular importance. Although feline and canine brains have been examined for decades most other mammals are only occasionally scanned. Materials and Methods: Scans of anatomical of various mammalian species specimen were analyzed and retrospectively compared. Results: Metatheria (Marsupials): The corpus callosum (and septum pellucidum) is only present in eutherian mammals, kangaroos and other marsupial brains are remarkable for their lack of this and have a well developed rostral commissure that connect the hemispheres. In the opossum the cortical surface only shows two major furrows, one is the lateral rhinal fissure, separating a large ventrolateral rhinencephalon from the neocortical part and the other is the orbital fissure at the rostrodorsal end of the hemisphere. In higher marsupials, as the kangaroo a gyrencephalic brain is present. The olfactory bulbs are huge and solid, without an olfactory recess. Lagomorphs: In hares the surface of the pear shaped brain is smooth with only the lateral rhinal fissure and a dorsal longitudinal fissure, the marginal sulcus. Neocortical and older cortex divisions almost equal in size and the large ammons horn parallels the neocortical surface over a large extent. In contrast to higher mammals, the paleocortex is less developed, thinning rather in its dimension when running caudad. In the cerebellum, the vermis is straight and the most pronounced structure is the ventral flocculonodular lobe that extends far laterally. Ungulates: The scans nicely reflect the important characteristics of artiodactyl brain, e.g. the dominant position of the visual and olfactory systems. This is seen in the large diameter and size of the eye, optic nerve/chiasm, and rostral colliculus respectively, as well as in the impressive dimension of the olfactory bulb and tracts as well as of the piriform lobe. The cortex of ungulates telencephalic hemisphere is densely folded and the major sulci show a high variability in ramification. most pronounced in the horse. The corpus callosum is well developed in artiodactyls. In the diencephalons a large pituitary gland and pineal gland are striking. Discussion/Conclusion: For the identification of structural deviations from normal anatomy in species other than cats and dogs a thorough understanding of brain anatomy is necessary. Introduction/Purpose: Gliomatosis cerebri is a neoplastic condition of the central nervous system characterized by widespread diffuse infiltration of the brain and spinal cord by neoplastic glial cells with preservation of the brain architecture. The condition is rare in all species and has been confirmed histologically in dogs. A case report of a single dog described MRI findings as hyperintense and ill-defined in T2-w and FLAIR sequences within one hemisphere of the forebrain and extension to the brainstem and cerebellum. The purpose of the present case series is the description of the MR features in five dogs and their correlation to histopathology.

The present case series includes 5 dogs, one Yorkshire Terrier and four Bearded Collies, presented because of neurological disorders including circling (3/5), central blindness (2/5) or decreased menace response. The disease was localized to the forebrain. MRI was performed at two different institutions: 3 dogs were examined using a 0,3T MRI unit and two dogs using a 1,5 T unit. Minimal sequences consisted of T2-w sequences, T1 sequences pre-and postcontrast in different planes and FLAIR (3/5). Results: MRI findings included diffuse hyperintense areas in T2 w sequences in all dogs. The ill defined lesions were hypointese (4/5) or isointense (1/5) in T1 w images. No contrast uptake was present in any case. The lesions were bilateral with one side always more severely affected and continued along the white matter extending partially into the gray matter with contact to the brain surface. In all cases the thalamus and cerebellum were affected. Mild to moderate mass effect and midline shift was present (5/5) and there were signs of increased intracranial pressure with flattening of the sulci, transtentorial herniation (3/5), herniation through the Foramen magnum in two of these dogs, and periventricular edema on FLAIR images (3/5). Other signs included syringohydromyelia (4/5) in the cervical spine. Additional hyperintense areas interpreted as infiltration within the spinal cord were visible in two cases. Histopathologically all cases revealed a glial cell population with homogenous morphology infiltrating diffusely the white and gray matter of both hemispheres, the thalamus, brainstem and cerebellum. In two out of three cases an infiltration of neoplastic cells in the spinal cord was found. Discussion/Conclusion: MRI features of all five dogs were very similar in distribution; signal behavior, lack of contrast uptake, had signs of mass effect and increased intracranial pressure. Descriptions in men are similar except mass effect, which might be explained by the later presentation of dogs and progression of disease. Distribution and extent of the lesion corresponded well to histopathological findings. On MRI interstitial edema could not be differentiated from cellular infiltration, however, hypointense signal in T1-w sequences in more severely affected areas could be a sign for brain edema (high water lesion), which was confirmed by histology. Because four of the five presented dogs were Bearded Collies we hypothesize that Gliomatosis cerebri in this breed may have a hereditary background.

C. Haarstrick, C. Niesterok, G. Oechtering, M. Alef, I. Kiefer. Department for Small Animal Medicine, University of Leipzig Introduction/Purpose: In small animals the soft palate can only be examined by oral inspection, such as pharyngoscopy enabling only examination of the organ surface. Investigation of the internal pattern is possible by CT or MRI, but the expensive equipment and required anaesthesia form limitations of performing these examinations. The examination of the palate gains importance due to the increasing popularity of brachycephalic dog breeds. Because of their anatomical conditions and problems during recovery from anesthesia, less risky procedures, such as ultrasonography, are needed. The aim of this study is to characterize the normal sonographic anatomy of the soft palate and to define predetermined landmarks for ultrasonographic measurements of the soft palate diameters in dogs.

The study was performed on seven canine heads. The dogs were euthanized because of various diseases not related to the head. Ultrasonography was performed via a submental approach. Depending on the shape of the ultrasonic probe also intraoral ultrasonography was performed. The studies were performed in three or four probe positions in sagittal and transversal direction. Thickness of the soft palate was measured at two defined locations. Subsequently the detected anatomic structures were reexamined by submental approach in a water bath. Afterwards the single anatomic layers were dissected starting with the submental layers working up towards the soft palate. After every single preparation a sonographic examination was carried out. Finally the isolated soft palate was analyzed in the water bath by ultrasound. The produced images were assessed regarding recognizable structures and analyzed for image quality by a subjective score system.

The transition between the hard palate and the soft palate as well as the middle section of the soft palate was distinguished from the surrounding tissue by submental approach in all cases. Evaluation of the echogenicity of the soft palate parenchyma was difficult. Also visualization of the caudal part of the velum depended strongly on testing conditions. The differentiation between cutaneous and subcutaneous tissue, sublingual musculature and tongue is easily achievable. Measurement of the thickness of the palate was done at two defined locations. The first was defined one centimeter caudal to the transition between the hard and the soft palate. The second more difficult location was determined at the caudal soft palate adjacent to the presumed palatopharyngeal arch which is identified as hyperechogenic line caudodorsal to the soft palate. Discussion/Conclusion: In this study the sonographic appearance of the canine soft palate using cadavers was described. The submental approach enables the differentiation between intraoral structures. Evaluation of the echogenicity and length or thickness of the soft palate is limited, due to influence of surface pressure by the investigator for instance. 

The normal echographic anatomy of the esophago-gastric junction and the gastro-duodenal junction has not been studied in the normal cat. The aim of the present prospective study is to describe the normal aspect of the pylorus, proximal duodenum and the cardia of the stomach.

A prospective study was performed in 15 fasted adult cats (11 purebred, 2 crossbreed and 2 domestic shorthair cats) free of gastro-intestinal disease and deworming. Cats were scanned in dorsal recumbency, without sedation. A linear probe (12 or 18 MHz) was used for better resolution during the study. Measurements were obtained for caudal esophageal wall thickness just cranial to the cardia (from inner mucosa to external serosa in longitudinal section: Ew), cardia wall thickness (Cw, in longitudinal section), pyloric wall thickness (Pw, in longitudinal section), muscularis of the pyloric thickness (Mp, in longitudinal section), length of the thicker part of the cranial duodenum submucosa (Dl), gastric lymph node thickness (GLNt). Freezed images and videoclips were recorded for retrospective review. Several cats free of digestive symptoms were excluded of the study because of: incomplete emptying of the stomach, abnormal gastric thickening, abnormal digestive lymphadenopathy, incomplete alignment of the gastroduodenal junction, agitated cats.

Images obtained during the prospective study were reviewed by two ECVDI diplomates (LC and DR). The most caudal part of the esophagus and the gastro-esophageal junction were identified in 10/15 cats (66.7%). Submucosa of the most cranial part of the duodenum was thicker in 13/15 cats (86.7%). In all cats, the muscularis part of the cardia and the pylorus was thicker. The former was strip shaped and the latter had a short triangular shape. On the 15 cats studied, only 3 measurements could be systematically performed: Pw, Mp, GLNt. Measurements (1st number correspond to mean, minimal and maximal values are in brackets, confidence intervals of 95%) were as follows: Ew: 3.1 mm (2.3-3.8 mm); Cw: 4.7 mm (3.6-5.1 mm); Pw: 4.2 mm (3.1-5.2 mm); Mp: 2,5 mm (2-3.6 mm); Dl: 7.4 mm (3-13.7 mm); GLNt: 4.1 mm (3.4-5.1 mm). Discussion/Conclusion: Linear probe seems to be an efficient tool to identify the pylorus in cats but limited for the cardia and abdominal esophagus especially in fatty cats. The description of the normal echographic anatomy of the ß borders ý of the stomach could help further identification of cardia and pyloric disease in cats.

Olivier Taeymans, DVM, PhD, Dipl ECVDI, Dominique Penninck, DVM, PhD, DACVR, Dipl ECVDI. From the Department of Clinical Sciences, Foster Small Animal Hospital, Cummings School of Veterinary Medicine, Tufts University, North Grafton, 01536 MA, USA Introduction/Purpose: Few reports exists on normal 1,2 and pathological ultrasonographic appearances of the structures at the ileocecocolic junction (ICJ) in the cat. During routine abdominal ultrasound of feline patients, we observed that it is not uncommon to find changes in this area particularly in cecal wall thickness, cecal contents, mesenteric root fat, and regional lymph nodes. We hypothesize that these changes are not incidental, and may be representative of gastrointestinal (GI) disease in cats. Materials and Methods: The hospital information system of TCSVM was searched for ultrasound records of feline patients containing a description of abnormalities at the level of the cecum, ileocecocolic area, ileocolic junction, ileocecal (cecal) lymph nodes, or colic lymph nodes between January 2002 and June 2008. Patients having gastrointestinal neoplasia were removed based on available patient outcome. This resulted in 29 cats, 14 males and 15 females. Mean age was 87 months. Results: There were 18/29 cats with enlarged cecal lymph nodes. Focal hyperechoic mesenteric fat was noted in 18/29 cats. Mild fluid accumulation at the ileocecocolic junction was seen in 7/29 cats. Fluid content was noted in 7/29, and foreign cecal material in one cat. Six cats showed a rounded shape of the cecum. The cecal wall was thickened in 19/29 cats, 6/19 showing wall alterations. The ileal wall was mildly thickened in 6 cats, and 4 of these had altered wall layers. 18/29 cats had no ultrasonographic evidence of GI changes in other portions of the GI tract, and 17 of these 18 cats were symptomatic for GI disease. 8 cats had fine needle aspirate or biopsies. Histopathology confirmed mild colitis/enteritis in one case, and mild reactive lymph nodes in another case. Cytology showed reactive lymph nodes in two cases.

Our results show that ICJ changes was the only GI finding in 62% of our population, and that all except one of these cats were symptomatic for gastrointestinal disease. These results support that ultrasonographic changes noted at the level of the ICJ may have clinical relevance, and may represent typhlitis. Introduction/Purpose: Comparison of the kidney long axis dimension to the second lumbar vertebra (L2) on ventrodorsal projections has proved to be an accurate method for assessment of kidney size on radiographs. In the adult dog, the kidney length has been reported to be 2.5-3.5 times the length of L2. 1 The aim of the study was to test the hypothesis that the suggested maximum normal kidney size obtained from radiographs is too high and also to evaluate whether breed type (brachycephalic, doliocephalic, mesocephalic), age, gender, weight and body condition of the dog have an impact on kidney size.

Abdominal radiographs of 100 dogs with no evidence of concurrent disease that might have an effect on renal size and that had no clinical or ultrasonographic evidence of renal disease were included in the study. All animals had normal renal blood profile and electrolytes. The animals were radiographed in right, left lateral and ventrodorsal recumbency although not all animals had all projections performed. The number and type of projections depended on necessity, and the reason for the clinical investigation. Radiographic length of the kidneys was measured on digital images and compared to the length of the second lumbar vertebra. The ratios of kidney length to length of L2 (kidney/L2) of each dog in all available views were calculated from measurements obtained from radiographs.

The length of kidneys of 3.5 times the length of L2 was above the 97.5th percentile of the distribution for all the ratios except the left kidney/L2 measurements on VD views. Significant differences (P ¼ 0.008) in kidney/L2 ration between the groups were present, especially between the brachy-and dolicocephalic groups. The brachycephalic group appeared to have the highest kidney/L2 ratio. The dolicocephalic group never reached a kidney/L2 ratio of 3.5. There was no significant difference between genders. Higher kidney/L2 ratios were observed in intact animals compared to neutered animals. The weight did not have a significant impact on kidney/ L2 ratios although there was very close to a significant difference between weight categories (o10, 10-30, 430 kg) with the largest difference between the categoryo10 and the other two weight groups. Discussion/Conclusion: Mild renomegaly should be suspected if the kidney/L2 ratio is equal to or larger than 3.5 in dogs other than brachycephalic breeds. The kidney/L2 ratio of 3.5 in our study was found only within the brachycephalic dog group. All other groups had slightly lower ratios. It is suspected that this may be due to shorter vertebrae rather than longer kidney length in brachycephalic dogs but further studies are needed. Introduction/Purpose: Granulomatous meningoencephalitis (GME) is an acute progressive and often fatal inflammatory disease of the central nervous system (CNS) mainly affecting small and toy breed dogs, often young and middle-aged. A definitive diagnosis of GME can only be achieved after postmortem and histological examinations. The purpose of this study was to describe the ultrasonographic findings, using color Doppler mapping and spectral analysis in dogs with postmortem histologically confirmed GME.

Retrospective study of transcranial Doppler sonography (TCDS) examinations performed between August 2005 and September 2006. During this period, 116 dogs with clinical signs of CNS disease were submitted to TCDS. But only 11 dogs were selected with GME diagnosis confirmed at postmortem examinations. B-mode images were taken to evaluate brain parenchyma anatomy and echotexture. Color Doppler mapping was performed to identify the circle of Willis on both hemispheres. After identifying the vessels, pulsed Doppler was accomplished on each one, obtaining resistive index (RI).

Seven female (64%) and 4 male dogs (36%) presented GME diagnosis confirmed at postmortem and histological examinations. The mean age was 6 years (range: 3-15 years). Decreased parenchymal echogenicity was observed in 9 dogs (82%); 8 dogs presented ventriculomegaly. Focal lesions were detected in 6 dogs (54.5%). Color Doppler study revealed conspicuous vessels in the circle of Willis of 10 dogs (91%). Pulsed Doppler detected the six major cerebral arteries in 7 dogs. In 2 dogs only one artery was not depicted and in one dog, two cerebral arteries were not outlined. Pulsed Doppler showed normal and high RI values in the outlined arteries. Discussion/Conclusion: TCDS findings were correlated with pathologic examination diagnosis. There was a positive correlation between ultrasound and pathology findings concerning the focal lesions. Color Doppler study showed conspicuous intracranial vessels using minimum color gain, possibly due to the increased diameter. Vascular occlusion and stenosis caused by space-occupying masses or inflammatory infiltrates may be the possible cause of undetectable vessels. CSF pressure is normal or high in GME patients. The dogs showed normal or high RI values. Focal granulomas and vascular changes in different locations and various degrees of presentation can modify microvascular impedance. Sonographic findings in this study corroborate previous GME pathology reports. Although it is known that an in vivo diagnosis is difficult, TCDS can be a useful instrument to evaluate CNS when GME is considered. Introduction/Purpose: In recent years, with the new development of high resolution, electronic broad band transducers, ultrasound (US) is being considered as an optimal image technique to evaluate normal anatomy and abnormalities of the peripheral nerves. To the author's knowledge, information regarding the appearance and ultrasonographic approaches of peripheral nerves in the cat has not been documented. The aim of the present study was to determine the ultrasonographic appearance and the approaches to assess the sciatic nerve (ScN) in cats. Materials and Methods: Anatomical nerve study: Eight limbs from four feline cadavers were used for anatomical dissections. Another two fresh feline cadavers were frozen and cut in cross sections to determine the anatomic landmarks. Ultrasound nerve study ''in vitro'': Eight pelvic limbs from four fresh feline cadavers were employed. All limbs were scanned using a 4-13 MHz linear transducer (MyLab 70, Esaote s ). The examinations were performed in three regions: glutea, femoris and poplitea. The accuracy of the ultrasound localization was demonstrated by injecting ink around the targeted nerves which were, then, immediately dissected. The obtained ultrasonographic images were compared with the anatomical cross sections to facilitate their interpretation. Ultrasound nerve study ''in vivo'': Five healthy adult experimental cats were sedated to perform the ultrasonographic examination of the ScN using the technique described for the ''in vitro'' study.

The ultrasonographic images of the ScN showed a good correlation with the anatomical nerve study. The ScN was easily identified on glutea, femoris and poplitea regions. On the longitudinal view, the ultrasonographic appearance of the ScN was a tubular hypoechoic structure with internal echoes, outlined by hyperechoic lines. On the transversal view, it was observed as an ovoid hypoechoic structure surrounded by a hyperechoic rim. There were no differences between the ultrasonographic appearance of the ScN in fresh cadavers or healthy cats. Discussion/Conclusion: This study shows the usefulness of ultrasound to evaluate the ScN in cats, from its origin until its division into the peroneus communis and tibialis nerves. The ultrasonographic appearance of the ScN was as a hypoechoic structure with internal echoes, outlined by hyperecogenic lines.

Benigni, L., Corr S. A., Lamb C. R. Ultrasonographic assessment of the canine sciatic nerve. Vet. Rad. Ultrasound. 2007. 48 (5) : 428-433. Echeverry D., Gil F., Laredo F., Ayala M. D., Belda E., Soler M., Agut A. Ultrasound-guided block of the sciatic and femoral nerves in dogs: A descriptive study. Vet J. 2009 (In press). Introduction/Purpose: Ultrasound examination performed by and experience sonographer is becoming the elective exam when suspecting of a gastrointestinal obstruction, replacing contrast radiographic studies for confirmation of these. To the author knowledge only hyperechoic interface foreign bodies with acoustic shadowing have been described in the literature. This report describes the appearance of hypoechoic foreign bodies, specifically rubber pacifiers.

Three dogs with gastro-intestinal clinical signs where referred for an ultrasound. In all of them, a throughout exam of the gastro-intestinal tract as of the remaining abdominal cavity was made (Sonosite Titan with microconvex transducer from 5-8 MHz). A gelatin model (collagen based) was made containing 3 rubber pacifiers (1 latex and 2 silicone). An ultrasound scan was performed to access the ultrasonographic appearance of the pacifiers without the interference of gas or ingesta from the gastrointestinal tract. Results: Three foreign bodies were detected ultrasonographically. One foreign body was detected in the stomach, 2 were detected in the small bowel (jejunal loops). Only two animals had ultrasonographic signs of mechanic Ileus, the ones that were detected in the small bowel, the one visualized in the stomach was only partially obstructive. All of the foreign bodies shared the following ultrasonographic appearance: hypo to anechoic band with a hyperechoic capsular interface. Their shape varies according to the plane sectioned as well as the part of the pacifier scanned. The teats had an oval shape, were composed of and hypo to anechoic band with strong hyperechoic contours, they were open on one side, allowing for the content of gastrointestinal tract to enter it. All of the foreign bodies detected were surgically removed. The pacifiers visualized in the gelatin model had low or null acoustic impedance, and therefore were anechoic with a hyperechoic contour. Discussion/Conclusion: Rubber pacifiers, latex and silicone based, are present in many households with children and seem to be a commonly ingested foreign body by our pets. Their specific ultrasound appearance makes their diagnosis easy by ultrasound, since its ultrasonographic features are patognomonic. Rubber pacifiers size and position as well as the size of the animal, will determine if they are completely or partially obstructive. Nonetheless even in partially obstructive cases, their surgical removal is advised. Introduction/Purpose: Recently, a fatal syndrome characterized by mesenteric lymphadenopathy and hypergammaglobulinaemia has been recognized in domestic ferrets. Histological lesions are similar to those of the non effusive form of feline infectious peritonitis (FIP), and with immunohistochemical techniques coronavirus antigen in several affected organs have been detected. A recent mutation or shift in the ferret enteric coronavirus (FECV) is thought to be the cause of this syndrome (known as ferret systemic coronavirus infection). This report describes radiographic and ultrasonographic findings observed in the abdomen of seven ferrets with confirmed diagnosis of systemic coronavirus infection. Materials and Methods: 7 domestic ferrets (2 female and 5 male; 8-18 months) were presented at the Veterinary Teaching Hospital with a history of lethargy, hyporexia or anorexia, and weight loss. Diarrhea was recorded in 3 and vomiting in 1 of the ferrets. Abdominal masses were palpated in 4 animals. Pale mucous membranes, abdominal enlargement, and abdominal pain were observed in 2 of the ferrets. Faecal examination, complete blood count (CBC), serum biochemical profile, abdominal radiographs (right lateral and ventrodorsal views) and abdominal ultrasound (linear high frequency, 11-13 MHz, transducer) studies were performed in all the ferrets. Hyperglobulinaemia with monoclonal gammopathy (7/7), anaemia (5/7) and severe neutrophilic leucocytosis (4/7) were detected. Ferret systemic coronavirus infection was diagnosed based on histological findings (pyogranulomatous inflammation of the viscera and peritoneum) and on immunohistochemical confirmation of coronavirus antigen presence in affected tissues.

Radiographically, signs of emaciation (loss of lumbar musculature) and decreased serosal detail were recorded in 5 animals. Presence of mid abdominal soft tissue mass was observed in 4 ferrets. Other radiographic findings include moderate splenomegaly and mild to moderate dilation of the intestinal tract with gas (2/7), pendulous ventral abdominal wall and left kidney enlargement (1/7). Sonographically, a mild amount of echogenic effusion and thickening of the peritoneum, omentum, and mesentery with a hyperechoic and hyperatenuated aspect was the most common finding (6 ferrets). Heterogeneous, ill-defined soft tissue masses, surrounded by hyperechoic fat and mesenteric lymphadenopathy were imaged in 5 animals. Mild to moderate splenomegaly was observed in 3 patients and inguinal inflammation accompanied by signs of vascular thrombosis was recorded in 1 ferret.

Although definitive diagnosis of systemic coronavirus infection is based on histological and immunohistochemical results, a tentative diagnosis can be made on the basis of the animal's history, clinical signs, and clinicopathological, radiographic, and ultrasonographic findings. Imaging signs of peritonitis and the presence of soft tissue masses are highly suspicious of systemic coronavirus infection. Aleutian disease, lymphoma, and chronic inflammatory bowel disease should be included in the differential diagnosis.

Susanne Stieger-Vanegas, Christopher Cebra, Jason Wiest. Department of Clinical Sciences, College of Veterinary Medicine, Oregon State University, Corvallis, Oregon, USA Introduction/Purpose: In llamas and alpaca colic symptoms can be obscure. Additionally, the cause of colic symptoms can be difficult to localize due to the limited availability of diagnostic techniques evaluating the abdomen. Colic symptoms caused by bacterial enteritis or enteritis caused by Eimeria macusaniensis can be very difficult to differentiate from colic symptoms caused by intestinal strangulation, entrapments, or intraluminal obstruction. The lack of the ability to differentiate between medical and surgical causes of colic symptoms affects treatment and therefore the survival rate and outcome of New World camelids with colic symptoms. Computed tomography has been performed in llamas 18 hours after barium was given with an orogastric tube.1 However, 18 hours can be a long delay time for medical or surgical treatment in a patient with colic symptoms. The purpose of our study was to evaluate a new gastrointestinal imaging protocol using iodinated contrast media diluted in water. The objective of this study was to evaluate the feasibility of this new CT protocol to evaluate the gastro-intestinal tract in clinically normal New World camelids.

Three hours prior to the computed tomography study, three alpacas and two llamas without clinical signs indicating gastro-intestinal disease received iodinated contrast media diluted in water via an orogastric tube. Following sedation, the animals were positioned in sternal recumbency on the CT table and axial, sagittal and coronal images were obtained. To enhance the definition of the organs and gastrointestinal wall intravenous iodinated contrast was applied and images were obtained with a 60 sec delay.

In all animals the first, second and third compartment of the stomach (C1, C2, C3), small intestine, spiral colon and ascending colon were identified. In all animals the cranial and caudal glandular saccules of the stomach were observed. The ampulla of the duodenum was well seen and could be separated from the remainder of the duodenum. The walls of the intestine could be identified. In all patients the contrast reached at least the cranial aspect of the spiral colon. Introduction/Purpose: Although radiographs are essential in the evaluation of thoracic diseases, the findings may not be specific or limited. In dogs and cats, the use of ultrasonography in the diagnosis of thoracic extracardiac diseases demonstrates an important role, providing images in real time and providing additional information regarding the location, size, extension and nature of the injuries. This study aims to compare the radiographic and B-mode and Doppler ultrasonographic findings of the thorax, excluding the heart. Materials and Methods: Twelve dogs and one cat suspected of thoracic disease undergoing thoracic radiography with formation of potential acustic window (caused by fluid, lung consolidation, mass lesions and/or pulmonary nodules) where submitted to ultrasonography of the thorax between March 2009 and March 2010 in the Veterinary Department at the Federal University of Viçosa, Minas Gerais, Brazil. Four cases had the post mortem examination.

In 12 cases the radiographs showed effective acoustic window. In 10 cases the thoracic ultrasonography added important information such as: location of the lesion in the pulmonary parenchyma because of the sliding sign (9), best estimative of the amount of free pleural fluid (3), visualization of structure not detected during radiographic exam [cranial mediastinal lymph nodes(1) and atelectatic lung (2)]. In 10 cases the lesion was suitable to Doppler evaluation: mass lesions and/or pulmonary nodules showed peripheral vascularization (9) and it was possible to separate pulmonary vessels from fluid filled bronchi. In 3 cases ultrasound did not add to the diagnosis. Discussion/Conclusion: In this study the B-mode and Doppler ultrasonography played an important role in better characterizing of pulmonary, pleural and mediastinal conditions. Besides it is an easy, fast, inexpensive, safe method. When available it should be considered to be done to help to reduce the list of differential diagnoses.

Christian Niesterok, Eberhardt Ludewig, Gerhard Oechtering, Michaele Alef, Ingmar Kiefer. Departement of small animal clinic university of Leipzig Introduction/Purpose: Both CT and radiography are commonly used for detection of metastases or primary lung tumors as well as for staging of neoplastic disease. On the other hand, they play a minor role in detection of benign round-shaped nodes, being differential diagnoses. The aim of this study was to estimate the value (sensitivity/specificity) of plane radiographs as a first line diagnosticum compared with computed tomography (MSCT Philips Brilliance MX 8000) as gold standard.

In our study, we were looking for animals which were diagnosed in CT with a nodular lung pattern. In a second step, we then compared those results with those findings from the radiographs. Altogether there were 55 dogs and 24 cats that met our inclusion criteria, undergoing both thoracic radiography and CT between June 2005 and Dec. 2009. CT Diagnostic was performed with a 6 line-spiral-CT ( Philips Brilliance), with a slice thickness of 2 mm, collimation of 1.5, Pitch of 0.9 and a reconstructing matrix of 512 Â 512. Finally, statistic analysis was performed on order to determine the radiographic sensitivity.

Overall in 56 of 79 animals (71%), nodular changes already were detected during the radiographs and confirmed later on by the results of the computed tomography. In detail 39 of 55 dogs (71%) and 17 of 24 (71%) cats showed nodular round herds undergoing plane radiographs. The remaining percentage, which we missed in the radiographs mostly consisted of either very small changes or -even more often -were the result of highly changed lung lobes (like pneumothorax etc.), that lead to an overall lung opacity, providing only poor contrast. Therefore, in these cases, soft tissue interpretation was difficult or even impossible. Discussion/Conclusion: Radiography still plays an important role as a first line diagnosticum in detecting and differentiating nodular lung lesions. However, one needs to be aware that quite a high percentage of nodular lung changes can be missed with radiographs showing an overall sensitivity of 71% with no differences between dogs and cats. Furthermore our study showed, that plane radiographs are of poor diagnostic values wherever concurrent problems (e.g. pneumothorax, pleural effusion etc) exist, that lead to an increased lung opacity. Introduction/Purpose: Oesophageal pathology is not always easy to detect on survey thoracic radiographs and the exact nature of the pathology may be difficult to determine. Endoscopy and CT are reliable diagnostic methods to address these deficiencies but are not always available in general practice. It is hypothesized that pneumo-oesophagraphy may be a cost-effective/affordable and readily performed technique to provide additional information on oesophageal pathology.

Thirty dogs with endoscopically confirmed Spirocerca lupi nodules were selected from a larger prospective spirocercosis trial. Each dog had RLR and DV survey thoracic views. Dogs then had endoscopy under general anaesthesia immediately followed by RLR, LLR, DV and VD pneumo-oesophagraphy thoracic views (pneumo-views) made whilst still anaesthetised. Endoscopic nodule number, location and characteristics were recorded and were used as the gold standard. The oesophagus was inflated with a Ambu bag s through a cuffed endotracheal tube placed in the oesophagus. Radiographs were evaluated for degree of oesophageal distension (graded 0-4), nodule/mass visibility (yes/no), nodule surface characteristics and their mural attachment location. Data were analysed in Excell using paired t-test and chi-square tests and significance was set at Po0.05. Results: Dogs had a mean age of 52 months (range 9-125) and weight of 19.6 kg (range 3.6-41.4 kg) and a variety of breeds were represented. Lateral pneumo-views had more oesophageal distension than the DV/VD pneumo-views. DV survey and RLR pneumo-views detected significantly more nodules than the RLR survey and pneumo-DV views respectively. If all four pneumo-views were used 90% of nodules found on endoscopy, could be detected and if this was combined with the 2 survey views this became 93%. There was no significant difference in apparent length of oesophagus affected between the views or the modalities used. In six of nine malignant nodules an irregular mucosal surface, (indicative of neoplasia) could be seen on at least one of the pneumo-views. The mural attachment location could readily be seen on pneumo-views. The associated gas filled stomach gave the best visualization of the cardia on the RLR view. Discussion/Conclusion: This study indicates that pneumo-oesophagraphy is a useful and simple technique to use in a clinical environment to detect oesophageal pathology and its characteristics. Introduction/Purpose: Pulmonary edema is the most common complication of congestive heart failure in dogs. Interstitial pulmonary edema, which usually precedes more severe alveolar edema, may not be radiographically detectable. In human medicine, transthoracic ultrasonography is used to detect narrow repetition artifacts, or comet-tail artifacts, that indicate the presence of edema in the lung. The hypothesis of this study is that comet-tail artifacts can be identified using ultrasono-graphy in dogs with cardiogenic pulmonary edema, and that these artifacts occur in greater numbers and are more widely distributed in dogs with congestive heart failure than in normal dogs.

Transthoracic ultrasonography was performed in a normal group of clinically healthy dogs and dogs with heart failure with radiographic and echocardiographic signs of congestive heart failure. Physical examination, two-view thoracic radiographs, two-dimensional and Doppler transthoracic echocardiogram and heartworm antigen test were performed on all dogs. During ultrasonography, dogs were gently restrained in a standing or seated position. The thorax was divided into 4 quadrants on each side, and video clips and still images of each quadrant were recorded. Distribution, characteristics and number of comet-tails were determined using Qlab software (Philips Healthcare, Bothell, WA). Statistical analysis was performed using SAS software version 9.1.3 (SAS Institute, Cary, NC). Results: 31 dogs were recruited in the normal group and 6 dogs were included in the heart failure group. Transthoracic ultrasonography was easily and quickly performed on all dogs, with an average examination time of ten minutes or less. Comet-tails were detected in 32% of the normal dogs, with a mean of 2 per dog. No statistical association was found between the variables age, sex, side and distribution using a phi correlation coefficient. Subjectivley, in dogs with cardiogenic pulmonary edema, comet-tails occurred in greater numbers and were more widely distributed. In severe cases, comet-tail artifacts were so numerous that they became confluent and obscured the borders of the pleura. The presence of artifacts was consistent with the distribution of edema on radiographs. Discussion/Conclusion: Artifacts are rare and isolated in normal dogs, but numerous and widely distributed in dogs with congestive heart failure, as has been described in human patients. Further studies are needed to determine if the number and distribution of comet-tail artifacts in dogs with pulmonary edema correlates with the severity of clinical and radiographic signs. Thoracic ultrasonography is a fast, easy, and effective method to characterize pulmonary edema in dogs with mitral valve disease. It may provide additional information in combination with traditional diagnostic tests for the assessment of dogs with heart failure, the most important being the detection of early signs of heart failure and monitoring of the response to treatment. Introduction/Purpose: To evaluate multiobserver diagnostic performance and reader agreement for meniscal tear detection using low-field MRI (lfMRI) in dogs with suspected cranial cruciate ligament (CrCL) disease, with arthroscopy as the reference standard. Materials and Methods: Fifty consecutive stifles with clinical and radiographic signs suspicious for CrCL insufficiency underwent MRI using a 0.5 Tesla Philips Gyroscan T5 NT magnet with a human quadrature knee coil. The protocol contained sagittal und dorsal PD-and T2-, sagittal and transversal T1and sagittal fat suppressed T2-weighted sequences with images of 3 mm slice thickness and a total scan time of 32.5 min. After MRI, diagnostic arthroscopy was performed by 1 surgeon (P.B.) unaware of the MRI findings. MRI images were read independently by 15 observers, unaware of historical and clinical data, asked to identify presence of a meniscal tear, which would necessitate invasive verification of meniscal integrity probably requiring meniscal surgery. Diagnostic performance of each individual observer was analyzed. The impact of the individual observer's experience in stifle MRI on the observer's diagnostic performance was estimated by correlating the level of experience (5 self reported levels) with all parameters of test performance.

The overall diagnostic performance for the lateral and medial menisci expressed as median and associated interquartile range (IQR) was: sensitivity 69.2 (61.5-73.1), specificity 91.9 (85.6-94.6), PPV 72.7 (62.4-81.8), NPV 87.7 (86.0-90.3), accuracy 85. 6 (80.8-87.6 ) and diagnostic odds ratio 24.5 (14.7-45.0). The level of experience in stifle MRI did not influence the diagnostic performance of the individual observer (P40.05). Discussion/Conclusion: In a previous study with a single observer, using 42 data sets out of the 50 data sets from the present study, sensitivity and specificity of lfMRI for detection of meniscal tears were 0.56 and 0.97, respectively, with a PPV and NPV of 0.88 and 0.84, which might have been related to the low oberserver experience. These results with 15 observers of various levels of self reported experience in stifle MRI show that the level of experience did not impact on the performance of the individual observer, with a median diagnostic performance similar to the first study. Introduction/Purpose: The temporomandibular joint (TMJ) is formed from the temporal fossa of the temporal bone and the condylar process of the mandible. Investigation of TMJ diseases such as osteoarthritis and dysplasia requires a clear diagnostic image. While advanced imaging such as computed tomography allows clear imaging of the TMJ, superimposition of the TMJs on conventional lateral radiographs creates difficulties in visualization and interpretation. Oblique projections are generally used to separate the TMJs on a radiograph, but these are technically challenging to achieve. The aim of this study was to compare 5 different oblique radiographic views with the head in lateral recumbency, assessing the clarity of visualization of the normal TMJ anatomy.

The views under investigation were the laterorostral-laterocaudal oblique at a 101 and 201 rotation of the head (''nose-up'' view), lateral view with a rostro-caudal x-ray beam angulation of 101 and 201, and a parallax view with the beam centered over C2 and collimated to include the TMJ region, using the divergence of the x-ray beam to project the TMJs separately on the radiograph. The views were performed on both TMJs of thirty canine cadavers with no breed specification although brachycephalic dogs were excluded. The radiographs were graded independently by experienced (board certified radiologist) and inexperienced (undergraduate veterinary student) observers. Grading was performed on the temporal fossa, condylar process, joint space, retroarticular process and the overall TMJ, and was based on a four point scale, with 0 indicating the structure could not be visualised and 3 indicating clear visibility and definition of the structure in question. Mean grades for each component and for the overall joint were compared for each observer and each projection.

Mean grades were significantly (Po0.005) higher for the ''Nose-up'' projections than the angled beam or parallax projections, and both observers showed significantly higher (Po0.05) mean grades for the 20o ''Nose-up'' angulation than the 10o ''Nose-up'' angulation. Cohen's Kappa was used to assess interobserver agreement, and was also highest for the ''Nose-up'' projections. Discussion/Conclusion: These results suggest that a latero 20o rostral-laterocaudal oblique gives the best representation of the anatomy of the TMJ of the dog of the projections assessed, and should be considered when investigating clinical cases of TMJ disease. Introduction/Purpose: Foot pain is a common cause of lameness in horses. When horses with foot pain have no radiographic abnormalities, soft tissue assessment remains a diagnostic challenge without magnetic resonance imaging (MRI). The purpose of this study was to retrospectively review ultrasonographic (US) findings in the front feet in a series of clinical cases with lameness abolished by palmar digital nerves block but without radiographic bone abnormalities.

Medical records of horses that underwent a complete US examination of the digit were collected. History, clinical findings, radiographic and US results were retrospectively reviewed.

The records of 39 horses satisfied the inclusion criteria. Thirty of the 39 horses had lesions affecting the deep digital flexor tendon (DDFT), 27 horses had abnormalities in the distal interphalangeal joint of which 6 had a visible abnormality in one collateral ligament. Distension of the podotrochlear bursa was seen in 22 horses and abnormalities in the ligaments of the navicular bone were detected in 2 horses. Irregularity of the navicular bone flexor surface was found in 8 horses. In only 3 of the 39 horses the DDFT was the only structure affected. The other 36 horses had US abnormalities in more than one anatomical structure. Discussion/Conclusion: US diagnosis of soft tissue damage in the foot was possible in cases where radiographic diagnosis was unsuccessful. US offers an easily available diagnostic tool for an evaluation of the digital soft tissues and may decrease the need for MRI when this modality is difficult to apply. Introduction/Purpose: Trochanteric bursitis is said to be common cause of lameness in horses and is usually diagnosed based on clinical signs and findings during physical examination. The purpose of this study was to describe the anatomical features of the trochanteric bursa and to develop a technique for centesis of the bursa.

The trochanteric bursae of 2 cadavers were injected with latex to demonstrate their boundaries. Four techniques for centesis of the trochanteric bursa were evaluated by attempting to inject dye into the bursae of 20 horses which were scheduled to be euthanized for unrelated reasons. Following euthanasia, the region of the bursa was dissected to determine the location of the dye.

Injection was successful in 8/10 bursae when the foot was situated in a Hickman block without using ultrasonographic guidance, in 6/10 bursae when the foot was situated in a Hickman block using ultrasonographic guidance and a needle guide, in 3/10 bursae when the needle was inserted without ultrasonographic guidance and without placing the foot in the Hickman block, and in 9/10 bursae when the foot was situated in a Hickman block and the needle inserted using ultrasonographic guidance without using the guide. Discussion/Conclusion: Placing the foot in a Hickman block facilitates centesis of the trochanteric bursa.

Warren-Smith CMR, S. Danika, N. Saunders, L. Sambrook, T. Bouts, J. Hutchinson, R. Weller. Royal Veterinary College University of London, Zoological Society, London (Sambrook L. and Bouts T.) Introduction/Purpose: Foot problems are considered the most important ailment seen in captive elephants and can involve the integument and/or the musculoskeletal system, their aetiology can be infectious, traumatic or degenerative. Radiography is currently the most commonly used modality in the clinical investigation of musculoskeletal problems in elephants. In this study we described the radiographic and computed tomographic (CT) changes observed in a range of musculoskeletal pathology in captive elephant feet. We also established the intra and inter observer agreement regarding the topographic location and type of these disorders on both radiography and CT.

Radiographs and CT scans of fifteen cadaver foot specimens from one African and fourteen Asian elephants (euthanased for various reasons unrelated to this study) were acquired. In addition 24 radiographic sets were obtained from six live Asian elephants from a single zoological collection. A total of 22 front feet and 17 hind feet were evaluated. Radiographic and CT assessment was performed by two observers according to an evaluation grading scheme developed based on the reported radiographic and computed tomographic findings in elephant foot disorders. Each observer performed the reading of all images twice.

Pathological changes were observed in 50% (observer 1) and 90% (observer 2) of cases with pathology equally divided between front and hind feet (contrary to reported studies where forefeet have been more commonly involved).Changes were seen in all five digits but most commonly in digits III and IV. Pathological changes included osteitis (39.2%), enthesiopathies (17.5%), infectious arthritis (14%), osteoarthritis (7%) and soft tissue infections and defects (21%). Intraobserver agreement ranged from ''good'' to ''moderate'' for topographic location, from good to substantial for radiographic diagnosis and was good for severity of radiographic lesions (k: 0.656 to 0.029). However interobserver agreement was only ''moderate'' for diagnosis (k ¼ 0.595). Both intra and interobserver agreement for CT agreement was ''poor'' at all levels including topography and diagnosis (k ¼ 0.000 to 0.198. Discussion/Conclusion: Radiography offers an effective way of identifying and differentiating pathological changes both in vivo and post mortem in the elephant's foot, allowing diagnosis and treatment plans to be generated. However CT appears not to offer any advantage over radiography at this time. Significant differences in inter observer agreement were seen likely reflecting different levels of expertise in reading radiographs of elephant feet. Further large scale studies are warranted to evaluate the different pathologies seen and their respective causes. Introduction/Purpose: Foot related problems are the most common cause of euthanasia in captive elephants and at the moment radiography is the only imaging modality that is available for their assessment. The aim of this study was to (1) develop standard protocols for front and hind legs that are practical to use and allow the thorough examination of all digits and (2) to describe the radiographic anatomy of the elephant feet on those projections. Materials and Methods: Fifteen elephant cadaver feet of different ages and sizes were radiographed using a portable, digital radiography system in a variety of projections for which a custom-made block was designed. In addition computed tomographic (CT) scans were acquired before the cadavers were finally dissected. Anatomical features and the pathological changes which were observed on CT scans and gross inspection were identified on the different radiographic projections. The projections that allowed the best visualisation of the most common changes were subsequently trialed on eight live animals using a custom-made block. 3D reconstructions were created from the CT images and registered onto the radiographs to demonstrate the radiographic anatomy of the various skeletal elements of the elephant foot.

The dorsoproximal-palmarodistal and the plantarodistal-dorsoproximal projections were found to be best for the examination of the most common sites of pathology. A custom made block allowed the accurate positioning of the foot without endangering the personnel involved. Ideally each digit should be radiographed individually and the angle adjusted accordingly, however it was found that the time needed to do this exceeded the patience of any live elephant (and their keepers). Hence the number of radiographs was reduced to two per foot, resulting in a total acquisition time of less than 30 minutes for all four feet, which was deemed to be acceptable, but may still be too long in less well-handled individuals. The radiographic anatomy of the elephant foot itself is not overly complex however pathological changes are very common and often result in severe changes in the shape of whole bones. Discussion/Conclusion: The combination of a DR system and our custom-designed block facilitates radiographing elephant feet considerably. The proposed standardised protocol will not only aid in monitoring the development of foot pathology in individual elephants but will also allow the comparison between elephants in different collections. The detailed description of the normal radiographic anatomy provides a reference against which suspected pathological changes can be compared. Introduction/Purpose: Infiltration anaesthesia of the alveolar branch of the mandibular division of the V th cranial nerve at the mandibular foramen offers superior pain relieve for invasive dental procedures of the mandibular cheek teeth (1) . Precision of infiltration is enhanced, if needle placement is guided by computed tomography (2) . This pilot study explores the feasibility of the technique in horse cadavers as well as the possibility of a catheter placement. Materials and Methods: Isolated heads of nine euthanized warmblood horses were scanned using a single slice 3rd generation CT scanner as well as a multislice scanner. All heads were placed in dorsal recumbency within the CT (according the positioning during a head ''real'' scan). The mandibular foramen was located on transverse images and marked with a radiopaque marker or a hypodermic needle using the laser light. The nerve block-simulation was performed with a spinal needle (quinke type point, 18 G, 152 mm). The needle was introduced through the skin medially to the mandibular notch and advanced between 13-15 cm perpendicular to the table surface in the direction of the mandibular foramen. Subsequent needle adjustment was guided by CT scans until correct position of the needle's tip at the mandibular foramen was confirmed. Diluted contrast medium lopamidol (Scanlux r 370 mg J/ml) was injected in eight horses and imaged using a control CT scan. Methylene blue was injected in one cadaver and verified by anatomic dissection.

The assumed complete coverage of the mandibular nerve could be demonstrated by injected methylene blue during dissection. Final positioning of the needle required only a few additional minutes following the first scout CT-scan of the head. One to three corrections in needle placement with subsequent CT scans were necessary to reach the final location. The placement of a catheter (Perifix r epidural catheder, 1000 mm, 20 G) through the needle is possible and stays in place after the removal of the needle. A proper fixation on a living horse has to be evaluated. Discussion/Conclusion: CT guided infiltration anaesthesia of the mandibular nerve in horses is feasible and can be performed immediately after diagnostic CT scans. Precision of local anaesthetic deposition at a location that is not readily accessible can be enhanced using this technique. Further pain management could be improved by the positioning of a catheter next to the mandibular foramen. 

Metal is a known source of image artefacts in MRI by creating image distortion and large areas of signal void. Radiography has not been validated for metallic screening, nor has the relationship between particle size and resulting MRI artefact size been established. Purpose of this study was to gather sensitivity and specificity data for radiographic screening procedures and to create a scoring system for metal particles for their resulting artefact size. Materials and Methods: 14 cadaver horse feet were radiographed (DPrPaDiO, LM views) and a standard foot protocol MRI (T1W GRE, T2 Ã W GRE, STIR FSE) was performed. Stainless steel ball bearings with known diameter (1, 2 and 3 mm) were placed into the most dorsomedial and palmarolateral nail hole of the foot, and DPrPaDiO and LM radiographs were again obtained. The MRI foot protocol was repeated with the metal particles in situ. Radiographic and MRI studies were evaluated by one board certified radiologist. Radiographs were scored for presence, diameter and location of metal particles. On MRI studies the maximum signal void size was measured in all sequences.

Radiographic identification and accurate measurement of all metal particles was possible using both projections. On MRI metal artefacts were seen as an area of signal void, often with a high intensity rim and distortion in adjacent tissues. Sequences in transverse and sagittal planes that contained the entire particle volume within the slice thickness showed the biggest signal void. The size of the MRI artefact was approximately 10 Â larger than the original metal particle in FSE and 15 Â larger in GE sequences. T1W gradient echo sequences showed void signal areas 5-10 mm larger than FSE sequences. Discussion/Conclusion: Radiography is a sensitive and specific screening tool for MRI-artefact-inducing metal particles of diagnostically relevant size. Two orthogonal radiographic views are necessary to correctly localize metallic debris. If a metal particle cannot be removed, the diagnostic quality of a foot MRI exam can be predicted by plotting the predicted artefact size on the anatomical area of clinical interest. GE sequences should be avoided due to their higher artefact sensitivity. 

The magic angle effect is a magnetic resonance phenomenon responsible for increased signal intensity in tendons imaged at 551 relative to the magnetic field using short echo time pulse sequences. The magic angle effect has previously been considered as a deleterious artifact responsible for false positive diagnosis of tendinopathy. Recently it has been suggested that intentionally imaging tendons at 551 relative to the magnetic field may enhance the detection of some lesions. The purpose of our study was to assess the benefits of magic angle imaging for the characterization of lesions in a diode laser-induced tendinopathy model. Materials and Methods: 6 isolated equine limbs were imaged in the study. A total of 16 lesions were created using a surgical laser. In 2 limbs, 2 lesions per tendon were induced post-mortem both in the deep digital flexor tendon (DDFT) and in the superficial flexor tendon (SDFT). In the other 4 limbs, 2 SDFT lesions per limb were induced in vivo 3 weeks prior to euthanasia. The limbs were imaged immediately after euthanasia with a 1.5T MR scanner using a surface general-purpose flexed coil. The limbs were imaged parallel to the magnetic field (01) using sagittal T1-w and transverse T1-w, T2-w and STIR sequences. Imaging with T1-w sagittal and transverse sequences was then repeated with the limbs positioned at 551 relative to the magnetic field. The conventional (01) and magic angle (551) images were reviewed independently for presence and pattern of lesions. All lesions were examined histologically.

All lesions could be identified on the magic angle images. The post-mortem induced lesions were not apparent on any of the sequences with the limbs positioned at 01. On magic angle images, the post-mortem induced lesions appeared as a hyperintense core surrounded by a hypointense rim, whereas the ante-mortem induced lesions appeared as mildly hypointense relative to the normal tendon. On 01 images, the ante-mortem induced lesions were hyperintense relative to normal tendon on all sequences. Discussion/Conclusion: Magic angle imaging allows detection of lesions that are not detected using conventional imaging. The differences observed between ante-mortem and post-mortem induced lesions may be explained by the presence or absence of an inflammatory process. With the postmortem induced lesions, in the absence of an inflammatory response, changes in the tendon result from a modification in the organization of the collagen fibers, which can be detected with magic angle imaging. These results suggest that magic angle imaging could be clinically relevant to detect reorganization of collagen fibers in chronic tendinopathy in the absence of an inflammatory response. Introduction/Purpose: experimentally induced tendinopathies represent an accurate means of evaluating the efficacy of new treatments using blinded studies. In these investigations, imaging modalities are frequently used to assess tendon healing. The purpose of this study was to compare ultrasonographic (US) and magnetic resonance imaging (MRI) evaluations to assess the healing of induced superficial digital flexor tendon (SDFT) lesions. Materials and Methods: bilateral metacarpal SDFT lesions were induced surgically in 10 standardbred adult horses (week 0). US evaluations were performed before induction and at week (w): 3, 7, 11, 15. At each time, the maximal tendon cross section area was measured and the echogenicity and architecture were graded on a 0-4 scale (0 ¼ sound tendon). Low field (0.2 T) MRI was conducted under general anesthesia at w11. Ultra-high field (7T) MRI was acquired post-mortem at w15. For both examinations, the lesion was graded subjectively using the same 0-4 scale. Quantitative data were also measured using a homemade Matlab image analysis software for the 0.2T data and ImageJ routines for the 7T ones. These included: tendon and lesion volumes, T1-weighted signal values, T1 and T2 (only for 7T) relaxation times of the lesions. Correlations were calculated using Pearson coefficients and statistical differences were evaluated using paired Student t-test. In both cases, statistical significance was set at Po0.05. Results: US parameters changed significantly over time: the mean tendon grade decreasing from 2.8 AE 0.2 at w3 to 0.9 AE 0.2 at w15. No significant differences were found in the MRI grades of the lesion between the 0.2T (2.6 AE 0.5, w11) and 7T (2.7 AE 0.7, w15) examinations. These grades were significantly higher compared to US ones. Significant correlations were found between 0.2 and 7T values for the tendon volume (r ¼ 0.92), lesion volume (r ¼ 0.78), T1-w lesion signal (r ¼ 0.55). In contrast, the lesion volume, expressed in percentage of the tendon volume, was significantly greater in 7T (62 AE 15%) compared to 0.2 T (47 AE 11%). A positive non significant correlation (r ¼ 0.31) was found between the T1 lesion relaxation times at 0.2 T (362 AE 16 ms) and 7T (1749 AE 86 ms). Compared to 0.2T, 7T MRI allowed to calculate the T2 lesion relaxation time (17 AE 1 ms) and to assess, based on high resolution images, the fascicular architecture of the injured tendons through the visualisation of the connective septa dividing the collagen fiber bundles.

This study provides reference data that can be used for future assessment of new treatment efficacy. It confirms the interest of US for evaluating the healing process, particularly in the first stages. Despite a significant US improvement, lesions remained severe on MRI which may indicate the interest of MRI to evaluate the late healing stages. High correlations found between 0.2 and 7T data validate the interest of low field MRI, even if 7T MRI permits a more accurate evaluation, particularly through the assessment of the tendon fascicular architecture. Introduction/Purpose: Lung diseases occur commonly in captive snakes, especially when snakes are housed under inadequate conditions and/or when errors of management are made. The assessment of the snake's lungs by computed tomography (CT) has already been reported but not compared with radiographic examination. The purpose of this study is to describe and compare the radiographic and CT findings of normal snake lungs and lungs with bacterial pneumonia. Materials and Methods: Six healthy snakes (4 common boa constrictor (Boa constrictor), a brown water python (Liasis fuscus) and an African rock python (Python sebae)), and four snakes (a carpet python (Morelia spilota) and three Indian python (Python molurus)) suspected of respiratory diseases based on clinical examination, transtracheal wash and culture, were used for this study. Radiographic examination including a lateral view of the pulmonary area was performed without sedation. CT was performed using a high resolution CT (Somatom 16, Siemens, Germany, Erlangen) and obtaining transverse reformatted images of 1 mm thickness. Also for the CT examination, the snakes were not sedated and held with plastic drainpipe and straps. Results: Radiographically, the normal lung appeared as a radiolucent, cylindrical shadow with a ''honeycomb'' weft. On CT images, the healthy respiratory alveolar parenchyma was a circular, hypoattenuating, homogeneous area with a ''honeycomb'' weft around a central lumen. The plexus of blood vessels is seen as a circular hyper attenuating line. In Boidae (boas) the left lung appeared to have less volume then the right lung. This difference was less evident in Pythonidae (pythons). Radiographic findings of the diseased lungs included one nodule of soft tissue opacity in 1 patient, several small nodules in 2 patients, and 2 thick lines of soft tissue opacity perpendicular to the lung's axis in 1 patient. On CT images the following findings were made: the respiratory parenchyma was in all patients more heterogeneous with several focal hyper attenuated areas in 2 patients, or with more extended hyper attenuating areas in 2 patients. These lesions were more obvious in the cranial part of the right lung. The inner margin of the parenchyma appeared focally hyper attenuated in 3 cases. When comparing the lateral radiographs to the CT images the extent of the pulmonary lesions could be assessed in detail more on the CT images. Discussion/Conclusion: Even though the radiographic findings of bacterial pneumonia in snakes were less obvious than on CT images. Pulmonary changes could be diagnosed in most cases in this series. CT images could show early changes and may be useful for monitoring the disease process during treatment. Introduction/Purpose: The production efficiency in hatching Senegal sole (Solea senegalensis) is highly influenced by the development of bone deformities during the rearing process. The presence of vertebral deformities may be caused by a variety of factors such us genetic, nutritional, environmental conditions in the tank, or possible infections. The present study is a radiological description of the meristic characters of Solea senegalensis and the skeletal bone malformations present in juvenile specimens coming from farms in northern Spain.

For this study, 30 specimens of Senegal sole (Solea senegalensis) with 481 days after hatching (140-220 mm long) having macroscopic signs of spinal malformation were selected. After a morphometric study, the specimens were radiographed (dorsoventral projection) using mammography films. From the images obtained, the number of vertebrae and their associated spines in the different body regions were observed. Metric measurements of vertebrae of each region were taken and the vertebral anomalies and of their associated elements were counted and described. Finally, the severity of the injuries was assessed. Results: SKELETAL CONFORMATION: The number of vertebrae ranged from 42 to 45, including 6-8 abdominal, 31-34 caudal and 3 in the caudal complex. 60% of specimens showed variations in the number of rays present on the caudal fin. Variations in other regions were less frequent. SKELETAL ABNORMALITIES: 96.7% of individuals had skeletal abnormalities, and they were severe in 30% of them. The most frequent abnormalities detected were abnormal vertebral bodies, either isolated or as a part of more serious lesions including deviations from the axis (60%) or spinal fusions (23,3%). 70% of specimens showed abnormalities in their neural or hemal elements. The most commonly affected region was the caudal complex with 90% of the animals affected mainly involving the rays of the caudal fin (86.7%).

The discrimination of juvenile specimens based on their appearance is an effective method of selection. It is suspected that slight initial bone changes might be accentuated in the adult stage and its presence may be an early indicator of suboptimal rearing conditions. Therefore, new studies are necessary to determine possible causes, and thus to obtain better performance in captive breeding. Introduction/Purpose: Diagnostic imaging techniques are being utilised more frequently for the diagnosis of disease in wild animals. However, there is a lack of literature describing the normal appearance of the gross and imaging anatomy The purpose of this study was to describe the normal anatomy of the grey seal using radiography, ultrasound, CT and MRI.

A 26 days old male grey seal was admitted for treatment of a humeral fracture. Radiographs of the humerus and the chest where taken with the patient under general anesthesia. Given the difficulty in obtaining standard views of the humerus and to completely characterize the fracture CT of the humerus was performed and the chest was also scanned. Surgery was attempted to reduce the fracture, but it was found that the fracture site was infected and the seal was euthanized. Postmortem CT scans of the whole body, with bone, lung and abdominal windows depending on the area was performed to obtain images of normal anatomy. Abdominal ultrasound with a 5-8 MHz probe and MRI of the brain including T2 weighted sagittal, dorsal and transverse sequences and T1 transverse sequences using a 1.5 Tesla scanner were also performed.

CT images provided images of the skeletal structures and thoracic cavity. Abdominal contrast however was poor and the different organs were difficult to individualize-this was likely caused by the young age of the patient. Ultrasound of the abdomen allowed the identification of the different organs. These had similar appearance to those in the dog or cat with the exception of the kidneys which showed a heterogeneous echostructure with a hyperechoic parenchyma and multiple small irregular shaped-anechoic areas. This structure is thought to represent the kidney composition of many closely adherent renculi surrounded by pericapsular plexus and interlobular veins in the seal. MRI of the brain allowed identification of the brain and other head structures and organs. The seal underwent post mortem examination and the gross anatomy was compared with imaging findings. Discussion/Conclusion: Having normal references for anatomy is important especially when imaging species unfamiliar to the radiologist. CT provided good quality images for the skeleton and the thorax of the seal. Ultrasound and MRI allowed identification of abdominal organs and brain, respectively. Description of normal imaging anatomy of more individuals and different ages would be useful to aid in recognizing disease. Introduction/Purpose: The most commonly presented problem in reptiles are abscesses. Subcutaneous abscesses, superficial, or abscesses localized near the oral cavity are well noticeable and readily palpable on clinical examination, thus easy to diagnose. Internal abscesses remain a diagnostic challenge owing to their different locations, which in turn, determine clinical manifestations observed.

A male 5-year-old Boa constrictor, 800 g bodyweight was presented with deep cutaneous inflammatory lesions due to burning, caused by a heating cable 5 months ago. Since the burning event, the snake refused any food. The clinical evaluation showed emaciation with concomitant severe dehydration and decreased responsiveness to external stimuli. The boa was assigned to strict quarantine in a terrarium for observation and diagnostic procedure. The microbiological study of the skin lesions revealed the presence of Pseudomonas aeruginosa. The blood examinations indicated elevated count of total white blood cells TWBC (18 10 3 /mL) and heterophils (78%). Due to poor body condition, the snake underwent ultrasonographic evaluation. The study was performed with 7.5 MHz linear transducer.

A nodular lesion of 2 Â 2,5 cm dimensions, localized caudally from the thinned-wall, anechogenic gallbladder was visualized. The tumor-like lesion showed heterogeneous echogenicity. Centrally located an irregular anechogenic area of 1,3 Â 0,9 cm was visible. The ultrasonographic appearance suggested abscess or neoplasia with the presence of large focus of tumor lysis. The other abdominal organs did not show any changes in the ultrasound image. Other diagnostic procedures and surgical intervention were not possible to perform due to poor animal condition. Despite therapy the snake died after 6 days. The necropsy confirmed the presence of a well formed encapsulated abscess located in the duodenal wall. The lesion caused marked narrowing of the intestinal lumen. The microbiological study of the abscess contents showed the presence of the same bacteria that were detected on the skin at the earlier bacterial culture. Discussion/Conclusion: Infectious factors producing internal abscesses may be identical with those developing epidermal lesions. Most probably they spread via the hematogenic route 1 Introduction/Purpose: In order to establish the venous pattern of the body cavity in the Common Buzzard (Buteo buteo), a venographic study is carried out by means of a postmorten technique. Materials and Methods: Six Common Buzzards (Buteo buteo), coming lifeless from a wild animal recovery center, are kept freeze at À201C until the moment of the venographic study. For the contrast medium application, the right and left jugular veins are located. A lead oxide jelly solution with a 150% concentration of Pb3O4 is injected at 501C. The filling process is controlled under radioscopy. After the radiographic study, the anatomical dissection of the body cavity is performed.

Pictures of the anatomical dissection, radiological projections and the tomographic images with their corresponding schemes are shown.

This postmortem technique is a simple technique, and provides fairly good vascular images, allowing appropriate study of the vessels in the body cavity. The venographic studies and the anatomical dissection permit the accurate identification of the main veins and its branches and allows to establish the venous pattern for this species. Introduction/Purpose: Ectromelias are congenital deformities characterized by total or partial absence of one or more limbs. Total absence is named amelia while partial absence is anamed transversal or longitudinal hemimelia (''congenital amputation''). These malformations are reported sporadically in children and in several mammalian species. Herein we describe the radiographical and CT pattern of hemimelia and amelia in a group of Mediterranean Italian buffaloes. Materials and Methods: 14 Mediterranean Italian buffaloes, 8 males and 6 females, from one day to six months old, affected by limb malformations of the thoracic or pelvic limbs or both, from different farms in Campania, were studied. Radiographs were performed on the euthanized animals or on the isolated limbs, using the two orthogonal views. In two buffaloes affected by amelia of the pelvic limbs, CT study of the pelvis was also made. Results: 7 buffaloes had unilateral transversal hemimelia of the left pelvic limb; 3 buffaloes had bilateral transversal hemimelia of the pelvic limbs; 1 buffalo had bilateral amelia of pelvic limbs; 1 buffalo had transversal hemimelia of both the pelvic limbs and of the left thoracic limb; 1 buffalo had bilateral pelvic amelia and longitudinal hemimelia of the right thoracic limb; 1 buffalo was affected by transversal hemimelia of all the limbs. In all the buffaloes the malformed limbs presented more or less developed outlines of claws. In 2 animals affected by transversal hemimelia of both the pelvic limbs the bones amputated were exposed. The transversal hemimelia showed different degree of amputation. In the case of longitudinal hemimelia the third digit was absent. The two buffaloes affected by amelia of pelvic limbs showed small calcified oval-shaped nuclei near the left acetabular rim. The CT study demonstrated a cartilaginous bud of the proximal femur. Discussion/Conclusion: radiographs were valuable to characterize all the bones malformations distal to the carpus or the knee. CT demonstrated to be more accurate in cases of amelia since it demonstrated that clinical lack of a limb does not always imply the complete absence of the underlying bones. The more frequent radiographical pattern of the transversal hemimelia was represented by a ''wedge-shape'' of the bone affected. Our findings show that in Italian buffalos pelvic limbs are more frequently affected than the thoracic ones and this is in contrast with the reports in human medicine. Since the true prevalence of these malformations may be underestimated and since a possible toxic teratogenic toxic etiology (e.g. dioxins) is possible, our report could also have an epidemiological significance. Introduction/Purpose: The equine tarsal joint is an anatomically complex region and has already been thoroughly described with radiography and ultrasonography.1-3 However in clinical conditions these techniques are sometimes inconclusive and computed tomography (CT) is recommended. This study was established to describe a detailed CT reference of the normal equine tarsal joint. Materials and Methods: CT was performed on 3 normal equine limbs from patients euthanized for reasons unrelated to this study. In all limbs CT was also achieved after intra-articular injection of the tarsocrural, centrodistal and tarsometatarsal joints with 60 mL, 3 mL and 3 mL contrast (30 mg iodine/mL) respectively. 1,3 mm and 0.6 mm slices were made (120 kV and 352 mAs), dorsal and sagittal planes were reformatted. The CT images were compared with corresponding anatomical slices.

The tibia, talus, calcaneus, central, fused first and second, third and fourth tarsal bones could be clearly visualized. The long digital extensor (with tarsal extensor retinaculum), superficial digital flexor, deep digital flexor (with tarsal flexor retinaculum), gastrocnemius, peroneus tertius and tibialis cranialis tendons and long plantar ligament could be clearly identified. The lateral digital extensor, medial digital flexor, splitted peroneus tertius and tibialis cranialis tendons and collateral ligaments could be localized but not always clearly identified. Some of the numerous small tarsal ligaments could be identified: the plantar, lateral and interosseus talocalcaneal ligaments, interosseus talocentral-centrodistal-and tarsometatarsal ligaments, proximal and distal plantar ligaments and talometatarsal ligament. The articular cartilage at the non weightbearing surfaces of the distal tibia and proximal talus could be clearly assesed on the postcontrast images, the articular cartilage at the weight-bearing surfaces of the distal tibia and proximal talus was more difficult to examine and the articular cartilage at the small tarsal joints was not visible. Discussion/Conclusion: CT of the tarsal joint can be of great value when radiography and ultrasound are inconclusive and in pre-operative planning of complex fractures. With CT having a superior anatomic resolution and detailed bone and soft tissue visualization this technique can increase our knowledge of the pathology in the equine tarsus. Introduction/Purpose: The aim of this study was to provide a detailed anatomical description and reference values of the structures of the thorax in clinically normal goats by means of CT for use by radiologists, clinicians, researchers and veterinary students.

Twenty-six clinically healthy, anesthetized Saanen goats (median age 4.0 years; SD 1.1 years) were positioned in sternal recumbency. Transverse images were obtained with a 40-slice CT scanner (120 kV, 350 mA, 1 s tube rotation and a slice collimation of 5 mm reconstructed to 1.5 mm). Subsequently, 12 animals were euthanized, frozen and sliced into 10-15 mm transverse sections, of which the caudal aspect was photographed. The study was approved by the Animal Ethics Council of the Canton of Zurich. Corresponding CT and anatomic images were compared. Absolute and relative measurements of thoracic structures were obtained.

Corresponding thoracic structures on CT and anatomic sections were well identified. Speciesspecific features, like the synostosed sternum, Bronchus trachealis and large caudal mediastinal lymph node were consistently identified. Contrary to the dog, in all goats the lung lobes could be differentiated, except the right middle lung lobe. The thymus was inconsistently seen (12/26). Minimal to moderate interstitial pneumonia or bronchopneumonia (24/26) and pleural thickening (5/26) were diagnosed in the caudal and dorsal lung fields. Mean values of thoracic structures included: angle between trachea and spine (16.51); ratio of tracheal to thoracic height (0.13); vertebral heart score (8.0); ratio of thoracic aorta to caudal vena cava (1.0); ratio of maximal height of cranial vena cava to the length of the thoracic vertebrae at the level of the tracheal bifurcation (0.8); mean cross-sectional area of the tracheal lumen just cranial to the thoracic bifurcation (2.1 cm 2 ). Further, the square area of the principal bronchi, the caudal lobar bronchi and the concomitant pulmonary arteries and veins was measured. Discussion/Conclusion: With the exception of small vessels, nerves and certain small lymph nodes, all the structures of the thorax that were identified on the anatomic sections could be identified on the corresponding CT scans. Subclinical lung disease in the majority of goats, most likely attributable to parasitic infection (Protostrongyloides), was documented in CT and confirmed with fecal examination and macroscopic pathology. Introduction/Purpose: Contrast enhancement in computed tomography (CT) plays an important role in the detection of several diseases. In human medicine, use of a saline solution injected intravenously immediately after the contrast material bolus, also known saline chaser, has been reported to allow a significant reduction of the contrast material volume in helical CT. However, there was no study about the reduction of the amount of contrast material using a saline chaser in veterinary medicine. The purpose of this study was to assess the degree of dose reduction of the intravenously injected contrast material without decreasing vascular enhancement using a saline chaser in multidetector-row CT of cattle.

Dynamic CT was performed in the head of six normal Holstein. In a crossover study design, group A was administered only contrast material (600 mgI/kg) whereas group B and group C were given contrast material at a reduction of 30% and 40%, respectively, followed by 50 ml of saline solution. In all groups, a power injector was used with an injection rate of 4 ml/s and a pressure of 4.7-5.7 kg/cm 2 . Attenuation values were obtained from the right and left maxillary arteries and dorsal sagittal sinus.

There was no significant difference in the maximum enhancement value (MEV) of the maxillary arteries and dorsal sagittal sinus between groups A and B. MEV of the maxillary arteries and dorsal sagittal sinus in groups A and B was significantly greater than that in group C (Po0.05). Discussion/Conclusion: Results indicate that use of a saline chaser allowed 30% reduction in the dose of contrast material of without decreasing vascular enhancement. Introduction/Purpose: Subchondral bone damage has been diagnosed using Magnetic Resonance (MR) in the equine distal limb as areas of reduced signal intensity in T1 weighted sequences and increased signal intensity in STIR sequences 1 . A recent study showed no correlation between degree of lameness and radiographic changes in the tarsus 2 . The purpose of the present study is to describe MR findings in horses with acute onset of severe lameness localized in the tarsal region.

Horses were selected for inclusion that had acute onset of severe lameness localized to the hock by local analgesia and the results of radiographic and ultrasonographic examination did not conclusively explain the degree of lameness. MR images were acquired by use of a 0.2T imaging system, with the horse under general anesthesia; 3D T1 gradient echo, T2 fast spin echo and STIR sequences were acquired on sagittal, dorsal and transverse planes.

We included 3 horses (used for show jumping, age range 6-12 years) showing 3/5 unilateral pelvic limb lameness localized to the hock. Radiographic examination showed enthesiophyte formation on the dorsal aspect of the central and third tarsal bones and dorso-proximal aspect of the third metatarsal bone in case 1; case 2 showed mild degenerative joint disease of the distal intertarsal joint and no radiographic changes were observed in case 3. MR images showed marked irregular thickening of the subchondral bone plate and abnormalities of the adjacent spongiosa of the third and central tarsal bones in all cases. Increased signal on STIR sequences involved the dorso-medial aspect (Case 2 and 3) or axial portion (case 1) of the third and central tarsal bones. In addition, case 3 showed a linear area of hyperintense signal in each sequence on the dorso-medial aspect of the central tarsal bone, interpreted as an incomplete slab fracture. Therapeutic plan included distal intertarsal joint medication (case 1) and extracorporeal shockwaves (case 2). Case 1 and 2 were rested for 1 and 8 months respectively before return to the previous competition level. Case 3 is still in the rehabilitation process. Discussion/Conclusion: Increase signal intensity in STIR sequence could be related to different pathologic situation such as hemorrhage, edema, bone necrosis or fibrosis 1 . In the present report, horses showed acute onset of lameness suggestive of bone trauma, microfracture of the subchondral bone and fluid accumulation within the bone as the more likely cause of the abnormal signal observed. Subchondral bone trauma can involve the small tarsal bones and MR is crucial for lesion identification and therapeutic plan when radiographic and/or ultrasonographic findings are equivocal. Introduction/Purpose: The purpose of study was to register the reason for extracting the incisors in upper and/or under jaw and to examine the consequences of extraction of all incisor teeth by taking radiographs regularly for 1-2 years post operatively. Materials and Methods: Four horses, 14 to 32 years old, were presented at the University Clinic for Large Animals. The horses had problems with eating including weight loss, related to dental disease. In all four cases treatment included extraction of all incisors in the mandible and/or maxilla. The horses were monitored using radiography prior to tooth extraction, and at regular intervals over a period of one to two years post-operatively. Images VD and DV for the mandible and maxilla respectively were taken using intraoral position of the cassettes and developed in a digital method (CR-system).

The underlying pathology detected in these horses included both infectious and non-infectious diagnoses. Specific clinical diagnoses included tooth resorption, periodontitis, dental-fractures and malocclusion. All horses showed signs of oral cavity related pain and gingival inflammation.

All horses commenced eating within 24 hours post surgery. After eight to 10 weeks the alveolar sac was filled with regular trabecular bone. Areas of sclerotic bone on the preoperative images remained slightly visible for some months after surgery.

In cases where there was severe bone resorption associated with the initial dental disease, there was no significant re-growth of bone, so that the outer bone margins in these cases remained irregular in outline. All horses improved after surgery, gained weight and returned to normal use as pleasure riding horses. Discussion/Conclusion: No problems were evident with grazing; eating apples and carrots however presented problems unless these items were broken into small pieces. In general these horses became slower at mastication post surgery and this should be kept in mind if they are kept in groups with normal horses where feeding is from a common source.

All horses showed a dramatic improvement after removal of the incisors. There were no side effect or long term special requirements for these horses other than ensuring that they are fed individually if housed in large groups. The tongue was occasionally visible in some of the horses. The experience reported in the abstract suggests that removal of all incisors is a viable option in cases where these teeth have significant irreversible and major disease.

L. Evrard à , S. Rabba à , G. Bolen à , D. Verwilghen Ãà , V. Busoni à . University of Liège -University Veterinary Clinic. à Diagnostic Imaging Section. Ãà Surgery Section Introduction/Purpose: When horses with foot pain have no radiographic abnormalities, the soft tissues have to be assessed. However, some radiographic changes in the foot (e.g. enlarged synovial fossae, distal sesamoid bone fragments) are commonly seen in horses without lameness and their significance may be difficult to correlate to the clinical presentation in some clinical cases. The aim of this study was to review US findings in horses with lameness abolished by palmar digital nerve block but with inconclusive radiographic examination.

Medical records of horses that underwent a complete radiographic and US examination of the digit were collected and reviewed. Horses with a clinical history of infection or direct trauma and horses without radiographic abnormalities were excluded. Results: Twenty-three horses fulfilled the selection criteria. On radiographs, 16/23 horses had enlarged synovial fossae, 8/23 had distal sesamoid bone fragments, 1 had a suspected irregularity of the flexor surface. At US distal interphalangeal joint (DIPJ) synovitis was seen in 16 horses, of which 6 had a visible abnormality in one of the DIPJ collateral ligaments. Lesions of the deep digital flexor tendon (DDFT) were seen in 9 horses and distension of the podotrochlear bursa in 14. The ligaments of the navicular bone were abnormal at US in 3 horses. At US the flexor surface of the navicular bone was irregular in 11 horses. US abnormalities in more than one anatomical structure were seen in 20/23 horses. Discussion/Conclusion: In horses with foot pain but inconclusive radiographic examination, US may reveal useful complementary information about soft tissue involvement and integrity of the navicular bone flexor surface.

K Peremans, S Vermeire, A Dobbeleir, I Gielen, E Vandermeulen, J Saunders, K Dik, H van Bree. Medical Imaging Department, Faculty of Veterinary Medicine, University Ghent, Belgium Introduction/Purpose: Canine orthopedic problems in the forelimb frequently originate from the elbow. The origin of pain can be caused by developmental, traumatic or degenerative processes requiring different treatment strategies. Bone scintigraphy is included in the diagnostic frame work when clinical/ radiographic examination is inconclusive or non-localizing. Determination of the exact localization of pathology within the elbow is hampered by resolution limits of conventional gamma camera systems. The use of conventional micro-SPECT systems is precluded by limited gantry opening.

The aim of the study was to investigate the value of HiSPECT (multi-pinholes mounted on a conventional gamma camera) in the evaluation of elbow pathology in the dog. Materials and Methods: 5 dogs with elbow disease were included (7 elbows). Radiographic examination was negative (N ¼ 1), inconclusive (N ¼ 3) and not performed in 1 case. Two dogs had undergone arthroscopic interventions for an elbow problem, but lameness had switched to the other leg. All animals underwent planar bone scintigraphy. Micro-SPECT was performed using a conventional triple head gamma camera (Triad, Trionix), adapted with 3 multi-pinhole collimators (6 holes, 3 mm diameter, resolution 2.5 mm) (Bioscan). Data were acquired in step-and-shoot mode (10 steps, 361 angular step, 120 sec per step). Images were reconstructed using a dedicated ordered subset-expectation maximisation (OSEM) algorithm (Scivis). Structural imaging and/or arthroscopy were performed to confirm pathology.

The lame leg was correctly identified based on intensity of uptake on planar scintigraphy but no localization within the elbow could be determined. On the HiSPECT images, pathology could be confined to anatomical regions. Four elbows showed high uptake in focal regions. Pathology was confirmed with structural imaging or arthroscopy and included: insertion desmitis of the flexor carpi ulnaris (N ¼ 1), detachment of the medial coronoid process (N ¼ 2) and a cystic lesion in the anconeal process (N ¼ 1). In one elbow, several areas of increased uptake were noted in the humerus and radius representing severe subchondral erosion on arthroscopy. In the previously operated elbows residual mild increased uptake in one elbow was focally at the level of a removed fragment of the coronoid process (scar tissue on arthroscopy) and more generalized in the other (erosion of the joint cartilage on arthroscopy). Discussion/Conclusion: With the HiSPECT system increased uptake was discernible in localized elbow areas correlating with pathology found on structural imaging modalities and/or arthroscopy. HiSPECT improves anatomical localization of pathology and may allow a more accurate choice of subsequent structural imaging modalities and therapy. Introduction/Purpose: Idiopathic pulmonary fibrosis (IPF) is a chronic, progressive parenchymal lung disease described mainly in West Highland white terriers (WHWTs). High resolution computed tomography (HRCT) is used for visualizing pulmonary parenchyma and diagnosing IPF in man. The study objective was to investigate HRCT findings in WHWTs with IPF and in healthy control WHWTs. Materials and Methods: HRCT was performed under general anesthesia to 7 WHWTs with IPF (mean age 11.6 years) and 11 control WHWTs (mean age 8.5 years). A CT-scanner with axial high-resolution algorithm, 1.0 mm slice thickness and 7.5 mm table movement was used. The findings were evaluated and classified according to previously published criteria 1. Quantitative CT-values were measured from the nondependent parts of the lung. The radiologist was blinded to the dog's disease status.

The radiologist was able to distinguish all sick WHWTs from healthy controls. The HRCT findings of IPF dogs were ground glass opacity (n ¼ 7), traction bronchiectasis (n ¼ 5), parenchymal bands (n ¼ 5), consolidation (n ¼ 5), peribronchovascular interstitial thickening (n ¼ 3), honeycombing (n ¼ 2), subpleural lines (n ¼ 2), and subpleural interstitial thickening (n ¼ 1). Control dogs had only 1-2 focal lesions in the dependent part of the lung. The CT-values were significantly higher in IPF (mean À731 AE SD 51 HU) compared with controls (À821 AE 36 HU), Po0.001. No correlation with PaO2 was detected. Discussion/Conclusion: HRCT was accurate in differentiating dogs with IPF from healthy controls. IPF causes a diffuse increase in radiodensity and lesions which are not visible in healthy lungs. Unlike in man, ground glass opacity is common but honeycombing only rarely seen in canine IPF. Introduction/Purpose: Feline Injection-Site Sarcoma (FISS) is a locally invasive and rapidly growing tumor. Computed Tomography (CT) is widely used in the preoperative evaluation for clinical staging. However, CT can also demonstrate the volume of the mass. The aim of the present study was to retrospectively analyze the CT features of 200 cases of FISS and to assess whether there is a statistical relation between tumor volume and local invasiveness of FISS. Materials and Methods: CT images and medical records of 200 cases of histologically confirmed FISS were reviewed. Diameter and volume of the mass were measured on CT (Tomographic Diameter, TD, and Tomographic Volume, TV) and compared to the diameter and volume recorded during previous clinical examination (Clinical Diameter, CD, and Clinical Volume, CV). Statistical analyses were performed on SAS 9.1 software to evaluate the influence of tumor variables (site, first presentation or recurrence, aspect of the mass, number of involved muscles, involvement of underlying bones, presence of skip metastasis, of lymphadenomegaly, of radiographic or histological evidence of lung metastasis), and animal variables (breed, gender, age and body weight of the cat) on CD, CV, TD and TV. Values of Po0.05 were considered significant. Results: No significant difference was observed between clinical and tomographic tumor measurements. A statistical association between the mean diameter (mean of the CD and TD) and each tumor variable was observed. A statistical association was also observed for mean volume (mean of the CV and TV) and each tumor variable. A greater diameter and volume was associated with an interscapular localization, first presentation, heterogeneous aspect, greater number of involved muscles, involvement of underlying bones, and lymphadenomegaly. Smaller diameter and volume were associated with radiographic or histologic evidence of pulmonary metastasis. No statistical association was observed between any animal variable and mean diameter/mean volume. Discussion/Conclusion: The volume of FISS is related to the local invasiveness. Larger tumor mass is more invasive locally, involving a greater number of muscles and underlying bone structures. A larger volume is also associated with an increase of lymph node volume, suggesting further diagnostic steps, but not necessarily proving metastatic disease. In this study smaller tumor volume was associated with a higher occurrence of pulmonary metastatic disease. However, 35% of the animals with lung metastasis were patients with recurrent tumor growth. Furthermore, histological confirmation was not always available. Introduction/Purpose: Elbow dysplasia(ED) is common in Labrador retrievers and is frequently found in radiographic screening. Grade 1 (mild osteoarthritis) is the most common screening result of affected individuals in Finland. Regardless of extensive screening based mainly on secondary radiographic signs of the disease, progress in decreasing morbidity has been modest. The aim of this prospective study was to compare CT findings and radiographic findings of mildly affected dogs to find out if Finnish protocol using one mediolateral flexed radiograph with emphasis on osteoarthritis is sufficient to reach a diagnosis of ED in this breed, or if signs of the primary lesion should carry more weight.

Thirteen clinically healthy Labrador retrievers with screening results 0/1 or 1/0 were studied. Their age ranged from 16 to 28 months (mean 21 months). Age range at the time of the original screening was 12-19 months (mean 14 months). Mediolateral 451 flexed and oblique craniocaudal radiographs were made and a CT study was conducted on both elbow joints. Mediolateral radiographs were first graded according to Finnish Kennel Club's regulations, where main emphasis is on osteophyte formation at the dorsal border of the anconeal process. Secondly, the mediolateral and craniocaudal radiographs and also the original mediolateral radiographs were evaluated by taking into account also blurring of cranial edge of medial coronoid process (MCP) and ulnar trochlear notch sclerosis. CT was used as golden standard for diagnosing ED.

On CT nine dogs had FMCP, three dogs were affected bilaterally. One dog had bilateral osteochondrosis of the trochlea humeri. Ununited medial epicondyle was diagnosed in one joint, which was excluded from the study. On radiographs 12 of 14 joints with ED and two of 11 joints free of ED had blurring of the cranial edge of the MCP. Sclerosis of ulnar trochlear notch was seen in seven joints with ED and in three joints free of ED. In the radiographs taken for the original screening five joints with ED had blurring of the cranial edge of the MCP and seven joints had subtrochlear sclerosis. In the joints free of ED one joint had both signs. The most reliable radiographic sign in the present study was osteophytes distal to the medial epicondyle seen in the craniocaudal view. They were visible in all dysplastic joints but in none of the healthy joints. In the present study, evaluation based on secondary radiographic signs was correct in 21/25 joints (sensitivity 79% and specificity 91%), and in 16 of 25 joints in the original screening (sensitivity 64% and specificity 64%). Discussion/Conclusion: Accuracy of screening for ED in Labrador retrievers using a single flexed mediolateral projection based either on secondary radiographic signs or radiographic signs of the primary lesion was only moderate in the original screening. However, accuracy based on either of these signs was noticeably higher in the present study. Screening for ED at the age of 24 months instead of 12 months and including the craniocaudal view in the radiographic protocol could be considered in this breed.

Skrzypczak P., Atamaniuk W., Kiezbowicz Z., Biez ' yński J., Nicpoń J. Department and Clinic of Veterinary Surgery, Faculty of Veterinary Medicine, Wroczaw University of Environmental and Life Sciences, Poland Introduction/Purpose: Vascular ring anomalies in 95% of affected dogs are represented with persistence of the fourth right aortic arch. Double aortic arch is a very rare anomaly of the vascular ring that results in esophagal compression [1] . The use of computed tomography angiography (CTA) has not been described in a dog with such an anomaly. In humans, double aortic arch is the least common but most symptomatic form of vascular ring anomalies.

A ten-week-old Czechoslovakian Wolfdog (ceskoslovensky vlcak), male, 18 kg, was admitted to the Department and Clinic of Surgery with a suspected vascular ring anomaly. The dog regurgitated regularly shortly after feeding. There were no changes upon clinical examination. The dog was smaller than the littermates. During laboratory examination, there was mild hypoalbuminemia and neutropenia. RX, esophagoscopy and subsequent CTA were carried out. For this purpose, contrast material (Omnipaque s 240 mg/ml) was injected into the saphenous vein at a rate of 2.5 ml/s for 30 s. Transverse images were obtained using helical CT (GE s , light speed VCT) scanner (5 mm thick slides, kVp 120, mAs 4200 matrix 512 Â 512, pitch1.2). Images were acquired 8 seconds after the start of injection. Following data acquisition, 3D image reconstruction was performed using vascular software.

Radiography and esophagography examination revealed dilatation of the esophagus with no signs of displacement at the level of the aortic arch. This was confirmed with the contrast study. The initial diagnosis was the presence of a persistent right fourth aortic arch, although leftward displacement of the trachea was not observed. Computed tomography angiography was performed to further classify the anomaly. 3D CT reconstructions of the patient depicted the presence of second hypoplastic aortic arch. After surgery, the CT angiography procedure was repeated.

The initial diagnosis of the presence of a vascular ring was confirmed for this case. CT angiography was used to better characterize a suspected vascular ring anomaly [2] . It fills the gap between standard diagnostic modalities, such as conventional radiography or echocardiography, and angiography, which is more complicated and hazardous. Non-selective CT angiography is a reasonable choice in the diagnostic process for vascular ring anomalies. Moreover, 3D reconstruction provides an excellent illustration of the problem for the owner or for students who are not involved in diagnostic imaging. Introduction/Purpose: Ultra High Field MRI (UHF MRI) has recently become available in a few research units. Beside the experimental use of spectroscopic and functional MRI applications, it is also used to obtain high spatial resolution images of the cerebral system of humans. The aim of this poster is to present the MRI cerebral images of a clinical case with neurologic disease obtained post-mortem at 7 Tesla MRI and to compare them to macroscopical and histopathological examination.

A 9 year old Yorkshire Terrier was administered to the critical care unit of the Veterinary School of Liège. The dog was in shock and had difficulties to maintain his head in equilibrium after he had fallen from stairs. The shock animal was finally euthanized in comatose state. A CT-and MRI-scan were performed post-mortem. CT revealed fracture of the skull with cerebellar and cerebral trauma. The MRI scan was performed using a Philips 7 T MRI with a 1-channel cylindrical RF-transmitting coil and a 16-channel receiver coil. A T2-weighted GRE -sequence (FOV of 70 Â 70 mm, TR: 60 ms, TE: 25 ms, FA: 15, slice thickness: 0.3 mm, scan time: 58 minutes) and a T1-weighted GRE sequence (FOV of 70 Â 70 mm, TR: 7,7 ms, TE: 3,5 ms, FA: 8, TFE-factor: 219, TFE dur. Shot: 1754,4 ms, TFE dur. Acq: 1685,3 ms, slice thickness: 0.55 mm, scan time: 9 minutes) were performed. After scanning, the head was fixed in a 10%formaline solution, the brain was removed from the skull and a microscopic study using HEstaining of histological sections was performed.

On MRI, there was significant haemorrhage visualised in the left cerebral and cerebellar hemispheres, which were confirmed by macroscopic and microscopic examination. MR image quality was influenced by susceptibility artefacts of blood and air. High resolution T2-weighted images of the left and right cerebellar hemisphere allowed delineation of anatomical areas corresponding to zones of high neurons concentrations on histology. These areas correspond to the areas identified in human neuroanatomy as nucleus dentatus, fastigii, emboliformis and globosus. In humans these structures are responsible for the equilibrium, the muscular tonus and the coordination of movements. Haemorrhage and traumatic damage of these areas may explain the clinical signs of this dog. Discussion/Conclusion: UHF MRI is a promising tool for identification, illustration and description of small anatomical structures in dogs. Post-mortem application of this technique can be a potent help to neuro-anatomists and neuropathologists for describing anatomical details and addressing microscopic examination in the light of abnormalities detected. Secondly, if scan times and cost can be reduced in the future, using such systems in a clinical way can make diagnosis and prognosis more precise and appropriate due to the possibility to visualise neurological structures that can not be visualised at 1.5 or 3 T, because image quality and spatial resolution increase with magnetic field strength. 

A 5-year-old 36 kg intact male German shepherd dog was referred to the Veterinary Teaching Hospital with a 15 days history of right hind limb lameness without previous history of trauma. On physical examination, the right thigh muscles were atrophied and signs of pain and swelling were noted on the right stifle. The rest of the examination and blood analysis were unremarkable. Radiographs of the right stifle were taken.

Multiple foci of lysis with amorphous periosteal reaction were observed within the patella. There was an increase of periarticular soft tissue opacity. Differential diagnoses included primary neoplasia of the patella and osteomyelitis (bacterial or fungal). An echo-guided aspiration of the joint fluid was performed. Cytological results suggested a mixed inflammatory and neoplastic lesion, but infectious etiology could not be ruled out with cytology. Samples from a biopsy of the patella were submitted for histopathology and microbiological culture. Numerous poorly-stained yeast-like organisms were isolated and identified as Cryptococcus neoformans. Discussion/Conclusion: Survey radiographs are the primary tools for investigating osteomyelitis. Radiographs may reveal foci of lysis, periosteal reaction, sequestra and moderate soft tissue swelling. Primary or secondary neoplasias are considered the main differential diagnoses of osteomyelitis. Patellar osteosarcomas have been rarely described in dogs, and radiographic findings are very similar to those found in fungal osteomyelitis. Ultrasound examination can detect fluid collections and surface abnormalities of bone, whereas CT scan can reveal small areas of osteolysis in cortical bone, small foci of gas and minute foreign bodies. MRI has great sensitivity, specificity and accuracy for the detection of osteomyelitis. Although imaging studies help to distinguish osteomyelitis from other differentials, histopathology and microbiological culture continue to be the gold standard for definitive diagnosis.

In conclusion, cryptococcosis should be considered as a differential diagnosis of osteolytic patellar lesions. Introduction/Purpose: The purpose of this study was to evaluate and compare the radiographic efficacy and safety of a non-ionic dimeric and isotonic iodinated X-ray contrast medium Iodixanol [Visipaque320 mg I/ ml (Amersham Health Cork, Ireland)] and a non-ionic monomer contrast medium Iohexol [Omnipaque 300 mg I/ml (Nycomed, Spain)] in feline cervical myelography. Materials and Methods: Five adult healthy cats were studied after injection of Iodixanol (0.5 ml/kg) and Iohexol (0.5 ml/ kg) into the cerebellomedullary cistern through the atlantooccipital space. Ventrodorsal and left to right lateral radiographs of cervical, thoracic and lumbar vertebrae were obtained before and 10, 20, 40, 60 minutes after injection. After completion of the study, the radiographs obtained were evaluated and compared independently by a radiologist who was unaware of the types of contrast agents. Myelographic quality was assessed by numerical scoring method [0 (no contrast) to 4 (excellent myelogram)]. Statistical analysis (nonparametetric test) was performed by SPSS package.

Diagnostically adequate radiographic examinations were obtained with both agents. Adequate opacity in thoracic and lumbar vertebrae was obtained after 10 and 20 minutes post injection for both contrast agents. After 20 minutes contrast agents had reached the end of lumbar vertebrae column. There was no statistical difference between both contrast medias. Evaluation of each of the radiographs showed good to excellent opacification. No adverse effect occurred up to one week after end of the study. Discussion/Conclusion: In conclusion, Iodixanol and Iohexol proved to be safe and effective contrast materials for myelographic studies in cats. Introduction/Purpose: Interpretation of thoracic radiographies is difficult for beginners. Image and computer graphic processing with quantitative analyses of image features could be of help. The goal of this study was to evaluate the accuracy of two semi-automatic analyses: the VHS 1 and the NCA. Materials and Methods: Forty sets of thoracic radiographs of Poodles were divided in two groups: (I) images from dogs with no clinical signs of cardiopulmonary disease (n ¼ 10), and (II) images from patients with proven cardiac disease (n ¼ 30), confirmed by echocardiography to be caused by degenerative mitral valve disease. 74% had severe cardiac morphological changes (LA and/or LV enlargement), 6% had moderate cardiac morphological changes, 17% had mild changes and 3% (only one dog) had a normal cardiac silhouette size. VHS and NCA measurements were performed in right lateral view with computing techniques, by one experienced veterinary radiologist (A) and two trainees (B and C). All the observers had prior information of each group composition. To determine NCA (v 2 ), the contours of the heart and of a square, which base corresponds to one fifth of the distance between five vertebrae (from T 4 to T 8 ), were drawn, and the ratio of both areas was obtained. The interobserver measurements were compared using simple linear regression analysis. The accuracies obtained from the specificity and sensitivity functions were considered.

Group I VHS measurements of all observers had statistically comparable values and the general average could be used. Group II VHS and area related NCA measurements of group I and II had significant different observer values, in these cases only observations of the experienced radiologist were processed. In groups I and II the VHS had normal distributions, ranging respectively from 9.18v to 10.70v and from 9.60v to 14.10v, and having a mean value (AESD) of 10.07v (AE0.44) and 11.69v (AE1.31). The optimal threshold value of VHS was 10.7v which corresponded to an accuracy of 85%. In groups I and II the NCA had normal distributions, ranging respectively from 16.51v 2 to 21.20v 2 and from 19.03v 2 to 42.16v 2 , and having a mean value (AESD) of 19.13v 2 (AE1.58) and 27.87v 2 (AE6.58). The optimal threshold value of NCA was 21.20v 2 which corresponded to an accuracy of 91.7%. Discussion/Conclusion: The accuracy of normalized cardiac area was approximately 7% greater than the one of the VHS approach. Therefore, the NCA appears to be more accurate than semi-automatic VHS method to classify the cardiac silhouette from Poodles as normal or enlarged. The use of NCA measurement showed more difficulties and variable results in observers of trainee group (B and C.). 1 J.W Buchan, J. Bücheler. Vertebral scale system to measure canine heart size in radiographs. Journal of the American Veterinary Medical Association, v. 206, n. 2, p. 194-199, 1995 The aim of this study was to assess the appearance of the feline spleen on abdominal radiographs and to examine intra-and inter-observer agreement.

Right lateral and ventrodorsal projections of the abdomen of 50 cats with no evidence of splenic disease were independently assessed by 4 individuals (2 specialist radiologists and 2 specialist surgeons) blinded to patient details. Assessment criteria comprised borders (not visible, visible but not sharp, sharp), visibility (head, body and tail visible), and shape (not visible, triangular/ folded over/double thickness, rectangular/single thickness). Both intra-and inter-observer agreement was calculated using Cohen's kappa and the intra-class correlation coefficient. The effect of patient and radiographic variables on the appearance of the spleen was tested using the chi-squared test.

The consensus score of the lead assessor assessed that, on the VD and lateral radiographs respectively, the borders were invisible (4 vs 17), visible but not sharp (20 vs 22) and sharp (25 vs 10). The visibility of parts of the spleen were: head (43 vs 21), body (39 vs 14) and tail (34 vs 10). The spleen was triangular/double thickness or rectangular/single thickness for the head, (16 þ 23 vs 20 þ 0), body (19 þ 20 vs 0 þ 14) and tail (4 þ 29 vs 8 þ 2). The shape of the spleen was not visible for head: (7 vs 29), body (11 vs 36) and tail (16 vs 40). There was 1% total disagreement in the consensus score. Kappa values ranged from 0.324 to 0.864. Of the 7 variables assessed on VD radiographs, agreement was moderate (9) and good (12) and on lateral radiographs, agreement was fair or moderate (9) and good or very good (12). Kappa values for inter-observer agreement ranged from À0.006 to 0.852. Of the 7 variables assessed on VD radiographs, agreement was poor or worse (6), fair or moderate (29) and good or very good (7) and for lateral radiographs, agreement was poor or worse (6), fair or moderate (24) and good or very good (12), with no difference in these categories between projections (P ¼ 0.409). There was a significant association between age and the visibility of splenic borders (P ¼ 0.047) and the visibility of the head (P ¼ 0.021) on the lateral projection for assessor 3. There was a significant association between the body condition score and the visibility of parts of the spleen on the VD projection for assessors 1 (body, P ¼ 0.001) and 2 (head, P ¼ 0.001; body, P ¼ 0.049; tail, P ¼ 0.015) Assessor 4 found the borders of the spleen less visible in sedated/ anaesthetised animals on the VD view (P ¼ 0.048) and assessor 2 found the head more visible on the VD projection (P ¼ 0.030) in sedated or anaesthetised animals.

The spleen is commonly visible on radiographs, although its appearance may vary between individuals and according to assessor.

J. Hall, K. Lee, S. Priestnall, C.R. Lamb. The Royal Veterinary College, University of London Introduction/Purpose: Pseudoaneurysm is a blood-filled cavity adjacent to an artery following disruption of the arterial wall as a result of iatrogenic or traumatic puncture. A patent channel to the artery lumen persists and blood continues to flow into a pseudoaneurysm, hence rupture may lead to severe bleeding. There are few reports of pseudoaneurysm in animals. Pseudoaneurysm has been observed most frequently in humans after femoral artery catheterisation (Ahmad et al. 2008 ). Pseudoaneurysm following a cat bite puncturing the radial artery of an elderly woman has been reported (Levis & Garmel 2008). Materials and Methods: Case report Results:

An 8 year old male neutered Maine Coon cat presented initially with two puncture wounds on the medial aspect of the carpus, suspected to be the result of a cat bite. The wounds were treated conservatively but there was a persistent, focal 3 Â 1.5 cm fluctuant subcutaneous swelling on the craniomedial aspect of the right distal radius. The manus was diffusely swollen. There was no palpable thrill or bruit. There was no lameness, evidence of pain on palpation or regional adenopathy. No bone lesions were found radiographically. Aspiration of the fluctuant swelling produced fluid consistent with sterile whole blood. Ultrasonography of the antebrachium demonstrated a cavitary swelling with a thin regular wall and content characterised by a sedimentation, producing an anechoic upper layer and a uniformly echogenic lower layer. On the distal aspect of the swelling, pulsatile blood flow with a communication to the adjacent radial artery and swirling of the blood within the cavity were visualised. Non-selective angiography of the right antebrachium showed extravasation of contrast from the radial artery into the fluctuant swelling. Good collateral blood flow to the distal limb was evident. Diagnosis was radial artery pseudoaneurysm. The lesion was exposed surgically and resected after ligation of the radial artery. Histopathology demonstrated haemorrhage and fibrin surrounded by a thick fibrous granulating capsule of mature fibroblasts and focal areas of inflammatory cells (lymphocytes, plasma cells and macrophages) consistent with a pseudoaneurysm. At follow-up examination 12 days postoperatively, all swelling had resolved. Discussion/Conclusion: As in the present case, ultrasonography usually enables prompt, non-invasive diagnosis of pseudoaneurysm. When the ultrasound findings are equivocal or the anatomy is not well defined, angiography or computed tomography may be useful. In humans, a pseudoaneurysm may resolve spontaneously without treatment or may require compression, ultrasound-guided thrombin injection or surgery. In the absence of any veterinary reports on which to base treatment in the present case, surgical resection was considered a safe and definitive option. 

C.M.R. Warren-Smith, K. Roe, B. de la Puerta, K. Smith, C.R. Lamb. The Royal Veterinary College, University of London Introduction/Purpose: Tumor seeding is the spread of viable tumor cells as a direct result of an interventional procedure such as surgery or biopsy. Tumor seeding along needle tracts in animals is an infrequent event, most often associated with transitional cell carcinoma (1) . Seeding from a pulmonary adenocarcinoma has been reported in a cat (2) . Materials and Methods: Case report Results:

A 7 year old, female spayed Cairn terrier had cardiac pacemaker implantation as a treatment for sick sinus syndrome. Thoracic radiographs obtained at follow-up examination 6 months later showed a soft tissue mass in the left caudal lung lobe. On computed tomography (CT) the mass appeared as a well-defined rounded structure, approximately 1.7 Â 2 cm diameter with a uniform density. No other masses were detected by CT. Cytology following ultrasound-guided fine needle aspiration (FNA) of the mass was diagnostic of pulmonary carcinoma. The tumor was subsequently resected surgically. No signs of metastasis were seen at surgery. Histopathological examination of the resected specimen confirmed adenocarcinoma. At follow-up examination 12 months later a 2.5 cm calcified mass was found radiographically within the soft tissues to the left of the 7th thoracic vertebra. A repeat CT scan showed a large calcified mass affecting the soft tissues of the thoracic wall dorsolateral to the left 5-9th ribs. There was moderate enhancement of the mass following intravenous contrast administration. The location of the mass corresponded exactly to the previous FNA tract. Repeat FNA of this mass was again diagnostic of adenocarcinoma with similar cytological features as the original pulmonary mass. The owners declined further invasive treatment. Palliative meloxicam was administered. At 12 month follow-up the dog had developed a cough and mild exercise intolerance. Repeat thoracic radiographs showed a marked increase in the size of the mass and lysis of the adjacent ribs. The dog continues to receive palliative meloxicam. Discussion/Conclusion: As in the present case, tumor seeding in needle tracts usually becomes evident clinically 2-16 months following the original FNA (3). In the present case, there was no sign of intrathoracic recurrence or metastatic spread of the pulmonary tumor following surgical resection, hence the needle tract seeding was a significant complication that assumed a prognostic importance. Although the risk of tumor seeding following FNA is small, it seems prudent to consider this possibility when planning FNA or biopsy and to warn owners of this possibility when such procedures are considered a necessary part of a diagnostic work-up. Introduction/Purpose: MRI is the best imaging modality for diagnosis of lesions of the axillary region, but due to limited availability, high costs and necessity for anesthesia less applicable. The aim of the study was to characterize the ultrasonographic appearance of soft tissue masses in the axillary region and to determine the value of ultrasonography (US) in diagnosing tumors of the brachial plexus.

A retrospective analysis of soft tissue masses in the brachial plexus region in dogs referred to the Small Animal Surgery at the Justus-Liebig University Giessen between 1989 and 2009 was conducted. Dogs with progressive forelimb lameness, neurologic deficits, muscle atrophy, pain, or a combination of these signs, were examined clinically, neurologically and ultrasonographically. Linear/convex 7.5-15 MHz transducers were used in conscious and sedated dogs. Diagnoses were confirmed by FNA, biopsy, necropsy or MRI.

In 286 patients the axillary region was examined (31 bilaterally) of which 199 axillary regions appeared normal on ultrasound. In 118 (41%) a mass was identified of which 26 showed typical ultrasonographic appearance of lipomas. In the remaining cases, cytology/histopathology revealed 11 fibrosarcomas, 8 metastases of mammary gland carcinomas, 8 undifferentiated sarcomas, 4 histiocytic sarcomas, 3 nerve sheath tumors, 3 carcinomas, 2 lymphosarcomas, 1 melanoma, 1 hemangiosarcoma, and 1 mastocytoma. Other diagnoses were inflammatory lymphadenopathy (11), hematoma (6), abscess and traumatic avulsion of the brachial plexus (each 2). Twenty nine cases were not available for histopathology. In one case the diagnosis of a peripheral nerve sheath tumor was verified by MRI. Within the limits of this study, and the available histopathology, there appeared to be no clinically utilizable relationship between the ultrasonographic appearance and the histological interpretation. Discussion/Conclusion: US detected a lesion in $40% of the examined animals with clinically suspected alterations in the region of the brachial plexus, of which lipoma, fibrosarcoma and inflammatory lymphadenopathy were the most common. US allowed detection of non-clinically evident masses and further evaluation of palpable masses. In addition, US-guided FNA or biopsies may allow accurate diagnosis of the lesion, are minimally invasive, and can safely and easily be performed without harm to nearby vessels/nerves. Disadvantages of US are the limited evaluation of proximal parts of the nerves, nerve roots and spinal cord due to anatomy. Furthermore, a general overview of the region is not possible like on MR images. False negatives can be caused by tumor size, localization and number of involved nerves, experience of the examiner, and quality of the equipment. False positives may occur due to confusion with normal axillary lymph nodes, vessels or artifacts. Thus, we conclude that US is an effective and practicable method to examine the axillary region for soft tissue masses.

Piotr Debiak, Wojciech Lopuszynski 1 , Anna Lojszczyk-Szczepaniak. Department and Clinic of Animal Surgery Laboratory of Radiology and Ultrasonography, 1 Department of Pathological Anatomy, Faculty of Veterinary Medicine, University of Life Sciences in Lublin, Poland Introduction/Purpose: Among the companion animals and livestock, testicular tumors are most common in dogs. The most frequent are interstitial cell tumor (ICT), Sertoli cell tumor (SCT), and seminoma (SEM) 1 . The objective of our study was the imaging modality-based assessment of the frequency and localization of metastases in dogs with testicular tumors.

The study involved 42 dogs of different breeds, aged 6-13 years, whose ultrasound evaluation of testes identified focal lesions in the parenchyma. Abdominal ultrasound evaluation and thoracic radiographs (in 3 projections) were performed to exclude/confirm metastases. On the basis of diagnostic imaging, dogs without metastases were neutered. Dogs with metastatic lesions found in organs were euthanized and necropsy procedures were performed. Tissue samples of testicular lesions and altered internal organs obtained during necropsy were divided for routine histology.

On the basis of the histopathologic examination, the animals were assigned into 4 groups: I-15 dogs with ICT, II-11 dogs with SEM, III-9 dogs with testicular SCT, IV-7 dogs with two histopathologically different testicular tumors (3 dogs with ICT/SEM, 4 dogs with SEM/CST combination). The presence of metastases from the testicular tumors was detected by imaging procedure and confirmed histopathologically in 2 cases (4,8%), namely: one dog with SCT (group III), and one dog suffering from SEM/SCT (group IV) with seminoma metastases. Abdominal ultrasound examination visualized in these dogs irregularly enlarged lymph nodes of the iliosacral lymph center. The dog with SEM/SCT tumors had enlarged inguinal lymph nodes and those from the lumbar lymph center. The enlarged lymph nodes were of oval shape, nonhomogeneous opacity with no hyperechogenic hilus. The radiographic evaluation of the dog with SCT showed the presence of pulmonary metastases. The necropsy revealed micronodular lesions along the testicular cord spreading onto the omentum in both cases. In the case of SCT the histopathological assessment has confirmed metastases to the lymph nodes, omentum and lungs. Discussion/Conclusion: Our study indicates a relatively low risk of metastases from testicular neoplasms in dogs. We observed rare metastases from SCT and SEM tumors, whereas metastases from ICT did not occur. However, focal lesions in testicular parenchyma require thoracic and abdominal imaging to exclude metastases. A diagnostic ultrasound examination of SEM and SCT metastatic altered lymph nodes does not indicate inflammatory reactions. Unfortunately, micronodular lesions of the testicular cord may not be diagnosed by ultrasound evaluation. Introduction/Purpose: In humans, duplication of the ureter is the most common unrinary tract anomaly while it appears to be very rare in domestic animals. In the present report we describe the presence of a duplicated ectopic ureter in a 9 year old Labrador.

A 9 year old male neutered Labrador presented with a history of chronic urinary tract infections, malodorous urine and occasional dribbling of urine noted over the past 2 years. Work up prior to referral had shown a recurrent E.coli infection. Treatment with antibiotics alleviated the symptoms. On clinical examination no obvious abnormalities were detected and haematology and biochemistry were unremarkable. Abdominal ultrasound and intravenous urography (IVU) were performed.

Abdominal ultrasound showed a tubular fluid filled structure with peristalsis measuring up to 2 cm in diameter. It was extending from the cranial pole of the left kidney towards the bladder neck and no insertion into the bladder was evident; the tubular structure was still visible to the left side of the bladder neck just cranial to the pelvic inlet. This was consistent with an ectopic ureter. The right kidney appeared normal. A contrast urinary study was performed one month and a half later, following resolution of the urinary infection. A pneumocystogram performed previous to the IVU showed an air filled tubular and tortuous structure of up to 3 cm in diameter extending from the region of the prostatic urethra to the left kidney. During intravenous ureterography the left renal pelvis appeared slightly distended compared to the right. The right ureter was normal. A left ureter, with only slightly larger diameter than the right and with normal insertion in the bladder was also seen. The distended air filled structure did not show detectable filling with contrast during the IVU but some contrast reflux could be seen at its caudal aspect during a retrograde urethrogram. A duplicated ureter was suspected. Surgery confirmed the presence of a duplicated left ureter with both left ureters terminating at the bladder wall but only the smaller one could be directly catheterised via the ureteral papilla, confirming thus the ectopic termination of the dilated one. The distended ureter was dissected and resected. The dog recovered uneventfully. Discussion/Conclusion: Duplicated ureters are common urinary tract anomalies in humans. But only 2 reports in dogs and 1 in a cat could be found in the literature. In one of the dogs an incomplete ureteral duplication was found while in the other dog a renal and ureteral duplication with normal termination at the bladder was present. So the present report describes for the first time a duplicated ectopic ureter in a dog. Ultrasound allowed identification of an ectopic ureter while the pneumocystography and IVU showed the presence of a normal left ureter along with the ectopic one. The combination of both techniques was important to reach the diagnosis. Introduction/Purpose: Detection and characterization of tumor vascularity by using contrast ultrasonography is useful for differential diagnosis of liver tumors. Sonazoid is a new ultrasonic contrast agent for detection of liver tumors. In ultrasonic contrast imaging in livers, the arterial vessels including tumor feeding artery are contrasted immediately after Sonazoid bolus injection. It is not easy for doctors to distinguish the inflow time of arterial vessels and tumor vessels because of the narrow time window. We therefore proposed a new imaging technique to visualize inflow-time of contrast agents into vessels for further tumor diagnosis.

In this study, we established a localized tumor model by inoculating VX2 tumor into rabbit livers. A female Japanese White rabbit inoculated with VX2 by intramuscular injection was supplied by Sankyo Labor Service Corporation (Tokyo, Japan) as a tumor source. Seventeen female Japanese White rabbits, weighing 2-2.5 kg, were used as recipient animals. For tumor implantation into the liver, a small midline incision was made in the recipient animals and the tumor was injected into the exterior left lobe of the liver (Sasaki 2006). Contrast ultrasound was done on the 14th, 21th, 28th day after tumor implantation. VX2-bearing rabbits were fasted for 24 hours prior to experiments and were anesthetized with a mixture of ketamine and medetomidine. During the experiments, anesthesia was maintained by oxygen and isoflurane. The liver of each rabbit was carefully examined through the skin using a linear diagnostic imaging probe (EUP-L65, Hitachi Medical Co., Ltd., Tokyo, Japan) operated by an ultrasonic scanner (EUB-7500, Hitachi Medical Co., Ltd., Tokyo, Japan). The contrast images with pulse inversion were recorded at a rate of 15 frames per second immediately after Sonazoid bolus injection for 60 seconds. In order to prevent the breathing movement of the liver during recording of contrast images, breathing was suppressed by pancuronium bromide i.v. infusion. Animal experiments were conducted in accordance with guidelines for the care and use of laboratory animals, Faculty of Agriculture, Tokyo University of Agriculture and Technology.

To compare the inflow-time of Sonazoid into tumor vessels and the other vessels, the time course of grayscale intensity of each pixel was measured and the time (frame) which reached 80% of maximum intensity was calculated by using software (MATLAB, MathWorks, Natick, MA). Then the inflow-time mapping was constructed by coloring each pixel according to the each inflow time. By inflow-time mapping, the tumor feeding artery and liver artery can be separately observed in VX2 tumor bearing rabbit liver. Portal veins and liver veins are also distinguished because of the difference of inflow times. Discussion/Conclusion: The inflow-time mapping constructed according to the different inflow time of contrast agent into each pixel would be useful to visualize tumor vascularity and helpful for differential diagnosis of liver tumors. References: Sasaki K, Medan MS, Azuma T, Kawabata K, Shimoda M, Umemura S: Effect of echo-guided high-intensity focused ultrasound ablation on localized experimental tumor. J. Vet. Med. Sci. 68 (10) In present study, kidney dimensions and volume were measured using radiography and ultrasonography and compared with real measurements immediately after necropsy to determine correlations between radiographic, ultrasonographic and real measurements (real size of kidney).

The study was performed on 10 intact mongrel dogs referred to the veterinary teaching hospital for euthanasia for reasons other than abdominal disease. Excretory urography was performed with Iodixanol (Visipaque 320, Amersham Health Cork, Ireland) and the renal length and width were measured on ventrodorsal radiographs. Length, width, height and volume of kidneys were determined by ultrasonography, in two different positions (lateral and dorsal recumbency) for each kidney. Length, width and height of kidneys were measured during celiotomy and after necropsy. Volumes of kidneys were measured by water displacement. All of measurements were compared with linear regression analysis (SPSS package). The gold standard was real dimension of kidneys, measured during celiotomy and immediately after necropsy.

There were positive and significant correlations between ultrasonographic and real measurements of renal length, width, height and volume. Correlation indices were 0.853 and 0.773 for right and left renal volume, respectively. Sonographic dimensions and volumes were smaller compared to the real size during celiotomy and after necropsy. There were also positive and significant correlations between radiographic and necropsy measurements of renal length and width, while radiographic measurements of length and width were greater than the necropsy measurements. There were no significant differences between two positioning of renal ultrasonographic measurements. Discussion/Conclusion: It is concluded that renal measurements were underestimated using ultrasound and overestimated using radiographs. It seems there is no other study to show correlation between renal volume measured by ultrasonography and real one which compare effect of positioning on renal mensuration in ultrasonography.

Introduction/Purpose: All processes of bone healing depend from blood supply and the revascularization process. Related to injury of blood vessels haemorrhage, initial hematoma, and blood clotting cascade and fibrinolysis cause local changes of the physiological blood flow and contribute to initiation of local inflammation. Angiogenesis and the significant increase of blood supply in bone fractures of limbs is a continuous subject to research. Unsuccessful healing has been a motivator for many scientists and clinicians dealing with this problem to investigate the process of normal bone healing and search for new methods to stimulate the healing process. The objective of this study is to identify and describe changes of blood flow parameters in popliteal artery after experimental osteotomy and osteosynthesis of the tibia in sheep using external bone plate fixation in comparison with blood flow of untouched limb. Materials and Methods: Study group consisted of 8 sheep of both sexes and same breed. The mean age was 3 years (AE0.25) and the study duration was 4 weeks. The measurements were performed in 4 sheep after tibial osteotomy and osteosynthesis with ZESPOL external bone plate method and 4 control sheep that were kept under same conditions. Duplex Doppler examinations of popliteal artery right before ramification into cranial and caudal tibial artery were made using Philips EnVisor C ultrasound system. Radiographic examinations were performed simultaneously with no correlation to the final results as a supplementary study of changes in fracture healing. The study has been approved by the ethical committee of the Wroclaw University.

Changes of resistivity index (RI) depending of time elapsed after surgical procedures were noticed. In the first 24 h obtained values were slightly higher than before surgical procedures reaching values of 1.13 to 1.34 and increasing about 0.01 to 0.19. In the next 48 h RI significantly decreased 0.1 to 0.49 in comparison with presurgical results reaching values of 0.67 to 1.10. Peak systolic velocity (PSV) values were highly different in each sheep and oscillated between 40 and 91.3 cm/s. Discussion/Conclusion: Results are presented with reference to the values of blood flow in the same vessels of healthy sheep since normal values and their changes in bone fracture and surrounding soft tissue disruption are lacking. Initial increase of RI values and its later decrease correspond accordingly to the initial (right after injury) phase and inflammatory phase of fracture healing. In the last 2 weeks of study normalization of obtained values was noticed corresponding to the callus formation and bone remodeling phase of fracture healing. PSV values did not follow a specific pattern and no correlation to the time elapsed was found. Introduction/Purpose: High-resolution ultrasonography is considered an optimal image technique to assess normal anatomy and abnormalities of the peripheral nerves. Ultrasound has been used for imaging of peripheral nerves in dogs. However, to the authors' knowledge, information regarding the appearance and ultrasonographic approaches of peripheral nerves in the cat has not been documented. The aim of the present study was to determine the ultrasonographic appearance and approaches to assess the brachial plexus in cats. Materials and Methods: Anatomical nerve study: The dissection of the brachial plexus was carried out in six forelimbs. The landmarks used for anatomical location were axillary artery, scapulo-humeral joint and transverse processes of C5, C6, and C7. Vascular injection and criosections: Red latex was introduced through the thoracic aorta in two cadavers, which were frozen at À201C for 24 hours and then to À801C for 24 hours more. Cross sections were made from cervical spine to elbow. Photographs of each slide were taken to provide the interpretation of the ultrasonographic images. Ultranosographic nerve study ''in vitro'': The brachial plexus were ultrasonographically evaluated using a 4-13MHz liner array transducer in four fresh feline cadavers. The accuracy of the ultrasound localization was demonstrated by injecting ink around the targeted nerves and immediately dissected. Ultranosographic nerve study ''in vivo'': Three live healthy adult experimental cats were sedated to perform the ultrasonographic examination of the brachial plexus using the same technique described in the ''in vitro study''.

The anatomical dissection of the brachial plexus allowed establishing the anatomical landmarks to locate the nerves by ultrasound. The cervical spinal nerve roots were visualized on paravertebral approach. The brachial plexus was localised using the axillary artery as anatomic landmark on the axillary approach. Musculocutaneous, radial, median and ulnar nerves were individually identified on medial humeral approach. The nerves appeared as hypoechoic tubular structures with an echotexture of discontinuous hyperechoic bands surrounded by a hyperechoic rim. Discussion/Conclusion: The axillary artery was the anatomical landmark used for localization of the brachial plexus. The different components of the brachial plexus were identified as hypoechoic structure with internal echoes and surrounded by a hyperechoic rim on medial humeral approach. References: Guilherme S., Benigni L. Ultrasonographic anatomy of the brachial plexus and major nerves of the canine torathic limb. Vet Radiol Ultrasound. 2008. 49 (6) Introduction/Purpose: Diagnostic ultrasonography has been used to define cutaneous layers and changes of skin thickness in relation to hydration status and fluid distribution. Recently, skin thickness has been measured in Shar Pei dogs, a breed considered unique because of its folding and thickening of the skin, with a linear 13 MHz transducer. 1 It was demonstrated that skin thickness in the Shar Pei is greater than in Beagles. This was caused by abnormal dermal deposition of mucin, a substance predominantly constituted by hyaluronic acid (HA). The aim of the present study was to understand how changes in dermal components could likely contribute to differences in dermal echogenicity and to relate them to the histological findings. Materials and Methods: 10 Shar Pei dogs from breeders and 10 Beagles from the Veterinary Faculty of UAB used as controls were enrolled in this study. Shar Pei group included 6 intact females and 4 intact males with a mean age of 2 years, all with the characteristic increase of skin thickness and bundles of wrinkles extending on the forehead, withers and hind legs. Beagle group included 7 intact females and 3 males of which one spayed, with a mean age of 5 years. Dogs were considered healthy on the basis of clinical examination, CBC count, total proteins analysis and urinanalysis. Ultrasonographic examination was made with a 13 MHz linear-array transducer. The ultrasound beam was maintained strictly perpendicularly to the skin surface on frontal, withers, sacral and metatarsal regions. Biopsy specimens were taken from the same skin areas and stained by haematoxylin and eosin s(H&E), Alcian Blue at pH 2.5 and Masson's thrichrome stains.

In all dogs a defined hyperechoic line was observed at the interface between the coupling gel and the skin, which corresponds to the epidermal entry echo. A homogeneous hyperechogenicity with a fine echotexture was observed in the epidermis-dermis interface in the Shar Pei group. In 6 out of 10 Beagles from the control group a double-layered appearance of the dermis were observed. In these dogs the superficial layer was more echoic than the deeper layer. By H&E, collagen fibers were demonstrated to be separated by a pale grey-pink substance corresponding to mucinous material in Shar Pei dogs. By Alcian Blue stain, in the dermis of Shar Pei scattered collagen fibers were between a network basophilic material, considered a characteristic aspect of acid glycosaminoglycans such as HA, whereas by Masson stain collagen fibers stained of a turquoise material. All these findings were not present in the skin of Beagles. Discussion/Conclusion: In Shar Pei, dermal hyperechogenicity, as in human studies, was related to HA-dependent deposition and its ability to form a viscoelastic network together with collagen and elastic fibers. The presence of a diffuse, intense dermal distribution of HA between collagen fibers was further confirmed by the histological results. In Beagles, the distinct bands observed with ultrasound in the epidermis-dermis layers were hypothesized to be related to differences in dermal fluid storage in the interstitium separating the collagen fibers, such as reported in previous studies. 2À3 In conclusion, ultrasound is a useful diagnostic tool to demonstrate dermal changes in Shar Pei dogs. High frequency ultrasonography could be considered a valid alternative to other more invasive methods of investigation such as biopsies, to measure the degree of HA dermal deposition. Introduction/Purpose: Non-cardiac thoracic changes are common in dogs and imaging is of importance in the evaluation of these animals. Ultrasound is a method that can be used to better understand the nature of these lesions, however, prior knowledge of the ultrasonographic anatomy of the region is necessary. This study aims to determine the non-cardiac thoracic structures that can be evaluated by ultrasonography in normal dogs.

Ultrasonographic study of non-cardiac thoracic structures of two normal dogs obtained in the Veterinary Department at the Federal University of Viçosa, Minas Gerais, Brazil. Radiographic examination of the thorax of each animal was obtained in order to exclude thoracic abnormalities. The ultrasonographic examination was performed with the animal in both lateral positioning and each intercostal space was evaluated.

During ultrasonographic examination of the thorax the following structures were visualized: skin, intercostal muscles, ribs, parietal pleura and pulmonary pleura. The image of normal lungs was a smooth hyperechoic line representing the pulmonary pleura and can be visualized as sliding line under the thoracic wall as the animal breathes (sliding sign). Because of the reflection of the sound beam after the pulmonary pleura, reverberation artifacts were visualized as a series of parallel hyperechoic lines. Discussion/Conclusion: The evaluation of the thoracic cavity is difficult because the normal lung, filled with air, does not allow transmission of sound waves due to the marked difference in acoustic impedance between soft tissue and air. Furthermore, the ribs produce a strong posterior acoustic shadow, making the exam more difficult. Thus, intrathoracic structures, such as the mediastinum and normal lung parenchyma are not visualized by ultrasound examination. If pleural fluid is present or a disease replaces the air in the lung parenchyma an acoustic window will be provided. Introduction/Purpose: Renal diseases are usual in small animals, especially in cats. There are few reports describing values for hemodynamic renal vasculature derived by duplex Doppler ultrasonography in normal cats. The purpose of this prospective research was evaluating normal dopplerfluxometrics parameters of renal arteries (RA), interlobar arteries (IA) and abdominal aorta artery (AO) in adult healthy Persian cats. Materials and Methods: Fifty renal units of 25 Persian cats (13 females and 12 males, age range 12-60 months, mean 30 months) were selected for normal by clinical and biochemical examination; for normal systemic blood pressure, and normal anatomy with B-mode ultrasonography. It was performed color mapping of renal vasculature and pulsed Doppler of both RA, IA in both kidneys at three sites (cranial, middle, caudal), and AO. All vessels were measured resistive index (RI). Early systolic acceleration (ESA) of RA and IA in both kidneys was obtained at spectral mapping. It was also obtained ratio indices between AR/AO, and AI/AR velocities.

Results of AO measurements were PSV 53.17 AE 13.46 cm/s and mean diameter of 0.38 AE 0.08 cm. The mean RA diameter of fifty renal units was 0.15 AE 0.02 cm. Considering velocimetric values in both renal arteries the mean obtained was PSV 41.17 AE 9.40 cm/s and RI 0.54 AE 0.07. ESA of RA mean obtained was 1.12 AE 1.14 m/s2. Superior reference value calculated was 3.40 m/s2. Renal-aortic ratio mean obtained was 0,828 AE 0,296. Results of IA measurements were PSV 32.16 AE 9.33 cm/s and RI 0.52 AE 0.06. ESA of all IA mean obtained was 0.73 AE 0.61 m/s2. Superior reference value calculated was 2.0 m/s2. Renalinterlobar ratio mean obtained was 1.45 AE 0.57. Discussion/Conclusion: This research was performed with selected cats as normal, healthy and same ranging of age. RI obtained was similar that referred in literature. Some conditions that leave to a lost of complacency and increase vascular resistance, can affect the waveform at Doppler spectral mapping but not RI. Literature suggests that RI have poor sensitivity for detection of renal disease. ESA parameter of renal and intrarenal arteries provides information of spectrofluxometric alterations of distal vascular bed. To our knowledge, no reports are available describing normal values of ESA to renal vasculature in Persian cats. It was established a new ratio between renal artery peak systolic velocity and interlobar artery peak systolic velocity. This index was developed based on well-known Doppler identifiable effects of stenosis, undependable of the cause. More studies would be necessary to verify hemodynamic behavior of this index in pathological conditions in cats as the effect of age, nephropathies or systemic pressure in these dopplervelocimetric parameters. Introduction/Purpose: Cerebrovascular diseases in dogs seem to be increasing in advancing age. These diseases are defined as brain abnormalities caused by blood supply impairment. Although the strokes are considered important causes of neurological symptoms in dogs, reports in veterinary literature are unusual. The purpose of this study was to describe the underlying causes and ultrasonographic findings of cerebrovascular diseases in small breed dogs, and comparing them to the postmortem findings. 1 

Medical records of transcranial Doppler sonography (TCDS) examinations performed between 2007 and 2008 in 512 dogs with clinical signs of central nervous system disease were reviewed. The criteria for selection of the 52 patients reported here were: small breed dogs with no predilection for sex, showing an acute onset of clinical signs 2 (nystagmus, head tilt, hemiparesis, sudden blindness) with no progressive focal cerebral dysfunction and diagnosis of cerebrovascular disease confirmed at postmortem examinations within 30-60 days after imaging studies. TCDS examinations were performed by the same operator. Dogs were restrained by the owner, without sedation. Hair clipping wasn't often required and acoustic coupling gel was applied above the zygomatic arch to obtain temporal and occipital windows in all dogs. TCDS examinations were taken to evaluate brain parenchyma anatomy and echotexture as well as identifying the ventricular system and detecting focal lesions. Color Doppler flow mapping and Pulsed Doppler spectral analysis were performed to identify the circle of Willis and obtaining the resistive index (RI).

Underlying causes of cerebrovascular diseases were subdivided as follows: 1) cerebral amyloid angiopathy (27%); 2) endocrinopathies (25%) such as Diabetes, Cushing Syndrome and hypothyroidism; 3) coagulopathy (19%); 4) granulomatous meningoencephalitis (11%); 5) Schnauzer hyperlipemia (6%); 6) tumors (6%). Transcranial ultrasonography detected focal encephalic lesions in 22 dogs; diffuse lesions were observed in 20 dogs and 10 dogs showed no abnormalities on B-mode scan. These dogs presented low RI values for one or more arteries studied. Duplex Doppler ultrasound detected higher RI values in 43 dogs. Only 4 animals showed normal Doppler spectral parameters, with RI values ranging from 0,45 to 0,55 3 . Discussion/Conclusion: Results showed that there is no sexual predilection in canine cerebrovascular disease, but older dogs were more often affected, mean age 9 years (range 15 months-18 years). Correlation between TCDS and histologic results indicates that focal lesions detected by TCDS were nonspecific findings and could be associated with several causes including hemorrhagic mass and connective tissue apposition. Diffuse decrease in echogenicity was related to inflammatory diseases and/or edema. Diffuse increased echogenicity was usually related to aging changes. Tumor and granulomatous meningoencephalitis detected by spectral analysis also determined arterial hemodynamic changes. Focal lesions were correlated to local artery hemodynamic changes. TCDS provided hemodynamic information of cerebral circulation adding to B-mode features. 1 

Department and Clinic of Veterinary Surgery, The Faculty of Veterinary Medicine

Cerebrovascular disease. Veterinary Clinics North America Small Animal Practice

Transcranial Doppler ultrasound analysis of resistive index in rostral and caudal cerebral arteries in dogs

Introduction/Purpose: In compare to other diagnostic techniques, ultrasonogeraphy is relatively inexpensive, noninvasive, and allows definition of ocular and retrobulbar anatomy and, with the advent of Doppler imaging, permits measurement of the blood velocity parameters of the orbital and ocular va-sculature1. The aim of this study was to evaluate feline ocular arteries using doppler imaging

Results: Mean PSV and EDV were 14.3 cm/s and 6.1 cm/s respectively for LPCA, 12.8 cm/s and 5.5 cm/ s for SPCA, 13 cm/s and 4.2 cm/s for ACA and 21 cm/s, 8.9 cm/s)for PRA. Discussion/Conclusion: Doppler imaging has the potential for determining no invasively and consecutively the blood velocity parameters found in orbital and ocular diseases including orbital inflammations and neoplasia, vascular diseases including systemic vascular disease (hypertension)vasculopathies, anemia

Doppler imaging of the ophthalmic vasculature of the normal dog, blood velocity measurement and reproducibility