key: cord-0038324-dqa5vdw6
authors: nan
title: Abstracts of Papers Submitted to the American Gastroenterological Association
date: 2019-04-28
journal: Gastroenterology
DOI: 10.1016/s0016-5085(72)80074-x
sha: d1859d613f931432a376e652953a1371b9ad89de
doc_id: 38324
cord_uid: dqa5vdw6

nan

Parenteral hyperalimentation (PH) is currently being used as nutritional support in a variety of clinical situations including pancreatitis. Since prevention of pancreatic stimulation is a principle of the treatment of pancreatitis, this study was undertaken to determine the effect of PH on resting and secretin-stimulated pancreatic exocrine function. Five patients with abdominal complaints but without pancreatic disease were studied as follows: (1) Secretin stimulation was carried out with secretin (Boots) at a dose of 1 U/kg body weight in 2 studies and 2 U/kg body weight in 3 others. (2) Following the 80 min. post-secretin observation, another hour collection of duodenal juice to allow pancreatic secretion to return to resting levels. (3) Next,PH was administered for two hours. The infusate consisted of 5% modified protein hydrolysate and 25% glucose administered at the rate of 200 ml/hr. (4) Secretin stimulation was then repeated while continuing PH. Duodenal juice and urine were collected throughout the study. Blood samples were drawn at the beginning of each period and at the end of the study. All samples of duodenal juice were examined for volume, bicarbonate, total protein,amylase and bilirubin. The blood was examined for glucose, BUN,electrolytes, amylase, osmolality, and the urine for amylase. PH depressed pancreatic secretory flow from a mean resting level of 0 . 45 ± 0.1 to 0.22 ± 0 . 1 ml/kg/60' during paThe mean bi~ rubin output rose from 5.7 ± 2 to 10.3 ± 4 ~g/kg/60' during PH. Bicarbonate, amylase and total protein output were not affected. After low dose secretin stimulatio~PH did not significantly affect the mean bilirubin output. However, the 3 patients who received the higher secretin dose, the bilirubin output rose from 65 ± 33 to 428 ± 154 pg/kg/80'. No consistent effect on the volume, bicarbonate or total protein output was noted. This study suggests that PH increases bilirubin secretion after adequate secretin stimulation. This probably cannot be explained by gall bladder contraction since identical doses of secretin were used during the two periods and no consistent changes m volume response to PH were seen. It would appear that PH does not stimulate pancreatic secretion under the conditions of our study and its use in acute pancreatitis would probably not be contradicted because of that possibility. The purpose of this study was to observe peristaltic activity in the oropharynx using the dL (Antonelle and Glass, Gastroenterology, 60:499-508, 1971 ) because of the advantage offered by its characteristic peristaltic complexes. Twenty patients , 12 normals and 8 patients with various esophageal diseases, were studied measuring manometri£ changes from the upper high pressure zone to the incisor teeth following deglutition. Two serial dL's were used to demonstrate wave propagation, the proximal opening of a second dL at the level of the distal opening of the first dL, and a standard lead at the proximal opening of the first dL. In the pharynx, peristaltic complexes, lasting from 0.6 to 1.2 seconds, were found using the dL. Demonstrable dL complexes resembling peristaltic waves were found in the mouth in some patients. At the junction of the mouth with the oropharynx, usually 6-9 cm from the incisor teeth, motor complexes were seen in both standard and differential leads which had durations ranging from 1.4 to 2 . 2 seconds. In the standard lead, these complexes were positive waves with notched peaks, while in the dL these complexes appeared to be peristaltic waves in continuous sequence. In 15 patients, the first portion of this junctional complex was not transmitted to the pharynx, whereas the second portion of this complex moved down the pharynx. In the remaining 5 patients, the first portion of this junctional wave was propagated down the pharynx. There was no correlation between underlying esophageal disease and the portion of the wave that was transmitted. From its location, this junctional complex may represent the transition point between the first or voluntary stage of deglutition and the second or reflex peristaltic stage of deglutition. The swallowing act in most patients would thus consist of the voluntary stage in the mouth , with the pressure gradient moving aborally, toward the faucial areas. At this point the first wave is followed by a typical peristaltic wave giving rise to these long junctional complexes. Only the peristaltic part of this wave is then transmitted into the pharynx with the voluntary portion of the wave not being transmitted. The transmission of first portion of the wave in some patients is not explained by our present data. To assess cell-mediated immune responsiveness in inflammatory bowel disease, in vitro cultures of washed peripheral blood lymphocytes were obtained from 16 patients with Crohn's disease of the small or large intestine , 17 with nonspecific ulcerative colitis , and 24 healthy individuals of similar sex and age distribution. Cultures contained 2 x 106 lymphocytes suspended in 2 ml of Eagle 's medium supplemented with 15% pooled normal human serum (Group AB Rh neg), and were performed in the presence of either saline (control cultures) or increasing amounts of phytohemagglutinin, conconavalin A or pokeweed mitogen. The resultant DNA synthesis (measured by incorporation of tritiated thymidine after incubation for three days) was assessed in duplicate cultures, and the mean results for each group of subjects were compared for each dose of mitogen. Lymphocyte stimulation by phytohemagglutinin was depressed significantly in patients with Crohn's disease in contrast to healthy subjects (p < 0.05), but only at certain concentrations. Thus, this phenomenon was seen in cultures containing 100 or 50 micrograms , but not 250 or 10 micrograms of phytohemagglutinin. The lymphocyte responses to conconavalin A and pokeweed mitogen were depressed in the Crohn's disease group, but not to a Significant degree. In ulcerative colitis, the mean lymphocyte responses to each of the three mitogens were not impaired significantly. These results suggest a defect of cellular immunity in Crohn's disease which is independent of serum factors and possibly affects only certain lymphocyte populations. Such a defect might be important in host reactivity to an unidentified agent in this condition.

While some coagulation defect is suspected in certain categories of massive G. I. bleeding, routine lab tests have failed to reveal any consistent abnormality. Recently, we have observed an abnormally low platelet adhesiveness with a mean of 17 2: 5% in a group of 16 patients with massive G. I. bleeding following trauma and sepsis. This is significantly lower than the mean of 402: 9% for normal individuals (p<. 01). Similarly, a significant reduction in platelet adhesiveness was noted in 10 patients bleeding following intake of aspirin and in 6 with a history of alcoholism and in 6 with cirrhosis and esophageal varices. The platelet count, however, was not critically reduced except in 6 patients with sepsis and in 2 with cirrhosis. By contrast, in 8 patients bleeding from chronic gastroduodenal ulcer, there was slight or no reduction in platelet adhesiveness which ranged from 26-37%. Eight of the patients with trauma and sepsis in whom this platelet dysfunction was not corrected continued to bleed despite a corrective procedure in 4 of them. They all died. In the other 8 patients , the platelet dysfunction was corrected by transfusion of fresh blood, platelets and frozen plasma. Bleeding stopped in all. Six of them survived and two died of recurrent sepsis later. Similarly in patients with history of aspirin intake and alcoholism and those with chronic gastroduodenal ulcer and esophageal varices, \tn platelet adhesiveness was improved spontaneously or corrected as above, bleeding stopped in all but one. A corrective operation was performed electively later in 7 patients with chronic gastroduodenal ulcer and in 4 with esophageal varices. This reduction in platelet adhesiveness which is essential in the initiation of hemostasis provides a plausible explanation for the persistence and severity of massive G. I. bleeding. It is recommended that by correction of this platelet dysfunction unnecessary risky operation may be avoided in patients with massive G. I. bleeding following sepsis and from aspirin and alcoholic gastroesophagitis and that operative intervention for esophageal varices and chronic peptic ulcer may be postponed to a more favorable time. It is hoped that this new approach would result in substantial decrease in mortality and morbidity from massive G. I. bleeding. THE PREVALENCE OF MACROAMYLASEMIA. D. Barrows, J.E. Berk, and L. Fridhandler. Department of Medicine, Univ. Calif., Irvine, Calif.

All studies reported to date on so-called macrocmylasemia have been concerned with patients selectively studied because of persistent hypercmylasemia. To determ ine the frequency of macroanylasemia in the population at large, a newly described rapid screening method for macroamylasemia (Clin. Chem. 17:423, 1971 ) has been used to examine randomly selected serum samples obtained from three groups: (I) 440 hospitalized patients with a variety of disorders; (2) 251 presumably normal persons represented by Blood Bank donors and a small group of volunteers; and (3) 200 newborn studied by means of umbil ical cord blood samples . A macroamylase was detected on initial excmination in the serum of nine of the 440 hospitalized patients (:20k) . In two of these nine patients, the large companent was no longer present in a second sample taken within a month of the first one. Only one of the 251 normal subjects (0.04%) exhibited a macrocmylase in their sera and none of the 200 cord blood samples contained a macrcmolecular amylase form. The findings suggest that macroamylasemia is probably much more frequent than suspected. The data also indicate that macroamylasemia may be a transient phenomenon and that a macroamylase companent may be present in serum whose total cmylase activity is within normal limits. Several investigators have reported that the fluid transported across the intestinal wall is isoosmotic with the mucosal solution when the mucosal solution is hypertonic. This isoosmotic transport of fluid occurs irrespective of the substance used to increase the osmolality of the mucosal solution. In the present study, we investigated the mechanism by which isoosmotic transport occurs when the osmolality of the mucosal solution is increased by substances of different molecular radii. Everted hamster small intestine was incubated either in isotonic (Krebs-Henseleit solution containing 10 111M glucose, 292 mOsm/kg) or in one of the 5 hypertonic (342 mOsm/kg) mucosal solutions used. The hypertonic mucosal solutions were prepared by the addition of the follOWing substances of diffe£ent molecularcradii to the isotonic so!ution: mannitol Q (4.0 X), erythritol (3.2 A), urea (2.3 A), ethylene glycol (2.3 A) or formamide (2.2 A). (From here on these substances will be referred to as "added solutes"). In order to collect the transported fluid without dilution, the serosal side of the intestine was not bathed. Results show that the transported fluid was isoosmotic with any of the mucosal solutions. The concentration of glucose (Yl) and Na (Y2) in the fluid transported from the hypertonic mucosal solutions decreased ~inearly with the decrease in molecular radius (1) of the "added solutes" (rl = 0.682, fi = 33.09 + 7.81X, P<O.OOl; r2 = 0.867, ~2 = 102.81 + 6.801, P<O.OOl). There was a negative cor· relation between the concentration of "added ~olutes" in the transported fluid (Y3) and their molecular radii (1) (r3 = -0.941, Y3 = 97.07 -23.661, P<O.OOl). These data indicate that, when the osmolality of the mucosal solution is increased by addition of poorly permeable substances such as mannitol and erythritol, the transported fluid becomes isoosmotic with the mucosal solution by increased concentration of glucose and Na but, when the osmolality of the mucosal solution is increased by substances of smaller molecular radii (urea, ethylene glycol or formamide), the transported fluid becomes isoosmotic with the mucosal solution, due to the presence of these "added substances" themselves. Glucagon is known to inhibit the gastric secretory response to food and to exogenous gastrin, but the effect of glucagon on endogenous gastrin release is unknown. We have studied the effect of intravenous glucagon on the serum gastrin response to a meal in 9 Heidenhain pouch dogs and 3 duodenal ulcer patients. Gastrin was measured by radioimmunoassay. Methods: In man, blood samples were taken for basal measurement of gastrin, and basal acid output was measured for 1 hour. A standard meal was given and on test days, 5 mg glucagon was infused over 2 hours. Multiple serum samples were taken for gastrin measurement. Studies in dogs were identical except that the dose of glucagon was doubled and Heidenhain pouch acid output was measured at 30 minute intervals. Results: The averoge basal serum gastrin concentration was 182 ± 65 picograms (pg/ml) in duodenal ulcer potients and 52 ± 5 pg/ml in dogs. In man, there was a 85% increase in gastrin concentration at 15 minutes after eating and 85% increase at 60 minutes. When glucagon was given, gastrin rose only 27% at 15 minutes and after 60 minutes had fallen to below basal levels. In dogs, there was a 75% increase in gastrin concentration at 15 minutes after eating and 76% increose at 60 minutes. Glucagon caused a highly significant diminution in the response to food (only 17% increase at 15 minutes and 23% increase at 60 minutes). The average Heidenhain pouch output for the first hour after eating was 1. 64 mEq. This response was nearly abolished by glucagon (0. 14 mEq). After the glucogon infusion was halted, acid output rose to above control levels. Conclusions: These studies provide direct evidence that glucagon suppresses the release of gastrin in man and dogs. Thus glucagon, like secretin, appears to be capoble of interfering with both the release of gastrin and the action of gastrin in the porietal ce II. Some healthy adults with low lactase levels (LLL) «2 units) develop symptoms of gas, cramps and diarrhea with the amount of lactose found in an 8 oz. serving of mi Ik. Others note symptoms only after a quart of mi Ik. The variation in response to different amounts of lactose was studied in 20 assay-proven LLL subjects. They were fed 3, 6, 12, 24, and 48 gm. of lactose in an electrolyte solution. Seventy-five percent were symptomatic with 12 gm. or less which is equivalent to 8 oz. of mi Ik.Twentyfive percent had symptoms with 6 gm. or less of lactose. Equivalent oral loads of glucose-galactose or of sucrose did not cause symptoms. In LLL subjects there was no correlation between the minimum dosage of lactose which caused symptoms and the degree of lactase deficiency. Six LLL subjects and 2 controls were intubated with a polyvinyl triple-lumen tube and fed PEG (4 gm%) as a marker and 6, 12 or 24 gm. of lactose in an electr~lyte solution simi lar to mi Ik. Specimens were aspirated from duodenum, mid-jejunum, and ileum. In all LLL subjects and controls a 2-3 fold dilution occurred in the stomach. Fluid became isosmotic within the duodenum. However, [Na+] was significantly higher and [K+] lower in the duodenum bf LLL subjects than of controls. In LLL subjects there was decreased net fluid absorption in the jejunum with as little as 6 gm. of lactose and net fluid production in jejunum and ileum with 12 and 24 gm. When glucose solution (60 mM) was added to lactose, net fluid production in the jejunum was significantly decreased (p<.OOI). Feeding lactose with a meal increased the amount tolerated in 13 of 14 LLL subjects. Conclusions: I) Over 2/3 of persons with low lactase levels were aware of symptoms after ingesting the lactose equivalent to 8 oz. of mi Ik (12 gm.); 2) Physiologic derangement of fluid absorption in the upper smal I bowel occurs with as I ittle as 6 gm. of lactose; 3) glucose added to lactose lessens the small intestinal fluid response to undigested lactose; 4) Lactose taken with a mea lis better-to I erated by persons with low I actase I eve Is. To evaluate the role of antacid and atropine therapy on gastroesophageal reflux in the post-cibal state, studies were performed on 25 subjects with histologically demonstrated esophagitis and 7 control subjects without esophagitis. All had radiologically demonstrated sliding hiatal hernia. Occurrence of gastroesophageal reflux was determined by continuous monitoring intraesophageal pKwith a pH electrode placed 5 cm proximal to the lower esophageal sphincter. All the patients with esophagitis and none of the controls showed a fall in esophageal pH below 4 within one hour after the test meal. The antacid significantly increased the reflux-free interval after the test meal and pretreatment with atropine prolonged this period even further . The effectiveness of both the antacid and the anticholinergic therapy was most marked in the patients who had clinical and histologic evidence of esophagitis but without endoscopic changes (mild esophagitis) than in subjects with advanced esophagitis with endoscopic evidence of ulceration and stricture formation (severe esophagitis). The prolongation of reflux-free intervals after the antacid and atropine therapy showed a significant correlation with the resting lower esophageal sphincter pressures. Two different patterns of reflux were noted: (1) prolonged reflux which was characterized by a fall in pH that persisted for 15 minutes or longer, and (2) the brief reflux in which the fall in pH lasted for 2 minutes or less. Prolonged reflux was usually seen in the patients with severe esophagitis and was frequently associated with manometric abnormalities in the distal esophagus. Neither antacid nor atropine therapy modified the pattern of reflux in any of the patients. It is concluded that antacids and atropine are beneficial in prolonging the reflux-free period in the post-cibal state and that their effectiveness is more marked in patients with mild esophagitis with moderate impairment of sphincter competence. However, these agents may have only limited value in the treatment of advanced esophagitis with marked impairment of lower esophageal sphincter pressure. ventricular response and atrial tachycardia resulted in marked diminution of superior mesenteric artery blood velocity with peak levels reduced to 20% of control values, dependent on preceding cycle length (r = 0.88, P < 0.01). During ventricular tachycardia, superior mesenteric artery blood velocity decreased markedly with some beats resulting in no forward flow.

Paroxysmal tachyarrhythmias were followed by blood velocity "overshoot" with the resumption of sinus rhythm. Short diastolic filling periods, reduced stroke output and increased mesenteric vascular resistance account for these alterations.

It is concluded that 1) an aggressive approach be instituted for the treatment of arrhythmias in patients with suspected mesenteric vascular insufficiency and 2) the Doppler ultrasonic catheter flowmeter system is a valuable technique for characterizing superior mesenteric artery blood velocity during cardiac arrhythmias. The purpose of this paper is to present the hypothesis that cholerheic enteropathy is a factor in the pathogenesis of diarrhea and death due to acute and chronic radiation enteritis and to examine the effect of cholestyramine on the prevention and treatment of the acute syndrome. Sullivan has shown in acute experiments that the intestinal absorption of bile salts in radiated rats is diminished. Therefore, it is likely that cholerheic enteropathy occurs after abdominal radiation and that it plays a role in the production of the acute radiation syndrome. This would explain the deleterious effect of bile on acute radiation diarrhea noted by Jackson and Entenman, Sullivan, and Archambeau and the excessive loss of sodium in the stool after radiation observed by Jackson. Similarly, patients with chronic radiation ileitis have bile salt malabsorption so that cramps and diarrhea in these cases may also be a manifestation of cholerheic enteropathy. Preliminary acute experiments in rats suggest that pre-treatment with cholestyramine reduces the incidence of diarrhea and death after abdominal radiation in comparison to rats given radiation alone. When 30 rats were given 4.0 Gms. of cholestyramine daily for 10 days prior to abdominal radiation of 1000 rads, 45% developed diarrhea and 7% died, compared to an equal number of rats given radiation alone in which 67% had diarrhea and 30% died. All of the rats succumbed when the radiation dose was increased to 1250 rads, whether or not cholestyramine pre-treatment was given. Cholestyramine therapy should be considered in patients with acute or chronic radiation enteritis when the diarrhea cannot be controlled by symptomatic therapy alone. The discovery by Levi. Gatmaitan and Arias of two hepatic cytoplasmic protein fractions which appear to be important in the transfer of bilirubin and bromosulfophthalein from the plasma to the hepatic cell may provide a better understanding of the physiochemical factors that determine the excretory pathway of the radiographic contrast materials. These proteins. designated Y and Z. are not present in tissues which do not preferentially extract bilirubin and bromosulfophthalein from the plasma since they are found only in liver. except for the presence of Z in small intestinal mucosa. In addition to a role in the elimination of bilirubin and bromosulfophthalein in bile. the proteins may also be important in the hepatic excretion of the biliary contrast materials. To evaluate the influence of Y and Z on the pathway of excretion of radiographic contrast agents. G75 Sephadex gel filtration experiments were performed with rat liver supernatant to determine the affinity of various radiographic contrast materials for the proteins in vitro. The results show that iopanoic acid and iodipamide. two of the cholecystographic contrast media. bind to the Y and Z protein. while iothalamate. a representative urographic contrast agent. has no affinity for the proteins. This indicates that these proteins may be an important determinant of the pathway of excretion of the contrast materials. Knowledge of this type is important for the future development of less toxic and more organ specific contrast materials for use in clinical radiology.

G. Bevan Department of Medicine. Center for the Health Sciences, Los Angeles, California.

In vitro lymphocyte transformation is impaired in some patients with chronic liver disease.

This impairment is likely to be a result of the disease process but the factors responsible for it are Wlknown.

In order to examine the possibility that bile salts may influence transformation lymphocytes from a nonnal donor were incubated in TC 199 medium with phytohemagglutinin in the presence of varying concentrations of bile salts -cholic, deoxycholic, chenodeoxycholic, lithocholic acids and the taurine and glycine salts of all these acids except lithocholic acid. 0.1 ml of a methanolic solution of a mixture of these substances in equal concentrations, or of each individual acid alone, was added to the lymphocyte cultures to give a final concentration in the medium ranging from 1 to 100 mg/IOO ml.

Blast transformation of lymphocytes was assessed by their ability to incorporate tritiated thymidine added to the culture media 4 hrs . before harvesting the cells.

The amount of tritium incorporated was measured either by autoradiography (bile acid mixtures) or by liquid scintillation counting (individual bile aCids).

Tritium incorporation was not significantly different from controls when the concentration of bile salt was less than 25 mg/IOO ml. Increasing concentrations above this level resulted in diminishing rates of thymidine incorporation until at 50 mg/IOO ml transformation was completely inhibited, although the morphological appearance of the untranaformed cells was normal and they excluded trypan blue stain normally.

Trihydroxy bile acids were significantly less effective in inhibiting transformation than dihydroxy bile acids. No difference was found between free and conjugated bile acids.

It is concluded that bile salt concentrations similar to those found in the serum of some patients with chronic liver disease will impair lymphocyte transformation in vitro. Increased excretion of the divalent cations Calcium (Ca)and Magnesium (Mg) in steatorrhea has previously been attributed to the formation of insoluble Ca and Mg soaps of fatty acids . However, as there is no correlation between total fat and total Ca excretion in the stool, the role such soaps play is unclear. While fecal Ca soaps have been measured in some patients, the total Ca and Mg and the total unesterified fecal fatty acids were not measured and thus a correlation between total unesterified fecal fatty acids and total Ca and Mg, or their soaps, could not be made. We have measured the 24 hr. fecal excretion of total fat, total unesterified fatty acids and Ca and Mg soaps. We also measured the daily fecal excretion of Ca and Mg by atomic absorption spectroscopy. On controlled Ca, Mg and fat diets, 7 normal subjects, 5 patients with Pancreatic Insufficiency and 5 patients with small bowel mucosal disease were studied and the data is summarized in the following 27.1 ± 2.77 8.5 ± 3.14 36 ± 12.0

The total fat and tot;l Ca and Mg excretion did not correlate as patients with Pancreatic Insufficiency excreted more Ca and Mg than those with mucosal disease even though the total fat excreted was much higher in the latter group. However, if one plots all values for the total fatty acid excreted per day versus the percentage of total Ca and Mg excreted as soaps, a straight least squares regression line is ob- The histologic similarity but different behavior makes the surgical management difficult for carcinoid tumors and islet cell tumors occurring in the duodenum. Carcinoid tumors, by metastases, and islet cell tumors, by producing gastric hypersecretion and peptic ulcer disease, are life-endangering. Carcinoid tumors of the duodenum with regional lymph node metastases are best treated by a radical pancreaticoduodenectomy whereas this operation may not suffice to control the ulcer disease for patients with an islet cell tumor associated with either metastases or other islet cell tumors. For the latter patients, total gastrectomy is advisable. In an experience with two patients with islet cell tumors of the duodenum associated with gastric hypersecretion and one patient with a malignant carcinoid tumor of the duodenum followed postoperatively for 12, 6, and 2 years respectively, individualization of the operative procedure performed was based on the presence or absence of gastric hypersecretion and the presence of regional lymph node metastases. The absence of gastric hypersecretion and the presence of regional lymph node metastases dictated the need for a radical pancreaticoduodenectomy. The presence of gastric hypersecretion justified total gastrectomy. However, in one patient with a large duodenal islet cell tumor and another patient with an isolated ectopic islet cell tumor near the duodenum, no further ulcer disease has followed a partial gastrectomy. The need for total gastrectomy for patients for whom a single isolated islet cell tumor has been removed can be determined later on the basis of radioimmunoassay for the level of serum gastrin. The effects of a halogenated anesthetic, halothane, on the rat liver, previously damaged by either carbon tetrachloride or by E. Coli endotoxin, were contrasted with those of diethyl ether and with controls. Microsomal detoxification of hexabarbital was reduced by 50 percent by carbon tetrachloride and was unaffected by endotoxin. Halothane caused a 50 percent reduction in hexabarbital metabolism in the carbon tetrachloride treated rats but had no effect in the endotoxin treated rats. Ether did not further depress hexabarbital metabolism in the carbon tetrachloride treated rats. These results demonstrate that the halogenated hydrocarbon anesthetic has a directly toxic action on the rat liver. This action, which could not be demonstrated in the normal control animals, may be significant in the choice of anesthetics for patients with demonstrated impairment in liver function. 

The observation is made frequently that diarrhea ensues in patients following the oral administration of certain antibiotics. The mechanisms by which antibiotic-induced diarrhea is effected in man are unknown. This study was done in experimental animals to see whether antibiotics eliciting gastrointestinal symptoms as major side effects, induce disaccharidase (DO) deficiencies, a well-recognized cause of diarrhea. DD deficiency of the small intestinal mucosa in man is a·complication of intolerance to several oral ingestants, most notably gliadin, cow's milk proteins, and neomycin. Four antibiotics were chosen for this study: neomycin (NE), erythromycin (ER), ampicillin (AM), and cephaloglycin (CE). Group-paired, healthy, growing Charles River Sprague-Dawley rats were fed ground chow mixes with antibiotics. The dosage of the latter was calculated on a weight-related basis commensurate to therapeutic doses for man. The antibiotics were administered over a period of 10 days. NE-containing feed was partially rejected during the first 3 days, but at no other time were there any changes in feeding habits. Weight gain was similar in all groups and changes at the end of 10 days were not significant between groups. Diarrhea did not occur in any animal group. After sacrifice. jejunal mucosal lactase, sucrase, and maltase activities were measured (Dahlqvist A, Analyt Biochem 22:99, 1968 ). The results showed the following: NE and ER produced no significant changes in DO activity. AM and CE produced unexpected increases in lactase activity (p<0.02) with a slight but not significant increase in sucrase and maltase activities. The drug-induced changes in DD activities could be related to changes in villus height and epithelial cell mass in AM and CE treated animals, but differences between treated and control animals were not significant. No pathological epithelial changes were discernible in all antibiotic treated animals. Although data obtained from animals cannot always be utilized to explain the pathophysiology of human disease, the results suggest a mechanism other than direct cellular injury responsible for antibiotic-induced diarrhea. It has been postulated that disruption of the normal gastric mucosal barrier is a consequence of the destruction of tight junctions between surface epithelial cells, increasing effective pore area and mucosal permeability. Validation of this hypothesis requires, in part, the demonstration of absorption by the gastric mucosa of a lipid insoluble substance with a large molecular radius to which the stomach is not normally permeable. In the present study, the absorption of phenolsulfonphthalein (PSP; MW 364; mol. radius ~6AO) from 7 canine fundic pouches before and after the topical application of 4M urea (an agent known to disrupt the barrier) was determined using a test solution consisting of 80 mM/L HCl, 80 nM/L NaCl, 1.25 ~/ml PSP, and 1 mM/ml PEG. Associated alterations in the net fluxes of H+, Na+, K+, Cl-, and H 2 0 were also assessed. Mean PSP absorption prior to urea instillation was 0.13 + 0.04 ~/30 minutes/lOO cm 2 mucosal area. Following urea, mean PSP absorption was 0.63 + 0.06 ~/30 minutes/lOO cm 2 mucosal area. Associated ionic and water flux rates + -SEM/30 minutes/lOO cm2 mucosal area «+)=pouch gain; (-)=pouch loss) = - These results suggest that, following urea application, mucosal permeability is increased and that increased effective pore area may be responsible. It is recommended that PSP not be used as a volume marker in studieo in which the integrity of the gastric mucosal barrier is being assessed.

Bozymski and Scott Y. Pharr, Dept. of ~1edicine, Univ. of North Carolina Sch. of Medicine, Chapel Hill, North Carolina. An effort was made to define more clearly the role of the esophagus in the speech processes of 16 post laryngectomy patients particularly with respect to intraluminal gastric and esophageal pressures especially in the sphincteric areas and to determine if this is an important factor in relation to intelligibility.

To test speech profiCiency, the patient's pronunciation of 24 random words from a standardized speech test was tape recorded. Ten judges, unaccustomed to esophageal speech, listened to the tapes and attempted to choose from four similar sounding words the word the patient actually read. % intelligibility (I) was then calculated by the following formula: I = Number of Words Correctly Chosen by Judges X 100 24 (test words) Esophageal and gastric pressures were measured using a triple lumen, constantly perfused catheter with lateral openings 5 cm apart. Pressures in the stomach and at various levels in the esophagus, the upper esophageal sphincter (UES) and lower esophageal sphincter (LES) were recorded as the catheter was withdrawn from the stomach to the pharynx .

The tracings were analyzed and % intelligibility was then correlated with intraluminal pressures in the stomach, throughout the esophagus and its sphincters and the pharynx .

All pressures in the stomach and esophagus of the good speakers were averaged and compared with those in the poor speaking group and analyzed using a two -tailed T test. Good speakers were found to have a 27% lower pressure compared to the poor speaking group. P>.Ol.

Pressure at comparable levels was always higher in the poor speaking group; and the difference was statistically significant at certain levels in the esophagus both at rest and with speaking, and in the stomach with speaking.

Other parameters relating to intelligibility were also studied and indicated that younger patients and those having ' more speech lessons spoke more intelligibly.

There is a correlation between pressures and speaking ability in these patients.

Peter Bright-Asare and Henry J. Binder. Yale University, New Haven, Connecticut.

A relationship between OHFA and the diarrhea that often accompanies steatorrhea has recently been suggested (Hofmann, JCI 49:44a, 1970) , in part based on the presence of both an hydroxy fatty acid, ricinoleic-acid (RA), in castor oil and increased fecal ORFA in some patients with diarrhea and steatorrhea. To determine the effect of RA on net water and eler~rolyte transport (~l or pEq/g dry wt/min), an in vivo perfusion technique using C -PEG as a non-absorbable marker was employed in the rat colon. Water, Na and Cl was absorbed from control isotonic saline solutions in 13 animals (183±43 ~l, l8.4±2.8 pEq and l6.5±7.8 pEq respectively). In contrast, significant net secretion of water, Na and Cl (-73±20 pI, -10.4±3.8 pEq and -14.±4.9 pEq respectively) in 12 animals was caused by a 2 mM RA solution solubilized with 5mM Na taurocholate (TC). Saline solutions with 5 mM TC alone did not decrease absorption. Histologic examination of the RA perfused colon demonstrated uniform depletion of goblet cell mucous without other histologic changes. RA induced secretion was partially reversed during 4 hour saline perfusions. The measured net secretion induced by RA is not secondary to absorption of the marker, since recovery of C 14 _PEG was similar in both saline and RA perfused animals (99.6±0.7 vs 99.l±0.9%). Further, RA caused a significant decrease in intraluminal potential difference (PD) (-5.8±1.1 mY, lumen negative) compared to controls (-14.6±1.2 mY). In additional perf us ions unidirectional movement of Na was determined with Na 24 • During saline perfusions Na insorption was 23.l±4.2 pEq and exsorption was 11.8±1.S pEq. During RA induced secretion insorption decreased significantly (14.5±3.3 pEq, p<O.Ol) and exsorption increased slightly (17.2±1.2 pEq). Another series of perfusions demonstrated that oleic acid, which differs from RA only by the absence of the hydroxy (OH) group, did not produce secretion. Thus we conclude that the effect of OHFA on Na absorption is related to the OR group. These studies demonstrated that RA causes net secretion of water, Na and Cl in the rat colon. The mechanism of this net secretion is consistent with an inhibition of Na insorption. This interpretation is supported by the unidirectional Na flux data and the decrease in intraluminal PD. The diarrhea observed in some patients with steatorrhea may be caused by OHFA. Two years ago we int:roduced the concept of the artificial gut, a system for selfadministration of parenteral nutrients. We believe the artificial gut system can do for the patient with severe ch:ronic bcwel disease what the artificial kidney does for the patient with end-stage kidney disease. The artificial gut system consists of the follCMing components: 1) Use of one of 3 :routes of circulatory access: (a) an all silastic indwelling cuffed right atrial catheter that has p:roven :remaI'kably free of cOJJq)lications, including sepsis; (b) creation of an A-V fistula at the wrist with infusicn into the large dilated veins; (c) infusion th:rough a sidearm placed in a standard A-V shunt. All 3 IlEthods provide access to areas of high blood flCM for dilution of hypertonic solutions; 2) A small portable pump to cont:rol infusion rate; 3) Nutrient solutions specially bottled in 2-liter containers to facilitate mixing by the patient at hOIlE; 4) AA alarm device that warns that the bottle is nearly ~ty. Eight patients completed training and 7 were established at home. All patients gained weight during continucus infusion and were able to IIEintain their desired weight with inte:nnittent overnight infusion. Complications directly attributable to the IlEthod were minor and in general were handled by the patient at hOIlE. One unexpected finding was the develop1IEnt of hepatomegaly with mild enzyme elevation and BSP retention in 11 patients, with large fatty livers derronstrated at autopsy in two. This p:roblem is reversible by t~orarily discontinuing infusion of nutrient solutions. The advantages of long term home parenteral nutrition for chronically malnourished patients have been an avoidance of repeated hospitalization costs, a great imp:roVerra1t in patient independence and general state of well-being, improvelIEnt in gast:rointestinal sYJlq)tomatology, and p:robably imp:rovelIEnt of the underlying disease p:rocess. Increased net absorption of water, sodium, and glucose have been demonstrated in remnants after massive gut resections, but the changes in unidirectional fluxes which cause this net change are not known. Accordingly, in vivo ileal perfusions were done on 24 pair-fed sham operated and jejunectomized~ats four weeks after surgery. Net water, sodium and glucose absorption, and unidirectional sodium fluxes were determined by measuring changes in PEG, glucose, sodium, and 24sodium in the perfusate. Jejunectomy caused a 330% increase in water absorption and a 250% increase in sodium absorption, and the addition of glucose to the perfusate increased absorption even more (all p values less than 0.61). The increase in net sodium absorption was due entirely to increased insorption, since exsorption was unchanged. These findings suggest increased active transport of sodium from mucosa to serosa in the remnant, an hypothesis which is further confirmed by in ~ studies of similar remnants in Ussing chambers which showed significant increases in potential difference (0.56 millivolts control to 0.73 millivolts experimental) and in short circuit current across the membrane. The data represent the first direct demonstration of increased unidirectional active transport in adapting gut remnants. We compared pigs experimentally infected with a specific corona virus at age 23 days with match-fed litter mates. Profuse diarrhea occurred within 16 hours of infection charac terized by massive elec trolyte and water losses, alkaline, sugar free stools with normal fat content. Stools (24 hour) from infected and control pigs weighed 261.5 % 74.4 vs. 36.3 % 9.4 g; contained 22.0 % 6.0 vs. 0.9 % 0.6 mEq Na+; 14.2 % 3.1 vs. 1.5 % 0.5 mEq K+; 19.0 % 5.3 vs. 0.4 % 0.3 mEq Cl-respectively. Light and electron microscopy (scanning and transmission) showed structural abnormalities in the mucosa of the proximal jejunum. In the proximal jejunum activities of Na+, K+ sensitive Mg++ATPase (1.40 % 0.13 vs. 0.93 % 0.09 units/mg protein), Mg++ATPase (1.13 % 0.09 vs. 0.76%0.10 units/mg protein) and lactase (21.21 % 5.17 vs. 2.60 % 0.88 units/g protein) were significantly less in infected pigs than controls, whereas adenyl cyclase activity was unchanged. The diarrhea was further characterized by perfusing specific 30 cm segments of proximal jejunum, mid jejunum, distal ileum and the total colon using triple lumen tubes with polyethylene glycol as the nonabsorbed marker. In infected pigs significant changes occurred only in the proximal jejunum. Net flux into the lumen of water (+15.5 ml/hr/30 cm vs. -1.4 ml/hr/30 cm) , Na+ (+1.86 mM/hr/30 cm vs. -0.29 mM/hr/30 cm) and Cl-(+1.50 mM/hr/30 cm vs. -0.40 mM/hr/30 cm) increased compared with data from the same pigs immediately before infection. The unidirectional flux of Na+ into the lumen (using 22Na) increased significantly in the same segments. Suspensions of epithelial cells isolated from the perfused segments of infected pigs had a significantly greater total 22Na efflux rate constant (13.78 % 0.55 vs. 6.74 % 0.18 hr-l ) than cells from controls. Diarrhea in T.G.E. appears to result from a net flux of electrolytes and water into the lumen of the diseased small bowel. We suspect that a massive increase in unidirectional movement of Na+ into the lumen is important in causing these abnormalities. 

The gastric hypersecretory state which characterizes human duodenal ulcer disease might be due either to increased stimulation or to decreased inhibition of normal stimulation or both. To assess possible inhibitor mechanisms, 11 patients with chronic duodenal ulcer disease were studied. In each patient the effects of I.V. saline (control); I.V. secretin (Boots, 2 units/Kg in 4 divided doses); and duodenal perfusion with 0.1 N HCl (7ml/min) on broth-stimulated gastric acid secretion were evaluated. Gastric acid concentration (tested both by titration to pH 4 ("free acid") and to pH 7 ("total acid"» was significantly inhibited (t test, p < 0.001) by either secretin or duodenal perfusion. Total acidity decreased similarly. Neither the amount of NaOH required to titrate from pH 4 to 7 ("combined acid") nor the volumes of gastric juice differed significantly, indicating that neutralization by reflux of alkaline duodenal contents was not responsible for the observed effects. Furthermore, elimination of periods with greater than 10% reflux, assessed by use of a non-absorbable marker in the duodenal perfusate, failed to alter the effect of duodenal perfusion. The inhibitory effect of duodenal perfusion with acid was significantly less marked (p 0.01 ~ 0.05) than that of secretin. The decreased effect might be due to greater variability in the response to duodenal perfusion than to secretin, but an analysis of variance could only demonstrate significant patient variation without identifying a poorly-responding subgroup. The results of this study indicate that failure of end-organ (gastric mucosa) response to secretin does not appear to playa role in the hypersecretion of duodenal ulcer, but that relative failure of release of either secretin or of other inhibitor substances might be a contributing factor. THE VAGAL CONTROL OF CANINE GASTRIC ELECTRICAL AND MECHANICAL ACTIVITIES. J. Carrigg and S. Kohatsu. Stanford U. and VA Hospital, Dept. of Surgery, Palo Alto, Ca.

In order to separate the local (food, distention) and vagal effects of feeding, 4 dogs were prepared in the following manner. The stomach (vagi intact) was transected just distal to the esophagogastric junction. The distal end was closed and the proximal end was anastomosed to the jejunum. The dogs were later instrumen~ed with stainless steel electrodes implanted in the gastric wall and extraluminal strain gages to record the electrical (slow waves) and mechanical activities. Recordings were obtained during fasting, during and following feeding, ' IV insulin, and feeding after insulin stimulation. Following the tests, transthoracic vagotomies were performed. The tests were then repeated. The electrical or mechanical recordings were analyzed for rate changes, and the mechanical for amplitude changes. The pre-vagotomy fasting slow waves were regular with rates of 4.5 to 5.3 cpm. The mechanical activity varied from a pattern of weak omnipresent rhythmic contractions (usually in the body) to a pattern with periods with no contractions followed by bursts of 3 to 10 contractions (usually in the antrum). Insulin hypoglycemia produced a significant increase in electrical rate above fasting within 10 to 20 minutes. Contractions decreased or ceased for 10 to 20 minutes immediately following insulin injection and was followed by rhythmic, strong contractions greater than fasting. Following feeding, in about half of the tests, the slow waves became irregular with a rate decrease for about 5 minutes. Within 10 to 20 minutes post-feeding, there was a significant increase in rate above fasting. Contractions became weaker or absent for 20 to 45 minutes after feeding followed by rhythmic strong contractions. The inhibition of contractions with feeding was great enough to inhibit the strongly contracting stomach of the hypoglycemic animal. The post-vagotomy fasting slow waves were irregular in 3 of 4 dogs, and the contractions were weaker and less frequent. Insulin or feeding produced no electrical or mechanical changes. We conclude that the vagus nerves have multiple effects on the gastric electrical and mechanical activities as vagal stimulation may: 1) increase slow wave frequency; 2) decrease slow wave frequency; 3) increase the frequency and amplitudes of contractions; 4) decrease the frequency and amplitudes of contractions or abolish contractions. Patients with partial gastrectomy and simple gastric achlorhydria may have low serum B12 levels despite adequate amount of intrinsic factor and normal BIZ absorption tests. These findings prompted investigation into the B12 metabolism and gastric acid secretion in 82 subjects with various clinical disorders who had normal B12 absorption .

Serum B12 (Euglena gracilis assay)~ B12 absorption (plasma absorption test) and gastric acid secretion after Histalogw stimulation were measured. Most studies of ionic flux across canine gastric mucosa have utilized closed system instillation and withdrawal of acid solutions into antral or fundic pouches. An experiment was designed to measure ioniC flux across the gastric mucosa in the intact canine stomach by a perfusion technique devised to establish a steady state, obviating the need for calculation of pyloric losses . A method of measuring the residual pool of fluid in the stomach using 57CoB12 as indicator was initially validated in calibrated cylinders . At laporotomy in each of two dogs a perfusing tube was positioned in the cardia and a collecting tube at the pylorus; both tubes were introduced perorally . A second collecting tube was placed immediately distal to the pylorus; this was introduced transduodenally . Perfusion of this system demonstrated a recovery of indicator of 100% and no pyloric losses. Cervical esophagostomies were created in a separate group of dogs; polyethylene perfusing and collecting tubes were positioned flurosco.pically at the cardia and pylorus, respectively. Isotonic 0 . 16 N HC1 labeled with 57COB12 (5 Ilc;1) was continuously perfused and collected at a rate of 42.5 cc/min. for 15 min. periods. Absorption of H+ ion in dog 1 was 962 ± 162 (SEM) ,PEq/15 min . and Na+ entry 758.6 ± 80,PEq/15 min. For dog 2, H+ absorption was 1,957 ± 139 "v Eq/15 min . and Na+ entry 1,178 ± 101~Eq/15 min . There was little fluid entry . Use of methylene blue as perfusate enabled us to trace the area of gastriC mucosa perfused (after sacrifice of the animals) . Absolute H+ loss and Na+ gain was greater than those found in pouch experiments , reflecting a greater surface area of mucosa available for ioniC flux . This technique provides a more physiological means of evaluating the gastriC mucosal barrier in the intact canine stomach.

( Gastroenterology, 56:92, 1969) with normal saline introduced at 15 ml./min. in the caecum. In addition to the measurement of electrolytes transport, water movement, transit time and volume in the colon, continuous recording of potential differences (PD) and -intraluminal pressure in the caecum was done. In two additional studies, the electromyogram of the caecum was also recorded through a suction silver electrode The quality of electric contact was observed by a continuous recording of the impedance. Potential differences increase markedly with the introduction of sodium chloride in the colonic lumen and reach a steady state where variations in time are still considerable (maximum range of variation l2 ±2 mV -mean:!: ISE). Variations of small amplitude of PD do not coincide with the electromyogram in the experimental conditions. Analysis of the curves showed a good correlation (Z=0.99 p< .001) between the average time potential and the true mean potential -TMP (TMP from -8 to -23 mV. lumen negative with respect to blood). There was no evidence of correlation between TMP and pressure variations. Absorption of Na+, Cl-,and water, secretion of K+, transit times and colonic volumes were within the normal range. There was no correlation between sodium absorption from the entire colon and TMP in the caecum (p> .05). The present study provides a tool for the study of colonic physiology which takes into account several parameters simultaneously. It demonstrates the complexity of the phenomena which occur during colonic perfusion. In particular, it suggests that caution should be exercised in the interpretation of PD measurements in the human colon. Ethanol in a dose of 4.4 gm/kg. as 40% solution was administered daily through esophagostomy except for Saturday and Sunday. The dogs received well balanced diet in same quantity daily before and after daily administration of ethanol was started. During the pre-ethanol period the mean maximal acid output (MAO) from the stomach in response to histamine was 18.7 mEq. in the 6 dogs studied. At the end of first month of daily ethanol administration, the MAO increased in all 6 with a mean of 26.3mEq. (40.6%~). The mean MAO remained increased at a similar level for 4 months. The mean maximal acid concentration, however, did not change. Neither the mean MAO nor the mean maximal acid concentration changed in 3 control dogs. In order to explain the increase in MAO following daily ethanol administration, parietal cell mass fCM) and size of parietal cell were determined. In 5 dogs including 2 control dogs and 3 healthy dogs, the mean PCM was 133.5 x 106/kg. and in 4 dogs which received ethanol for 3 months, PCM was 151.7 x 106/kg. (13.6%1'). The mean diameter of parietal cells was 12.8 u in 5 healthy dogs and 25.0 u (95.3%~) in 4 ethanol treated dogs. No changes were observed in zymogenic cells in the alcohol fed dogs. Electron-microscopic examination of the fine secretory tubules in alcohol fed dogs revealed an apparent increase in number, although the tubule lumens were collapsed. Mitochondria appeared nearly doubled in size relative to control animals . Mitochondria had blebs, and others had ruptured christae. Adjacent zymogenic cells and mucous cells contained normal mitochondria and other organelles. These observations suggest strongly that the increased MAO resulting from daily excessive ethanol intake in dogs is attributable to hypertrophy of parietal cells and to some extent, increased total parietal cell mass. These data further suggest that ingestion of large doses of ethanol in nutritionally replete dogs may be causally related to the hyperplasia and toxic changes observed in parietal cells. We observed previously that while cholecystokinin (CCK) stimulates small intestinal motility, secretin inhibits CCK stimulated intestinal motility in man. This observation prompted US to use the 2 hormones for radiographic examination of the small intestine. For the small intestinal examination, 4-6 ounces of barium sulfate mixture were infused into the distal duodenum or proximal jejunum through a Rehfuss tube. After intravenous administration of CCK in a dose of one Ivy dog unit/kg, barium column was followed fluoroscopically and spot films were taken. Cine-radiography was performed also in select patients. Sixty five studies were performed in 57 subjects. The transit time from upper jejunum to ileocecal junction was less than 15 minutes in 54 subjects. 37 of 57 had a variety of small bowel disorders. The examination using CCK revealed abnormal radiographic findings in 15 of these 37 in whom either conventional small bowel examination failed to demonstrate the abnormality or the findings were equivocal. Moreover the examination using CCK demonstrated more detailed mucosal abnormalities and clearer delineation of the extent of diseased bowel, and was able to differentiate between stenosis and spasm. Secretin can be used to inhibit CCK-stimu-1ated motility when one desires to examine a specific segment of the bowel more thoroughly. Secretin is used for hypotonic duodenography also in 40 subjects. The advantages of secretin over an anticholinergic for hypotonic duodenography are 1) prompt inhibitory action, 2) relatively short duration of inhibitory action, usually 10 minutes and 3) no known side effects or contraindications. The radiographic examination of the small bowel using these two hormones, then, can be very revealing when small bowel disease is strongly suspected clinically but the conventional examination does not show abnormality, when the radiographic findings are equivocal, or when one desires more thorough examination of the small bowel which has already been shown to be abnormal. In 3 conscious dogs with gastric fistula, effects of C-terminal octapeptide of CCK (octapeptide) on balloon-stimulated antral motility was studied by constant intravenous infusion of the octapeptide for 20 minutes in doses of 0.1, 0.2, 0.4, 0.8, and 1.6 ug/kg/hr. In each dog, 3 experiments ~ere performed at each dose level. The mean percentage changes of motor activity produced by different doses of the octapeptide were as follows: Dose (ug/kg/hr) 0. The data indicate 1) the maximally effective dose of the octapeptide for the inhibition was 0.4 ug/kg/hr, 2) the amplitude index was more markedly affected than other 2 parameters, indicating that the octapeptide causes partial inhibition instead of complete inhibition as produced by secretin (Chey et aI, Am. J. Physiol. 217:848, 1969) and 3) as the dose of octapeptide was increased beyond 0.4 ug/kg/hr, the inhibitory effect was much less than that produced by the dose of 0. 4 ug/kg/hr. In order to investigate the effect of octapeptide on unstimulated stomach, in 3 anesthetized dogs, a sensitive mini-transducer was placed on the antral wall and pyloric sphincter. The motilities of the two areas were recorded in response to continuous intravenous administration of saline alone and of octapeptide in doses of 0.25, 0.5, 1.0 and 2.0 ug/kg/hr. At the dose level of 0.5 or 1.0 ug/kg/hr., the motilities of both antrum and pyloric sphincter had increased more than 50% of control motilities obtained by saline infusion. These observations indicate that C-terminal octapeptide of CCK stimulates the motor activities of both antrum and pyloric sphincter in unstimulated stomach. The octapeptide, on the other hand, causes partial inhibition when the antral motility was stimulated by balloon. Diphenylhydantoin (DPH) has been used to treat spastic colon patients. The present study was designed to see if DPH would alter the phasic motor activity and contractile state of the in-situ segment of the ascending colon and terminal ileum of the pentobarbital anesthetized dog. The contractile state was indicated by the compliance of the intestine (Am. J. Dig. Dis. 12:1189 , 1967 . A flaccid balloon, connected to a pressure transducer, was introduced into a segment of ileum (33 gm) and colon (40 gm) and the pressure was determined at various balloon volumes (0-160 ml for colon and 0-60 ml for ileum). Compliance was calculated by dividing the change in volume by the change in pressure. When the compliance of both segments reached a steady value, A) DPH (125-250 mg i.v.) was given and followed 1 hour later by neostigmine (N) (0.25-0.5 mg, i.v.) (n=7) or B) N was given followed by DPH (n=5 Thus, the wall of the terminal ileum and ascending colon became more compliant following DPH regardless of whether the wall was in its resting condition or was made less compliant by N. The study also shows that N can decrease compliance of both the ileum and colon made more compliant by DPH. In addition, DPH also abolished or attenuated the phasic motor activities of the colon occuring spontaneously and following distension or deflation of the lumen. The phasic activity of the ileum was not consistently affected by DPH. DPH also inhibited the increased motor activity of colon induced by i.v. infusion of BaC12' These studies thus indicate that intravenous infusion of DPH at the dosages used, decreases the contractile state of both ileum and colon making them more distensible and also decreases the phasic activity of the colon. Its action on colon appears to be longer and stronger than that on the ileum. We sought local effects of the probable laxative component of castor oil, ricinoleic acid, on the electromyogram of circular muscle of the colon. Colons, taken from adult barbiturate-anesthetized cats, were opened along the antimesenteric border, pinned flat to a table of stainless steel mesh with the mucosa up, and put in a 2liter bath of Krebs solution, gassed with 95% 02-5% C02 at 36.5-3B.OoC. Removal of the mucosa exposed circular muscle from 0.5 to 3.5 cm below the ileocecal valve. 6 silver-silver chloride glass pore electrodes, aligned 5.B mm apart, were put on the exposed muscle in the long axis of the colon. Electrical potential differences between these electrodes and a common silver-silver chloride reference electrode were recorded separately through AC amplifiers on a 6-channel ink-writing polygraph. In 43 experiments, records were made for 90 minutes before and 90 minutes after the addition of either sodium ricinoleate or sodium oleate to make bath concentrations of 10-B, 10-7, 10-6 and 10-5 M (of the salt). There were 4 experiments at each concentration and 11 control experiments with nothing added to the bath. Records were read at 5-minute intervals to determine proportions of time with phase lock of slow waves throughout the region studied, frequency and apparent velocity of slow waves. Ricinoleate reduced the phase lock of slow waves, and the effect increased with concentration (p<0.05). The ratios of % time with phase lock in the first 90 minute period to % time with phase lock in the second period were: no fatty acid: 1.34; 10-BM:1.19; 10-7M:1.B2; 10-6M:2.13; 10-5M:2.66. Oleate increased the phase lock, but less significantly (0.10<p<0.20). The corresponding ratios for oleate were: no fatty acid: 1.34; 10-8M:1.30; 10-7M:1.03; 10-6M:0.94; 10-5M:0.B5. Neither fatty acid changed slow wave frequency or apparent velocity. Ricinoleate appears to increase the number of pacemakers within the region studied. The mechanism is local and is probably related to hydroxylation of the fatty acid. It resembles the change in slow waves associated with spontaneous and castor oil-induced diarrhea. Unanesthetized, restrained rats absorb intraduodenally infused triglycerides at rates dependent on the infusion rate, up to clearly definable maxima. The in-vivo rate-cont1slling mechanism is presently notl~own. During steady state infusion of triolein-C maximal uptake rates of lipid-C per unit weight of intestine did not differ greatly in different regions of rat intestine. In the lumen, unabsorbed lipid-14 C was hydrolyzed to the same extent in all regions beyond the first 2/10 during ~aximal rate infusion. Thereafter the percentage distribution of luminal lipid-14 c between fatty aCid, mono-, di-and triglyceride resembled that of a closed system in equilibrium, with lipase in excess of triglyceride substrate. Product removal (i.e. mucosal uptake of lipolytic products) was therefore controlling lipolysis throughout most of the intesihnal lumen in this system.

When steady state maximal triolein-C infusion was prolonged from 2 to 3 hours the total mucosal concentration of lipid-l 4c increased despite a constant rate of mucosal uptake, indicating the presence of an intramucosal reaction whgse average rate was slower than intestinal uptake rate. Intestinal tissue lipid-l C concentrations differed in different regions during steady state maximal absorption. Lipid-14 c concentrations were relatively low in the proximal 4/10 and terminal 2/10 of small intestine and did not increase when i~e infusion period was extended from 2 to 3 hours. The transport time of lipid-C was therefore less than 1 hour in these regions. In the third quarter of small intestine, prolonging t&e steady state infusion from 2 t~ 3 hours resulted in increased tissue lipid-l C levels. Serial measurements of clinical, biochemical, serological and histopathological changes were determined in 50 patients with hepatitis, during both the acute episode and for 13 to 52 weeks thereafter, to assess the relation of these findings to prognosis. While the presence of HAA in 26 patients was associated with a greater predictability of the histological finding of acute viral hepatitis with more strikingly abnormal biochemical tests of hepatocellular function, paralleling the increase in HAA titers and decrease in serum C3 levels, these features in the initial illness did not correlate with prognosis. After 12 weeks, 46 patients were judged to have recovered completely, while four patients went on to develop chronic aggressive hepatitis (CAH), showing either persistence of HAA in the serum beyond 13 weeks or the presence of LE cells in the serum during the acute stage of hepatitis without HAA. All patients with CAH were treated with prednisone with little evidence of clinical improvement. In the two patients with HAA (+ve) CAH, the HAA titers either increased or remained unchanged by steroids and both patients died within six months after the onset of acute illness; the patients with LE cells "telescoped" their clinical course into CAH, lost the LE cells during steroid therapy and one of these died within six weeks. Observations in this study support the conclusion that the initial clinical, biochemical, serological and histopathological changes in patients with acute hepatitis bear no relationship to the ultimate prognosis; patients with HAA that persists beyond 13 weeks after the acute illness or LE cells at the onset of the illness will develop CAH, and in this latter group the course of the disease is accelerated. Achalasia was long regarded to have relatively uniform functional impairments. Manometry initially demonstrated lower esophageal sphincter (LES) "achalasia" (normal pressure. absent relaxation) in all cases. lIIe rsported in 1965 varying LE:S hypertension with incomplete relaxation in 15 cases studied by perfUSion manometry. Additional physiological variations noted subsequently led to this study of 64 patients to define these variations and their implications. Clinical. radiological and manometric appraisal was obtained. roanometry in ell was by rapid perfUSion using 3 to 5. lcm spaced. radially oriented. fixed catheter openings. Testing with betamethacholine done in all subjects was positive though variabla in degree. LE:S pressure .as Significantly higher than for normals in most subjects. In some. LES pressure wae normal or lower than normal. these patiente having mild. atypical symptoms and on radiography minimal to absent LES narrowing. holdup of barium and body of esophagua (BE:) dilatation. In the overall group there were differences in the LE:S swallowing response. Some had no relaxation on awallowing. Incomplete relaxation occurred in most. Completa relaxation occurred in some. The BE ehowed variations in tha axial extant of aperiatalsis and amplitude of waves. IIIhila the BE was entirely aperistaltic in most subjects, peristalsis was present in the uppsr several cm in some, and in the dietal eeveral cm in a small number of others. BE: wave amplitude was low or absent in most. but normal or excessive in some. ProgreSSion of described manometric impairment of ths LE:S and/or BE: was insignificant in subjects with follow-up studies. These stable physiological diversities imply that the causitivs factor must vary in intenSity and occur at ons point in time. A ganglion cell nsurotropic viral infection recently has been hypothesizsd as a cause of this disease. Different degrees of loas of both agyrophilic neurons (supervisory or motor controlling) and of agyrophobic neurons (cholinergic. muscls-firing) has besn idsntifisd. Varying severity and extent of viral infsction and of fallout of nsuron typss is consistant with the range and admixture of cardiospasm, achalasia, cholinergic supsrssnsitivity and aperistalsis described fDr this disorder of eSDphageal contrac-tUe function. Most experimental work concerning the effectiveness of antacid preparations has been done in fasting subjects or subjects fed unusual test meals. Subsequent sampling of ~stric contents has been performed by multiple nasogastric intubations, which may stimulate gastric acid secretion. The Corning pH endoradio capsule (successor to the Heidelberg capsule) permits constant intragastric pH monitoring without gastric intubation or gastric stimulation. Fifteen normal volunteer subjects were studied. All had a usual American breakfast consisting of two scrambled eggs, two pieces of buttered toast, one cup of coffee, and one small glass of orange juice (i.e. 40 g carbohydrate, 18 g protein, 20 g fat, 412 total calories). One hour after finishing breakfast the patient swallowed the precalibrated Corning pH capsule. Baseline pH values between pH 2 and pH 3 were consistently obtained. The subject then swallowed 30 cc of an aluminum-magnesium hydroxide preparation. Intragastric pH levels rose to between 5 and 7. The time interval between the ingestion of antacid and subsequent fall of intragastric pH to pH 3.5 was recorded. The same study was performed on different days using 60 and 120 cc of the antacid preparation. Analysis of the duration of effectiveness (i.e. maintenance of intragastric pH greater than 3.5) for the three amounts of antacid medication showed that in normal patients 30 cc of an aluminum-magnesium hydroxide preparation is as effective as doses of 60 cc and l20 cc. Ca rbenoxolone is a triterpenoid which accelerates heal ing of gastric ulcer in ambulant patients. Since bi Ie reflux into the stomach may be important in the pathogenesis of gastric ulcer, we have examined whether Carbenoxolone prevents the m ucosa l damage which normall y fol lo~s conta~t with bi Ie. Three dogs with 'Heidenhain' po uches were used. The fluxes of Hand Na in the pouch .were measured for three consecutive 30 minute periods, (I) before, (2) during, and (3) after contact with either a 10 mM solution of human bi Ie at pH2 or a simi lar control solution without bile+ A standard acid solution was used in periods (I) and (3). Back diffusion of H from, and gain of Na+ by the pouch were increased in periods 2 and 3 by bi Ie (p ~ 0.01). However, when the pouch had been previousl y ~repared with 100 mgm. of Ca rbenoxolone over a period of 5 days, back diffusion of H was reduced after exposure to bi Ie (p <: 0.01); Na+ gain was the same in both prepared and unprepared pouches. Experiments were repeated four times in each dog and results examined by analysis of variance. In two animals, experiments were repeated after preparing the pouch for only one hour on the day of the test; results were simi lar to those with more prolonged pouch preparation. Carbenoxolone increased production of vis ible m ucus by the pouch. Si nce Carbenoxolone is excreted in high concentration in bi Ie, the effect of a solution of human bi Ie obtained by T-tube from patients receiving Car benoxolone wa s examined in two dogs. This bi Ie was equally damaging to the unprepared pouch despite its Carbenoxolone content. ~arbenoxolone, by a direct action on the mucosa, reduces the back diffusion of H across gastric mucosa after contact with bi Ie; changes in Na+ movement were uninfluenced by the drug. To study the potential pathogenetic role of LAH activities in UC we determined spectrophotometrically optimal assay conditions. kinetic properties and specific activities of 3 LAH activities. acid phosphatase (AP) . S-glucuronidase (SG). arylsulfatase (AS) and a non-lysosomal marker. alkaline phosphatase (ALKP). in colon mucosa from 12 control subjects . In 5 UC patients specific activities of mucosal AP and ALKP were significantly increased (p <O.02.0 . 05 respectively). whereas AS and SG activities were significantly decreased in UC (p <O.05.0.0l). Histologically normal colon mucosa from UC patients had normal LAH activities. Differences were not due to changes in binding of activities to particulate cellular fractions. different kinetic properties. or dissociable inhibitors or stimulators of activities. AP activity varied inversely with both SG and AS from patient to patient. Although mean ALKP activity was increased in UC mucosa. a significant negative correlation (r= -O.78.p<O.05) existed between AP and ALKP for each patient. Additionally L-phenylalanine inhibited ALKP activity significantly less in UC than normal mucosa. suggesting heterogeneity of increased mucosal ALKP activity . LAH activities of isolated circulating lymphocytes from both UC and normal patients assayed under conditions optimal for colon tissue demonstrated no significant activity. Incubation of normal colon mucosa with lymphocytes isolated from patients with UC reproduced LAH differences noted in UC mucosa when compared to incubation of normal mucosa with normal lymphocytes. In conclusion 3 LAH activities of colonic mucosa from UC patients differed significantly from normal. Although individual LAH activities varied. LAH profiles were consistent from patient to patient . Changes in mucosal LAH activity in UC are confined to diseased bowel and are not due to lymphocytes in inflammatory exudate. Sensitized lymphocytes from UC patients can promptly reproduce LAH changes detected in UC in normal colon. suggesting LAH changes may occur early. concomitant with cytopathogenetic changes (GE 51:985. 1966) . LAH changes appear to be divorced from changes in non-lysosomal~KP . It has been shown that bile secretion is dependent upon efflux of bile salts from the liver. Little is known about the kinetics of hepatic bile salt transport. We studied the transport of sodium taurocholate (TC) in stable hepatoma cells maintained at 37°C in growth medium. Cells were incubated in the presence of 1.1 x 10-7 M 3H_ TC at 37 0 C and at 4°C. At 37°C influx of 3H-TC reached steady state within 30 minutes whereas at 40 C influx proceeded exponentially requiring 2 hours to achieve steady state. When the cells were first saturated with TC by preincubation in high concentrations of unlabeled TC (1 x 10-2M) and then exposed to trace amounts of 3H-TC (1.1 x 10-7 M), influx of 3H-TC was more gradual, requiring 1 hour at 370 C and 4 hours at 40 C to reach steady state. Efflux from cells at 37 0 C in the presence of high and low concentrations of TC in the medium was exponential for 4 hours. In contrast efflux was completely eliminated when the cells were incubated at 4 o C. We conclude that:

1. Influx of TC is a passive process . 2. Efflux of TC is an active process which is completely inhibited at 4 o C.

3 . Efflux appears to be energy dependent and rate limiting in the transport of sodium taurocholate by the liver cell. TWenty-two patients were studied by esophageal perfusion within 7 days following esophagoscopy, using acid-pepsin solution or 0.1 N hydrochloric acid as contrasted with saline placebo solution. Sixteen had definite esophagitis documented by photography. Four had a normal esophagus and normal stomach, 2 had a normal esophagus with "superficial gastritis" of the gastric mucosa at the cardio-esophageal junction. Random numbers determined the use of placebo perfusate or acid perfusate for the first half of the test so that the solution was unknown to both the physiCian and the patient. The second half of the test employed the alternate solution. Response to the saline placebo solution provided a baseline for comparison so that the physician was subsequently able to assess the esophageal symptoms which were related to the acid perfusate. Results of the double-blind perfusion study identified the presence of esophagitis (positive test in 15/16) and was negative in 3/4 with no disease present. The test was positive in 2/2 with the presence of inflamed gastric mucosa at the cardioesophageal junction. The recognition of significant heartburn and the good correlation between the induced symptoms and the spontaneous symptoms suggests that a mucosal component exists in this symptom complex. The best correlation existed between the objective evidence of mucosal inflammation and the double-blind perfusion tests. The influence of suggestion by the physician is eliminated by this technique and more meaningful results are thus achieved. We reported CEA positivity in approximately 85% of patients with carcinoma of the pancreas and approximately 5~1o of selected patients with severe alcoholic liver disease. We have done CEA assays (method of Thomson eta~ on the serums of 38 patients with acute and/or chronic pancreatitis -32 alcoholic and 6 nonalcoholic. Sixteen (42%) were positive, all in the alcoholic group. Nine of the 16 had clinical and/or laboratory evidence of liver disease: 7 did not. Four of the 7 were malnourished and also debilitated with, respectively, pancreatic insufficiency, osteomy1e1itis, diabetes, and complicated peptic ulcer. Six of the 7 were under 50 years of age. Neither occult gastrointestinal cancer nor liver disease accounted for the CEA positivity in these 7 patients. Neither chronicity of pancreatitis alone, severity of acute episode, amylase elevation, or reduced serum albumin correlated with CEA positivity. cautious interpretation of CEA positivity in patients with alcoholic pancreatitis is warranted. An ileal condui t used as a urinary bladder in man provides an unusual opportunity to follow the morphologi c and funct i onal changes that occur in a segment of small intestine when its environment is mod ified and its function is changed . In the present study tritiated thymidine was administered to a patient with terminal bladder car cinoma who had an ileal conduit constructed nine months previously. Biopsies of the conduit were obtained fo r intervals up to 148 hours with the dual purpose to define the cell proliferation of t his tissue as well as to describe the wide spectrum of histologi c changes seen in an individual patient. Varying degrees of atrophy was present. Areas with blunt and broadened villi lay adj acent to areas completely devoid of villi. A labeled mitosis cur ve revealed an S phase of approximately 20 hours and a Gl phase of greater than 2 1 hours. The replacement time for an atrophic area of mucosa based on the labeled index at 1 hour of 1.4 cells per crypt was 71 hours or three days while the estimated replacement time based on the observed mitot i c freque ncy of 0 .4% was 250 hours or 10 days. The histological and kinetic information obtained is compared to the normal human ileum and colon as well as clini cal condit ions characterized by atrophy of the small intestinal mucosa. serum against gastric cancer juice appears as a strong dominant precipitin line on double diffusion in agar gel. Both on immunodiffusion and by immunoelectrophoresis, the prevalence of gastric cancer antigen (GCA) over others, shared with noncancer juices, is most conspicuous when the immune serua is absorbed with noncancer juices. After absorption with normalhuman plasma, the GCA appears as a sharp single line having an electrophoretic mobility greater than Gold's CEA. GCA has an immunologically reactive carbohydrat$ moiety, and is different from blood group substances. It can be isolated from gastric secretion in a highly purified state by a phenolethanol extraction. The antigenic material was not detected in the tumor itself but appeared only in the secretion. The corresponding precipitin line disappeared when the immune serum was absorbed with gastric cancer juice, thus establishing its pathognomonic significance. The detection of GCA may be crucial for early diagnosis of gastric cancer particularly in younger patients, still salvageable, when there is free HCl on gastric analysis. 1969 and Sept. 1,1971 and investigated according to a similar protocol, 12 patients, age 2-31, were found to have less than three defecations per week on a high residue diet for 28 days and no known etiology for the abnormality. Colonic perfusion (G.Devroede, Gastroenterology, 56:92, 1969) was performed in all 9 subjects aged 15 and more and 26 controls with isotonic solutions to determine transport of water and electrolytes. transit time and volumes in the colon. The group of constipated su~ects had a longer mean transit time at 10 ml./min. (83.:t18 vs 40 ±5 min. p <.Ol)(x:!:SE) and 16 ml./min. (53:!.7 vs 29 ± 5 min. p ~ .01) of perfusion, a greater colonic volume (485;!:. 45 ml. and 604 ± 56 ml. vs 351 ± 54 and 397 ± 59 ml. p" .05) and a greater absorption from sodium chloride rich solutions of sodium and water (p<.05). but proportionate absorption per minute of contact with the colonic mucosa was less (p~.05) in constipation. Swallowed radioopaque markers were retained longer in the colon of the diseased group (17.4 ± 1.2% and 6.5.t.2% after 3 and 7 days vs 6.2:t 2.4% and 0)( p < .005). For diagnostic purpose. all patients have undergone an anorectal myotomy extending 6 cm. proximal to the dentate line. 5 patients had no neurone, 1 abnormal ones and 6, neurones close to the proximal line of resection. They all had more than 3 defecations per week on a high residue diet (p~.OOl) from 30 to 60 postoperative days. In one patient, mean colonic transit time during perfusion decreased from 93' to 87' at 10 ml. and 82' to 48' at 16 ml./min. after myotomy. Retention at 8 days of markers decreased from 100% to 55%. Colonic transport of water and electrolytes was different (p<.OOl) as it was in 2 out of 3 controls but the patients' absorption returned to the normal range (H 2 0. 5.64 ~ 0.08 to 4.33:!:0.17 ml./min. Na, 910±14 to 728:!:24 ~q/min. Cl, 1045±15 to 959±32 pEq/min. at 10 ml. and H20. 4.53±.0.22 to 2.25±0.14 ml./min. Na, 822.t.38 to 511±.19 ~q/min. Cl. 1098±. 26 to 805±. 24 ~q/min. at 16 ml.) In conclusion. the increased absorption in idiopathic constipation is secondary to an outlet obstruction and it is possible to select patients for myotomy with a simple Objective clinical protocol, when it is difficult to distinguiSh between short segment Hirschsprung and idiopathic constipation with classical diagnostic tools. The reported incidence of CEA positivity in colonic cancer varies widely from series to series. Using the method of Thomson et al we performed assays on 51 patients with proven colonic cancers . The test was positive on 19 (48%) of the 40 patients whose tumors were considered resectable at operation and on all 11 of those considered not resectable. Preoperative CEA levels were correlated with the surgical and pathological findings. The concentrations of CEA in the sera of individual patients were markedly influenced by the extent of the primary tumors and the presence or absence of metastases as follows:

Extent of tumor Noo ~EA positive* CEA 10ng/ml or 1. Localized only to bowel wall 16 2. Extending to pericolic tissues with or without regional lymph node invasion 17 3. Metastatic to distant tissues 18 *CEA level of 2.5 ng/ml or greater in 3 (19%) greater 1 (6%) 9 (53%) 4 (24%) 18 (1000,6) 11 (72%) each of the duplicate tubes.

We conclude: (1) high preoperative serum CEA (10 ng/ml or greater) correlates with disseminated tumors and implies reduced likelihood of curative resection, (2) negative preoperative CEA correlates with localized tumors and implies curative resectability, (3) incidence of positivity depends upon the relative proportion of early to advanced cancers. Two miniature Schnauzers, one Fox Terrier and one German Shepherd with so-called "Achalasia" were studied. With the animals fasting and unrestrained, esophageal motor activity was recorded by multi-lumen side-opening polyvinyl catheter assemblies which were continuously perfused with water at a rate of 0.5 -2.0 ml/min. Cine fluoroscopy after the instillation of barium into the esophagus was also performed. The disorder in dogs exhibits differences when compared to the human disorder. 1) Symptoms invariably appear shortly after birth . 2) The disorder in dogs involves striated muscle rather than smooth muscle. 3) When present, motor activity is seen in the lower 1/2 to 1/3 of the esophagus but absent in the upper 1/2 to 2/3; the motor acti vi ty is usually "peristaltic". 4) With body motor activity, the lower esophageal sphincter (LES) relaxes completely after a swallow ~d is closed by a sphincter contraction in sequence with contraction in the esophageal body. 5) The body of the esophagus does not show hypersensitivity to S.C. Mecholyl given in a dose of 0.2 mg/kg. Other features noted were: 1) Frequent failure of swallowing to produce body motor activity. 2) Relaxation of the LES with swallowing or following balloon distension in the esophagus in the absence of body motor activity. 3) Normal activity in the upper esophageal sphincter. These findings suggest that "Achalasia" in dogs is different than in humans and may represent a congenital immaturity of at least motor innervation to a portion of esophageal body. The studies also provide evidence that LES relaxation and body rotor activity are separable entities. If, as we believe, there is validity in drawing an analogy between chronic active hepatitis and inflammatory bowel disease (IBD) , then an extensive search for the presence of viral invasion is warranted in IBD . Only the epithelium has been examined in detail by electron microscopy in IBD. We have performed an in-depth analysis of all histologic areas of the colon in IBD. Surgical and punch biopsies were obtained from 6 patients with Crohn's disease of the colon and from 5 patients with ulcerative colitis. Control surgical specimens of the colon were obtained from 6 patients without IBD. These specimens were prepared by routine methods and 1,000 electron micrographs were obtained at varying magnifications. The epithelium was unimpressive other than for a remarkable prominence of multi vesicular bodies (MVB). Plasma cells were prominent, contained an endoplasmic reticulum (ER) massively distended with secretory product, and were wedged into ganglion cell clusters, between smooth muscle cells, and in other areas where they are not normally found. Capillary endothelia were uniformly normal. There was a striking membranous change in smooth muscle consisting of the presence of myelin figures, replacement of the cytoplasm by smooth ER, presence of great numbers of MVB's and condensation of dense bodies. Neural fibers showed a prominence of neurotubules and an increased number of neurofilaments. Centrioles were prominent in all cell elements. Mitochondrial dense bodies were prominent and most unusual in appearance. Mast cells were prominent. Fibrin deposits were frequent within the lamina propria as was presence of elastic tissue. Free bacteria were found within the submucosa in a single biopsy specimen. We studied the effects of increased intraabdominal pressure on esophageal motor function in 20 subjects. The manometric catheter consisted of 4 fused tubes (ID 1.6 rom; tips 5 cm. apart) infused wi th water, 1.6 ml/min. Different types of sv1allowing modalities were studied: wet s,~allow (WS) -8 ml ,~ater bolus; "dry swallm~" (,'OS") -infused fluids not aspirated; and dry "dry s"1allow" (D''DS'') -during esophageal suction. Abdominal compression (AC) , with a pressure cuff, in increments of 25 rom Hg caused about 4 rom Hg increases in intragastric pressure. During AC-50 rom Hg, intragastric pressure rose 7.4 ± 3 Hg, lower esophageal sphincter (LES) pressure rose 9.1 ± 6 rom Hg and u~ esophageal sphincter (UES) pressure rose 6.2 ± 10 rom Hg. AC caused the LES to move 1 ID 1.5 cm. proximally. During UES and LES relaxation, pressure fell to the same level as intrathoracic and intragastric pressure, respectively. The duration of LES relaxation for a WS was significantly greater (10.2 ± 1.7 sec.) than for a ''OS'' (8.7 ± 1.6 sec.).

During AC-50, the duration of LES relaxation increased 2 sec. for WS in contrast to 1 sec. for ''OS''. In the distal esophagus, peristalsis associated '.lith a WS had a greater amplitude and complex duration and a slm~er v78ve speed than a D''OS''. Intermediate values were obtained for ''OS''. For WS and ''OS'', graded increases in intraabdomina1 pressure resulted in progressive increases in pressure complex duration and slowing of '~8 speed in the distal esophagus. Effects on D''OS'' Here negligible. This study demonstrates that increased intraabdominal pressure: (1) results in augmentation of LES and UES resting pressure (2) prolongs sphincter relaxation ,~hen a bolus is present, and (3) is associated with prolongation and slm~ing of the peristaltic wave in the distal esophagus for WS and ''OS'', but not D''OS''. Al though unknmm, the mechanism of these effects is probably mediated by neural reflex arcs, perhaps initiated by esophageal stretch or abdominal pressure receptors. The l'.lagnitude of response may be related to bolus volume. In any case, "'hen a bolus is presen t, the esophagus performs vlork to move the bolus from a 1m. pressure cavity (thorax) to a cavity of higher pressure (abdomen). The pre~ sure gradient bea.een these cavities is increased during AC and quantitative changes in peristaltic variables in the distal esophagus are proportional to the gradient. This study e valuates some of the variables which affect the fidelity of manometric recording. Polyvin yl tubes (ID 1 .6 mm) with side holes were used. In an invitro model (modified from Pope) , test pressure complexes were varied in amplitude (50 to 200 mmHg) and duration (1 to 8 sec. ) . Infusion rate (IR) was also varied (0 to 12 ml/ mi n). Results from studies done on 6 esophageal specimens indicated that pressure recording accuracy was: (1 ) in versely related to the pressure amplitude tested, (2) directly related to duration of the pressure complex, and (3) directly related to IR. For example, an IR of 1.6 ml/min . accurately recorded a pressure wave 50 mmHg amp ./8 sec. dur. , whereas an IR of 12 ml/min. was needed to accurately record a pressure complex 100 mmHg amp . /l sec . duro Esophageal pressure complexes were recorded in 6 hu man subjects using IR's of 0 to 12 ml/min. In the distal four-fifths of the esophagu s , the recorded pressure plateaued at an IR of 1.6 to 3.2 ml / min. In the proximal esophagus, the recorded pressure kept in~reasin9 as the IR was increased to 12 ml / min. Pressure amplitude ranged from 60 to 90 mmHg in the distal two-thirds of the esophagus; a pressure trough of 30 to 40 mmHg existed at the level of the aortic arch; and a mean am plitude of 110 mmHg was recorded in the proximal esophagus. Duration of peristaltic complexes averaged 2.5 sec. in the proximal esophagus, compared to 4 to 4 . 5 sec. in the rest of the esophagus. Peristaltic pressure complexes were also recorded simulta neousl y at identical levels by infused catheter tips and subminiature strain gages (Mo del 31 Probe, Honeywell Biomedical). Pressure amplitudes recorded by the strain gages were not affected by var ying the IR in the adjacent catheter tips and showed the same esophageal pressure profile as that recorded by open-tipped catheters i n fused at rapid rates. This study indicates that (1) peristaltic pressure amplitudes are highest i n the proximal esophagus, (2) an IR of 1. 6 to 3.2 ml/min . permits accurate recording of pressure amplitude in the distal four-fifths of the esophagus, whereas an IR of 12 ml/min. is necessary in the proximal esophagus, and (3) a strain gage probe is suitable for high fidelity recording of pressures . A rapid infusion manometric technique, due to resultant gagging, is not feasible for recording pharyngeal pressure. In 6 subjects He recorded pharyngeal motility using a strain gage recording system (Model 31 Probe, Hone)",lell Biomedical). The probe consisted of 3 silicon semiconductors mounted 5 cm. apart within a silas tic tube. Three catheters (ID 1.6 mm), "Iere tied to the silas tic tube so that the side-hole recording tip of each catheter Has at the same level as one of the strain gages. After nasal passage of the recording assembly into the esophagus, the cathe ters I~ere infused ,d th water (1.6 ml/min) and esophageal peristalsis vias recorded . The assembly was then vlithdrawn in 0.5 to 1.0 cm. increments through the upper esophageal sphincter (UES) and pharynx. TI~o dry swallo\ols ,~ere obtained at each station. Infusion of each catheter was discontinued as its tip entered the pharynx. The UES pressure recorded by the Model 31 hobe (66 ± 23 mm Hg) did not differ significantly from that recorded by the infused cathe ters (65 ± 20 mm Hg). The duration of UES relaxation, ho,~ever, recorded by the strain gage ,~as less (1.7 ± 0.5 sec.) than that recorded by the infused catheter (2.1 ±O.6 sec.) and the UES augmentation amplitude was considerably higher, 112 ± 64 vs.

27 ± 8 mm Hg. The pharyngeal pressure profile revealed peristaltic pressure amplitudes ranging from 180 to 420 mm Hg in the hypopharynx, a pressure trough of 30 to 100 mm Hg in the oropharynx, and amplitudes of 80 to 220 mm Hg in the velopharyngeal channel. The uninfused catheters recorded pressures of 10 to 65 mm Hg. Atmospheric pressure existed in the nasal canal. Duration of pharyngeal peristaltic complexes decreased aborally, measuring 0.5 to 0.7 sec. in the velopharyngeal channel and 0.2 to 0.3 sec. in the hypopharynx. Peristaltic wave speed ranged from 9 to 25 cm/sec. Compared to esophageal peris talsis, peak peris tal tic amplitudes in the pharynx are subs tantially higher, the duration of the peristaltic complex shorter and the wave speed much faster. The strain gage probe is a suitable method for accurately measuring the peak pressure transients generated during pharyngeal peristalsis. We believe that accurate determination of pharyngeal peristaltic variables I~ill result in precise characterization of normal pharyngeal function and better evaluation of pharyngeal motility disorders. The ulcerogenic effect of exogenously administered serotonin has led to an investigation of a possible role for this amine in the pathogenesis of acute stress ulcer. Stress, restraint and cold induced ulcers were studied in rats and the glandular stomachs were assayed for serotonin fluorometrically, and also for two of the enzymes involved in the formation and catabolism of serotonin, 5-hydroxytryptophan decarboxylase and monoamine oxidase. Although the gastric activities of these enzymes did not differ in stressed and unstressed rats, the concentration of sepotonin was significantly higher in the stomachs of stressed rats (3.77 ~ 0.29~ gig) compared to control rats (2.66 ± 0.4 ~g/g). Vagotomy offered no appreciable protection against restraint-induced ulcers, and had no effect on the serotonin concentration of the stomadE of either stressed or control rats. Neither the antiserotonin compounds methysergide and cinanserin, nor the inhibitor of serotonin biosynthesis, p-chlorophenylalanine protected rats against stress ulcer, although p-chlorophenylalanine lowered the serotonin concentration in the stomach by 40%. Thus, although serotonin concentration is apparently increased in stomachs of stressed rats, its concentration does not seem to be related to the severity of stress ulceration. Gastric ulcer was also induced in rats by means of hydrocortisone administration (steroid ulcer). There was no difference in gastric serotonin in these rats as compared to untreated controls. Vagotomy, however, offered marked protection against steroid ulcer in contrast to its effect on restraint induced ulcer. These data suggest different pathogenetic mechanisms for steroid induced ulcer as compared to restraint or cold induced ulcers and neither of which involves serotonin. In the present study some evidence is provided that the stimulative effect of histamine on gastric acid secretion is mediated by cyclic adenosine-3', 5' -monophosphate (cyclic AMP). In male Wistar rats (180-200 g b. w.) the content of cyclic AMP in unstimulated gastric mucosa was 1. 01 ± O. 17 nmole / g wet wt, obtained by an isotope dilution procedure with purification of the cyclic nucleotide and its subsequent determination by enzymatic cycling. The gastric acid output was determined employ· ing LAI's perfused rat stomach preparation and was shown to increase with increasing doses of histamine diphosphate (HDP) given subcutaneously; up to a dose of 6.0 mg HDP /kg, the rise in acid secretion was correlated with a dose-dependent increase in the cyclic AMP content of gastric mucosa (correlation coefficient r=O. 972; 3 degrees of freedom). In order to elicit temporal correlations between effects of HDP on gastric mucosal levels of cyclic AMP and the secretory response, HDP (0.75 mg/kg) was applicated rapidly by intravenous injection. Subsequently, the cyclic AMP level in gastric mucosa increased within a few minutes preceding the secretory reaction. -In adrenalectomized rats, subcutaneously administered HDP produced only minor rises in the gastric acid output and in the cyclic AMP content of gastric mucosa, whereas HDP was fully stimulative in animals while on replacement therapy with dexamethasone. In rats with intact adrenals, dexamethasone did not enhance histamine-stimulated gastric secretion or cyclic AMP production. -According to these results, histamine is suggested to stimulate gastric secretion via an increased formation of cyclic AMP, which is rendered possible by the permissive role of glucocorticoids. -This study was supported by grants from the Deutsche Forschungsgemeinschaft, Bad Godesberg, West Germany. (Stifel F.B. et al B.B.A.170 : 22l 1968) and active transport (McManus J.P.A. and Isselbacher K,J., Gastroenterology 59:214, 1970 ) is well established. However the effect of dietary manipulation on intestinal dipeptidase activity is unknown. Decreased activity of certain jejunal dipeptidases in the presence of normal small bowel morphology has been observed in undernutrition (Kumar V. and Chase H.P., Clin.Res. 18:204 1970) . The purpose of the present study was to determine whether the intestinal mucosal dipeptidase, glycyl-proline hydrolase is regulated by specific intra-luminal substrates. Intraduodenal canulae were inserted via a gastrostomy in 2 groups of 5 male Sprague Dawley rats weighing from 150 to 200g. The animals were restrained in Bollman cages and infused with either 6mM glycyl-proline or a mixture of 6mM glycine and 6mM proline at a rate of 1 ml/hr for 48 hours. Ten further control animals were infused intraduodenally with a physiological electrolyte solution for 48 hours. The animals were then sacrificed and scrapings of intestinal mucosa were obtained from both the upper jejunum and the lower ileum. The activity of the mucosal dipeptidases, glycyl-proline and valylproline hydrolase were measured by the method of Josefsson and Lindberg (B.B.A.l05:149 1965) and expressed as U/mg protein. The jejunal glycyl-proline hydrolase activity (mean + S.E.M.) in the glycyl-proline group(5.5+0.l3) was significantly greater than either-the electrolyte(4.l±0.2l)or the glycine ;nd proline(4.3~0.2l)groups(p<0.005). The adaptive changes appeared to be more marked in the ileum. In those animals infused with the intact dipeptide, the mean ileal glycyl-proline hydrolase activity was 5.1+0.15 compared with 3.5+0.21 and 3.4+0.49 in the electrolyte(p<0.005) and in the amino acid (p(O.Ol)groups respectively.-There was no statistical difference in valyl-proline hydrolase activity in either of the 3 groups. These data would suggest that intestinal glycyl-proline hydrolase activity is regulated by specific intraluminal substrates. Vagal stimulation in man using insulin-induced hypoglycemia evokes a variable response in serum gastrin levels; some workers have concluded this stimulus does not result in significant gastrin release. The latter hypothesis, however, ignores both the inhibition of gastrin release by antral acidification as well as the non-gastrin mediated stimulation of gastric acid production by insulin hypoglycemia.

We investigated the effect of antral pH on gastrin release by studying the serum gastrin response to insulin induced hypoglycemia in normal subjects and in patients with pernicious anemia. We further evaluated the effects on serum gastrin levels of antral acidification in the achlorhydric patients and of antacids in the normal subjects.

In 9 normal subjects, insulin induced hypoglycemia produced a mean increase in serum gastrin of 85 picograms per ml (pg/ml). The administration of antacid sufficient to maintain antral pH above 5.5 resulted in serum gastrin elevations of 100 to 150 pg/ml; addition of insulin hypoglycemia produced further elevation of 200 to 350 pg/ml.

Patients with pernicious anemia showed increases of 300 to 900 pg/ml following insulin administration. Antral acidification produced a prompt decrease in serum gastrin and the gastrin response to hypoglycemia was completely abolished.

Conclusion: Vagal stimulation is a potent stimulus of gastrin release in man; such release may be abol ished by antral acidification. Differences in antral pH may explain the variable response in normal subjects. 37 patients with duodenal ulceration were studied with esophageal manometry pre-operatively and then between 6 and 12 months later. 28 had vagotomy and either a pyloroplasty or gastroenterostomy pe rformed. 9 had a Polya p a rtial gastrectomy with or without a vagotomy. Using non-perfused water filled open tips (int. diam. 1.12 rom.) the mean maximum pressure above intragastric was 9.8 ~ 6.4 cm. H 0 and the zone of elevated pressure (ZEP) was 4,4 ~ 1.1 cm. long. The difference 2between mean maximum and intragastric pressure by open-tip was unchanged following vagotomy and draina~e but reduced significantly after partial ~astrectomy (3.2 ~ 2.5 cm. H 2 0 p < 0.05). The ZEP was shortened in both groups t3.5 ~ 1.0 cm. and 2,9 ~ 0.8 cm. p < 0.05). These parameters were also significantly different when measured by balloon. The mean intragastric pressure (open-tip only) was increased in both groups post-operatively from 17.9 ~ 5.0 to 22.6 ~ 4.7 cm. H 2 0 p < 0.05. These manometric findings suggest that the gastroesophageal sphincter is not impaired following vagotomy and drainage operations and by these techniques there is no increased incidence of hiatal hernia following either type of operation. No attempt was made to measure reflux, but following partial gastrectomy this is more likely to be due to a weak sphincter than a hiatal hernia. The low pressures following a partial gastrectomy may be due to removal of the antrum which produces gastrin. --ulCerative colitis (UC) is a mucosal disease of unknown etiology in which epithelial cell renewal may be altered. Some investigators have reported increased whereas others have described decreased epithelial cell proliferation in UC using short-term (1-2 hr) in vitro DNA labeling methods. We describe studies of epithelial cell renewal in UC and normal rectal mucosa using an organ culture method which has been shown to maintain mucosal viability and continued epithelial cell renewal in small intestine for at least 24 hr (J Clin Invest 48:1423 , 1969 . Suction rectal biopsies were obtained from 11 patients with UC and 5 normals. After incubating 2 biopsies from each subject over thymidine-3 H-containing medium for 6 hr, 1 specimen was removed from culture. The other specimen was allowed to continue in culture over isotope-free medium for an additional 18 hr. Light and electron microscopy revealed excellent preservation of mucosal architecture and epithelial cell morphology after 24 hr of culture. To quantitate epithelial cell renewal, radioautographs of l~ sections of Epon-embedded tissues were prepared. After 6 hr of culture, a mean of 12.47% of epithelial cells in the lower 2/3 of the crypts were labeled (labeling index) in UC whereas a mean of 5.15% were labeled in normals. Moreover, twice as many labeled epithelial cells were seen in the upper 1/3 of the crypts in UCs (0.42/crypt) than in normals (0.2l/crypt). Occasional surface epithelial cells were labeled after 6 hr in UCs (0.62/histologic section) whereas no surface epithelial cells were labeled in normals. After 24 hr, labeled cells in the upper 1/3 of the crypts had increased to 1.3 cells/crypt and labeled surface cells to 2.l9/section in UCs compared to only 0.54 labeled cells/crypt and only 0.07 labeled surface cells/section in normals. These studies indicate 1) that human rectal biopsies remain viable in vitro in organ culture for at least 24 hr, 2) that epithelial cell proliferation and migration are accelerated in ulcerative colitis as documented by a more than two-fold increase in labeling index at 6 hr and migration of many labeled cells to the upper crypts and absorptive surface after 24 hr of culture and 3) that organ culture may be applied to the study of metabolic processes in ulcerative colitis and other rectal mucosal disease. mYSIOLOGIC CHARACTERISTICS OF THE HUMAN PYLORIC "SFHINCTER". Robert Fisher and Sidney Cohen. Hospital of the University of Pennsylvania and VA Hospital, Hrl.la., Pa.

It has been generally accepted that the human pylorus has no independent sphincteric properties. However, this conclusion was based upon manometric studies perfonned with methods now considered to be insensitive. To define the p,ysiologic characteristics of the human pylorus, infused, open-tipped catheters were slowly withdrawn from duodenum to stomach in 23 normal subjects. Mucosa to skin KCl potential difference (PD) was monitored for precise localization of the gastro-duodenal junction. In the basal state, a high pressure zone 1.6±O.1 em (mean ± S.E.) in length was detected. This zone had a pressure of 5.5iO.5 DDll Hg above intra-abdominal pressure and was closely associated with a PD change of -34.7j).1 D1V. A decline in pyloric pressure preceded antral peristalsis. Duodenal acidification with O.lN HCl (48 meq/ hour) significantly increased pyloric pressure to 18.1±1.4 DDll Hg (P<.OOl) while saline perfusion caused no significant change. Using phenol red as a marker there was 34.2±9.3% renux of duodenal infusate into the stomach during perfusion with saline as compared to only 1.6±O.8% duodeno-gastric renux during acidification (PcC.001). An elevation in pyloric pressure was associated with a decrease in pyloric renux. Endogenous hormone released by duodenal instillation of olive oil caused an increase to 12.6±1.3 DDll Hg (p.( .01) while amino acid infusion produced an increase to 15.2±2.3 DDll Hg (P<.Ol). Finally, the intravenous administration of cholecystokinin or secretin produced significant elevations of pyloric pressure. In response to cholecystokinin infusion (l.Ou/kg/hr) pyloric pressure increased to 10.4±1.9 DDll Hg (P<.02) while in response to secretin (l.Ou/kg/hr), the pressure increased to 10.3± 1.6 DDll Hg (P< .005). The elevations of pyloric pressure in response to individual exogenous hormonal administration were less than those caused by endogenous hormonal release. These studies indicate that: (1) the human pylorus is associated with a zone of elevated pressure that relaxes preceding antral peristalsis; (2) this high pressure zone increases its tone in response to both endogenous hormonal release and exogenous hormonal administration; and (3) an increase in pyloric pressure diminishes duodeno-gastric renux. From these data, we conclude that the lruman pylorus functions as a hormonally-regulated, ~siologic sphincter. Our previous studies have indicated that permanent mucosal damage is not responsible for the transport changes when acidic test solutions are introduced into human jejunum. To determine the importance of protonation of the amino acid side chain as a mechanism for the reduced absorption, intestinal absorption rates of free amino acids with long (leucine) and short (glycine) side chains were investigated in the presence or absence of excess H+ in the lumen. Utilizing a constant in situ perfusion technique, test solutions containing 40 mM leucine, 130 mM NaCl, and having the pH of either 6.90 or 2.25 were infused into a jejunal segment. On separate occasions, test solutions containing So mM glycine, 110 mM NaCl, and having the pH of either 6.90 or 2.25 were infused in the same segment. The mean intraluminal pHIs during the infusion of neutral and acidic test solutions were 6.90 and 3.49 respectively. While acidification modestly decreased leucine absorption rate from 7.04 to 4.S4 umoles/min/cm it severely reduced the glycine absorption rate from 15.63 to 1.10 umoles/min/cm (p~ 0.001). Since enzymatic digestion is markedly retarded by the acidic milieu, the following studies were performed to determine whether jejunal disappearance rates of glycylleucine and glycylglycine are also altered when the intraluminal pH is considerably reduced. Glycylleucine is an excellent and glycylglycine is a poor substrate for the membrane-bound enzymes. The concentration of each dipeptide solution was 40 mM. The pH of each test solution as well as the changes in intraluminal pH was similar to the free amino acid studies. Acidification of intraluminal fluid decreased the disappearance rate of glycylleucine from S.OS to 4.92 and of glycylglycine from 7.S2 to 3.64 umoles/min/cm. However, there was considerable glycine absorption (6.72 umoles/ min/cm) from the acidified glycylglycine solution. Our data indicate that: I} Introduction of a positive charge in the side chain may account for the reduced rate of disappearance of free amino acids and dipeptides in human intestine. 2} In the acidic milieu where there is a profound reduction in the absorption rate of glycine the rate could be considerably increased by presenting glycine in dipeptide instead of free form to the intestinal mucosa. Intestinal surface glycoproteins and mucus may be elaborated, or secreted, in response to external stimuli, but these stimuli are presently unknown. Since cyclic AMP mediates many hormonal actions, studies were initiated to determine whether stimuli, known to raise intracellular cyclic AMP levels, would affect intestinal glycoprotein synthesis. Intestinal slices were incubated for up to 60 min in a modified Krebs I medium with 1.0 uc [l_14C] glucosamine under 95% 02' 5% C02. Glycoprotein synthesis was estimated as radioactivity incorporated into acid precipitable protein. Incorporation was enhanced by theophylline (27%), isoproterenol (37%) and epinephrine (28%). Combination of either isoproterenol or epinephrine with theophylline resulted in greatly increased incorporation (isop. & theoph. + 114%; epineph. & theoph. + 60%), suggesting that maximal intracellular cyclic AMP levels had been attained by simultaneous stimulation of adenyl cyclase and inhibition of phosphodiesterase. Dibutyryl CAMP, 10-3M, stimulated incorporation to comparable levels (+ 90%). Cycloheximide, 10-2M completely abolished the stimulatory effect of isoproterenol and theophylline. Subcellular fractionation of intestinal slices after incubation with Dibutyryl CAMP showed that synthesis of all cellular glycoprotein fractions was stimulated equally. Intestinal glycoprotein synthesis therefore responds to P -adrenergic stimuli, through mediation of CAMP.

Both the glycocalyx and mucus secretions may be subject to rapid hormonal modification . Four have subsequently died. Five to ten year follow-up studies are available on ten patients. Most have been known to have cystic fibrosis since infancy, but in three the disease was discovered after age 16. Nutrition ranged from poor to fair and pulmonary involvement ranged from mild to severe. Most have gross steatorrhea, but four have essentially normal fecal fat and fecal nitrogen. Pancreatic replacement and its effect on nutrition and metabolic balance will be discussed. Intraduodenal enzyme levels following pancreatic replacement will be discussed. The course of the disease, both pancreatic and pulmonary aspects, will be discussed. The irritable colon syndrome includes a variety of colonic disturbances occurrinq in the absence of organic disease. In addition to colonic motor dysfunction, clinical studies have suggested that the irritable syndrome may also encompass disordered qastric and small bowel motor function. Twelve intraluminal oressure studies were oerformed on five male and one female oatients. Patients were selected with the typical clinical criteria for irritable colon syndrome (lonq history of intermittent and recurrent abdominal cramps and/or diarrhea, normal ohysical examination, sigmoidoscopy, barium enema examination, and GI and small bowel x-rays.) Intraluminal pressure studies were oerformed through an assembly of three oolyvinyl tubes each havinq a 3 mm. oval aperture throuqh which intraluminal pressures were recorded. The apertures were 2 cm. apart. Pressures were recorded· with Sanborn 267A transducers on a Polyviso recorder. Studies nerformed included: a) basal studies after an overniqht fast, b) qastric and SB reactivity after inqestion of a standard elemental, low residue liquid diet and c) qastric and SB reactivity to 0.25 mg. of prostigmine, s.c. Results were compared to 20 oreviously studied normal patients. Results demonstrated (1) a marked increase in total basal activity in 4 of 6 patients studied. Type III activity was siqnificantly greater than in normal patients.

(2) Following the ingestion of a standard elemental low residue diet, irritable bowel patients showed greater reactivity with predominance of Type I waves.

(3) Marked pharmacologic hyperreactivity to prostigmine with a dramatic increase in Type III activity in SB and Type II activity in stomach were noted. It is suqgested that the irritable bowel syndrome universally affects the gastrointestinal tract with hyperreactivity of gastric and small bowel motor function as a part of the syndrome. The current (lsc) required to nullify the transmural electric potential difference of jejunum stripped of its muscularis has been observed previously to be less than the Isc of stripped ileum. This suggests that a quantitative or qualitative difference in ion transport exist between these two segments of intestine. In the present study, isotopic Na and Cl fluxes were measured simultaneously across short-circuited, stripped jejunum obtained just distal to the ligament of Treitz and mounted in Lucite chambers. Both sides of the tissue 2 were bathed with a HC0 3 -Ringer' s solution. Results are given as mean ~Eq/hr cm ± S.E. The unidirect10nal mucosal (m) to serosal (s) flux (J) of Na was less than Jsm, and resulted in a secretory net flux (Jnet) of 1.9 ± 0.6 (p < 0.01, n=lO). Net Cl transport was negligible, 0.0 ± 0.6 (n=lO).

Bidirectional Na fluxes measured during two flux periods in control tissues indicated that Jms, Jsm and Jnet did not change significantly (n=lO). Following a baseline flux period which showed net Na secretion, glucose, 5 mM, was added to the mucosal bathing solution. Jms increased by 1.7 ± 0.6 (p < 0.02), Jsm did not change (p > 0.5) and Jnet was negligible (p > 0.2, n=lO). The increase in Jms, or change in Jnet, was not significantly different from the increase in Isc after glucose (p > 0.5).

Theophylline, 10 mM, added to the serosal bathing solution caused a transient increase in Isc and did not alter net Na (p > 0.2) or net Cl (p > 0.5) fluxes (n=lO). No change occurred in both net fluxes of control tissues (n=lO). Secretion of Na and negligible net Cl transport across proximal jejunum are in contrast to previous observations of absorption of both ions across terminal ileum studied under identical conditions. The effects of glucose on Na fluxes and Isc of jejunum suggest that there are two independent and oppositely directed Na transport processes in the presence of negligible net Na transport. One of these Na transport processes, however, is coupled to metabolite transport. In contrast to its negligible effect on jejunum, theophylline has been shown previously to stimulate secretion of Na and Cl across ileum. We have reported that near physiologic doses of pancreatic glucagon (Gl) decrease net Na and H 2 0 absorption as well as flow velocity in the hu man jejunum during steadystate perfuslon. In view of a recent report (Science 174, 422, 1971 ) that Gl causes a marked increse in secretion by the canine small bowel, the question arose whether the observed decrease in net absorption in our studies was actually due to increased intestinal secretion. This hypothesis was evaluated by "slow marker~ intestinal perfusion: Isotonic 5% PEG 4000 plus D-mannitol infused at 1.1 ml/min at the ligament of Treitz with sampling from jejunum and ileum every 15 min. Data were obtained fro m 3 normal subjects during 32 hours of perfusion. After a 4-6 hour control period, 57 2.4 ~g/min of Gl was given intravenously for 6-8 hours and sampling continued. Co vitamin Bl? dye-dilution curves were obtained before and during intravenous Gl infusion. Gl nad no effect on flow rate (ml/min) and composition of fasting intestinal contents (rIC) in jejunum and ileu m. By contrast, the flow velocity (cm/min) of rIC decreased from 2.2 to 0. 63 cm/min in jeju num and from 2.33 to 0.93 cm/min in ileum. The calculated volume of rIC contained in the small intestine increased 2.5 -fold, from 461 ml to 1155 ml, during Gl infusion. These changes are similar to those observed when the effects of parenteral atropine were studied by the same method. The results do not support the hypothesis that Gl-induced decrease in net absorption is due to increased intestinal secretion; rather, they suggest that glucagon interferes with absorptive mechanisms directly. Of 38 patients diagnosed during the past two years as having adenocarcinoma of the stomach, 20 had lesions originating at or including the cardio-esophageal junction. In several of these, cytology and endoscopic biopsy had failed to prove the etiology of the esophageal stenosis, which was discovered only at laparotomy. Esophageal manometry was performed pre-operatively on eight patients with fundal adenocarcinomas. Perfused catheters with side openings 5 cm apart and a glass pH electrode attached to the distal catheter were used. Results of two of these studies were normal. The other six patients had entirely aperistaltic swallow waves of diminished amplitude. In two of these, sphincter pressures were increased; in two, they were markedly diminished; and in two, the assembly tube could not enter the stomach. Mecholyl tests, performed in three patients, were all positive. In three out of the six patients the manometry was the only test to suggest that the patient's stricture might be malignant. One patient with an apparently benign stricture had aperistaltic swallow waves of low amplitude but a negative Mecholyl test.

Adenocarcinoma of the cardio-esophageal junction is a common lesion. Its preoperative diagnosis may be suggested only by esophageal manometry demonstrating the motility pattern of achalasia, a positive Mecholyl test, and a variable sphincter pressure. Manometry, therefore, may be useful in the diagnosis of lesions at the cardio-esophageal junction. Increased plasma levels of mitochondrial enzymes have been reported frequently fol lowing the ingestion of ethyl alcohol, but little information is avai lable concerning the release of lysosomal hydrolases. Arylsulfatases (arylsulfate sulfohydrolases 3.1.6.1) which catalyze the hydrolysis of arylsulfates, are widespread in mammalian tissues, especially liver, kidney, pancreas and suprarenal glands. Assays of arylsulfatase activity in tissue extracts and plasma were carried out in this study with a highly sensitive method (Rinderknecht, H. et al., CI in. Chim. Acta 29: 491, 1970) using methylumbel iferone sulfate as substrate. Units of arylsulfatase activity are expressed as ~ moles methylumbeliferone I iberated per minute from the substrate. Arylsulfatase was determined in extracts of human liver (26,300 + 11,200 ~U), kidney (26,100 + 5,020 ~U), pancreas (15,300 + 8,020 ~U), heart (4,070:;:-1,300 ~U) and skeletal muscle (777 + 386 ~U). Concentrations of enzyme in these organs represent the mean of 5 different extracts and are expressed ~U/g wet tissue. The plasma va I ues expressed as ~U/m I (mean + S. D.) were 53 + 15 for 30 norma I subj ects and 125 + 65 for 53 patients with acute alcohol ic intoxication. Comparison of the two groups showed a highly significant difference (P < 0.001, by the student's test.) These findings are simi lar to those reported for b-glucuronidase activity (Geokas, M. C. et al., Gastroenterology 60:665, 1971 ) and may constitute additional evidence of increased permeability of intracellular membranes, effected by alcohol, in various organs, especially the liver. Enzymes like other proteins are antigenic and their catalytic action is usually reduced by the corresponding antibody. In this stuny, specific antisera were obtained by immunization of rabbits with bovine a-chymotrypsin and trypsin in Freund's adjuvant. The isolation of the yG globul in fraction from normal and immune rabbit sera resulted in the effective el imination of non-speci fic antiprotease activity and allowed the study of inhibition by antibodies in the presence of two types of substrate: 1. Remazolbri 1 liant Blue-Hide (RBBH), high molecular weight protein substrate, for both chymotrypsin and trypsin. and 2. LOl'l molecular weight, Benzoylarginine p-nitroanilide (BAPNA) for trypsin and acetyl-L-tyrosine ethyl ester (ATEE) for chymotryosin. Complete inhibition of proteolysis on RBBH was observed for a, b, and ychymotryosins by low levels of homologous antibody whereas a-chymotrypsin required three times that amoun t. In th ATEE the i nh i bit i on for all chymot ryps i ns did not exceed 85-90% in the zone of antibody excess. Whereas complete inhibition of proteolysis by trypsin on RBBH with low levels of homologous antibody was also observed, the amidase activity of trypsin on BAPrlA was decreased only by 10 per cent. Normal rabbit yG at concentrations, 10 times above those needed for complete inhibition by immune yG, showed negl igible inhibition for trypsin and chymotrypsins with RBBH. HO\~ever, normal yG enhanced enzyme activity on ATEE and BAPNA by 20-30 per cent. All four chymotrypsins as wei 1 as trypsin were completely inhibited by stoichiometric quantities of a2-tlacroglobulin (1 mol of a2M/2 moles of chymotrypsin or trypsin) with RBBH as substrate. The Cl2t1-chymotrypsin complex showed practically unimpaired activity on ATEE. The activity of a2M-trypsin complex on BAPNA was only 10 per cent less of that of the free enzyme, which is the same with the inhibition of trypsin by its antibody. It is concluded iliat the inhibition of these peptidases by thei r antibodies or a2M is di rectly related to the size of the substrate and the inhibitory effect is the result of steric hind rance. Ileal and jejunal function in regional enteritis was evaluated prospectively in a large series of patients by performing Schilling tests, 72 hour fecal fat determinations and xylose excretion tests (with a two hour serum level) after a 25 gm dose. In addition, jejunal perfusions measuring absorption of folic acid (3H-PGA), glucose, sodium and water, jejunal cultures and jejunal biopsies were performed in selected patients. The results were correlated with clinical, laboratory, x-ray and surgical data and the patients were divided into four groups: no surgery (NS), surgery with no recurrence (SNR), surgery with recurrence (SR), and short bowel (SB). Jejunitis was present in only four patients aside from the SB group and was not a significant factor. The There were significant differences between NS and SR for Schilling (p < .01) and fat (p < . 02) and between SR and SB for fat (p < . 05). Excluding the SB group with its gross abnormalities, fecal fat excretion for the rest appeared to correlate better with Schilling than with xylose. Of 10 patients with steatorrhea in whom Schilling test was performed, 9 had low B12 absorption while of 12 with steatorrhea who had xylose tests, only 3 had low absorption. Jejunal perfusion revealed normal absorption in aIlS patients studied (1 NS, 1 SNR, 2 SR, 1 SB). Jejunal biopsies taken from 4 patients, 3 of whom had low xylose excretion were all normal . Jejunal cultures taken from 5 patients, 3 with low xylose, 2 with steatorrhea yielded no significant bacterial overgrowth. These studies indicate significant impairment of B12 and fat absorption especially after surgery and recurrence. Steatorrhea appears to be associated with ileal rather than jejunal dysfunction and jejunal function is preserved in most patients though malabsorption of xylose, as yet unexplained, does occur in some. Salmonella enteritis was studied by injecting 2 strains of Salmonella typhimurium into ligated loops of rabbit ileum. Strain TI1L was fully virulent causing death of mice and guinea pigs and diarrhea in Rhesus monkeys; strain M206 lacked these attributes. Rabbits were sacrificed at 3,5,7,12 and 18 hrs, heart blood, liver and spleen cultured, loop reaction noted, fluid volume measured, and tissue examined with light and fluorescent microscopy. Also, HeLa cell monolayers were infected and examined at 1 to 24 hrs for invasion and cell death. Both strains caused positive loop reactions with secretion of fluid as soon as 3 hrs. There was no difference in dose-response or rate of fluid secretion between strains, as few as 100 organisms causing fluid secretion. With the fully virulent strain TI1L, positive liver and spleen cultures were not regularly seen until 18 hrs while none were noted at any time interval with M206. Focal morphologic lesions were seen at 3 hrs, i. e. collections of polys in the crypts and villi with a small area of ulceration or degeneration of overlying epithelium. These lesions involved only 1 to 5% of villi, the intervening villi being normal. With increasing time the proportion of affected villi increased with most villi affected at 18 hrs. Numerous salmonellae were seen in these areas as well as within epithelial cells, the lamina propria and Peyers patches. The latter was associated with lymphocyte blast transformation. The full blown lesion demonstrated destruction of villi, proliferation of crypts, massive infiltration with polys and widespread antigen staining. In HeLa cell culture, both strains invaded as early as 3 hrs in comparable numbers; however, niL began killing the monolayer at 7 hrs with virtually all cells killed by 24 hrs while M206 caused minimal cell death. Thus, Salmonella typhimurium causes a positive rabbit ileal loop test similar to other enteric pathogens. Invasion of the mucosa occurs and elicits an acute inflammatory reaction with ultimate destruction of the mucosa. The difference between strains in ability to spread beyond the bowel is presumably related to the known inability of M206 to multiply and survive within macrophages. Studies are underway to determine whether an enterotoxin may participate in the morphologic and secretory abnormalities cited. Two groups of children were investigated to compare finger pulp circulations (as an index of autonomic function), in response to stress induced by the cold pressor test. There were 14 patients aged 5-17 years, and 20 controls aged 6-18 years. A photoelectric hemodensitometer was applied on the thumb to measure the peripheral circulatory responses to 30 second immersion of the opposite hand in cold water (50 + 2°C). The degree of change was expressed as a percentage of the baseline measurement withthe patient at rest. Means were calculated for each group. Maximum reduction in flow was similar in both groups, viz. controls 66% and children with recurrent abdominal pain (RAP) 67%. There was no significant difference in the rate of achieving maximum reduction. Similarly the initial time of recovery to baseline was the same, 88 and 89 seconds respectively. There were highly significant differences in augmented flow in the finger after initial recovery; controls 11% and children with RAP 42%. These findings reflect instability in the control of the peripheral arteriolar circulation and are consistent with previous studies showing unstable pupil reactions in response to stress. This affords further evidence that autonomic nervous function is disturbed in children with RAP. The existence of auto-immune mechanisms in the course of gast~ic atrophy and pernicious anemia is suggested by the presence of circulating parietal cell antibody (PCA) 3nd intrinsic factor antibodies (IFA) in these conditions . In view of the importance Jf the antigen-antibody-complement immune complexes in the development of the tissue lnJury, we investigated whether immune complexes form in gastric mucosa in vivo, similarily to other organs, and whether they participate in the development of gastric atrophy by auto-immune mechanisms. Using indirect Coons' test and fluorescein labelled anti-IgG and anti-complement (C 3 ) sera we demonstrated, first in vitro, the fixation of C 3 in parietal cells of man, following the union of microsomal antigen with PCA. Thereafter, we demonstrated similar complement fixation in parietal cells of man following the union of IF with IFA in vitro. Then, by direct Coons' tests with fluorescein labelled anti-IgG or anti-C3 complement sera, we demonstrated formation of immune complexes in vivo, between microsomal antigen, PCA and C 3 in parietal cells of rats, injected daily for 6-8 weeks with human PCA. These rats showed marked thinning of gastric mucosa and reduction of parietal cell mass and size, .as well as that of HCl output. Finally, in other rats injected daily for 8-12 weeks with homologous or heterologous IFA against rat resp. human IF, we demonstrated formation of immune complexes in vivo between IF, IFA, and C3 in peptic cells, which produce IF in this species. The rats injected showed marked thinning of gastric mucosa and reduction of peptic cell mass and pepsin and IF outputs. As a clinical correlate of complement fixation in atrophic gastric mucosa we found decreased serum complement titer in about 1/ 3 of pernicious anemia patients and in 1/ 4 of those with atrophic gastritis and gastric atrophy. All this adds to the evidence, that immune complexes form in gastric mucosa in vivo in the presence of circulating PCA and IFA. These complexes may decrease parietal cell and peptic cell masses, reduce the outputs of HCl, pepsin and IF and result in at least a partial gastric atrophy.

Soberman and Frank J. Veith. Montefiore Hospital and Albert Einstein College of "ledicine, 111 East 210th Street, New York, N. Y. 10467. Our experience with clinical segmental pancreatic transplantation indicated that a simple method for drainage of the exocrine pancreas was needed. A clinical segmental pancreatic transplant with ligated pancreatic duct developed considerable peri-pancreatic fluid that was high in ~lase. This occurred without evidence of ductal rupture or a source of leakage and accompanied a rapid increase in size of the transplant. The innumerable advantages of a retroperitoneal drainage method not requiring transplantation of duodenum or opening of the peritoneal cavity for a pancreatic-enteric anastomosis led us to investigate the possibility of anastomosis of the pancreatic duct to the ureter. To evaluate the effect of this procedure on the pancreas and urinary tract, 7 dogs underwent right nephrectomy, mobilization of the dorsal pancreatic duct, and end-to-end anastomosis of the duct to the cut end of the ureter. A polyethylene stent was placed through the anastomosis. Postoperatively serum ~lase levels rose markedly (12,000-18,000 Somogyi ul 100 cc) and returned to control levels in 7-10 days. Urinary ~lase rose markedly and remained high (50,000-260,000 Somogyi ul 100 cc). Serum electrolytes, urea nitrogen, blood sugar and complete urinalysis remained normal. Two dogs continue to survive without distress or ill effects 8 and 5 months later. Examination of the other dogs in the series up to 3 months after transplantation revealed no evidence of pancreatitis, pancreatic atrophy, or damage of the urinary tract. All dogs had patent anastomoses; solid healing had occurred beyond one week. One diabetic patient has received a bodyand-tail of pancreas transplant with a duct to ureter anastomosis as well as a renal transplant. Her blood sugar had remained normal without insulin for 2 months. The transplant has not appreciably swelled. Although she excretes daily as much as 500 ml. of pancreatic juice, she has no dysuria or other evidence of ureteral, bladder or urethral damage. Her urinary ~lase has been over 1 million Somogyi ul 100 cc.

Thus pancreatic duct to ureter anastomosis appears to be a Simple, safe and effective means of providing exocrine drainage of the pancreas. Med. , 285:138, 1971 ) was modified by eliminating the steps of preparation of the plasma (extraction and dialysis procedures) prior to the assay. The total reaction volume was reduced to 1 ml, required 0.1 ml of plasma and was incubat ed overnight at 4 0 C as compared to 10 ml, 1 ml of plasma and 45 0 C for 1 hour respectively. The modified assay was reproducible and less complex for routine use . Fifty-five coded plasma specimens were assayed for CEA by both techniques with good agreement (r = 0 . 8). The modified assay was then employed in investigating plasma CEA levels in health, histologically proven malignancies, inflammatory bowel disease and biopsy proven chronic active hepatitis. In health (42 individuals), plasma CEA was 2 nglml or less. We set a plasma CEA of > 2.0 nglml as abnormal. In carcinoma of the colon and rectum (114 patients), 67% were positive and CEA was significantly higher (p < 0.01) in those with unresectable or metastatic lesions (70% positive) than in those with resectable lesions (56% positive). In other gastrointestinal cancers (68 patients), including primary carcinomas of the esophagus, stomach, pancr eas and liver, 39% were positive . The test was also positive in 32% of 44 patients with various extraabdominal primary cancers. In 74 patients with inflammatory diseases of abdominal organs, the test was positive in chronic active hepatitis (30%) and ulcerative colitis (35%), but negative in all of 13 patients with diverticulitis . W e conclude that 1) our simplified technique is reliable; 2) CEA is frequently but not always elevated in "early cancer" of the colon and rectum; 3) levels of CEA may relate to the extent of the malignant disease; 4) CEA may also be elevated in primary cancer of other organs; 5) the test must be interpreted cautiously i n the presence of inflammatory disease of the liver or intestines; and 6) the test usually differentiates between these various conditions and health. Cooper, et al. (Science, 172:1238 (Science, 172: , 1971 ) demonstrated that calcitonin secretion from the porcine thyroid is stimulated by pentagastrin. To relate this to man, we studied the interrelationship of these hormones simultaneously in subjects with high endogenous secretion of each. Fasting plasma immunoreactive human calcitonin and fasting serum immunoreactive human gastrin were measured by specific radioimmunoassays (Clin. Res., 20 '3, 1972) in 16 healthy individuals, 8 patients with histologically proven Z-E and no thyroid disease and in 7 patients with histologically proven MCOT. All patients were normocalcemic except for one with MCOT. The mean and range for these groups are shown below. , 52: Feb . 1972, in press ). The association of Raynaud's phenomenon and esophageal motility abnormalities in some of these patients suggests that the motility changes are related to impaired esophageal blood flow. Esophageal blood flow may be indirectly assessed by measuring the time for intraluminal esophage~ temperature to return to normal following ice water ingestion. Our previous study (Zboralske, et al : Clin. Res., 20:184, Feb. 1972) , confined to scleroderma, has been extended to several CTD and modified to utilize single 5 ml swallows of 0.5°C ice water. The present study involved 4 patients with scleroderma, 2 with Raynaud's disease, 2 with MCTD (Mixed Connective Tissue Disease), and 5 normal volunteers. Esop~ ageal motility was studied using cineroentgenography and intraluminal manometry. Fingertip rewarming curves were obtained for all subjects following 10 minute total body exposure to 40°F air temperature. With the subject sitting, 2 small thermistors (time constant 0.3 sec) were inserted into the esophagus; one located in the distal 1/3 and the other in the proximal 1/3 near the junctional zone of smooth and striated muscle. Thermistor rewarming curves were recorded and the amplified signals collected by the AMCE (Advanced Computer for Medical Research) system for storage and analysis . Thermistor curves were analyzed (deg/sec) for the total period of rewarming, the first half temperature rewarming, the last half temperature rewarming, the last 1, 3, and 5 degree temperature rewarming, and the time (sec) for the first half rewarming. A significant difference existed between the slopes of the first half rewarming in CTD patients and normals in the lower esophagus (P< 0.05). Although of lesser magnitude, a significant difference also existed between the slopes of the first half rewarming in the upper esophagus (P<0.05). Preliminary evaluation suggests that the time for first half rewarming (sec) may also be significantly different between the two groups. These data provide additional support for the hypothesis that esophageal blood flow is decreased in some CTD and may be related to their manifest motor dysfunction. Man is dependent upon his liver to metabolize an increasing variety of amine medications. Yet , too few studies have been concerned with amine drug metabolism in human liver tissue, during disease states, and by pathways other than N-demethylation. Because an alternate pathway involving the N-oxidation of tertiary amines was recently demonstrated in porcine liver, we measured not only the N-demethylation (DEM) of aminopyrine to formaldehyde at pH 7.4 , but also the N-oxidation (NOX) of N,N dimethylaniline to its oxide at pH 8.4 in homogenized portions of 113 percutaneous or surgically obtained liver biopsies from living patients . Histology and serum enzymes were normal for 15 patients with mean NOX activity of 1 . 98±1.08 nmoles oxide/min./mg. protein, and mean DEM activity of 0.54±0.46 nmoles CH20/min . /mg. protein . Significant differences were not found in biopsies grouped by the histological and clinical classifications: normal histology -abnormal serum enzymes (19), viral hepatitis (22), cholangitis (22), fatty change (12) , alcoholic liver disease (9), metastatic tumor (5), and nonspecific inflammatory reaction (6). NOX was 1 . 06±O.42 in 3 cirrhotic patients (p<0.05). Activity diminished with severe necrosis or fibrosis but was not related to a specific disease process , age, sex or biopsy size . In 5/7 patients with NOX> 3. 0 (pc( O. 001) the increased rate of N-oxidation appeared related to use of tranquilizers or amphetamines. In 3/4 with DEM~l.O (pc(O.OOl) barbiturates were given . In 2/3 with both NOX:>3.0, and DEM~l.O, drug abuse was suspected . Conclusions: Both N-oxidation and demethylation of amine substrates can be demonstrated in small homogenates oJ liver biopsies from normal and disease states. Enzyme activity is correlated more with disease severity than a specific pathological process. In humans Noxidation may be a more significant metabolic pathway than previously suspected. The gctivities of these pathways may be altered by drugs . I.V. c 0 ec.ysto n n nstea 0 atty meal in oral cholec.ystography produced painless reduction in gallbladder (GB)volume of 50-80%, with visualization of common duct (CD), in 17 control subjects. Over a 10 yr period, 40 pts with biliary pain, visualized and acalculous GBs were studied. Partial obstruction of cystic duct(c.ystic duct syndrome) found in 26 pts was indicated by GB volume reduction of 5-45%, minimal or no filling of CD and reproduction of spontaneous pain. All but 3 were women, ages 19-55, with mean pain duration of 22 mos. All but 1 required cholecystecto~ for pain relief. Anatomic causes of partial obstruction included extrinsic adhesive bands, sharp bends of c.ystic duct and intrinsic fibromuscular strictures. Obstruction at sphincter of Oddi (SO) was found in 8 women aged 35-54. Their CCK responses were similarly abnormal but CD was prominent (~7 rom ¢) and main hepatic ducts filled with dye. Of this group, 5 responded to nitroglycerine and/or anticholinergics by egress of dye into the duodenum and pain relief, suggesting SO spasm. The remaining 3 pta were found at operation to have SO stenosis relieved by sphincteroplast,y and cholecystecto~. ~eractive responses to CCK (over 75% volume reduction and pain) in 4 women and 2 men suggested qyperplastic cholecystoses of Jutras; 5 of these had cholecystect~ for pain relief. Histology of all resected GBs showed overlapping findings between groups ranging from minimal to moderately severe chronic cholec.ystitis, to features of cholecystoses incl. Rokitansky-Aschofr sinuses, adeno~omas, cholesterol polyps and muscular qyperplasia. The results suggest that 1) CCK cholecystography is an important tool in diagnosis and characterization of painful acalculous GB conditions; 2) CCK responses permit differentiation between normal GB contraction, qypercontraction and impaired contraction with pain, the latter caused by partial obstruction either at level of c.ystic duct or SO. Drug responses further separate SO group into spalJlll and stenosis; 3)effective treatment can be based on response to CCK; 4) chronic acaleulous cholecystitis, cholecystoses and cystic duct syndrome may represent overlapping expressions of acalculous GB disease. 

Adequately controlled studies of the treatment of irritable bowel syndrome (IBS) have not been reported. The purpose of this study was to develop and test a model for the clinical evaluation of therapeutic agents in this condition. Dilantin was tested because previous uncontrolled studies had indicated possible benefit in IBS. Prestudy patient evaluation included sigmoidoscopy, barium enema, lactose tolerance test and psychological tests. Twelve patients who satisfied the criteria consented to participate in a double-blind crossover study lasting 20 weeks. After a control period on placebo, each patient received in random order placebo or Dilantin, 300 mg. daily, in identical capsules. The alternate agent was given for a similar period followed by another control period on placebo. Throughout the trial, the patients' usual treatment regimen remained unchanged. Clinical benefit was assessed in two ways: 1) by scorecards on which patients tallied daily ratings of six indices and 2) by a physician interview at each biweekly visit. Serum levels for Dilantin also were determined during both test periods of the study. The results were analyzed by analysis of variance. There were no significant differences (p<0.05) between treatment effects on stool frequency or consistency, frequency or severity of abdominal pain or overall disability as scored by the patients. Physician evaluation of overall well-being also did not differ. Moreover, no correlation existed between the indices and neurotic tendency assessed by psychological testing. Half of the patients had frequent and severe abdominal pain which was unrelated to stool frequency. Although day-to-day variations occurred in each index, striking consistency was observed in all indices when these were averaged over intervals of several weeks or more. This finding suggests that such an approach may prove useful in the evaluation of agents used for the treatment of this syndrome. In this study Dilantin was found to be ineffective. The psychotomimetic effects of marihuana are due most probably to hydroxylated metabolites of tetrahydrocannabinol (69THC), the major active alkaloid of the Cannabis plant. Because ingested marihuana "brownies" are psychotomimetic, we hypothesized that the human small intestinal mucosa is able to hydroxylate 69THC, a metabolic activity previously found in rodent liver. We took 15 to 30 jejunal biopsies from informed normal volunteers. Biopsies were incubated in oxygenat. ed flasks with purified l4C-69THC (1. 7 to 2.5 )Jg/ml) in a nutrient medium. Activity of "fresh" biopsies was compared with that of control "boiled" biopsies heated to 100 0 for 3 minutes prior to their addition to the flask. After incubation and extraction, unchanged l4C-69THC (Rf 0.7 to 0.9) and l4C-69THC metabolites (Rf 0 to 0.2) were chromatographically separated and quantified. Three biopsies (31.0 -37.4 mg total wet weight) from each of 6 normal subjects converted 6.1 to 8.8% (net) of the 69THC to polar metabolites in 2 hours. In other experiments, metabolite formation was approximately linearly related to both duration of incubation and to tissue weight (maximum, 21% net conversion with 94 mg tissue). Rat intestinal mucosa was also able to convert 69THC to polar metabolites. The polar metabolites formed by human intestinal mucosa consisted of one major compound (62%) and several minor ones (38%). The major metabolite had the chromatographic mobility of authentic II-hydroxy THC in 3 solvent systems.

To define the route of transport of absorbed 69THC and its metabolites, we fed 3H-oleic acid and l4C-69THC intraduodenally to each of 2 rats equipped with thoracic duct fistulas. Over 24 hours, 67.5 and 82.3% of the 3H, but only 1.1 and 1.4% of the l4C, were recovered in lymph. Urine contained 12.5 and 11.4% of the fed 14C, indicating that transport of absorbed 14C was by another route. Thirty minutes after an intact rat was given a similar meal, the ratio of portal blood radioactivity to inferior vena cava radioactivity was 1.06 for 3H, compared with 3.48 for l4C.

We conclude that human and rodent small intestinal mucosa can hydroxylate 69THC. In the rat, these cannabinoids are then transported from the intestinal absorptive cells by the portal venous system. This study did not involve marihuana administration to human volunteers.

NOli-ABSORBED HIDICATORS FOR CLINICAL STUDIES USING TEST r1EALS: Cm1PARISON OF POLY-ETHYLENE GLYCOL (PEG), PHENOL RED (PR) AND 51CRCL3 (51CR). Alan J. Greenwald, Duane L. r1iller, Barhara Reinken, and Harold p. Schedl, University of ICMa, 101·la City-;-1Ow'a.

To measure calcium absorption at various sites along the small intestine we gave test meals and collected intestinal contents by using a transintestinal tube and collecting samples from both nasal and anal ends. Consecutive collections were made and a new sample ~Ias begun as soon as the tube dead space volume was obtained. The 600 ml test meal contained calcium chloride 24 mr4, sodium chloride, 150 mM, and indicator. During each study as the test meal bolus entered the sampling site, there was an initial set of samples with a high ratio of initial to final indicator concentration (Ci /-Cf) followed by a set of samples having a lower Ci/Cf. The latter samples were collected at a rapid rate while the bulk of the test meal was passing the sampling site. Indicator comparisons are from these samples. In g studies from 84 to 218 cm from the nose in 2 normal men, ages 30 and 38 the PEG ratio, 1.26+0.02 (mean + SE), was significantly l~ler (p < 0.001) than the ~lCr ratio 1.50+0.02 l38 samples,-paired analysis), When several indicators are compared, the indicator~est recovered is that giving the highest final concentration (Cf). This gives the l~lest indicator ratio (Ci/Cf). Hence, PEG recovery was significantly greater than 51 Cr. In 7 of these studies (32 samples) PR ~Ias also analyzed and the ratio, 1.25+0.02, I' las the same as the PEG ratio but less than the 51Cr ratio, 1.50+0.02 (p < 0.0011. When PEG and PR ratios were compared in a total of 13 studies (70-samples), the mean PEG ratio, 1.25+0.02 was significantly less than that for PR, 1.31+0.03 (p < 0.001). In a patient with active granulomatous colitis, three studies (20 samples) at 158 to 270 em, the same solution without PEG gave a mean PR ratio of 0.90:,0.05 which was significantly 10~I;r than the 51Cr ratio, 1.03+0.05 (p < 0.001). We showed previously that PEG, PR and lCr instilled into the stomach at pH 1 gave the same recoveries (Ivey and Schedl, Gastroentero1., 59:234, 1970 ). The additional variables in this study include the test meal at pH 6.9 and possible small intestinal effects. Our studies show that although PR and 51Cr may be useful indicators for intestinal perfusion or measuring intragastric volumes at pH 1, they could be inferior to PEG when given with a test meal. The validity of PEG itself as an i ndi cat~r cannot be es tab 1 i shed by these experiments. The immediate effect of vagal section on the gastric submucosal and deep mucosal microvessels and mucosal blood flow was studied in the rat. The exteriorized stomach of the anesthetized rat (Na pentobarbital. 40 mg/kg i.p.) was transilluminated by a cool light transmitted from a high intensity light source by means of a quartz rod. Careful removal of the serosal and muscle layers from a small area permitted excellent visualization of the submucosal and deep mucosal microvessels. Photomicrographs were taken at 100X. 200X. and 300X magnifications before and within 10 minutes after bilateral subdiaphragmatic vagal resection. Successful removal of nerve trunks was confirmed histologically in all instances. Measurements of the diameters of the same vessels were made from the pre-and post-vagotomy photographs. Statistical analysis (t test for paired differences) revealed no significant effect of vagotomy on the caliber of any of the vessels. Average pre-and post-vagotomy diameters were as follows: submucosal artery 31.8 -30.3 v. mucosal artery 12.6 -13.8 v. deep mucosal capillary 7.8 -7.5 v. collecting vein 33.1 -34.2 v, and submucosal vein 37.0 -34.4 v. Preliminary studies on mucosal blood flow in the anesthetized rat using the aminopyrine clearance technic revealed no significant alteration following vagotomy. Average mucosal blood flow before vagotomy was 0.16 ml/min and after vagotomy 0.18 ml/min. Conclusion: In the anesthetized rat vagotomy has no acute effect on either the gastric submucosal and deep mucosal microvessels or mucosal blood flow. It has been reported that in viral hepatitis with positive titers for "hepatitis associated antigen" (HAA), a serum-sickness-like syndrome with arthritic manifestation may be caused by a circulating immune complex. In another study it was reported that rheumatoid factor (RF) was present in 16 of 17 patients with HAA positive hepatitis .

The purpose of the present study is to determine the interrelationship of rheumatoid factor, arthritic manifestation and HAA in acute viral hepatitis.

Serum from 32 patients with hepatitis was tested for HAA using immunoelectrophoresis. Rheumatoid factor was determined by the Hyland RA-Test. The results are presented in tabular form. Studies on macroamylasemia due to binding of amylase to immunoglobulin G are in progress to identify the binding site for amylase on the immunoglobulin.

A sedimentation coefficient between 7.4 and 11.3 S for the macroamylase would be consistent with a soluble 1 : 1 complex which is not the usual antigen-antibody ratio.

The y-globulin fraction from the serum of a patient with macroamylasemia was precipitated with 30% saturated ammonium sulfate. The precipitated protein was dissolved in normal saline and separated on a column of agarose gel taking advantage of the adsorption of amylase to the gel to remove the bulk of the y-globulin . The binding protein was recovered by elution of the macroamylase complex with pH 3.4 glycine.HCl on Bio-Gel P-150 . The binding protein was concentrated to 1%. To obtain FAB and FC pieces, the binding protein was incubated with papain.

The digested material was eluted through polyacrylamide gel P-150 to remove whole y-globulin from the FAB and Fe pieces. Through the use of specific antisera to FAB and Fe pieces , precipitation studies suggested that the binding site for amylase is in the FAB piece .

Further studies with DEAE-cellulose to purify FAB and FC are in progress. A catheter-type semiconductor beta radiation detector system (Catelix) has been developed which we have used for serial studies of esophageal perfusion. Because of its multiple vascular supply, standard techniques for measuring esophageal blood flow (e.g. flow meters) are not applicable and the usual indicator fractionation technique cannot be used, because the animal is killed after each determination. Utilizing increasing volume Lucite cylinders, the isoresponse curve of a Catelix detector system to a 1 uCi/ml solution of 86RbCl was determined. Fifty-five percent of the maximum counting rate resulted from radioactivity within 1.5 mm of the detector's outer surface and 94% of the maximum counting rate lay within 4 mm of the detector. A variation of the indicator fractionation technique allowed serial determinations to be performed: A Catelix detector was passed into the distal esophagus of an anesthetized cat. Two millicuries of 86Rb were injected intravenously and continuous records of Catelix counts versus time recorded over a ten minute period. After ten minutes, another 2 mCi of 86Rb were injected and counts again recorded. After serial determinations, a final 2 mCi of 86Rb were injected and the animal sacrificed 60 seconds later. We have found at least two tumor specific antigens in gastric, pancreatic,. colonic, and rectal adenocarcinoma and their metastases. It was considered unusual when perchloric acid extract from liver metastases of a gastric adenocarcinoma, purified by Sepharose 4B gel column chromatography and repeated polyacrylamide gel electrophoresis, contained no detectable CEA using highly specific antiserum with the Ouchterlony gel double diffusion technique. Accordingly, an explanation for this negative finding was sought and speCifically the masking of antigen by naturally occurring antibody was investigated. As CEA is soluble and immunoglobulins or CEA bound to goat antibody are insoluble in O. 6 N perchloric acid, the percipitate from this extraction was treated with 15% NaCI or acidified to pH 3 to dissociate the glycoprotein antigen from antibody. Antibody was then precipitated by 50% saturated ammonium sulfate. After exhaustive dialysis and concentration, CEA was then found in the supernatant in both instances showing 2 precipitin bands with highly specific aatiserum. Also, treatment of an initial isotonic saline homogenate of the tumor with 15% NaCI or controlled acidification to pH 3 with HCI, followed by 50% NH4S04 precipitation, resulted in the presence of 2 antigens in the supernatant. Subsequent purification of these antigens by Sepharose 4B gel column chromatography and repeated polyacrylamide gel electrophoresis, showed them to have similar chemical characteristics, as well as lines of identity with antigens extracted by Gold's method from hepatic metastases of colonic adenocarcinoma. It is clear from these studies that 1) the presence of CEA in this cancer was masked by a naturally occurring substance, likely an antibody and 2) perchloric acid extraction did not dissociate CEA from this substance. Also, it is possible that CEA may be masked by a naturally occurring substance, likely an antibody, in serum giving negative or falsely low values, and dissociation procedures should be included in methodology. 

Gastric emptying may have important effects on drug ahsorption in man. We studied absorption of acetaminophen given to hospital patients as 3 X 500 mg tableff and correlated absorption with gastric emptying measured by scintiscanning using 3mIn DTPA chelate (GUT. 12:611, 1971). Emptying rates were expressed as half-times (T 1/2). Absorption of the drug was assessed from the magnitude and time of occurrence of the peak plasma concentration, and from cumulative urinary excretion. In 8 patients gastric emptying and acetaminophen absorption were examined simultaneously 4sin9 a 400 ml water test meal. To simulate clinical drug administration more precisely. absorption was assessed in 14 patients after ingestion of the drug with 50 ml water. A separate study of gastric emptying was necessary for this group and "las performed using a standard test meal of cornflakes, sugar and milk. In both groups the time of occurrence of the peak plasma acetaminophen concentration correlated significantly with log T 1/2 gastric emptying (p < 0.05) and log peak concentration showed a siqnificant negative correlation with T 1/2 (p < 0.05). Twenty-four hour ul1inary recovery of the drug was not related to T 1/2. In 6 of the group of 8 patients. gastric emptying and acetaminophen absorption were restudied by the same method following IV injection of propantheline. Gastric enptying was significantly retarded and the peak level of acetaminophen vias Significantly diminished and delayed. Conversely. IV injection of metaclopramide increased and speeded attainment of peak levels in a further group of 6 subjects to whom the tablets were given with 50 ml water. the results indicate that gastric emptying significantly affects the rate of absorption of acetaminophen given in tablet form and that normal inrlividual variation in emptying can be responsible for substantial differences in rate of absorption of the drug. Gastric emptying may be equally important for other drugs under the conditions of their clinical administration. The results . of replacing the esophagus in 54 consecutive patients, using a pedicle tube constructed from the greater curvature of the stomach, are reported. Of 34 patients who had carcinoma, in 27, the carcinoma was in the middle or upper third of the thoracic esophagus. The carcinomatous esophagus, when resectable, was generally excised as a second stage procedure. The other 7 patients had carcinoma involving the cervical esophagus. In these cases, the reversed gastric tube was made long enough to be anastomosed to the pharynx at the base of the tongue. Twenty patients had benign lesions for which the entire esophagus was by-passed but not resected. This benign group includes 12 patients with stenosis or bleeding caused by esophagitis, 7 with lye stricture and one with esophageal constriction due to an enlarged heart. Forty-three patients have been followed up to 5 years, and 8 patients for 5 to 13 years after operation. There were 3 postoperative deaths, a mortality rate of 5 . 5%, which includes the first patient in whom the operation was attempted. Complications will be described. Reversed gastric tube esophagoplasty has proved to be a satisfactory method for replacing the esophagus, enabling patients to eat a regular diet in a normal manner. Intestinal brush border (BB) catalyzes the hydrolysis of a large number of small peptides but the number of enzymes involved and their substrate specificities remain uncertain. In the present study papain solubilized rat BB enzymes were chromatographed on DEAE-cellulose by elution first with buffer then with a linear NaCl gradient. Three well separated peaks with peptide hydrolase activity were isolated. ~~o of these (III,IV) were very active while the other (II) had minimal activity. An additional peak with apparent enzyme activity proved to be due to free amino acids released by papain digestion. Quantitative assays for phe-gly and gly-phe hydrolase activities revealed that phe-gly was split most rapidly by III and gly-phe by IV. Ratios of phe-gly to gly-phe hydrolase activity in peaks III and IV were 9:1 and 1:3, respectively. Specific activities of phe-gly hydrolase in III and gly-phe hydrolase in IV were 10-and l20-fold greater, respectively, than their specific activities in purified BB. Using a high-voltage paper electrophoresis screening technique, peaks III and IV were shown to have additional differences in substrate specificity. Enzyme III rapidly hydrolyzed dipeptides containing basic amino acids (gly-lys, phearg, lys-lys, arg-phe) and also all tripeptides tested while enzyme IV split these substrates poorly if at all. Enzyme IV was more specific for gly-tyr, met-gly, glyleu and glyNH2-gly. Other peptides tested were hydrolyzed approximately equally by the t{~O enzymes. All substrates hydrolyzed by intact BB were also split by either peak III or peak IV or both. The 29 substrates tested were hydrolyzed poorly or not at all by peak II and by all other column fractions not containing peaks III or IV. Acrylamide gel electrophoresis of concentrated protein from peaks III and IV showed 1 heavy band with peptide hydrolase activity in each peak plus 1 faint band consistent with a disaccharidase kno{~ to be present in both peaks. We conclude that intestinal BE contains at least 2 major peptide hydrolases, one of which is most specific for tripeptides and basic dipeptides and the other most specific for neutral dipeptides. To define the metabolism of the two moieties of cholyl taurine, cholyl-2-4-H taurine-35 S was synthesized and administered to 8 healthy volunteers. Bile samples were obtained for 4 days; conjugated bile acids were separated by chromatography; and the specific activity of the cholyl and taurine moieties determined. The daily fractional turnover of the taurine moiety (40±4%, M±SE) was similar to that of the cholyl moiety (43±4%) and fractional turnover rates in individual subjects were highly correlated (r=0.88), suggesting that cholyl taurine was conserved with little deconjugation during enterohepatic cycling. The daily fractional turnover of the taurine moiety was less than half that previously observed by us for the glycine moiety of cholyl glycin~ 3H from the cholyl moiety appeared progressively in cholyl glycine, indicating that the cholyl moiety liberated by bacterial deconjugation was absorbed and conjugated predominantly with glycine. 35S did not appear in dihydroxy taurine conjugates and was excreted predominantly in urine (43±4% of dose after 6 days); the fraction excretoo in 24 hours correlated highly (r=0.93) with the daily fractional turnover of the taurine moiety. Taurine in urine and plasma, determined by ion exchange chromatography, was labeled with 35 S, indicating absorption of taurine after deconjugation. In additional experiments, taurine-35 S was absorbed when instilled into the ileum and cecum but less than 1% of taurine-35 S radioactivity was incorporated into cholyl taurine 12 hours after intestinal or intravenous administration because of dilution by an endogenous taurine pool. The sulfate/taurine ratio of urinary 35 S was lowest when taurine-35 S was given intravenously, higher when given orally, and highest when instilled into the ileum or given as cholyl taurine-35 S, suggesting a bacterial origin for urinary sulfate derived from taurine. We conclude: 1) cholyl taurine is deconjugated considerably less than cholyl glycine during enterohepatic cycling in health; 2) after deconjugation, there is little reincorporation of either moiety into cholyl taurine: the cholyl moiety enters the cholyl glycine pool and the taurine moiety enters a large taurine pool; and 3) taurine liberated by deconjugation is absorbed as taurine, or as sulfate produced by intestinal bacteria. Micellar solubilization, which greatly increases the concentration of lipolytic products in the aqueous phase of intestinal contents, is considered to be an important step in fat absorption in man. Although studies in rats have shown that fatty acid in micellar form is absorbed more rapidly than non-micellar fatty acid, this comparison has not been made in man. To do this, we compared the rate of absorption from the proximal jejunum of micellar oleic acid with that of emulsified oleic acid by a perfusion technique in 4 healthy volunteers. A 5 lumen tube with proximal occlusive balloon was positioned fluoroscopically. Endogenous bile was aspirated from a site proximal to the balloon. Test solutions were perfused at 15 ml/minute for one hour from a site just distal to the balloon; collection sites were 25 and 50 cm distal. Both solutions were used in random order in each subject. Each solution contained 5 ~ oleic aCid-3 H in dilute phosphate buffer, pH 6.4. In the micellar solution, the oleic acid was present in micellar form in 10 mM taurocholate; in the non-micellar solution, as an emulsion in 10 mg/ml gum acacia . Both solutions contained PEG_14C and the rate of absorption was assessed by changes in 3H/14c ratio. Absorption from the m icellar solution was several fold faster than from the emulsion . These experiments show that micellar FA is absorbed more rapidly than non-micellar fatty acid in man. ~ speculate that in diseases associated with impaired micellar dispersion that the rate of fatty acid uptake will be reduced markedly causing the locus of fat absorption ~ extend throughout the small intestine. Glycine is utilized for conjugation of bile acids from a) de novo synthesis of primary bile acids, b) intestinal absorption of primary and secondary bile acids after bacterial deconjugation and c) intestinal absorption of deoxycholic acid following bacterial dehydroxylation of cholic acid. To calculate the amount of glycine used for these three conjugation requirements we synthesized the three major glycine conjugated bile acids (3H in the steroid moiety and 14C in glycine), administered each to 8 healthy volunteers and determined the specific activity decay curve for each moiety. To partition glycine utilization into its component parts we defined Nconj -that fra~ tion of the total bile acid returning to the liver which is conjugated. Assuming the usual model of the enterohepatic circulation -a single well mixed pool in steady state with respect to synthesis and loss -an expression for Nconj can be derived in terms of the fractional turnover of the steroid moiety, kst, the glycine moiety, kgl y ' and the recycling frequency, C, assumed to be 6/day. For the primary bile acids Nconj = (C-kgly)/( C-k st ) and the daily glycine requirement for conjugation is the sum of that for synthesis CP (l-fabs) and that for reconjugation CP fabs (l-Nconj). P is pool size and fabs the fractional absorption from the intestine, defined for each bile acid. The secondary free bile acid, deoxycholic, originates by deconjugation and dehydroxylation; hence Nconj = (C-kgly)/C and the daily glycine requirement involves only reconjugation -CP (l-Ncon j). Using these equations, we can characterize fully the metabolism of glycine conjuga ted bile acids in man.'" In heal th more free bile acids requiring conjugation are formed in the intestine than in the liver. The calculations used here, should allow the estimation of glycine utilization for bile acid conjugation in diseases associated with either increased synthesis, increased deconjation or both.

Glycine Vol. 62, No.4 Absorption of sodium by the rat small intestine is reduced by increasing the intraluminal hydrogen ion concentration, suggesting that sodium absorption and hydrogen i~n secretion may be linked. We have studied the influence of intraluminal sodium concentration on bicarbonate absorption by perfusing the jejunum of each anesthetized rat with 2 isotonic solutions which contained either NaCl or choline Cl buffered with NaHC0 3 (25 mM), and gassed with CO (5.4%). J HCO, J H, and final PCO were less 'n rats perfused with cholinet J etH?O, aR8ttra~smuPal PO also differe6, however (Table I , Study I). In Study II, we Rssessed the effect of water movement by perfusing each rat with 2 solutions of buffered NaCl, one of which was hypertonic (350 mOsm/ Kg): J net HC0 3 ,J etH, and final PCO? did not differ significantly despite a significant difference in In H 0. The influence of transmural PO was determined in Study III by replacing NaCl Q'th 2 Na isethionate, which increased lumen negativity by a mean of 7 mV: significant differences occurred in J net H 2 0 and transmural PO, but not in I n tHC03 JnetH, or final PC0 2 • Twenty patients with duodenal ulcer (DU) and 20 non-DU subjects were given pentagastrin by intravenous infusion in graded doses (2.7,8.2,74,223,667, 2000 , and 6000 ng/kg-hr) . Each dose was given sequentially in the order listed for three 10 min periods. Peak 10 min acid output was determined for each dose tested . Fasting serum gastrin was measured by radioimmunoassay in all DU subjects and in 17 non-DU subjects. Calculated maximal response (CMR), and dose of pentagastrin required for one-half maximal response (050) were determined by a computer program based on minimizing the sum of squares of observed minus predicted response in the equation: Y = VX/(K+X). Y = response, X = dose, V = CMR, and K = 050. In the DU group mean 050 was significantly (p < 0. 01) less and mean CMR significantly (p < 0.01) greater when compared with the non-DU group. Mean serum gastrin was not significantly (p > 0.2) different between the 2 groups . There was a significant negative correlation (r = -0.511) between gastrin and CMR in the non-DU subjects. There was no correlation between 050 and gastrin in either group, nor was there a correlation between gastrin and CMR in the DU group.

Mean ( ± SE) 050 CMR Gastrin ng/kg-hr mEq/10 min pg/ml DU 93.3 ± 28.9 9.3 ± 0. 8 55.0 ± 5.7 Non-DU 312.6 ± 70.3 5.9 ± 0.7 67.8 ± 7.8 It is concluded that, compared to non-DU subjects, DU patients not only secreted more acid in response to pentagastrin, but also were more sensitive to pentagastrin . The differences in acid secretion were not related to serum gastrin concentrations . Nine subjects with repetitive belching or post prandial bloating wer~ studied with cinefluoroscopy and esophageal manometry.

In five subjects who complained of constant belching, just prior to the belch, air was seen to suddenly distend the esophagus down to the lower esophageal sphincter.

This was followed by a belch at which time the esophagus was completely emptied of air.

Esophageal manometry revealed that with the entry of air into the esophagus, there was a simultaneous, sharp decrease in upper esophageal sphincteric and esophageal pressures.

With the production of a belch, a simultaneous spike-like increase in esophageal and gastric pressure was recorded.

In four subjects complaining of post prandial bloating, fluoroscopic observations revealed that with each dry swallow the entire esophagus became distended with air which was carried into the stomach by peristalsis. Manometry revealed that simultaneous with each dry swallow, there was a deep inspiration followed by a peristaltic complex.

The repetitive belcher uses a deep inspiration to overcome the resistance of the upper sphincter and aspirates air into the esophagus.

A forceful contraction of abdominal and thoracic wall musculature forces the air out as a belch.

The post prandial bloater continue~ to inspire during a swallow and aspirates air into the esophagus wh1ch is then carried into the stomach by peristalsis. Cytotoxic effects of endotoxin are prominent in septic shock. The large molecular size of bacterial lipopolysaccharides makes it unlikely that these toxins enter susceptible cells to alter enzymatic processes, such as the increased hepatic glycolysis which occurs in endotoxin shock. However, membrane-bound enzymes would be vulnerable to endotoxin. We tested the hypothesis that endotoxin activates membrane-bound adenyl ~clase (AC), thereby increasing intracellular cyclic AMP (cAMP) upon which glycolysis depends. In 6 guinea pigs killed 3 hours after injection of an LD50 dose of E. coli endotoxin, hepatic AC activity was 30+4(S.E.) x 10-12M cAMP formed/mg protein/20 min compared with 6 control animal levels-of 15+2 x 10-12M cAMP/mg/20 min (p<.05). In vitro incubation of liver tissue fr~m 6 guinea pigs showed comparable results:incubated f~th endotoxin 35~5 x 10-2M cAMP/mg/20 min vs incubated without endotoxin 25+3 x 10-M cAMP/mg/20 min (p<.05). Dialysates of liver incubated with endotoxin falled to activate AC. Our conclusions are: 1) endotoxin activates the membranebound enzyme, AC, which may be one of the earliest cytotoxic effects in endotoxin shock; 2) the effect of endotoxin on liver AC is not mediated by substances released in other organs and is probably a direct effect (Supported by N.I.H. grant #HE14649-01). In order to establish the effect of circulating gastrin on concentrations of parathyroid hormone (PTH), two experiments were performed . In the first, five female sheep received intravenous infusions of human gastrin, 4 -10pG/Kg/hr . , thus increasing concentrations of serum gastrin (as determined by radioimmunoassay) at least sixfold. Postinfusion levels of PTH, 34 + 5pl-Eq/ml, did not change from baseline values, 35 + 6pl-Eq/ml. Secondly, levels of immunoreactive PTH were measured in five patIents with the Zollinger-Ellison syndrome. Two had normal levels of PTH, 41 and 26pl-Eq/ml, despite elevations of serum gastrin, 6500 and 85,000 pg/ml . A third had borderline PTH levels, before operation (66pl-Eq/ml) and after resection of a duodenal islet cell tumor (61pl-Eq/ml). The remaining two patients (gastrin 820 and 2300 pg/ml respectively) had significantly elevated levels of Circulating PTH, 90 and 102pl-Eq/ml. Following resection of the ulcerogenic tumors and partial gastrectomy, levels of both hormones declined, gastrin to 160 and 520 pg/ml and PTH to 48 and 54pl-Eq/ml respectively. We have previously found in sheep that metabolic alkalosis results in decreased calcium and , elevated levels of PTH.

Both Z-E patients with increased PTH had mild hypocalcemia (8.0 and 8.3 mg%) and massive gastric hypersecretion (>4500 ml and.~500 mEq HCl/day). From these experimental and clinical data, we would suggest that the ,relationship between gastrin and PTH does not seem to be a direct one. In patients without parathyroid adenomas, some indirect factor such as alkalosis or another humoral mechanism appears to be at play when both hormonal levels are elevated. Pepsin secretion is stimulated when hydrogen ion back diffuses through the damaged gastric mucosa. Whether acid stimulates pepsin secretion in the normal stomach has never been tested since H+ is usually present when pepsin secretion i~ stimulated. Pepsin secretion was studied in three dogs with vagally denervated pouches of the oxyntic gland region. Experiments lasted three hours and consisted of twelve IS-min collections of 50 ml of fluid which had been put in the pouch at the beginning of each period plus any secretions during that 15 min. A solution of 100 mN HCl increased pepsin secretion 3-fold over basal conditions during which the pouches were irrigated with phosphate buffer, pH 7.5. Solutions of 25 and 50 mN HCl produced gradations of pepsin secretion between basal and the level attained during the presence of 100 mN HCl. An intravenous infusion of atropine (100 ~g/kg-hr) and topical application of a 2% solution of Xylocaine blocked pepsin secretion. Intravenous infusion of 2 unit/ kg-hr secretin during irrigation of the pouch with buffer increased pepsin secretion 3-fold, approximately the same response found after switching from buffer to 100 mN HCl . The combination of secretin with pouch acidification, however, resulted in a l2-fold increase in pepsin output. These experiments provide evidence for a cholinergic reflex within the oxyntic gland mucosa containing a receptor sensitive to H+ which, when activated, leads to the stimulation of pepsin secretion. Furthermore, the studies with secretin indicate that this reflex may be one of the more important physiological factors regulating the humoral stimulation of the chief cells. Antral and duodenal pressures were studied in patients with and without dyspepsia by means of saline-filled open-tipped catheters. Simultaneous radiological evidence of duodeno-gastric reflux of barium-saline suspension was obtained by screening with an image intensifier. Three gastric duodenal motility patterns were associated with reflux (a) Rapid duodenal contraction with weak antral contractions (b) A single strong duodenal contraction not linked with an antral contraction (c) Variable duodenal contractions with absent antral contractions. Reflux was not observed when gastroduodenal contractions were both absent or closely linked. Metoclopramide is a drug with a unique action on gastric smooth muscle 1.2. which is effective in the treatment of flatulent dyspepsia 3 . When administered I. V. (lOmg), it was found to correct these abnormalities of motility by increasing the strength of antral contractions and increasing the linkage between antral and duodenal contractions. After Billroth I partial gastrectomy the drug was found to increase the strength of contractions in the gastric remnant but to have less effect on gastroduodenal linkage. It is likely to be useful in the treatment of delayed gastric emptying after this operation. Atropine has no effect on flow and composition of bile when the enterohepatic circulation is artificially maintained by infusions of bile salt at low dosages. The present study was done to determine the effect of atropine on the flow and composition of bile at both low and high rates of bile salt infusion. The effect of intravenous sodium taurocholate on gastric acid secretion was also studied. Four dogs were prepared with cholecystectomy, a gastric fistula and a duodenal cannula through which the common duct was intubated. In all tests sodium taurocholate Has infused intravenously at dosages ranging from 4.5 to 72 ~eq/min. The dose was doubled each hour. In some tests atropine was given in an intravenous dose of 0.08 mgm/kg over 10 min followed by a continuous infusion of 0.08 mgm/kg/hr throughout the experiment. Atropine significantly increased bile flow rate at taurocholate infusions above 18 ~eq/min but had no effect on bile salt output. Atropine caused significantly higher outputs of biliary bicarbonate, chloride and sodium at taurocholate doses above 36 ~eq/min but had no effect on bicarbonate or chloride concentrations. Atropine significantly reduced sodium and bile salt concentrations. Potassium concentration and output were significantly higher with atropine at taurocholate dosages above 9.0 ~eq/min. Gastric acid output increased progressively from 466 ~eq/30 min at a taurocholate infusion rate of 4.5 ~eq/min to 866 ~eq/30 min at a taurocholate infusion rate of 72 ~eq/min. This effect was blocked by atropine. The mechanism of the choleretic effect of atropine is unknown, but is not entirely due to parasympathetic blockade because in other studies vagotomy resulted in little alteration in bile volume and decreased bicarbonate output and concentration. Atropine produced alterations in bile volume and composition at high, but not low, taurocholateinduced flow rates. The use of a dose-response curve in this study has resulted in the first demonstration that atropine administration significantly alters bile flow and composition. Because 80-90% of total intestinal peptidase activity is found within the cell sap, emphasis has been placed on absorption of intact peptides and their intracellular hydrolysis. The role of surface (brush border) peptidases in protein digestion has been largely ignored. However, these surface peptidases have 10 times the specific activity of the cell sap enzymes and are biochemically distinct. Leucyl-B-Naphthylamide (LNA) proved to be an ideal marker substrate for the major rat brush border peptidase and for the analogous peptidase from human intestine. The rat and human surface peptidase was purified by zonal gradient centrifugation, Sephadex G-200 chromatography and (NH4)2S04 ppt. to a single band on disc electrophoresis. pH optimum = 7. 5 and Mg++ , Mn++, Zn++ did not alter activity. Assay with natural peptides (by L-amino acid oxidase method) showed competitive inhibition for LNA and revealed maximum specificity for leucine , phenylalanine and methionine at the amino-terminus of the peptide. All peptides are split avidly. Tripeptides are hydrolyzed faster and at lower substrate concentrations than dipeptides. Tetrapeptides, having the greatest affinity (lowest Km) and a Vmax approaching that of tripeptides, can be expected to be the initial preferred substrate for the brush border peptidase. Blocking of the N-terminus with CBZ or presence of an imino acid in the peptide eliminated activity. Hence, the major intestinal surface peptidase in rat and man is an amino-oligopeptidase having appropriate specificity and a strategic location to handle the products of pancreatic digestion . By virtue of its diffe.rential affinity for tetra-, tri-and di-peptides permitting sequential scission .of amino acids for transport, this cell surface peptidase is most likely essential for the terminal digestion of the majority of peptides known to be rich in neutral amino acid residues. Absorption of intact peptide is probably restricted to those with a predominance of glyc i ne or proline residues .

POSTPRANDIAL SERUM BILE ACID : A SENSITIVE TEST OF LIVER FUNCTION. Neil Kaplowitz , Engeline Kok , Norman Javitt, Cornell Medical College , New York , New York. Studies were performed to evaluate the sensitivity of fasting and two hour postprandial (2 hr pp) serum bile acid concentrations compared to standard li ver function tests in the detection of liver d i sease . Serum bile acid le vels were determined by gas liquid chromatographic methods. Five normal volunteers were found to have less than 1.0 fJ.g / m1. in fasting and 2 hr pp serum bile acid samples. In six patients with viral hepatitis and one with drug hepatitis , the increase in the postprandial bile acid concentration ranged from 100-500% . In three of these patients followed through the recovery phase , the 2 hr pp le vel was abnormal (> 3.0 J.l.g / m1.) at a time when the fasting concentration was normal « 1.0 IJg / ml) and the bilirubin, transaminases , alkaline phosphatase, and 5'-nucleotidase were normaL and the bromosulphthalein (BSP) retention was less than 10% . In ten patients with severe chronic li ver disease , elevated fasting bile acids were also followed by a considerable 2 hr pp increase ranging from 50-350% . Our findings demonstrate : (1\ The need for uniformity in performing and reporting serum bile acid levels , since values can vary several fold in relation to meals; (2) The enormous uptake capacity of the normal liver to remove the returning bile salt load from the enterohepatic circulation following a meal; (3) That a single serum bile acid determination two hours after eating appears to be a more sensitive indicator of hepatic dysfunction than bilirubin, transaminases , 5'-nucleotidase and may be at least as sensitive as a 45 minute BSP test. The purpose of this study was to compare endoscopy with conventional radiology in the diagnosis of acid-peptic disease of the proximal duodenum. The Olympus GIF Type D Fiberscope was used to examine 100 consecutive patients for indications which included acute and chronic upper GI bleeding, epigastric pain, and an abnormal gastrOintestinal series. The duodenal bulb was entered in 97 of 99 cases in which it was attempted. The second portion of the duodenum was entered in 89 of 97 attempts. There was no morbidity or mortality. In this series there were 22 cases of duodenal ulcer, 20 of which were seen endoscopically. 21 of these cases were studied with an upper gastrointestinal series and in only 10 did the radiologist diagnose a duodenal ulcer. Of particular interest, in 40% of the cases endoscopy showed more than one crater per patient while radiology failed to document more than one crater in any case. Inflammation of the duodenal bulb (bulbitis) was 769 noted in 80% of the patients with an active ulcer crater. Duodenitis without a crater was noted in 13 cases endoscopically and was associated with epigastric pain characteristic of ulcer disease in nine of these patients. Biopsy of active duodenitis showed characteristic inflammatory changes, including infiltration of the lamina propria with round cells and eosinophiles. The upper gastrointestinal series suggested duodenitis in only four of these 13 cases. This series emphasizes the limitations of radiology in the diagnosis of both duodenal ulcer and clinically significant duodenitis. It suggests that duodenoscopy should be a routine addition to the examination of the proximal duodenum in both the acute bleeder and in patients with chronic upper gastrointestinal complaints. A series of thirty-two postgastrectomy patients was studied with the Olympus GIF Type D fiberscope and a conventional upper gastrointestinal series for indications which included acute and chronic upper gastrointestinal bleeding, epigastric pain, vomiting, and the dumping syndrome. In this group there were thirteen Bilroth II and nine Bilroth I resections, nine patients with a vagotomy and pyloroplasty and one with a gastrojejunostomy. Of ten patients with an active ulcer crater, five Md the ulcer at the suture line, four a jejunal (marginal) ulcer, and one an acute gastric ulcer. In each case of suture line ulcer, a black silk suture was seen protruding from the base of the crater(s). Suture material was removed endoscopically from two patients, one with a BI resection and one with a vagotomy and pyloroplasty. Both had low gastric acidity. Of interest was the observation of bacterial growth on these sutures; in the single case where a culture was taken, from the patient with the vagotomy and pyloroplasty, a heavy growth of E. coli was obtained. Symptoms were present in all patients with suture line ulcers and ranged from epigastric pain and vomiting to acute upper gastrointestinal bleeding. In the two patients with high gastric acidity surgical intervention was required. A suture line ulcer was suggested in only one of the five patients by the upper gastrointestinal series. Suture line ulceration appears to be a distinct, symptomatic entity related to retained nonabsorbable suture material. This experience suggests that endoscopy is a necessary part of the examination of a symptomatic patient following gastric resection. It also suggests that in addition to the suture material, the pathogenesis of suture line ulcers may involve the variables of gastric acidity and bacterial colonization. An instrument developed for insertion through a standar d sigmoi doscope to obtain cytologic material from the visualized and more proximal colon has been previously described. To date 130 patients have been examined with this instrument and its prototypes and satisfactory spe ci mens were obtained in all. Final histologic diagnoses were obtained in 70 of these pat ients who underwent laparotomy. Of 38 pati ents with lesions within range of the sigmoidos cope , a 100% correlation existed between preoperative cytologic and final histologic diagnoses. The latter diagnoses included lymphosarcoma as well as adenocarcinoma. This group included 5 pat ients with strictures in whom preoperative surgi cal biopsy was not feasible, and 8 patients with preoperative surgical biopsy interpreted as benign lesion but whose cyt ology was positive for mal i gnancy, the latter conf irmed at lapar ot orr~. Of 16 patients with sigmoid and left colon lesions beyond reach of the sigmoidos cope , 75% of cancer were diagnosed pre operatively. The correlation of cytology and f inal histologic diagnoses was l ower in the right colon in a l i m ited series. The technique is suitable for the general population as well a s for those with a higher risk of m alignancy. This can be performed as an out-patient procedure with physician time and patient discomfort kept at a m inimum . This study was undertaken to determine the development after birth of intestinal glucuronidase activity in rat and man and to investigate the development and characteristics of the bacteria responsible for production of intestinal glucuronidase in the rat. Bacterial type beta-glucuronidase activity is absent in human feces at birth and increases in human and rat feces with age. In rats, comparison of quantitative cultures of intestinal content with glucuronidase activity suggested an association of the activity with development of the anaerobic flora. Antibiotics which reduced the anaerobic flora eliminated enzyme activity from rat cecal content in vivo and in cultures. Gross cultures of rat cecal content produced glucuronidase only under anaerobic conditions and enzyme production rapidly diminished in the absence of substrate (glucuronide). Isolation of individual organisms that released glucuronidase into the media was difficult and required maintenance of strict anaerobic conditions and substrate in the media. A variety of anaerobes were found to produce the enzyme. The effect of luminal glucuronidase on endogenous and exogenous compounds excreted in the bile as glucuronides conjugates should be considered in situations in which the anaerobic flora is altered.

Venereal diseases may affect the rectal orifice and may mimic inflammatory bowel disease. This report deals with experience gained in the diagnosis and management of four cases of gonorrheal proctitis (GCP) seen in homosexual men, two cases of lymphogranuloma venereum (LGV), and two cases of traumatic fissure-in-ano seen over an eight-month period.

In GCP disturbed bowel function, blood or mucopus in the stools, tenesmus and rectal discomfort along with proctoscopic findings of hyperemia, edema, and friability were seen and suggested ulcerative colitis (UC). A purulent exudate overlying diffuse mucosal abnormalities is characteristic of GCP but was also seen in patients with UC. In both GCP and ulcerative proctitis (UP) mucosal abnormalities were limited to the rectum. In UC abnormalities as seen on proctoscopy and barium enema extend beyond the rectum. When abnormalities are confined to the rectum, diagnostic confusion was resolved with bacteriological culture on Transgrow medium. Treatment with high doses of penicillin, tetracycline, or spectinomycin was successful.

The cases of LGV presented with bloody rectal discharge, rectal pain on defecation, altered bowel habits, and anal masses mistaken for hemorrhoids. Perianal ulcerations were present, and proctoscopy showed the rectal mucosa to be severely ulcerated, inflammed, and narrowed. Barium enema showed rectal stricture, recto-vaginal fistula, and discontinuous involvement of the colon beyond the splenic flexure. Crohn's colitis was considered after initial evaluation, but serological and skin tests confirmed the presence of LGV. Despite appropriate antibiotic treatment, the symptoms were unrelieved and abdominal-perineal resection was required.

Perianal ulcers and fissures may be the initial indication of Crohn's disease, but mUltiple fissures-in-ano may also be of traumatic or syphilitic origin.

It is important to recognize the venereal nature of the rectal diseases so that appropriate therapy may be given. Metastatic infections may arise from GCP, and the active partner in rectal intercourse may contract GC urethritis. Prompt treatment of the LGV proctitis may prevent the devastating consequences of the chronic disease. Hl1II&n serwa contains a soluble monoamine oxidase (MAO) which deaminates a yar1ety of monoamines. We have studied MAO yalues in noraal subjects and in patients with acute and chronic liver diseaae. Forty-two noraal subjects with an age range of 18 to 75 years had a mean value of )).) :t: 1.2 (:t: 1 SE) units. Thirty-five patients with clinical and hiatologic evidence of hepatic cirrhosis had a .ean MAO level of 44.8 :t: 2.9 unita, which was significantly greater (p<O.OOl) than the noraal level. By con-t~st, twenty-seven patients with acute hepatitia had a mean MAO level of )1.1 :t: 1.4 units. Thia waa not significantly different (p>O.)O) from the mean noraal MAO but nil significantly less (,<0.001) than the .ean MAO for the cirrhotic patienta. In Mdition, there was no ~oaitive correlation between serna MAO levels and atandard hepatic function tests (transaminasea, alkaline phosphatase, bilirubin, or prothrombin time) in the patienta with either acute or chronic liver disease.

These findings indicate that eleyated serna MAO is not related to acute liver in-, but aay reflect the development of hepatic fibrosia. This hypethesia is suprted by prel1ainary aniaal studies in our laboratory. During the preduction of experimental hepatic fibrosis with CCla in rabbit. pretreatment values of serum MAO ef 664 :t: )1.1 units rose to levels of 11)8 :t: 21.0 units while the yalue. for the control rabbits showed no significant change. Poat-mortem eyaluation. of both liver hi.tology and hydroxy-proline content confir.ed the aarked d~e of hepatic fibro.is p~sent only in the treated group.

Conclusions Serna MAO levels are eleyated in liver diaeaae only when hepatic fi-oBis is present, and are not related to abnormalities of other liver ensy.es. This unique enzyme aay be a yaluable early clue to the development of cirrhosis in .. n. In 1970, Lincheer described the severance of esophageal webs with a balloon catheter and mentioned the application of this technique to lower esophageal strictures. We have subjected four patients with lower esophageal strictures to balloon pull-through bougienage. Two patients, W.W. and E.H., had well-documented peptic esophagitis resulting in a thin, short, fibrous stricture. The third patient, M.H., had well-documented scleroderm.a with development of a thin, fibrous lower esophageal stricture. The fourth patient, M.D., had known hiatus hernia for several years and developed dysphagia rather suddenly. She also had a known carcinoma of the colon . Radiography and endoscopy demonstrated a stricture which was somewhat thicker than those in the other three patients. In all cases esophagoscopy, biopsy, and exfoliative cytology were performed to exclude the possibility of malignant disease. Under fluoroscopic control a fabricated elongated Foley catheter, a prototype pediatric Baker tube, or an adult Baker tube was passed into the stomach. The balloon was then filled with 11 cc of 20% hypaque and cine radiograms were taken as the balloon was pulled through the stricture. In all cases, the balloon pull-through was accomplished without difficulty and with no significant discomfort to the patient, Three of the four patients, W.W., E.H., and M.H., appreciated significant improvement following the procedure. Post-procedure barium swallows and cine esophagrams documented an increase in diameter of the lower esophagus. One patient, M.D., did not benefit from this procedure. Review of the cine radiograms during the procedure in this patient demonstrated that the balloon itself deformed during passage. Balloon pull-through bougienage is a safe, comfortable, and effective treatment for lower esophageal strictures. It seems most appropriate in patients in whom the stricture is fibrous and short. The original description, by Thompson, Krupey, Freedman, and Gold, of an assay system for carcinoembryonic antigen of the digestive tract (CEA) in serum and its relative accuracy in the diagnosis of colonic carcinoma suggested that this test might be especially useful in the screening of patient populations with a high risk of developing intestinal cancer. A radioimmunoassay employing rabbit anti-CEA and 1125 labeled CEA was used to measure serum CEA in 77 patients with chronic inflammatory bowel disease. Twenty patients had clinically or pathologically documented ulcerative proctitis, 31 had Crohn's disease, and 26 ulcerative colitis. Sixty-seven of these 77 patients (87%) had a serum CEA in the normal range (+2.8 ng/ml.). Two with proctitis, four with Crohn's disease, and four with ulcerative colitis had serum CEA levels greater than 2.8 ng/ml. Only one of the ten patients with an elevated serum CEA had a detectable carcinoma of the colon. There was no apparent correlation between the level of serum CEA and the clinical activity of the disease in these patients. Although only one of the patients with an elevated serum CEA was proven to have a carcinoma, the possibility exists that the test may be detecting clinically inapparent carcinomata or a pre-malignant state. Further studies are required to determine this and to evaluate the usefulness of serum CEA levels in the diagnosis of intestinal carcinoma in patients with chronic inflammatory bowel disease. The preliminary results suggest such studies would be worthwhile. Acute gastrointestinal ulceration and hemorrhage associated with intracranial disease has been known clinically since Rokitansky's description In 1842. However, the pathophys I 01 ogy of th i s cond i t ion rema I ns unc I ear. I n the present study, gastr i c acid output was studied in six adult female Rhesus monkeys before and during periods of intracranial hypertension. Fol lowing a twelve-hour fast, each animal was anesthetized with intravenous Nembutal. Trephine openings were made in the right (recording site) and left (epidural balloon) posterior parietal area. A nasogastric tube was inserted into the stomach and Its position confirmed roentgenographically. The gastric contents present were aspirated and discarded. Systemic arterial blood pressure and intracranial pressure were monitored with indwel ling transducers. Gastric aspiration was then begun and carried out every 5 minutes. Fol lowing a twohour control period, intracranial pressure was elevated to 75 mm Hg (ten fold) by inflating a bal loon in the epidural space. Gastric aspiration was then continued for another two-hour period. During the acute period of intracranial hypertension, the volume of gastric aspirate decreased from a mean of 2.05 cc. per hour in the control period, to 0.85 cc. per hour, p (.05, and total acid output decreased from a mean of 0.575 mEq/hour to 0.225 mEq/hour, p ~.05. During this period of intracranial hypertension, the mean arterial blood pressure rose from 87 mm Hg to 120 mm Hg. Since it has been shown that this rise in systemic arterial blood pressure during periods of increased intracranial pressure is due to peripheral vaso constriction, including the splanchnic area, it is postulated that a decreased blood flow to the stomach during the study period might account for the decrease in volume and in acid output. In 3 dogs with vagally innervated gastric pouches /PP/, gastric fistulas and pancreat5_c fistulas, half maximal acid response to pentagastrin /2 pg/kg-hr/ and to Urecholine /0.1 mg/kg-hr/ has been obtained. When gastric acid secretion reached a plateau either caerulein was infused intravenously in doses ranging from 30 to 480 ng/kghr or L-tryptophan and L-leucine mixture was instilled intraduodenally at the rate of 8.8 m moles/hr to release cholecystokinin /CCK/ endogenously. Caerulein caused a dose related inhibition of PP response to pentagastrin with the maximum of 72.3% reached at the dose of 250 ~/kg-hr caerulein. It produced a dose related increase of PP response to Urecholine with the maximum of about 10~G reached at the dose of 250 ng/kg-hr. Endogenous CCK producing the pancreatic protein output equal to that obtained with the dose of 120 ng/kg-hr caerulein, reduced PP response to pentagastrin by 54.5% and augmented PP response to Urecholine by 9~fo.

Following surgical conversion of PP to vagally denervated Heidenhaln pouches /HP/ endogenous CCK and doses of caerulein producing equal rate of pancreatic protein secretion were ineffective in the inhibition of pentagastrin induced acid secretion from HP. Stimulation by pentagastrin combined with a threshold dose of Urecholine increased the susceptibility of HP to inhibition by caerulein but not by endogenous CCK. In addition the sensitivity of denervated pouch to Urecholine was profoundly increased and both caerulein and endogenous CCK augmented the Urecholine induced acid secretion more markedly from HP than from PP. From these results we have tentatively concluded that the cholinergic innervation plays a direct role in the action of CCK on gastric acid secretion. A series of 185 sigmoidoscopic biopsies included 101 from clinical ulcerative colitis, 27 granulomatous colitis, 21 indeterminate colitis,S other colonic inflammations and 36 control biopsies removed for other reasons. These were analyzed to seek characteristic differences between diagnostic categories. Without prior knowledge of the diagnosis, counts made of an average 500 mucosal crypt cells per biopsy distinguished the mucous goblet and other epithelial cells; besides mitotic, pyknotic and karyorrhectic epithelial cells. Also 500 connective tissue cells of the lamina propria were counted per case, including the fibroblasts, macrophages, mast cells, lymphocytes, plasma cells, eosinophil and neutrophil polymorphonuclear leukocytes. Comparisons by diagnostic categories showed the positive findings to be fragmented epithelial cells, 3.9 times the control mean in ulcerative colitis; goblet cells were reduced to two-thirds of normal. Lamina propria mast cells were 2.5 times the control value, neutrophils 11 times normal, and eosinophils 44 times normal. Drug and steroid therapy affected these parameters. Rectosigmoid biopsies are helpful to distinguish between ulcerative and granulomatous colitis, based partly upon their cytologic differences. Reduced goblet cells, increased nuclear fragmentation and mast cells indicate ulcerative colitis, while granulomatous colitis lacks this combination of cytologic. changes. This approach serves to supplement the clinical and radiological distinctions between these two conditions. Blood removal rates for alcohol (A) and tolbutamide (T) were compared with enzyme levels in liver biopsies in 13 patients (pts.). Repeat studies were made in 8 pts. at 6 wks. of hospitalization, and 5 pts. at 10 wks. Patients were severe chronic alcoholics, free of drug use in the 3 mos. before study. All were free of cirrhosis by biopsy, and received only paraldehyde during alcohol withdrawal. No other drugs were used during hospitalization. Testing was begun 3 to 10 days after admission, after the pts. had begun to eat well. I.V. alcohol was given in a dose of 1.6 gm./kg. over 3 hrs., followed by 6 hourly blood samples (results expressed as mg.%/hr.). T blood clearances (half-life, min.) were measured by 6 hourly blood levels after a dose of 1 gm. I.V. Liver biopsies were studied histologically, and analyzed for alcohcl dehydrogenase (ADH in ~mol/mg./hr.) and microsomal ethanol oxidizing system (MEOS, m~mol/mg./min.).

Results of T tests on the individual changes in rates are given in the table (data shown are means ± 1 S.E.).

Week 1 N=ll) . Alcohol and tolbutamide metabolism slowed during the first 6 wks. and fluctuated randomly after that, however, it appears that MEOS decreased during the ten week interval. This may indicate that change in MEOS is unrelated to A or T metabolism. There was no evidence of correlation between changes in alcohol removal rates and changes in either ADH, MEOS, or (ADH + MEOS).

These data suggest that factors other than the activity of these two liver enzymes are rate limiting in the removal of blood alcohol. determine whether these drugs influenced the metabolism of each other. The rates of disappearance of alcohol, tolbutamide and meprobamate from the blood were measured both singly and in pairs, in the sequence A, T ,M, (A + T), (A + M), A, T ,M. Patients were studied 5 to 10 days after admission to the hospital. Alcohol was given I. V. in a dose of 1.6 gm.;1<g. over 3 hrs., followed by 6 hourly blood levels. Tolbutamide blood clearances were measured by 6 hourly blood levels after a dose of 1 gm. I. V. Meprobamate, 12-15 mg.;1<g., was given orally as crushed tablets and 7 blood samples taken 4-12 hrs. after administration. When alcohol and either tolbutamide or meprobamate were simultaneously given, tolbutamide was given at the peak alcohol blood level (198 + 5 mg. %), and meprobamate was given 3 hrs. before the alcohol infusion was begun. Lowest alcohol blood levels during the tests were 70 + 7 mg. %. Results of tolbutamide tests on the individual changes in rates are given in the The correlation between activities of intestinal alkaline phosphatase and/or a calcium-activated ATPase and intestinal calcium transport, as influenced by vitamin D, has suggested that microvillus phosphatases play a functional role in intestinal calcium absorption. This study was undertaken to determine the relationships between alkaline phosphatase activity, calcium ATPase activity and calcium transport from duodenum, jejunum and ileum in rats raised on calcium-deficient and control diets. Calcium transport was measured by an in vitro everted gut sac technique using 45Ca serosal/ 45 ca mucosal (S/M) concentration ratios as a measure of transport. Alkaline phosphatase and calcium ATPase activity were measured in purified brush border preparations. S/M ratios of calcium-deficient animals (mean ± S.E.) were: Duodenum 5.8 ± 1.1, Jejunum 2.2 ± 0.2, Ileum 2.1 ± 0.1 compared to control Duodenum 3.3 ± 0.6, Jejunum 1.0 ± 0.1, Ileum 1.1 ± 0.1. Alkaline phosphatase activities from calcium deficient animals (mean IU ± S.E.) were: Duodenum 286 ± 21, Jejunum 46 ± 14, Ileum 1.2 ± 0.3 compared to control Duodenum 174 ± 35, Jejunum 41 ± 11, Ileum 0.8 ± 0.2.

:alcium ATPase activities from calcium deficient animals (mean IU ± S.E.) were: ~odenum 0.68 ± .18, Jejunum 0.16 ± .04, compared to control Duodenum 0.70 ± .19, Jejunum 0.12 ± .07. No Ca ATPase activity was found in ileal segments. These results demonstrate a gradient of alkaline phosphatase activity and calcium ATPase which corresponds in part with transport data. Correlation between enhanced transport ~d elevated alkaline phosphatase activity produced by dietary calcium restriction was present in the duodenal segment only, suggesting that alkaline phosphatase may participate in intestinal transport of calcium by the rat duodenum but not in more distal segments. DIABETES AND HlTESTHlAL AMINO ACID TRANSPORT 0 sity of Iowa, Iowa City, IO~la 52240. Dheeraj La1 and Harold Sched1. Univer-Studies in diabetic rats shol'l increased intestinal gro~/th, hexose transport, and disaccharidase activity. Although the intestine shO\'/S increased growth in the diabetic, body growth does not occur. Since amino acids are required for growth and because like hexoses their transport is linked to Ua+. l'Ie studied amino acid absorption in diahetics using a-aminoisohutyric acid (AlB). which is not metaboliszed but has the transport behavior of neutral amino acids. ~Ie perfused the entire small intestine in vivo in diabetic (D) and matched control (C) rats at 5 and 12 days after injection or allOxan in D and sal ine in C. The perfusion solution contained 5mt1 AlB. tracer l4C-AlB. 159 mM NaCl. and phenol red. At the end of perfusion the entire small intestine was measured. The mucosa was obtained by scraping I~ith a microscope slide. Small intestinal measurements (g. all data mean +sE) l'lere the same in both groups at 5 days.

Gut Dry Wt. g -Absorption. umoles/hr *p<O.Ol but were significantly greater ln the diabetlcs at 12 days for both wet and dry weight of both the whole gut and the mucosa. Absorption was the same in the blo groups at 5 days. At 12 days absorption per rat was significantly greater in D but per unit of gut there was no difference. Thus. at 5 days when weight of the intestine was the same in C and D. there was no difference in amino acid transport. At 12 days the increased intestinal growth in D rats was associated with a proportional increase in total intestinal absorption (per rat). but 'no increase per unit of intestine. This fits with failure of the diabetic rats to grow and demonstrates selective control of transport in diabetes. The factors controlling hepatic blood flow are incompletely understood. In order to evaluate the differential effects of hormonal influence on liver blood flow, the clearance of radio-xenon from the hepatic artery perfusion bed (HA) and portal vein perfusion bed (PV), was studied after parenteral administration of pitocin, glucagon, and pentagastrin. Purified preparations of cholecystikinin and secretin are under study. A chronic surgical preparation was made by implantation of separate catheters in the splenic artery and splenic vein at the time of splenectomy in premates. After allowing for recovery from surgery, the animals were intensively studied over the next 2-3 weeks. Radio-xenon was injected via subcutaneous reservoir into the splenic artery or splenic vein catheter. Four minute clearance curves were obtained over the right lobe of liver using a 2x2 inch NaI detector and specific organ blood flows were calculated from the disappearance rate, obtained from best fit analysis using a PDP-iO computer. In almost all instances, the clearance was monoexponential over this interval. In the anesthetized primate, specific flow through the HA was 2/3 of the flow via the PV . Hormonal stimulation had widely different effects on the HA and PV flow. For example, doses of pitocin sufficient to cause a marked depression of PV flow, to about 50% of the control value with gradual recovery complete at 20-25 minutes, had much less effect on HA, causing an initial 10-15% drop in flow and recovery by 5 minutes after injection. A dose of glucagon, sufficient to cause a doubling of PV flow, increased HA flow only about 20-30%, with prompt return to control values. Pentagastrin had little influence on HA flow, while PV flow was increased about 25%. In conclusion, the hepatic artery and portal vein perfusion beds appear to react as discrete functional entities in response to hormonal stimulation. The study of changes in the pattern of metabolic end products in the urine brought about after ingestion of alcohol could assist in defining pathway alterations attributable to alcohol metabolism. We examined the excretion patterns of organic acids before and after alcohol in seven chronic alcoholic subjects and three healthy young male controls. Of the seven chronic alcoholics, four were cirrhotic and three had normal livers by physical, laboratory and biopsy criteria. Organic acids were extracted, derivatized and gas-liquid chromatographed (by the methods of M.G. and E.C. Horning) from fasting urine and from urine formed during the 5 hrs following ingestion of 1.2 gm ETOH/kg in 30 min. 76 acid components were seen with emergence times corresponding to 11-28 methylene units (MU). After alcohol, urines of non-cirrhotic alcoholi cs showed increases in 30 components (mean = 193 J.Ig/hr) and decreases in 25 components (mean = 14 J.Ig/hr), resulting in a net mean rise of 52 J.Ig/hr. Cirrhotic urines after alcohol showed increases in 25 components (mean = 49 J.Ig/hr) and decreases in 51 components (mean = 242 \Jg/hr) for a net mean fall of 193 J.Ig/hr (p = < .005).

The component which showed the most consistent contrary directional change between non-cirrhotic and cirrhotic subjects was parahydroxyphenyl acetic acid (PHPAA) (identified by mass spectrography) at MU 17.7. This substance rose after alcohol in noncirrhotics by a mean value of 13 J.Ig/hr and decreased in the cirrhotic subjects by a mean value of 14.7 J.Ig/hr (p = < .002) 2/3 of the normal subjects increased and 1/3 decreased output of PHPAA, suggesting that a decrease in this substance is not a consequence of cirrhosis. These findings suggest that a general effect of alcohol in cirrhotic subjects is to inhibit some of the oxidative processes which give rise to urinary acids. This influence is not pronounced in non-cirrhotics. Systematic studies are needed to measure the inhibition of various oxidative systems during the metabolism of alcohol in a larger group of normal subjects and in non-alcoholic ci rrhoti cs.

Willem G. Linscheer. Lemuel Shattuck Hospital, Tufts University School of Medicine, Boston, Massachusetts.

3 The purpose of this investigation is to compare rates of absorption of H-oleic acid (OA) solubilized in clear micellar Na-taurocholate (TC) solutions with equimolar concentrations of OA in mixed micellar solutions (OA present in particles, micelles and as free soaps). Experimental design: Two consecutive jejunal loops (20 cm.) in rats were simultaneously perfused with a control and a test solution containing equimolar concentrations of OA at random order. Perfusion time 3 hrs. TC was constant in the control sol., but varied in the test sol. (TC/OA: 0.5'" 3.0). Following completion of the perfusion the two loops were washed by isotonic saline followed by 10% albumin. Radioactivity was measured in bowel juice and in emulsified bowel segments. In addition, histochemical studies were done of bowel biopsies taken before and after perfusion. Results: At a constant concentration of OA (20 mM) and TC/OA ratios between 0.5 and 3.0, it was found (18 rats) that the lowest rates of absorption occured with clear micellar solution (TC/OA > 2.5). A gradual increase of absorption was found with decreasing TC concentration. Maximal absorption was obtained with the mixed micellar solutions (TC/OA = 0.5, rate of absorption = 218 mJ.lmol/min./cm. jejunum) and the minimum rate at the highest TC concentration (TC/OA • 3.0, abs. rate = 86 mJ.lmol. (p < 0. 001). These results were reproducable and depended only on TC/OA ratios. Similar curves were obtained with lower concentrations of OA and similar re/oA ratios (12 rats). No toxic effect of the test sol . was observed by light and :lectron microscop~ The effect was not due to adsorption of OA as evidenced by albumin lashing procedures or to accumulation of OA particles between villi (microscopy). Conclusion: OA is better absorbed from the jejunal loops if perfused in a mixed micellar solution as compared to a clear micellar solution. An inverse relationship was found between absorption and the bile salt/FA ratio indicating that the conditions for absorption of OA from TC-FA micelles are optimal when the upper limit for dissolving FA by TC micelles has been exceeded.

James F. Lons, Division of Biological Research, Schering Corporation, Bloomfield, New Jersey.

The purpose of the present study was to determine the effect of intravenously administered oleic acid on the gastric secretory response to a meat meal.

Four dogs with vagally innervated gastric pouches (IP) and 2 dogs with vagally denervated pouches (DP) were fed a 250 gm. meal of ground beef hearts and gastric juice collections were made at 30 minute intervals for 4 hours thereafter. Oleic acid was infused into the femoral vein at 8 uEq/kg/min for 60 min starting 30 min after feeding in a solution containing 5% glucose, 1.2% licithin and 0.5% polyethylene-polypropylene glycol at 1.6 ml!min.

Control studies were performed with feeding alone, with 5~ glucose infusion and with infusion of vehicle.

Oleic acid produced a signficant inhibition of gastric secretion starting 60-90 min after feeding and continued for the duration of the experimental period. During the periods 120 min through 240 min after feeding, gastric acid output was inhibited 90~ in IP dogs and 70i in DP dogs. Five percent dextrose did not alter gastric secretion, whereas the vehicle did produce some inhibition (~40i) in 2 of the 4 IP dogs. No rebound in secretion was observed.

These results suggest that oleic acid inhibits gastric secretion by a different mechanism than that recognized as existing within the mucosa of the small intestine, or that IV administered fatty acid can stimulate this specific inhibitory mechanism. Malakoplakia of the gastrointestinal tract is a rare disease, characterized grossly by soft yellow-brown mucosal plaques or nodules and, microscopically, by numerous histiocytes containing pathognomonic Michaelis-Gutmann bodies. Although infection has been implicated in its etiology, the pathogenesis and morphogenesis of the leSion, particularly in relationship to bacterial infection, has aeretofore not been elucidated . We have had the opportunity to demonstrate the bacterial pathogenesis and ultrastructural morphogenesis of ano-rectal malakoplakia in a two-month old infant who presented with diarrhea and anal polypoid lesions. Associated hypogammaglobulinemia was present and death resulted at the age of 8 months from gram-negative septicemia. Lesions with intracellular baCilliform bodies and pathognomonic inclusions were observed in ano-rectal tissue obtained from incisional and excisional biopsies at the ages of 5 and 7~ months, respectively, and also at autopsy. Electron microscopic examination performed on tissue fixed immediately in paraformaldehyde-glutaraldehyde fixative revealed numerous gram-negative bacilli in varying states of morphologic integrity within the histiocytes. All stages of phagolysosomal bacterial degradation were found; in their most advanced state, these phagolysosomal bodies represented the ultrastructural equivalent of the typical PASpositive granules observed by light microscopy. Ultrastructural focal crystalline deposits which showed morphological and histochemical characteristics of calcium phosphate were found, and various stages in the morphogenesis of these formations were finally traced to well formed concentric structures, constituting classic Michaelis-Gutmann bodies. The ultrastructural relationships of the intact bacteria, phagolysosomal products of bacterial diSintegration and intracellular precipitation with deposition of calcium phosphate salts in the pathogenesis and morphogenesis of the characteristic lesion and disease process have been presented . Evidence for the bacterial etiology of malakoplakia is demonstrated.

Foundation, Oklahoma City, Oklahoma.

LP-X occurs in almost all cases of liver disease associated with histologically or surgically proven cholestasis. It is therefore the most consistent parameter for the diagnosis of obstructive jaundice. A semi-quantitative immunoelectrophoretic method exists which detects LP-X, after its cathodal migration in agar, using a specific antiserum. The method unequivocably detects LP-X, but quantitation is difficult since it depends upon visual comparison of the test sample immunoprecipitate with a control LP-X sample. In addition, antiserum titer varies between batches and LP-X-positive control samples lose their activity after 2-3 months. The present method utilizes the cathodal migration of LP-X in agar electrophoresis and its constant (60%) phospholipid content. Following electrophoresis a disc of agar is removed, cathodal of the sample well, from the region LP-X would migrate into if it were present. The disc is then assayed for phosphorus by a standard method. Using serial dilutions of LP-X-positive sera or purified LP-X the procedure has proved to be simple, accurate and reproducible. The coefficient of variation of repeated (33) determinations of 7 different sera, containing >40 mg/IOO ml of LP-X is <9.88%. At levels of LP-X >0 <40 mg/IOO ml of LP-X the coefficient becomes higher due to the raised levels of very low density lipoproteins in these sera. Mean levels of LP-X, in mg/IOO ml, and percent positive LP-X tests (total numbers tested are in parentheses), found in various disorders with obstructive jaundice are respectively: extra hepatic biliary obstruction (23), 549 and 100%; primary biliary cirrhosis (8), 417 and 100%; post necrotic cirrhosis (10), 264 and 40%; acute and neonatal hepatitis (39), 205 and 64%; alcoholic cirrhosis (25), 152 and 55%; alcoholic steatosis (25), 178 and 40%; chronic h'patitis (17), 103 and 53%; hepatic failure (4), 27 and 25%. This quantitative assay for LP-X therefore eliminates the requirement for a specific antiserum, and allows better correlation of LP-X with the degree of cholestasis and clinical status. The mechanism of K secretion was investigated using a unilateral preparation of frog gastric mucosa bathed on the serosal side only with buffered Ringers containing K in varying concentrations (1-4 roM). Microsamples (lD-50 ul) of secretion were obtained at intervals and analyzed for H, Na and K. The potential difference (PD) was measured continuously using a K-free 0.1 M NaCl agar bridge in the lumen. Secretory K was linearly related to serosal K ranging from 3.5 ± 0.2 roM at serosal 1 roM K to 11.9 ± 0.6 roM at serosal 4 roM K. The slope of the relation indicated that K was concentrated 3 fold in secretion. The initial PD in the resting state averaged 39.7 ± 1.8 mv but fell rapidly to 21.4 ± 0.8 mv with the inception of secretion. The extent of the fall was predictable and closely approached the expected liquid junction potential between the contents, particularly H, of each sample (n = 26) and 0.1 M NaCl in the bridge. It was possible to calculate that the electrical gradient to which secretion was exposed (38.5 ± 1 mv) was similar to the initial PD (39.7 ± 1.8 mv) and somewhat higher than the K equilibrium potential. It was concluded that K had moved passively in each instance down the electrochemical potential gradient. In other experiments, control serosal solutions (4 roM K) were rendered hypertonic by addition of 25-75 roM NaCl. The procedure enabled K secretion to be dissociated from H secretion.

K concentration decreased in parallel with water flow while H concentration increased with serosal and secretory osmolality. These studies form the basis of a model for K secretion: secretory K originates in a labile intracellular compartment, traverses the l.uminal cell membrane through water channels in the process of osmotic equilibration and distributes in the secreted fluid at or near electrochemical equilibrium. The model provides a ready explanation for the characteristic early K transients and for the parallelism between H and K secretion. The evident similarities between in vitro secretion and mammalian in vivo secretion support the wide validity of this model. CCK-PZ coordinates contraction of the gallbladder and pancreatic enzyme secretion; surgical disruption of these mechanisms may impair digestion. We have therefore defined responses of each organ to the hormone in health and investigated the changes which follow denervation (vagotomy) or removal of the bile reservoir (cholecystectomy~ Three groups, each of five individuals, were studied; 1) healthy volunteers; and, more than a year after operation, either 2) vagotomy and pyloroplasty or 3) cholecystectom~ During steady state intraduodenal perfusion of normal saline containing PEG, total ou~ puts of bile acids and trypsin were determined for 1-hour periods--basa1, then after IV CCK-PZ was given in sequentially increasing doses (0.0039, 0.0078, 0.0156, 0.0310, 0.0625, 0.125, 0.250 CHR U/kg/min). Three consecutive 20-minute collections were analyzed at each dose level. In health, gallbladder contraction (peak bile acid output) occurred at 0.0625, whereas the pancreas was more sensitive to CCK-PZ. Trypsin secretion commenced at 0.0078; thereafter an exponential increase in output occurred until maximal secretion was established at 0.0125. Vagotomy had different effects on the gallbladder and pancreas. Sensitivity of the gallbladder to CCK-PZ was enhanced and contraction occurred at 0.0078. In addition, bile acid output much exceeded that in health (p < 0.01). By contrast, basal trypsin output was less than in health (p < 0.05) and submaximal doses of CCK-PZ evoked smaller responses (p < 0.05), although maximal enzyme output was the same. After cholecystectomy, bile acid output remained at basal levels with all doses of CCK-PZ; basal trypsin output and the threshold response to CCK-PZ were the same as in health. Concerning the actions of CCK-PZ, we conclude a) the pancreas is more sensitive than the gallbladder in health; b) vagotomy decreases sensitivity of the pancreas and increases sensitivity of the gallbladder, additionally increasing gallbladder capacity; and c) cholecystectomy predictably decreases bile acid output. Studies following cholecystectomy also validated our quantitative assessment of gallbladder contraction in health or after vagotomy and demonstrated no direct effect of CCK-PZ on hepatic bile acid output. In order to study the pathogenesis of chylous ascites in patients with cirrhosis a test meal containing medium chain triglycerides (MCT) and long chain triglycerides (LCT) was fed to the 3 alcoholic cirrhotics with chylous ascites (Group A), to 8 alcoholic cirrhotics with non-chylous ascites (Group B) and to 4 non-cirrhotic patients with chylous ascites due to lymphat~c obstruction (Group C). The test meal contained Triolein (20 m ~~les) labelled with H-oleic acid (C18) and Trioctanoin (20 m moles) labelled with C-octanoic acid (C8). The radioactive markers in and outside the chylomicron fraction were determined in serial samples of ascitic fluid and venous blood over a period of 72 hours. In addition, ratios of triglycerides (TG), phospholipids (P), cholesterol (C), were measured as they reflect CM size. Results: C18 in ascites: The time of appearance (1 hr')'3time interval from ingestion to peak levels (5 hr.) and half-life time (58 hr.) of H labelled C18 in ascites were similar in the three groups. However, 8% of the ingested Triolein was present in ascites in Group A during peak levels and only 0.6% in Group B and as much as 22% in Group C (p < 0.001) for all groups. The CM fraction of C18 remained constant, but it was much higher in chylous ascites (mean: 83.8% ± 0.8 SE) than in clear ascites (45 ± 0.8) (p < 0.001).

C18 in serum CM were 7.3 times lower than those in chylous ascites. All patients with chylous ascites had large size CM in contrast to patients with clear ascites and small CM. C8: Peak serum levels occured in all patients after 1 to 2 hours. C8 was primarily non-esterified in serum and ascites of the cirrhotic patients, but esterified in ascites of Group C. Peak serum concentrations in the cirrhotics were 5.3 times higher than in ascites. Conclusions: 1. Fat in chylous ascites in the cirrhotics is of exogenous origin and reaches the ascitic fluid via the intestinal lymphatics as TG incorporated in CM (primarily large size CM). 2. The intestinal lymphatics of cirrhotic patients with clear ascites are only permeable for small CM. 3. The blood vessels as well as intestinal lymph vessels contribute to ascites formation in cirrhotics with chylous ascites.

When the enterohepatic circulation is intact, chenodeoxycholic acid (CDCA) has a slower turnover rate than cho1ic acid (CA) (V1ahcevic et a1. Gastroenterology 61:85, 1971), the major component of the bile acid pool. After ileal resection CDCA is usually the major bile acid present. We measured the rates of fecal excretion and the coefficient of absorption, Y (Woodbury and Kern, J. C1in. Invest. 50:2531, 1971) of 14C-24-CDCA and 3H-2-4-CA simultaneously in five patients with ileal resection and one with pancreatic insufficiency. Since the biologic effects of fecal bile acids are mediated by bile acids in aqueous solution, we determined the percentage of total fecal radioactive CA and CDCA in the fecal water, prepared by centrifugation of fecal samples at 100,000 X G for 30 minutes. The three patients with the slowest excretion of bile acids, including the one with pancreatic insufficiency, excreted CA more rapidly than CDCA and the coefficients of absorption of CDCA were greater than those of CA. The differences in Y were small in two of them (two and three percent) but the half time of excretion of CDCA (T 1/2-CDCA) was considerably greater than T 1/2-CA in all three. In the other 3 patients there was essentially no difference in excretion rates of the bile acids. Two of them excreted both bile acids very rapidly: T 1/2 four to five hours. Fecal water contained a larger relative percentage of 3H than 14C in all patients (p=0.05). The concentration of 3H-CA and its metabolites in the fecal water varied directly with the stool pH {r=O.66) but the aqueous concentrations of l4C-CDCA and its metabolites were independent of pH. Three of the patients with ileal resection were studied a second time after equilibration on an homogenized liquid diet of egg albumin, corn oil and sucrose. There were no consistent differences in bile acid kinetics or the proportion of bile acid in solution in fecal water. These results show that in some but not all patients after ileal resection CDCA is absorbed more efficiently and excreted more slowly than CA and that more CA is in solution in fecal water. These findings help to explain the mechanism underlying the predominance of CDCA after ileal resection. Indirect experiments with amphibian stomach in vitro have implicated cyclic AMP (cAMP) as the intracellular mediator of acid secretlon. Since cAMP is synthesized by AC and degraded by PDase, we tested the hypothetical role of cAMP in mammalian stomach in vitro by measuring effects of secretagogues on canine gastric mucosal enzymes. AC was assayed by the method of Krishna et al. and PDase by the method of Butcher and Sutherland. We found: (l) histamine TTO=O to 10-1 M} did not activate AC, whereas sodium fluoride (NaF, 10-2 M) activated AC 10 fold; the combination of histami~e (l0-3M) and NaF produced no increase in activat'ion over NaF alone; (2) urecho~lne (5xlO-4 M) alone stimulated AC only slightly, but Significantly augmented the stlmulatory effect of NaF; (3) histamine blocked the additive effect of urecholine on NaF activation; (4) gastric PDase was unaffected by histamine. These results differ from data obtained in amphibian gastric mucosa; histamine neither activated AC nor.in~ibited PDa~e in the. dog stomach. It seems unlikely that cAMP plays a role in medlatlng the canlne gastrlc secretory response to histamine (Supported by N.I.H. Grant #AM 15977-01).

In previous experiments rats were subjected to a two-thirds proximal small-bowel resection and subsequent metabolic studies performed for four weeks prior to killing. Resected rats grew at the same rate as controls but manifested elevation of fecal nitrogen in the second post-operative week (33+6% of oral intake: controls 22+2%); this fell towards normal by the fourth week (27+2%). The fall in fecal nitrogen was accompanied by a two-fold proximal increase in tissue enterokinase, a brush-border enzyme. Since the apparent improvement in protein absorption could not easily be explained solely by the assumed increase in lumenal trypsin, the possibility of associated changes in the activity of peptide hydrolase, another brush-border enzyme, was studied. Particulate and soluble fractions of 5 intestinal segments in 6 normal and 6 four-week resected rats (2 segments) were examined. In normals, total and specific activities of particulate and soluble peptide hydrolase (leu-ala substrate) rose progressively from duodenum to mid-ileum. By contrast, when jejunal and mid-ileal homogenates were assayed for sucrase, total segmental and specific activities were lower in the distal segment (S.A. jejunum 90+10 u/g: S.A. ileum 24+3 u/g p<.OOl). Following resection total protein content and wet weight of each standard segment increased 1.5 to 2 fold. No significant change in the total or specific activities of sucrase occurred, but peptide hydrolase activities were markedly elevated in the two segments studied, the change being greatest in the second part of duodenum (particulate specific activity: controls 36+7 u/g, resected 84+8; p .001; total particulate activity per standard segment: controls 205±4 u/g, re;ected 946:!:,9 u/g; p <.001). Similar changes were noted distally. Increases in soluble enzyme activity, while marked, were less significant (total activity per duodenal segment: controls 2520±250 u/g, resected 59l3±125 u/g; p<.005: total activity per mid-ileal segment: controls 4670+840, resected 6938+754 u/g; p<.Ol). Zymograms of soluble and particulate enzyme fractions revealed identical patterns in both groups. These data suggest a specific role for peptide hydrolases in adaptation to the protein maldigestion which follows resection. Gl and Se have important actions on gastric and pancreatic secretions. Controversy exists regarding the effects of these hormones on jejunal water and electrolyte transport. Accordingly, we investigated the effect of iv administration of Gl and Se on the jejunal transport of water and electrolytes in healthy humans by the triple lumen perfusion technique. Net jejunal transport of water and electrolytes and unidirectional movements of cations were measured in 20 subjects. Serum electrolytes, glucagon, insulin, and glucose levels were measured. After these measurements during the control period (C), each subject was given GI either at 2.0, 5.0, or 20pg/kg body wt/hr by a continous iv infusion. 2 pg/kg Gl prod'Jced 50% reduction in absorption (p < .02, paired analysis). 5.0 pg/kg GI produced secretion of water and electrolytes (p <.05) but 20pg/kg GI produced less secretion (p <.02) (see table) . There were no consistent alterations in the unidirectional movements of cations. The randomization of C and GI periods did not alter the results. Comparable alterations of serum glucose or insulin levels do not influence jejutlal absorption. (Costrini, 1971) . Se (GIH) 4 unit~ kg/body/wt/hr also produced secretion of water and electrolytes (see table) . Exsorption of sodium was significantly increased (p ~.005) with no changes in insorption. These studies indicate that glucagon and secretin alter jejunal fluid transport a9d thus may playa role in diarrheal syndromes associated with some pancreatic tumors; *\-) minus sign = secretion. 

Normally small amounts of BA are deconjugated and absorbed in the colon. During the oral administration of BA or in ileal disease greater amounts of BA may enter the colon and be absorbed. We studied the rates of absorption of cholic (CA), deoxycholic (DCA), and chenodeoxycholic acids (CDCA) and tested the effect of pH on the solubility and absorption of BA. The colon of 42 subjects was intubated by triple lumen tube and after thorough cleansing was perfused with a physiologic electrolyte solution at pH 8, containing BA and polyethylene glycol. BA absorption was calculated under steady state conditions by mass analysis and disappearance of labeled BA. At 1, 2, 3, and 10 roM concentrations (concn) BA absorption was (mean + S.D.) 0.88 + .2, 1~73 ± 1, 2.9 ± 1, 9.6 ± .5 pmoles/min/colon respectively. At 1, 2, and 3 roM concn CDCA absorption was 7.34 ± 2.4, 14.8 ± 4, 24 ± 3 pmoles/min/colon respectively. At 1, 2, and 3 roM concn DCA absorption was 5.72 ± 1.9, 9.9 ± 1, and 9.1 ± 4pmoles/min/ colon. CDCA and ~fA absorption was greater than CA at all concentrations tested (p < .02). Mean CDCA absorption was greater that DCA at a11 concentrations but the differences are significant only at 2 and 3 roM concn (p < .05). The absorption of CA and CDCA is concn dependent and linear between 1-10 roM and 1-3 roM concn respectively. The lack of increased absorption of DCA at 3 roM may be due to the secretion of water that is produced by 3 roM DCA. The solubility of BA was tested in vitro and was found to be pH dependent and different for each BA. A sharp rise-i;-sorubility of CA occurred at pH 5.5 and at pH 6.25 for CDCA. DCA formed a gel between pH 6.25-7.00 and was in true solution only at pH 7.5 or above. Perfusion of the colon with BA at pH 7.0 produced a significant increase of absorption of 1 roM CA from .88 ±.2 to 2.1 ±.4 (p(.02). CDCA absorption at pH 7 did not change from that at pH 8. DCA absorption could not be tested at low pH values because of gel formation. It is concluded that 1) significant differences exist in solubility and the colonic absorption rates of unconjugated BA; 2) At physiologic concentrations and pH CDCA has the greatest absorption; 3) colonic absorption of BA, and CDCA in particular, could contribute to the maintainance or expansion of the BA pool. The dog is frequently employed as a model for studies of serum and pancreatic enzymes.

The appropriateness of this may be questioned since the concentration of amylase in dog serum is approximately 10 times that in humans.

Chromatographic separation of serum amylase on a column of agarose gel revealed 3 amylase peaks in dog serum and only a single low molecular weight peak in humans.

In the dog the first peak of amylase activity is eluted immediately after the large sized proteins whereas the third peak is eluted at the same volume as human serum amylase.

It appears that concentrations of serum amylase in the dog greater than 1000 Somogyi Units may be attributed to variations in the amount of large size enzyme.

Dog amylase is more labile to low pH than human; chromatography at pH 3.4 which dissociates macroamylase in the human cannot be applied to the dog. Porcine pancreatic amylase is often used to evaluate assay techniques for amylase.

Porcine serum amylase was found distributed in 3 different molecular size peaks by agarose chromatography.

Considerations of an experimental animal to study the contribution of the pancreas to total serum amylase should include both concentration and molecular size of the enzyme.

Neither the dog nor the pig appear to be appropriate animals. Numerous studies have shown that chronic administra tion of pentagastrin leads to parietal cell hyperplasia in the rat. This indicates that ga strin may have a proliferative as well as a secretory effect on gastric mucosa. Primary explants of rat and human gastric mucosa from the oxyntic gland region were cultured in T flasks with medium consisting of 80% Minimum Essential Medicum (Eagle) 10% NCTC 135 , 10% fetal calf serum (rat mucosa) or 10% human cord serum (human mucosa). The cultures were not disturbed until outgrowth of fibroblasts and epithelial cells occurred. This latent period lasted approximately seven days. Either pentagastrin (to a concentration of 0.5 ~g/ml) or an equal volume of saline was then added daily to the cultures and subsequent subcultures. Control (saline) cultures showed rapid growth of fibroblasts and ultimate loss of epithelial cells. Fibroblastic outgrowth was inhibited and epithelial cell proliferation stimulated in the cultures receiving pentagastrin. Light microscopy showed a typical mosaic epithelial pattern. Electron microscopic observations have defined a primitive epithelial populace presenting as a composite of the anatomic features seen in the differentiated gastric mucosal cell flora. These results imply that gastrin has a direct proliferative effect on gastric mucosal epithelium. Established features of Peutz Jeghers Syndrome (PJS) are a Mendel i an dominant i~ heritance , mucocutaneous melanotic pigmentat ion and G. I. polyposis. In 1970 one case led to a large affected clan in Costa Rica . Its founder died in 1902. In whole , 5 generations and II fami I ies became affected. The following groups outline :

(1) Ten dead cases who according to brothers or sons had oral pigmentation; three died-;Tth acute abdominal conditions; but no medical evidence is available.

(2) Three well documented cases died of PJS complications. Two females, 21 and 22 year -old , died of peritonitis after surgery for intestinal invagination . Resected ava i lable specimens show typical PJS polyps. A 45 year-old male died of mal i gnant degeneration of an intestinal polyp; the autopsy showed other polyps and metastasis to lymph nodes and liver. (3) Eighteen I iving cases with PJS, ranging from 2%; to 84 years of age, all with mucocutaneous pigmentation that in one family has a diffu se freckle-l ike facial distribution . Nine cases have pigmentation in hands and/ or feet . One case has viti ligo. Polyps have been demonstrated in 14 cases (gastric in 4, intestinal in 10 , colonic in 9), are absent in a 4 year-old boy and in 3 the radiologic exam i nation has not been completed. The four elder cases had had 8 former emergency procedures for intestinal i nvagination and/or occlusion . Gross G. I. bleeding has occurred in 4 cases. Five cases have had recent laparatomies and polipectomy , after bleeding, invagination or as complementary procedure (after Cesarean operation in one case with extensive appendix involvement). (4) Two normal girls , 2 and 4 month-old, daughters of affected mothers, are potential-cases as we have seen the oral pigmentation to appear as late as 2%; years of age. High fertil i ty and widespread medical care in Costa Rica have combined to produce what seems to be the largest PJS family reported. Although many clinicians advocate the use of fresh blood as replacement therapy in cirrhotic patients with gastrointestinal bleeding, the clinical evidence to support this approach is scanty. Obtaining fresh blood (FB) « 8 hours old) puts a strain on blood banks. Therefore, a prospective study was designed in which all bleeding patients were randomly assigned to one of three replacement schedules. 1) all FB (Group 1), 2) 2 units of bank blood (BB) for each unit of FB (Group 2), 3) 2 units of BB for each unit of a FB substitute consisting of 1 unit packed cells, 1 unit fresh frozen plasma and 2 units of platelet concentrate (Group 3). The site of bleeding was determined by a combination of endoscopy and radiologic evaluation. The patients were stratified according to the site of bleeding into 4 groups for the purpose of randomization: 1) bleeding esophageal varices, 2) hemorrhagic gastritis, 3) diffuse gastrointestinal oozing and 4) miscellaneous causes, including gastric and duodenal ulcers. Thirty-three patients with biopsy proven Laennec's cirrhosis were admitted to the study. There were 11, 12 and 10 patients in the 3 groups respectively. There were no significant differences in age, sex, clinical findings or liver function tests on admission to the study. Group 3 required more blood (x 7.9 units) and had a longer duration of bleeding (x 6.1 days) than either Group 1 or 2 (P<.05); there was no difference between the latter groups. The mean platelet and fibrinogen values in the first 7 days of observation were significantly lower in Group 3 as compared with either Group 1 or 2 (P<.OOl). The incidence of consumption coagulopathy, hepatic coma and the mortality rate was similar for the 3 groups. The data indicate that there may be a significant therapeutic advantage in the transfusion of at least some fresh blood in actively bleeding cirrhotics. The early use of endoscopy to determine the site of bleeding in patients (PTS) with L.C. and recent ethanol consumption has led to some disagreement about the frequency of esophageal variceal (EV) hemorrhage in this group. To assess this, a prospective study was planned in which endoscopy would be performed within 12 hours of the clinically recognized onset of bleeding. 30 PTS with histologically proven L.C. and a recent episode of ethanol consumption were admitted to the study. All PTS had esophagogastroscopy, followed by upper G.I. series. In 25/30 cases (85%), the endoscopic diagnosis was made by the same observer (S.M.). 20 PTS were bleeding on admission and the remaining 10 had the onset of hemorrhage during hospitalization. The following table summarizes the data, dividing the PTS into 2 groups: those PTS felt to be bleeding primarily from EV and those who were bleeding from another source (A.S. In 19/30 PTS (63%) EV was the major source of hemorrhage. Among the 20 PTS who had been recently consuming alcohol and were found to be bleeding on admission, 55% were bleeding from EV while only 35% had HG as the major source of bleeding. In 9 of 11 PTS in whom the source of bleeding was AS, non-bleeding EV were visualized. The presence of EV was confirmed in 93% of all cases by radiologic examination. The PTS with EV had a longer duration of bleeding (4.73 ± 0.78 days) than the PTS with AS (2.36 ± 0.47 days) (P<.Ol), as determined by clear aspiration from a nasogastric tube. The initial hematocrit and the transfusion requirement were similar for the 2 groups. There were no differences in age, sex, or initial liver function tests. In conclusion this study indicates (1) the most frequent site of bleeding in PTS with L.C. and recent ethanol ingestion is EV, (2) the incidence of HG is lower than has been reported by others, (3) bleeding from other sources offers a far better prognosis (9% mortality) than bleeding from EV (52.6% mortality). While rabbit ileum has been widely studied, there has been no overall model of electrolyte and H 2 0 transport. We have developed several predictive models, including single, double and quadruple ion-exchanges during both net absorption and secretion, and tested them in 8 rabbits by 5 hr perfusions containing (mM): were anerobically recirculated through isolated ileal segments and a Plexiglas reservoir at 37C. Samples were removed at 30 min. intervals and net fluxes (~) calculated for each period (n=76) with PSP volume corrections (mean PSP recovery: 95.2 ± 3.0(SE)%). Initial mean mucosal PD was -2.4 ± 0.5mv and thereafter remained in the -1 to -2mv range. ~Na and ~Cl (~moles/cm/30') were linearly related to ~H20 by the regressions: y=0.36 + 0.157x (r=0.90) and y=5.18 + O.133x (r=0.85), indicating isosmotic Na absorption (-) and secretion (+). Perfusate pHs ranged from 7.60 to 8.13 (never less than blood) while calculated pC02s ranged from 8 to 37 mm Hg (never exceeding blood). These data are incompatible with H/Na exchange, and greatly simplify predictive models for possible Cl/HC0 3 exchange. Thus, ~Cl should decrease 1:1 with ~Na during decreasing Na absorption (at ~Na=O, ~Cl=O), then increase 1:1 with ~Na during net Na secretion. An inverse relationship should obtain for HC0 3 , so that the ratio: ~HC03/~Cl should remain unity throughout Na absorption and secretion. These predictions were not fulfilled. While ~Cl was linearly related to ~Na during net Na absorption, mean ~Cl remained constant at -2.6 vmoles/cm/ 30' during net Na secretion. Also, while ~HC03 was linearly related to ~Na during net Na secretion, mean ~HC03 remained constant at +3.5 vmoles/cm/30' during net Na absorption. Net HC03 absorption and net Cl secretion were never observed. The ratio: ~HC03/~Cl varied curvilinearly from -0.2 to -6.2 with Na absorption and secretion, respectively. These findings appear incompatible with significant Cl/HC03 exchange. We propose that NaCl absorption and NaHC03 secretion occur at different sites, perhaps different cells. The absence of obvious anion or cation exchange suggests that the latter is species-related, and raises phylogenetic questions regarding the biochemical significance of ion-exchange, particularly in response to hormones and cholera toxin. The importance of liquid junction potentials (EJ) in gastric PD measurements merits emphasis. For example, using 150 mM NaCl as mucosal bridge, the change in E J as [H+] changes from 0 to 150 mM is about 33 mv, i.e. a change in mucosal PD from -40 to -7 mv could be entirely due to variation in EJ. This study is the first attempt to provide detailed estimates and measurements of E J in complex solutions over the concentration ranges found in gastric juice. We have devised a simple in vitro system for indirect EJ measurement. Two balanced Ag/AgCl electrodes (without:Liquid junction) were mounted in Plexiglas half-chambers separated by a 0.08 cm 2 window sealed with a Viscose membrane. Solutions were mixed magnetically and potentials recorded to 0.1 mv at 15 sec. intervals and at 25±loC with an Orion Research digital electrometer. Observed potentials (ET) represented the sum of E' (transmembrane Cl-activity potential) and E J (membrane diffusion potential Mechanism of gastric mucosal stress injury was studied in 50 combat soldiers admitted for management of the brain, thoracic and abdominal wounds to the 24th Evacuation Hospital, Long Binh, Vietnam. Ten, had gross bleeding from acute erosions or ulcers of stomach or duodenum, while 26 oozed 5 to 60 ml blood per 1 L gastric secretion during the first post-trauma week. 24 hour gastric collections were obtained on ice for a 4 day period, on the average, in each patient during this week, measured for volume and total acid, filtered, centrifuged, dialyzed, lyophilized and sent by air to New York, where they were assayed for heme, bile pigments, proteins, sulfated and non-sulfated glycoproteins, fucose, hexosamine, hexoses and sialic acid. Blood content, 24 hours outputs of each component, various variables and means with st. dev. were obtained from a computer. The mean volume of gastric juice was markedly depressed on day 1 post-trauma (340±55 ml) but gradually increased to 930±1 30 ml on day 5. The 24 hour acid output rose in bleeders from 22±7 on day 1 to 134±26 on day 5, and in non-bleeders from 21±8 to 92±3 3 meq. The 24 hour mean total protein output rose from 1.6 on day 1 to 5.2 g on day 3 in bleeders, and from 0.6 on day 1 to 2.0 g on day 5 in nonbleeders. Serum proteins in bleeders were maintained at a high level of 2.1±1.0 to 3.0 ±1.3 g on days 1-4 post-trauma, while blood admixture was here responsible for serum protein rise of 0.6 g/24 hours at the most. The 24 hour mucosubstances output markedly increased in bleeders on days 3-4: non-sulfated glycoproteins from 440±170 to l670±1035, sulfated glycoproteins from 11±5 to 36±26, fucose from 40±17 to 150±9 3 ,hexosamine from 96±40 to 258, hexoses from 136±68 to 305±165, and sialic acid from 17±8 t o 102 61 mg%. In non-bleeders this increase was much less and peaked on days 5-6 pom -trauma. The initial depression of acid output, followed by serum protein leak and g& stric hemorrhage or blood oozing was consistent with the concept of increased back diffusion of H+ ions intJ gastric mucosa after trauma. This, and bile reflux, seen in most of our patients would result in mucosal injury proven histologically in these cases by Lev and Stremple(in Press). The sharp increase in mucosubstances output in bleeders 3-4 days post trauma reflect s discharge of mucus from the damaged gastric mucosa. The finding that subdeterminants d and y of hepatitis B antigen (HBAg) are mutually exclusive raises the question whether subtypes d+ and y+ afford cross-immunity. Prospective study of hemodialysis patients at Los Angeles County-USC Medical Center provides pertinent observations. Seventeen such patients who acquired y+ antigenemia as a probable result of indirect intra-unit transmission required 392 units of HBAg -unscreened blood during a period of 6 months or more after becoming HBAg-positive, and were then monitored for 6 additional months. There was non-recurrence of antigenemia detectable by countere lectrophoresis in 4 patients (who received 62 units)j reappearance of only the y+ subtype in 6 patients (162 units) with intermittent antigenemiaj and persistence of the y+ subtype in 7 patients (168 units) with chronic antigenemia. An unrelated study (Redeker and co-workers, unpubl ished data) indicates an expected rate of 7.3 d+ infections per 1000 units from the same sources. Therefore the lack of d+ infections among our hemodialysis patients with prior y+ infections suggests protection.

If this immunity is reciprocal, as seems I ikely, then the occurrence of more than two apparently separate episodes of acute viral hepatitis remains unexplained on the basis of recognized types and subtypes. Since hypoxia and anemia are features of clinical stress syndromes that precipitate acute gastric lesions, we have studied the effects of hypoxia and acute anemia on the stomachs of rats subjected to wire mesh restraint stress for seven hours. Experimental rats were exposed to 10% oxygen-90% nitrogen for 7, 14 and 50 hours while paired control rats were kept in similar chambers with room air to breath. Awake arterial p02 in 15 rats breathing 10% oxygen for two hours was 36.1 ± 3.8(S.D.) mm Hg. Anemia was produced by a double 3.0 ml exchange of blood for plasmalyte solution which reduced the hematocrit from 42.5±5.2 to l8.l±6.8%. Hypoxia did not influence the number of gastric erosions or ulcers: control vs hypoxia: 7Hours 7.6 vs 8.0 erosions/stomach, p>.8 and 0.4 vs 0.7 ulcers/stomach. p>.4, 14 hours 8.8 vs 9.9, p>.5 and 0.3 vs 0.7, p>.5 and 50 hours 12.1 vs 12.3, p>.9 and 0.3 vs 0.0, p>.3. When epinephrine was given 0.4 mg/kg intraperitoneally prior to 7 hours of restraint and hypoxia, the number of gastric lesions was unchanged. When the dose was increased to 1.0 mg/kg, there was a significant increase in gastric erosions for the hypoxic rats compared to control rats both with(P<.OOI) and without(p<.OI) epinephrine, but the incidence of ulcers was unchanged. Acute anemia reduced the number of gastric erosions (5.7 vs 2.5, p<.02) but not ulcers. When anemia was combined with 7 hours of hypoxia, the number of restraint induced gastric erosions was still decreased (9.6 vs 4.0, p<.02). Decreased arterial partial pressure of oxygen alone or in combination with reduced red cell mass did not adversely affect stress ulcer formation in the rodent restraint model. An action of epinephrine on gastric blood flow together with decreased arterial partial pressure of oxygen were necessary to precipitate stress gastric lesions. The effects of bilateral nephrectomy and metabolic acidosis on the stomach of the rat subjected to restraint stress were studied to determine if these factors would influence stress ulcer formation independent of the trauma and infection that usually accompanies acute renal failure(ARF) in the patient with previously normal kidneys. Since the rats died between 40 and 54 hours after bilateral nephrectomy, studies were done at 24 hours when the arterial pH was 7.30±0.08(S.D.), urea 107±13 mg% and potassium 5.3±0.4 mEq/L. There was no increased incidence of gastric erosions or ulcers with either 7 (P>.05, >.1) or 12 hours (p>.8, >.4) of restraint . At 36 hours after ARF, arterial pH was 7.25±0.08 in the restrained rat and at 48 hours, the pH was 7.22 ±0.05. There was no change in gastric luminal p02 equilibration but there was a marked increase in 4 hour gastric acid production in the unrestrained (25±17 vs 206±100 pEq, p<.OOl) and restrained (139±93 vs 492±297~Eq, p<.001) rat with ARF. The increas -ed gastric acid production in the restrained nephrectomized rat was reduced by prior vagotomy and pyloroplasty (30l±94 vs l25±97, p<.OOl). ARF increased gastric tissue water (78.9±0.8 vs 80.5±0.5, p<.001). Five days after drinking a 1.6% ammonium chloride solution, arterial pH was 7.l9±0.11. There was a significant increase in the number of gastric erosions (P<.001) but not ulcers (P>.05) with restraint. Metabolic acidosis had no effect on gastric acid production or luminal p02 equilibration in either the unrestrained or restrained rat. With metabolic acidosis, gastric tissue water increased (79.8±1.1 vs 80.6±1.1, p<.02). Other factors (e.g. sepsis) present in patients with post-traumatic renal failure may be of greater etiological significance in stress ulcer formation. Since metabolic acidosis also occurred with ARF, some factor present with ARF may have mollified the effect of acidosis or a longer period of acidosis may have been needed to increase the incidence of gastric lesions.

Nakayama and E. Weser. The University of Texas Medical School at San Antonio, San Antonio, Texas.

After small bowel resection in the rat, mucosal hyperplasia and an increase in 789 cell turnover occurs in remaining intestine. Cellular proliferation and nucleic acid synthesis is limited to intestinal crypt cells. Pyrimidine biosynthetic enzyme activity is primarily localized to the crypts and should parallel the rate of cell proliferation. Male Sprague-Dawley rats underwent resection of 50 cm of proximal or distal intestine or sham operation. One and 6 mos. after surgery, the intestinal remnants were removed. Homogenates were prepared from whole mucosa and the crypt layer after removing villi with an intestinal planing apparatus. The following enzymes were assayed: aspartate transcarbamylase (ATC), dihydroorotase (DHO), and uridine kinase (UK). Specific enzyme activity was greater in normal proximal than distal mucosa. One month after proximal resection, total enzyme activity per cm distal remnant increased significantly over controls as follows: ATC 47%, DHO 39%, UK o29%. After distal resection, increases in the proximal remnant were: ATC 28%, DHO 22%, UK 26%. Specific enzyme activity of the crypt layer (umoles/min/g protein) follows:

Proximal shams (5) Proximal remnant (6) ATC 9.0 ± 0.4 10.7 ± 0.4 DHO 2.2 ± 0.1 2.7 ± 0.1 UK 3.8 ± 0.2 4.9 ± 0.2 All three enzymes were increased in proximal remnants (p( 0.02). H)-thymidine incorporation into DNA of crypt epithelium of proximal remnants increased from 68 ± 7.1 to 95 ± 6.5 cpm/ug DNA (p(0.05). Conclusion: In controls, proximal> distal gut activity of ATC, DHO and UK suggests more rapid proximal cell turnover. After resection, increased enzymes and incorporation of thymidine into DNA of remaining bowel indicate increased pyrimidine biosynthesis associated with greater cell proliferation. We have previously reported that hepatic encephalopathy with hyperammonemia occurs in germfree (G·F) dogs following portacaval shunt, and that hyperammonemia occurred in both conventional and G·F animals following ingestion of a blood meal. This work has been extended to encompass urea tolerance studies in conventional and G·F Eck fistula dogs. Fasting G·F and conventional Eck fistula dogs were fed a urea meal (40 gms urea dissolved in 200 ml water). A rise in bloOd ammonia averaging 416% above control values, was Observed 2 to 6 nours following ingestion of urea in conventional animalS. A rise of only 5.1i% in blood ammonia was recorded In G·F animals. The observation of little or no rise in blood ammonia following a urea meal in G·F t:ck fistula dogs while conventionals exhibited a massive increase favors tne concept that the hydrolysiS of urea to ammonia is completely dependent on enzyme systems derived from the bacterial flora of tne gastrointestinal tract. The hyperammonemia observed in G·F Eck fistula aOimals does NOT result from hydrolysis of urea. Ammonia production in the G·F dog probably results from digestive breakdown by non·bacterial enzyme systems of other nitrogen substances such as amino acids. Recent data frOIl this laboratory indicate that consistent decreases in lover esophageal sphincter (IES) pressure follow a fat aeal. Because of the recent obaervations that the horaone ~strin is the pr1aary deterainant of LiS cOllpetence we have studied the relationship of fat inhibition to the known stiaulatory effect of pentastrin on this sphincter. LES pressure was recorded continuously in e~ht Ilormal subjects using an illtused open-tipped systea.

Pen~strin, ia a dose previously showa to produce aax1aal increases ill LES pressure (3 W!,/}qf, suboutaneously), resulted in a aeaa peak pressure iIlcrease of 20.6 * 7.0 .. Hg (* 1 BE) when given to the subjects in the fasting state. The . . e dose produced a aean peak pressure increase of only 8.4 * 1.7 .. Hg when given following a fat aeal (25 al corn oil in 150 al). This increase in pressure is s~n1ticantly less (p(O.OOl) than that obtained. in the tasting state.

In order to invest~te the aechanisa of fat induced. inhibitiOJl of the lES, a dose response curve for the LES response to rapid. intrave.ous injections of pen~strin vas constructed. In the fasting state this ClIrVe vas s~oid in abape with a peak change in IES pressure of 38.0 * 3.7 .. Hg at 0.6 W!,/}qf,. Fifteen ainutes after ingestion of a tat aeal, anotber pen~str1n 40.e reaponse cwrve vas obtained. This curve was also s~oid i. shape "but shoveel a parallel shift to the r~bt. The aaxi-mUll sphincter pressure change of 34.2 * 6.6 .. Hg was not .~nificantly (,>0.6) different froa that obtained. in the tasting state "but occurred at 1.0 -ng/}qf, of pentagastrin.

Tbese stuie. indicate I 1) Fat-induced inhibition of tbe lES most likely acts throW!,h antagonism of ~.trill. 2) The aeehanism of this ~strill antagoni.. would appear to be one of ce.petitive inhibition. The major problem in the diagnosis of the stagnant loop syndrome is to determine whether bacterial overgrowth is the cause of steatorrhea in a patient with multiple pathologies. The purpose of this study was to determine the most useful diagnostic test in this situation. Three groups of subjects were studied , Groups A and B consisted of 15 subjects with an anatomical stagnant loop. The 9 subjects in Group A had steatorrhea attributable to bacterial overgrowth (4 of them had no other cause for steatorrhea; the other 5 had an alternative cause but fecal fat excretion decreased markedly following ampicillin). The 6 subjects in Group B did not have steatorrhea attributable to this cause (3 did not have steatorrhea; the other 3 had steatorrhea, an alternative cause for this, and no decrease in fecal fat excretion following ampicillin). Group C consisted of 12 subjects with no known gastrointestinal disease. Fasting and postprandial samples were obtained by intubation from various levels of the small intestine. Free and conjugated bile acids were measured enzymatically, following separation by thin layer chromatography. Samples were also cultured for aerobic and anaerobic bacteria. The other diagnostic tests were urinary indican excretion and Schilling test plus intrinsic factor. In upper jejunal samples, free bile acids were detected postprandially in 8 out of 9 subjects from Group A, but in none of those from Groups Band C. They were detected in the lower jejunal sample in the 9th sub-Ject in Group A, but also, at a lower concentration ,in 1 subject from Group B and in 2 subjects from Group C. Ileal samples were of no diagnostic value, since high concentrations of free bile acids were detected in normal controls from Group C. Bacteriological studies,urinary indican excretion, and Schilling test failed to show any clear demarkation between subjects from Group A and those from Group B. Conclusions : the screening of postprandial upper jejunal samples for free bile acids was the most useful of these diagnostic tests for assessing the significance of bacterial overgrowth as a cause of steatorrhea. In lower jejunal and ileal samples the presence of free bile ac ids was a normal finding. Rectosphincteric manometric studies were performed on six patients with proven, untreated anal fissures to investigate the associated motor disorder, the relative roles of the internal and external anal sphincters, and the physiologic effects of sphincterotomy, which is empirically accepted as effective therapy for this disorder. Air filled balloons recorded pressures from the rectum, internal anal sphincter and external sphincter via pressure transducers. In each patient the internal anal sphincter response was characterized by an initial normal relaxation, which was followed immediately by a strong contraction persisting for ~o seconds (range 40-70 seconds). The external sphincter response was normal. The overshoot reflects spasm of the internal sphincter and not the deeper bundles of the external sphincter, since it was present even in a patient who had ex~ernal sphincter paralysis due to a spinal cord lesion. Furthermore, Diazepam (Valium ) 10 mg IV abolished the external sphincter response, but had no effect on the internal sphincter overshoot. After a course of medical therapy the overshoot disappeared. It was also absent in one patient who underwent sphincterotomy without fissurectomy. These motility stlldies provide objective evidence of sphincter spasm associated with fissures, and indicate that it is the internal sphincter rather than the external sphincter which is responsible. This explains the good response to internal sphincterotomy with or without fissurectomy, and provides support for the concept of a vicious cycle in the pathogenesis of fissures. Not only does medical healing of fissures relieve associated internal sphincter spasm, but also, internal sphincterotomy or bTllsk operative dilatation induces healing of the fissures by abolishing sphincter spasm. In two patients with previously unexplained constipation, in whom existing fissures were initially missed by end on proctoscopy the diagnosis 1vaS made by the motility study and subsequently confirmed by side view anoscopy. Thus, in addition to providing information concerning the pathophysiology of anal fissures and their treatment, anorectal motility studies offer a diagnostic tool of high sensitivity. I ornla, an lego, a 0 a, I ornla. Liver regeneration is a remarkable phenomenon of fundamental biologic importance, and of clinical relevance in the treatment of hepatic disease. Our development of a technique of liver transplantation in inbred rats, free from the complicating influences of immunologic rejection and Immunosuppression therapy, has made it possible to conduct three detailed s~udies of the qualitative . and quantitative influences of blood supply on regeneration of 30% hepatic Isografts transplanted heterotopically to 286 rats. Liver graft regeneration was evaluated by measurements of changes in wet weight, dry weight, protein content, number of mitoses, a~ nuclea~ DNA synthesis through autoradiographic determination of incorporation of thymidine-H • The first study (62 rats) showed that grafts vascularized by the portal vein alone, or by both the portal vein and hepatic artery more than doubled their initial weight after two weeks, while grafts supplied by the hepatic artery alone, systemic venous blood alone, or a combination of hepatic arterial and systemic venous blood lost weight even when the total afferent blood flow was greater than normal. The second studr (117 rats) showed that liver transplants provided with a portal blood supply developed explosive cel proliferation and regenerated vigorously, whi Ie grafts deprived of portal venous blood fai led to regenerate under all circumstances. (107 rats), designed to evaluate the combined influences on liver regeneration of 7Cf' 1o host hepatectomy and routing the portal blood to either the host liver or graft, showed that the phenomenon of "inter-liver competition" was simply a contest for a finite amount of trophic humoral agent in portal blood. The results of these studies provide definitive evidence for a hepatotrophic portal blood factor (HPBF) that is essential for liver regeneration. The long-term effects of pancreas transplantation on diabetes mellitus, insulin secretion and pancreatic morphology, all matters of potential clinical importance, have not been determined. Previous studies of pancreas transplantation have involved partial pancreatic autografts or allografts between genetically dissimilar animals and, of necessity, have been complicated by graft rejection and the need for immunosuppressive therapy. Our development of a microvascular surgical technique for transplanting the whole pancreas and duodenum in inbred rats has made possible long-term studies of endocrine function and morphology. Diabetes mellitus was induced by alloxan (42 mg!kg) in 103 Lewis strain rats, and was proven by serial blood glucose levels above 200 mg %. The animals were divided into three groups. Group I was made up of diabetic controls, Group" consisted of sham-operated rats, and Group III contained rats with pancreaticoduodenal transplants. All of the rats with pancreas transplants had immediate and permanent reductions in their blood grucose levels to normal, while all of the control and sham-operated rats maintained high blood glucose levels for up to 1 year. Iii striking contrast to the other two groups, the transplanted rats gained weight normally, were active, and had normal fur texture. Serial morphologic studies of the grafts showed no pancreatitis or loss of acini, normal islets of Langerhans with abundant secretory granules. In a second study involving 65 diabetic rats divided into three groups as before, serial blood insulin levels were determined by radioimmunoassay. The animals with pancreas transplants had consistently normal blood insulin levels (mean of 1.87 ng/ml), while the control and sham-operated rats had markedly depressed levels (mean of .72 and .76 ng/ml, respectively). Finally, in 8 transplanted rats which had been relieved of diabetes, the pancreas grafts were removed after 4 to 6 months. Diabetes promptly recurred with blood glucose levels ranging from 382-848 mg. %. It is concluded that heterotopically transplanted pancreas isografts are capable of consistently producing immediate and lasting relief of alloxan-induced diabetes mellitus and its systemic manifestations. Six patients with various gastrointestinal disorders were maintained on parenteral nutrition for a period of 3-4 weeks. Except for one patient, who received 40 kcal/kg, all patients received 60 kcal/kg of body weight and electrolytes plus vitamins. Intravenous protein in the form of 5% Amigen with 15% sorbitol or glucose solutions was given between 0900 and 2100 hr. daily. The remaining 50% of the calories were given as I ntralipid between 2100 and 900 hr. Each patient received an equivalent of 0.25, 0.5 and 1.0 g/kg/day of protein as casein hydrolysate for a period of seven days at each level. Nitrogen balance studies revealed that positive balance was achieved at the infusion level of 1.0 g/kg. This was true with either glucose or sorbitol as the carbohydrate source. Whole blood aminograms were obtained on the seventh day of each infusion level at 0900, 1300 and 2100 hr. A sudden rise has been observed in blood levels of threonine, valine, methionine and tryptophan, which coincides with the positive nitrogen balance achieved with 1.0 g/kg of casein hydrolysate and suggests fulfillment of the amino acid requirements. At the end of the 1.0 g/kg/day intravenous feeding period three patients were fed by mouth for one week 1.0 g/kg/day of milk protein in the form of cottage cheese, while 60 kcal/kg of the non-protein calories were given intravenously. With the oral protein a greater positive nitrogen balance was achieved. We conclude that the daily amount of protein required in the form of casein hydrolysate for optimal parenteral nutrition of adults is approximately 1 g/kg of body weight when an adequate supply of calories is given. A close relationship exists between a rise in blood levels of some essential amino acids and positive nitrogen balance. Oral protein appears to be better utilized than intravenous protein.

14 healthy volunteers were assigned ad random to a control diet (Mead-Johnson -3200 AU) or an elemental diet (Mead-Johnson -3200 AS) (G. Bounous. Ann. Surg •• 166:312. 1967 ) similar in carbohydrate composition but with the protein fraction in hydrolysed form and partial SUbstitution of LCT by MCT. These diets were given exclusively for 12 days. Biochemical. hematological and microbiological data were obtained during the treatment period. Quantitative pancreatic secretion of enzymes was assessed by an established technique of duodenal perfusion before and after the dietary manipulation. Transit time in the intestine was followed by radio-opaque markers. Subjects in both control and elemental diet groups lost weight (AU: 147± 11 to 145:'11 and AS: 148±9 to 145 ±9 Ibs. GGM is a rare inherited disorder described once previously in an adult. G.O. is a 21-year-old man with severe diarrhea since birth. He survived infancy with a sucrose formula and subsequent development was normal . On his usual diet, stools were voluminous (to 4 l/day), acid (pH 4-5) and contained lactate (to 45 mM/l), glucose (to 13 mg/ml) and normal fats. Histology and disaccharidase concentration (lactase 4U/mg, sucrose 4 U/mg) of jejunal mucosa were normal. Tolerance tests confirmed the diagnosis . Diarrhea decreased dramatically by reduction of dietary sugars and with fructose supplements. Jejunal absorption was studied by perfusion, using isotonic sugarelectrolyte (Na 130-140 mEq/l) solutions (Table ; minus = secretion). Glucose was absorbed only down a concentration gradient, glucose and galactose absorptions were similar and less than that of fructose. Sodium and water secretion persisted even when sugar was absorbed. When sucrose and lactose were perfused, hydrolysis was confirmed. SUGAR CONTENT PERFUSATE ABSORPTION (per min) (mg/ml in saline) Na Potential difference of jejunum was normal (0-2 mY, mucosa negative) but was unresponsive to glucose or fructose. Test meal (Lundh); jejunal pH, trypsin, lipase, bile acids and micellar fat (corrected for dilution) within normal limits. Dilution of PEG in meal was excessive (6-7 fold). In GGM, secretion of sodium and water and lack of active glucose absorption suggests failure of coupled transports of sodium and glucose. The results are consistent with the hypothesis that active hexose transport is necessary for sodium absorption. The preoperative identification of obstructed extrahepatic bile ducts in infants with persistent cholestatic jaundice remains a diagnostic challenge. Needle biopsy of the liver, the conventional IRB excretion test, the vitamin E-based erythrocyte stability test and the test for LP-X (JR Poley et al, Gastroenterology 58:983, 1970) have been helpful in identifying severe cholestasis, but may fail to localize the site of obstruction (extra-vs intrahepatic) in the presence of severe intrahepatic cholestasis. We have therefore attempted to modify the IRB excretion test, since it does yield valuable information on hepatic and biliary excretory function. The modification was accomplished by the addition of CH because of its twofold action: it reduces liver injury in patients with patent extrahepatic bile ducts and it binds (mainly electrostatically) most of the IRB excreted into the intestine preventing its enterohepatic cycling. To test the usefulness of CH, 12 infants with severe cholestatic jaundice (total bilirubin > 9 mg%, positive LP-X, < 10% of IRB excreted in 96 hours by conventional test, liver biopsy), were re-studied with an IRB excreti on test after 2 to 3 weeks of oral CH, 4 gm/day. After CH, 8 infants excreted only 3-9% of the injected label in an 96 hour stool, whereas the remaining 4 excreted 19-36% . The difference between the means was significant at the p<O.OOl level. Results in the former group were consistent with an extrahepatic biliary obstruction and surgical exploration was recommended: in each instance an obstructive lesion (atresia etc) was documented. The latter group represented "neonatal hepatitis" and an operative cholangiogram demonstrated patent extrahepatic bile ducts in one; the remaining 3 were not operated upon and all lost their jaundice with continued CH treatment. It is therefore suggested that the IRB excretion test with CH offers an excellent opportunity for the correct preoperative identification of obstructed extrahepatic bile ducts. Determinants of biliary cholesterol (C) excretion are not clearly defined, but some control b~ extrahepatic factors has to be considered. To test the influence of human growth hormone (GH) on gallbladder bile (GB) composition, we have studied 4 GHdeficient patients (3 m, 1 f) and 3 controls with normal GH secretion (2 Turner's syndrome and 1 male with constitutional short stature). Treatment with GH followed the same dosage and schedule as previously described (JR Poley et al, Gastroenterology 60:707,1971) . GB samples were obtained by duodenal drainage following stimulation of GB flow by essential amino acids (VLW Go et al, J Clin Invest 49:1558 , 1970 ). The composition of GB bile obtained by stimulation is virtually identtcal to that obtained by needle aspiration of the gallbladder (ZR Vlahcevic et al, Amer J Dig Dis 16:797, 1971). Total biliary bile acids (BA), phospholipids (PL) and C were analyzed and the results plotted on triangular coordinates. Two GH-deficient patients showing good growth acceleration with GH treatment, progressively oversaturated their bile with C (crystals) making it lithogenic: the BA+PL/C ratios decreased markedly and bile composition moved out of the micellar zone. Cessation of GH treatment after one year was followed by a return of the GB composition to normal in one patient. By contract, 2 GH-deficient patients showing no growth accelerat10n did not develop lithogenic bile. The 2 patients with the Turner's syndrome had lithogenic bile prior to GH treatment. While exogenous GH slightly accentuated oversaturation of GB with C in one, GB composition returned to normal in the other. In the third control, initial excessive oversaturation of GB with C was abolished with continued GH treatment. Return of GB composition into the micellar zone was associated in each instance with a conspicuous rise in the glycine/taurine ratios of BA. The development of lithogenic bile may have to be entertained in GH-deficient patients upon long-term GH treatment. Sodiwn retention is an annoying complication in a small percentage (N15%) of ulcer patients successfully treated with carbenoxolone (Biogastrone) . Clinically, this side effect is manifested as weight gain, hypokalemia and alkalosis, a picture resembling hyperaldosteronism without "escape". Two possible mechanisms responsible for this side effect were evaluated using an in vitro membrane system, i. e . that carbenoxolone possesses inherent mineralo~rticoid properties, or that it potentiates aldosterone . Transepithelial sodiwn flux was measured as short-circuit current (scc) in matched quarters of ventral toad skin. After addition of aldosterone to the skin, scc declines for Ithrs. then rises abruptly to plateau after another ~hrs. The magnitude of this rise, designated ~ scc, was used to quantitate the mineralocorticoid action on sodiwn flux . Addition of 10-4 , 10-5 and 10-6 M carbenoxolone failed to induce a +~scc in this system, therefore, an inherent aldosterone-like action does not account for carbenoxolone's sodiwn-retaining effect . However, when lO-4M carbenoxolone was added either Ithrs. before or simultaneously with 10-7M aldosterone, + ~scc exceeded aldosterone alone by 12:5 p.A (+41%, p<. 05) and 36:15 pA (+12%, p<. 05) respectively. Addition of carbenoxolone lthrs. after aldosterone failed to potentiate. The synergism between carbenoxolone and aldosterone is rather specific since carbenoxolone pretreatment failed to potentiate vasopressin stimulated Na+ flux, nor did prior treatment with 17p-Estradial enhance aldosteroneinduced 6. scc. Two possible explanations for the observed carbenoxolone/aldosterone synergism include : 1) by altering the turnover of aldosterone-induced intermediates, carbenoxolone increases the utilization of metabolic substrate by the sodiwn pwnp, and/or 2) carbenoxolone increases the affinity or actual nwnber of available intracellular aldosterone receptor sites. The purpose of this study was to develop a technique for simultaneous and independent measurement of gastric emptying and small intestinal propulsion in rats and to determine the extent of difference in these parameters between fed and 24-hour fasted rats. Radioactive test solutions, containing Na 2 5lcr0 4 and/or l25I-polyvinylpyrrolidone, were introduced into the stomach by gavage or into a plastic tube which had been surgically placed in the duodenum . The stomach and small intestine were removed 0, 15,30 or 60 minutes later. The amount and distribution of radioactivity was separately determined for each isotope by passing the gastrointestinal tract under a collimated scintillation detector and graphically recording the count rate. When both isotopes were introduced simultaneously at the same site, there were no significant differences in the results. In fasted rats 90% of the isotope was emptied from the stomach in 30 minutes compared to 50% in 60 minutes for fed rats. The gastric emptying rate was ~ponential in fasted rats from 0-30 minutes, and in fed rats from 15-60 minutes. After intragastric administration of test solution both the leading edge and points of half-activity in the small intestine were further down the intestine in the fasted rats than in fed rats. In both groups, the activity moved down the small intestine as a bolus and the rate of movement was similar and decreased exponentially. The variability in results within the fed rats was not greatly different from those within the fasted rats. The double isotope technique offers a means of simultaneously measuring gastric emptying and small intestinal transit.

Based on anatomical dissect10ns to be presented elsewhere, we cons1der the bare area of the liver to be cons1derably larger than the class1c texts refer. It extends far to the lateral aspect than one would imagine read1ng the class1c texts.

Using vertical X-Rays, with centl'al ray tangential to the mid-axillary line, and with the pat1ent on supine position, dur1ng inspiration apnea, it is ease to determine the width ot the bare zone, limited in its cepha11c aspect by the diaphragmat1c shadow and in its caudal aspect by the per1to neal reflex10n. In 20 determinations its average extension was 4.5 cm, in accord with the anatomical determinations. The use ot a metallic marker in the skin identifies easily the intersection ot its midplane with the axillary 11ne.

At this intersect10n, under local anesthesia, using a Menghini needle, liver biopsy was done in 71 patients; 51 anicteric patients were used 8S controls "0 cases ot bronchogenic carcinoma being staged, and a miscellaneous group ot 21 pat1entsl. A group of 20 pat1ents with Jaundice, includ1ng 12 with clin1cal and laboratorial evidence ot cholestasis was studied. In this latter group tha total serua bi11rub1n had a range ot 2.6 to 56 mg per 100 al, and the alkaline phosphatase had a range of12.5 to 75 K.A. units per 100 ml. In no instance there was any local pa1n after biopsy and, as expected, no instance ot bile peritonit1s wal seen. In 5 pat1ents with ascites no l1quid was aspirated during the procedure.

There was an excellent correlation ot the final c11nical d1agnos1s and the histopathological exaa1nations. The 7 cases ot extra-hepat1c obstruction were all identitied as such 1n the biopsy, and in 17 instances of intra-hepatic cholestasis by h1stology, the clinical evolution was 1n aocord. Liver biopsy cytology was done 1n 45 instances and was positive in 5 casea of metastatic 11ver.

In conclus10n, the technique appears to be sate and is recommended in selected cases, specially those with chOlestasis,Jaund1ce or large ascites. The fact the per1toneal cavity is not entered aakes bile periton1tis a risk not involved. Proteolytic enzyme inhibitors are obtainable from the body wall of the nematode, Ascaris lumbricoides. They are low molecular weight crystalline proteins, derived by TCA extraction and ethanol precipitation, which combine stoichiometrically with pepsin, trypsin and chymotrypsin. Compared to a standard legume inhibitor, such as soybean, the Ascaris inhibitors are clearly superior. Soybean inhibitors possess no activity against pepsin, only slight anti-chymotryptic activity and only 1/15 the anti-tryptic activity of Ascaris inhibitor. The Ascaris inhibitors are equally effective against pepsin, trypsin and chymotrypsin. In this study, 3 dogs were each prepared with 4 isolated jejunal loop fistulae. After abdominal irradiation, the fistulae were injected as follows 3 times daily for 4 days: (1) Trypsin and chymotrypsin in physiologic concentration at pH 8~ (2) Normal saline at pH 8 -the control~ (3) Trypsin and chymotrypsin and crude soybean trypsin inhibitor at pH 8~ (4) Trypsin and chymotrypsin and Ascaris lumbricoides trypsin inhibitor at pH 8. At 96 hours, the animals were sacrificed and sections of the various loops taken for histologic study. Results were dramatic. Loop #1, the uninhibited loop, showed severe gross and histologic damage. Loop #3, the soybean protected loop, showed moderate damage.

Both loop #2, the control, and loop #4, the Ascaris protected loop, had minimal gross and histologic damage.

This study suggests that if pancreatic proteolytic enzymes are neutralized by a suitable inhibitor, the intensity of intestinal radiation damage can be markedly reduced. It also suggests the possible clinical use of the Ascaris inhibitors for the amelioration of enteropathies of diverse origin, where enzyme damage may be a contribu- Characterization and quantitation of the precise esophageal motor abnormalities in achalasia has undergone dramatic revision during the past decade. A hypertensive, i ncompletely relaxing, lower esophageal sphincter (LES) has been reported and the significance of the Mecholyl test has been questioned. We have studied 21 achalasia patients (15 untreated) using an infusion (1.6 ml/min) manometric technique and cineradiography. In the 15 untreated patients, LES pressure (25±11 mmHg; S.D.) was m oderately but significantly elevated from that of the 23 control subjects (15±7.0 mmHg)(p=.Ol). The regional profile of LE5 relaxation showed a mean per cent of LES relaxation after swallows of 65±29 mmHg in the distal half of the sphincter compared to 52±24 mmHg in the proximal half. In some instances the LES relaxed completely. LES. response to graded increases in intraabdominal pressure (G) caused by external abdominal compression was determined in 9 untreated and 6 treated patients. In the untreated group, the ~LES/~G ratio was greater than 1.2, whereas, in the untreated group the ratio was less than 0.9 (normal 1 or greater). A peristaltic stripping wave was identified in the proximal 3 to 8 cm (mean 4.7 cm) esophageal segment in 14 of the 21 patients . The upper esophageal sphincter (UES) functioned normally at rest , with swallowing and during abdominal compression. The Mecholyl test was positive in all achalasia patients. It clearly separated achalasia from 15 patients with other esophageal disorders. Mean pressure rise at six minutes after 5.0 mgm of Mecholyl in the achalasia group was 42±30 mmHg compared to 6±4 mmHg for the latter group (p=.Ol). Fluoroscopic monitoring revealed a concurrent lumen obliterating contraction of the distal esophagus during a positive Mecholyl test.

Our study shows 1) LES pressures significantly higher in achalasia patients as compared to normals. The numerical difference is not great, however. 2) partial but occasionall y complete LES relaxation after swallows, 3) maintenance of the LES barrier to esophagogastric reflux, 4) peristalsis in the proximal esophagus, 5) normal UES functiDn, and 6) reaffirmation of the clinical usefulness of the Mecholyl test. The effect of synthetic C-terminal octapeptide of cholecystokinin (OP-CCK) (Squibb Institute for Medical Research, New Brunswick, N.J.) on lower esophageal sphincter (LES) pressure was studied in 7 normal male subjects (mean age 29.6 years). Intraluminal pressure was recorded using 4 perfused polyvinyl tubes with side openings as sensors. The tubes were anchored so that the distal sensor was in the gastric fundus, the next sensor in the LES, and the 2 most proximal sensors in the lower esophagus. After a 10 min basal period, 0. 15 M NaCl (2 ml) was injected intravenously as a control. At 10 min intervals thereafter graded doses of OP-CCK (2.5, 5, 10, 20, and 40 ng/kg) were injected intravenously over 30 sec in the order listed . LES pressure was measured at 30 sec intervals. Mean (± S.E.) basal LES pressure was 14.9 ± 0.7 mm Hg. LES pressure during 1 min before each injection was compared with LES pressure from 0.5 to 5.0 min, and from 5.5 to 10 min, after injection.

Mean Change in LES Pressure (mm Hg) 0.5 to 5 min 5.5 to 10 min NaCl alone +0.5

OP-CCK 2.5 ng/kg -0.7 +2.5 OP-CCK 5 ng/kg -1.7

OP-CCK 10 ng/kg -2.6 -0.1 OP-CCK 20 ng/kg -4.0 -1. 3 OP-CCK 40 ng/kg -4 .6 -2.7 After each dose of OP-CCK the mean of the 2 lowest consecutive LES pressures was significantly (p < 0.05) less than the mean pre-injection pressure. Both magnitude and duration of decrease in LES pressure increased with increasing doses of OP-CCK. It is concluded that OP-CCK decreased LES pressure in man. The intestinal microvasculature has been implicated in the pathogenesis of shock. This possibility was explored in anesthetized dogs and monkeys subjected to endotoxic and hemorrhagic shock using the silicone rubber injection technique. In 4 dogs and 4 monkeys segments of jejunum were prepared 4 hours after the animals had received ~. ~ endotoxin (LDSO, I.V.). The changes in intestinal vascular architecture included the following: 1. villus height was reduced by 5~/', 2. the subepithelial capillary plexus and villus core vessels were coiled, 3. the subepithelial plexus was congested and dilated, 4. there was diffuse extravasation of rubber from the submucosal and core vessels. These changes are compatible with known actions of endotoxin on vascular smooth muscle and suggest an etiologic mechanism to account for the observed increase in mesenteric vascular resistance characteristic of canine endotoxemia. Similar segments of gut were analyzed in 4 dogs after 4 hours of hypovolemic shock and l~ hours following reinfusion of the shed blood. The appearance of the microvascular architecture 4 hours following hemorrhagic shock'was essentially unchanged from control. However, l~ hours following reinfusion the following changes were noted: 1. there was extravasation of rubber into the core of the villi, 2. many vi lli were entire ly fi lIed with rubber, 3. the subepi the lia 1 capi llary plexus was unremarkable. In contrast, the same studies performed in the subhuman primate revealed essentially no change from the control architectural pattern during either endotoxic or hemorrhagic shock. These findings indicate a marked species difference exists between monkey and dog in terms of the intestinal microvascular responses to experimental shock. In the canine model there are also significant differences between responses to endotoxic and hemorrhagic shock. Thus, there is little evidence in support of the thesis that the gut is a common pathway in the pathophysiology of irreversible shock in either dog or subhuman primate and presumably man as well. Generalized darkening of the skin is frequently seen in patients with chronic jaundice. Most often the underlying disorder is primary biliary cirrhosis, bile duct stricture or bile duct carcinoma. The skin darkening is due to increased melanin rather than to deposition of an abnormal pigment. Causative factors could include increased serum bile acid concentration, increased serum lipids or irritation from scratching. Primary importance of the latter is. suggested by the frequent presence in such patients of a butterfly-shaped area of decreased pigmentation over the upper back. The edges of this area correspond to the limits of where the patient can reach to scratch. For example, a much larger area was seen in one patient with concomitant arthritis and limited mobility of the arms. An easily visible butterfly sign was present in 11 of 17 patients with primary biliary cirrhosis and in occasional patients with bile duct stricture due to cholangitis or to bile duct cancer. It was not present in patients with early primary biliary cirrhosis without pruritus. It was most obvious in patients with the combination of primary biliary cirrhosis and scleroderma (Amer. J. Med. 50:302, 1971 ). Determination of trypsin activity of enterokinase-activated pancreatic extracts with two different substrates, Benzoylarginine p-nitroan1lide (BAPNA) and Remazolbrilliant Blue-Hide (RBB-hide) furnishes widely divergent values. Evidence obtained in our laboratories shows that this discrepancy is not due to a new inhibitor as postulated by Haverback et al. (Clin. Res. 19:393, 1971 ), but to the presence of varying proportions of isoenzymes of trypsin with widely differing activities. Electrophoresis of pancreatic extracts from human, bovine, porcine, canine, murine and rat pancreas resulted in each case in the separation of an anionic and a cationic species of trypsinogen. Activation of the zymogens yielded trypsin variants (isoenzymes) of strikingly different specific activities toward the two substrates. For example: with the ratio of BAPNA/RBB-hide readings set at 1.0 for crystalline bovine cationic trypsin, the ratio for bovine anionic trypsin was 20; for human cationic trypsin 2.6, anionic trypsin 5.2; for canine cationic trypsin 10.6 and anionic trypsin 4.2. Some of these findings were confirmed by electrofocusing techniques. No a2-macroglobulin was present in any of the fractions tested. These data demonstrate that cationic and anionic trypsins of the same species possess different specific activities for different substrates and that differences also exist between the activities of similarly charged trypsins of different species. Our results dramatically illustrate the difficulties in. interpreting measurements of enzyme activity in biological fluids and the dependence of such measurements on the type of standard and substrate employed. In order to clarify some mechanisms of the diarrhea frequently found in binge drinking alcoholics, the effect of acute administration of ethanol on intestinal motility was investigated in chronic alcoholics. Sixteen studies were performed, 11 in the jejunum and 5 in the ileum, using a polyethylene tube with a flaccid distal balloon placed under fluoroscopic control in the first jejunal loop or in the ileum at 270 cm from the incisors. After a 45 minute baseline recording, 0.8 g/Kg of 25% ethanol was administered either orally(s jejunal and all ileal studies) or intravenously (3 jejunal studies), and a continuous recording was obtained for 3 more hours. For comparison, jejunal motility was measured in two additional patients after oral urea or amino acids in concentrations isosmolar to the ethanol. After oral ethanol, significant decreases « 0.05 by analysis of variance for overall effects and the Newman-Keuls test for simple effects) in the frequency, amplitude, and motility index (product of frequency, amplitude, and duration) of type I waves occurred in jejunum within 30 minutes, with maximum suppression after 1 hour. With intravenous ethanol similar inhibition occurred in 15 minutes with maximum response at 2 hours. In both, effects were persistent at 3 hours. There was an inverse correlation between blood alcohol levels and motility index of type I waves. Type III waves remained unchanged. No significant changes were found in the ileum. Intragastric urea or amino acids produced immediate decrease (within one minute) in type I and III waves with maximum effect at 15 minutes and recovery toward normal by one hour. The motor inhibition produced by ethanol seems to represent a specific action rather than associated hypertonicity since: the time course for ethanol change was different from that of solutions of similar tonicity; ethanol in contrast to the other solutions inhibited type I waves only; motility was inhibited by I.V. ethanol. The selective inhibition of type I waves following ethanol ingestion suggests a mechanism for the diarrhea associated with acute alcohol binges, since type I (segmenting) waves serve to impede the forward progress of intraluminal contents, and are generally depressed in various types of diarrhea.

Recent studies on small bowel 1 ipid reesterifying enzymes that are involved with the process of lipid absorption and transport suggested that enzyme levels are related to the load of fat presented for absorption. To further investigate this acyl-CoA: monoglyceride acyl transferase (MG-acyltrans.) was assayed using microsomes prepared from the jejunum and ileum from rats fed one of 3 diets for 4 weeks: Gp 1, regular chow 4% fat; Gp 2, regular chow with added lard to 20% fat; Gp 3, synthetic liquid diet with 20% fat (lard). Results in nmoles/min/sample + were: Enzyme activity was greater in the rats fed a high fat diet but the site of increased activity in the small bowel appeared to be influenced by the consistency of the diet. Next the effect of the type of long chain fatty acids (FA) in the diet (saturated vs. unsaturated) was studied on jejunal microsomal samples. In addition to the above enzyme, palmitic acid-CoA 1 igase (P-CoA 1 ig.) and oleic acid-CoA 1 igase (O-CoA 1 ig.)

were assayed. Rats were studied after consuming one of the two synthetic liquid diets for 3 weeks. Both diets contained 10% fat with 1 ipid supplied by corn oil in one (13 . 8% saturated FA) and by lard in the other (43.5% saturated FA). Total jejunal enzyme activities + SD expressed as nmoles/min/sample were as follows : Thirty-eight patients with active, arterio-capillary gastrointestinal bleeding were treated by the selective arterial infusion of epinephrine or vasopressin. Vasoconstrictive therapy had a success rate of 80% in small-vessel bleeding. Complete control of bleeding was achieved in twenty, temporary control in three and no control in two of twenty-five patients with gastritis, superficial stress ulcers, the Mallory-Weiss syndrome or colonic lesions. Selective vasoconstriction had limited value in the management of large-artery bleeding from peptic ulcers, affording complete control in two (15%) temporary control in two and no useful control in nine of thirteen patients. Individual factors of importance in the conduct of treatment include: the patient's immediate viability, angiographic demonstration of the bleeding, the character and location of the bleeding lesion, the obtainable degree of infusion selectivity, the choice of vasoconstrictor and vascular response achieved, and the patient's clotting mechanism. It is believed that in appropriately chosen patients, selective vasoconstrictor therapy offers an improvement in the management of active arterio-capillary gastrointestinal hemorrhage. Because of confusing data in studies of the autonomic regulation of pancreatic secretion, we investigated the effects of specific autonomic stimulants and blocking agents. In dogs prepared with duodenal (Thomas cannula) and gastric fistulas, steady states of pancreatic secretion were obtained in response to secretin (0.03 u/kg/min.) or secretin plus pancreozymin (0.05 u/kg/min.). We studied (1) vagal stimulation -insulin 0.2 u/kg Lv.; (2) parasympatholytic --atropine 0.4 mg/kg Lv.; (3) alphaadrenergic --epinephrine 0.025 pg/kg/min.; (4) alpha-adrenolytic --phentolamine 0.5 mg/kg; (5) beta-adrenergic --isoproterenol 0.25 pg/kg/min.; (6) beta-adrenolytic -propanolol 0 ,.5 mg/kg 1. v.

Experiments were repeated after antrectomy of the animals. Vagal stimulation (insulin) stimulated enzyme output (800%) without affecting volume or HC03-; this effect was blocked by atropine. Atropine in the absence of vagal stimulation, inhibited pancreozymin-stimulated enzyme output without altering volume or HC03-. Epinephrine inhibited volume (50%) but did not change HC03-or enzyme concentration. Phentolamine produced a transient increase in enzyme output. Isoproterenol inhibited volume, HC03-and enzyme concentration profoundly but propanolol had no effect. Antrectomy markedly reduced the vagally mediated effect of insulin on enzyme output but had no influence on the atropine effect. Thus vagal stimulation primarily stimulates enzyme secretion, an effect dependent on the vagal release of antral gastrin. Atropine thus has a vagolytic effect, and inhibits the permissive effect of acetylcholine on the pancreas, independent of the antrum. There may also be a direct andrenergic effect predominantly on the enzyme fraction of pancreatic secretion. To further evaluate the role of gastrone (G), rabbits were immunized with gastrone B separated from pooled anacid human gastric juices (PAHGJ) as fraction (fr) 7 by gel filtration on Sephadex G-lOO. The anti-gastrone serum (AGS), formed a single precipitin arc by gel diffusion against 8 different PAHGJ as well as fr's 6 and 7 from several Sephadex PAHGJ fractionations, but not with 3 pools of acid human gastric juices (one reutralized in vivo) or Sephadex fr's 1-4 not containing ~ No precipitin reactions occurred between the most potent antisera and two normal human sera or Cohn fr II. The specificity of de AGS was confirmed by its ability to remove gastric inhibitory activity from G soluti~ An optimum reaction mixture of G and AGS was determined by a quantitative precipitin test. Accordingly, supernatants from the reaction products of 50 ~g G and 0.15 ml AGS after incubation overnight at 4°C and centrifugation were injected i.v. to chronic fistula rats. The effect on volume and acid output was compared with that of G mixed with normal rabbit serum treated similarly,as well as G alone and saline controls, in a total of 85 assays. The use of the chronic gastric fistula rat permitted comparison of individual rats in various experimental conditions. The assay of AGS treated G demonstrated more than 2 times greater acid secretion than that with G alone ~<O.Ol)and was similar to that seen in saline controls. This removal of G activity by AGS confirms antibody's biological specificity. Using AGS and fluoresceinated anti-rabbit IgG in indirect Coons' tests, the presence of immuno-reactive G was detected by immunofluorescence in the surface epithelium and pyloric glands of the antral human gastric mucosa and in the pykric and duodenal bulb glands of the dog. These localizations probably reflect the cellular derivations of G in these two species. Similar studies in the rat were inconclusive.Human and dog fundal gastric mucosae were negative for immuno-reactive G, as were the jmmunofluorescence tests using normal rabbit serum or saline controls. Thus, an in vitro method for gastrone B detection has been developed and the cellular derivation of gastrone B from antral epithelial cells of man and dog, and the glands of the canine duodenal bulb has been demonstrated. Anti-gastrone antibody may now be used to extend the study of gastrone's physiological role in gastric secretory mechanisms. Human jejunal glycolytic enzyme activities (GEA) adapt to dietary sugars and folic acid. These GEA (hexokinase, fructokinase, fructose-I-phosphate and fructose-l,6diphosphate aldolases, and pyruvate kinase) are higher on high carbohydrate diets than on carbohydrate-free diets. The role of this adaptation in gastrointestinal physiology is unknown. If adaptation is important physiologically, then failure of adaptation should be associated with gastrointestinal disease. We studied the adaptive response of jejunal GEA in 2 groups of patients: 1) 8 patients with a damaged intestinal mucos~ (celiac disease, tropical sprue, etc.), and 2) 14 patients with apparent "functional" gastrointestinal disorders (cramps, gas, lightheadedness, bloating, or diarrhea and negative gastroenterologic evaluations). In group 1, GEA were very low and there was loss of the adaptive response to dietary sugars. There was a prompt response to folic acid in tropical sprue but not in celiac disease. In tropical sprue the adaptive response to sugars returned toward normal after therapy. In group 2, 12 patients failed to adapt to all sugars tested while 2 brothers failed to adapt to glucose only. Restriction of dietary carbohydrate produced long term improvement (>1 year) in some of the patients, but only temporary improvement «6 months) in others. These studies demonstrate that a failure of jejunal GEA to adapt to dietary sugars (maladaptation) is associated with gastrointestinal disease. In additional studies, 3 normal subjects were fed a carbohydrate-free diet for 3 days to produce a non-adapted state. When challenged acutely with 100 gm of oral glucose each subject experienced cramps, bloating, and lightheadedness within a few hours. Sinceit takes 6-24 hours for an adaptive response, symptoms occurring in the first few hours after sugar ingestion are a physiological maladaptation. This model in normals, then, reproduces the symptoms of patients with failure to adapt. Therefore, we propose an "intestinal maladaptation syndrome" as a clinical entity; primary maladaptation in patients with a normal mucosa, secondary maladaptation in patients with a damaged mucosa. Cholic acid increases the turnover of intestinal cells in germ free mice (P.S.E. B.M. 138: 270, 1970) . HaVing recently shown that the in-vitro jejunal transport of 3-0-methyl glucose is increased in bile fistula rats (Nature 225: 1055 (Nature 225: ,1970 and decreased in bile ligated animals, it was of interest to see if these absorptive changes could relate, in some way, to altera 3 ions in mucosal cell proliferation and loss. One hour after the I.V. injection of H-thymidine (luCi per g. body weight) in Sprague-Dawley rats operated on 48 hours previously, the small bowel from the ligament of Treitz to the ileocecal valve was rapidly removed and its mucosa scraped off. The animals could not be distinguished on the basis of initial body weight, % weight loss, intestinal length, mucosal weight, protein and DNA content. However, the DNA S.A. (CPM X 10 3 /mg DNA) was significantly decreased (p < .01) in the 10 bile ligated animals (260±74) and in the 10 bile fistula rats (27l±73) when compared to the 11 sham animals (3l3±73). Cell loss was evaluated in 3 groups of 8 rats. Normal saline was infused for 2 hours at a rate of 2 ml/min. through polyethylene tubing tied into the jejunum at the duodenojejunal flexure; the washings were collected from the cannulated distal ileum. There was no difference between the initial weight of the animals, the % weight loss during the 48 postoperative hours preceding the study of DNA exfoliation by intestinal washing and between the dry weight of the small bowels following the study. The perfusate DNA content (ug) was significantly lower (p < .01) in the bile fistula (94.l±20.7) than in the bile ligated (146.9±13.2) and in the sham animals (143.0±11.2). The decrease in DNA specific activity in bile fistula and ligated rats is interpreted in terms of a reduction of the proliferative cell pool induced by interruption of the enterohepatic cycle of bile acids. The disproportionately large shedding of cells in relation to the proliferative activity in bile ligated animals may affect transport. Gastrointestinal disturbances, including malabsorption. have often been reported in chronic alcoholics. although the small intestinal mucosa has generally been considered normal by light microscopy. We therefore investigated in rats the effect of ethanol on the electron microscopic (EM) appearance of the mucosa of the small intestine. Rats were fed nutritionally adequate liquid diets containing 36% of total calories as ethanol (or isocaloric carbohydrate) for 6-12 months. By light microscopy the intestinal mucosa appeared normal. In rats fed ethanol. EM examination of jejunal and ileal villi showed an increase in the endoplasmic reticulum. which in many cells was dilated and vesicular. Focal rarefaction of the matrix and occasional myelin figures were present in some mitochondria. Autophagic vacuoles and lysosomes were increased. The cisternae of the Golgi apparatus were often dilated. Crypt cells appeared normal. These results in rats prompted us to attempt to verify the effect of ethanol on the small intestine in man. &nall intestinal biopsies were obtained in volunteers before and after 23-64 days of ethanol administration (46-60% of total calories) with adequate diets. No alterations were seen by light microscopy. EM changes similar to those in rats were noted in volunteers. except that mitochondrial abnormalities were more conspicuous. In many cells mitochondria were irregular in size and shape. and often contained dense osmiophilic material. Occasionally crypt cells displayed irregular cytoplasmic blebs which extended into the lumen. Biopsy specimens obtained 3 weeks after discontinuation of ethanol showed improvement towards normal. The data indicate that chronic ethanol ingestion produces ultrastructural changes in mucosal cells of the small intestine in the presence of an adequate diet. These changes are similar to those induced by ethanol in the liver and since they were present in the ileum they do not merely result from a high intraluminal ethanol concentration. The structural alterations shown in this study may explain. in part. the impairment of gastrointestinal function in chronic alcoholics. Plasma obtained from patients with chronic inflammatory bowel disease have been tested for the presence of circulating Carcinoembryonic Antigen (CEA) by a radioiinununoassay. Levels greater than or equal to 2.5 ng CEA/ml. were positive assa:/s. The random error in this series of 168 plasma was ± 0.2 ng CEA/ml. In a series of 74 patients with granulomatous disease, 29.8% were positive; those with ileocolitis showed the highest positivity in this group (41.5%) . Of 52 patients with ulcerative disease,30.8% showed circulating CEA; those with ulcerative proctitis showing ~e lowest positivity (14.3%). In general the levels of CEA were proportional to both the extent of disease throughout the gut and disease-activity levels. Although only 3 patients with polyps were tested, one was CEA-positive. All of our 39 ambulatory healthy controls were negative. CEA-levels in all nonactive or normal controls remained fairly constant. Rising levels of CEA not associated with inflammatory flare-ups may thus possibly be diagnostic for cancer. We have studied the in vitro responsiveness of lymphocytes to phytohemagglutinin-M (PHA) in 37 patients with Crohn's disease (CD), 12 patients with ulcerative colitis (UC), and 16 normal control subjects (NC). The patients manifested widely varying stages of disease activity, but none was receiving steroid, ACTH, or immunosuppressive therapy . Separately studied were 14 patients with CD and UC who were receiving daily steroid therapy (ST) with up to 40 mg of prednisone. FHA was added to triplicate lymphocyte cultures in 7 stepwise concentrations from 20 to 2000 pgm/ml, and lymphocyte responsiveness was determined at 3 and 5 days by measuring cellular incorporation of l25I-deoxyuridine (IUdR). All groups of subjects were comparable in percent incorporation of IUdR in unstimulated cultures (0.2-0.3%), in time required to achieve maximal stimulation (3 days), and in the concentration of FHA which elicited optimal responses (500 pgm/ml) . However, the groups differed in the magnitude of their peak responses . Median peak incorporation was 8.2% for NC, but only 4.1% for UC, 4.2% for CD, and 2.0% for ST patients. The distribution of maximal lymphocyte responsiveness in each group was bimodal, with 79% of all peak responses falling into either a low range (0-3.9%) or a high range (7-22%). The low range comprised only 13% of NC subjects, but included 42% of UC, 48% of CD, and 64% of ST patients. Conversely, the high range comprised 64% of NC, but only 25% of UC, 33% of CD, and 21% of ST patients. In repeat studies on 16 subjects, lymphocyte responsiveness varied on different days and did not correlate with clinical status nor with absolute lymphocyte counts. Sera from 10 CD patients with hyporesponsive lymphocytes did not suppress PHA-stimulation of normal lymphocytes. Similarly, stimulation was not impaired by addition of Azulfidine to the cultures in concentr-ations from 2 to 40 flgm/ml. We conclude that an intrinsic defect in lymphocyte responsiveness is a common though not a constant or specific finding in patients with both CD and UC.

( The gastric epithelium normally acts as an effective barrier to the back diffusion of hydrogen ions, by virtue of mucus secretion and because of a low proton permeability of the mucosal membrane of the surface cell. Aspirin, and other acids can, however, act as p, roton carriers across this membrane with possible damage to the cell, either induced by pH changes, or by aspirin itself. A sensitive index of cell damage is the measurements of the possible conductance pathways in the tissue. There are several such pathways in epithel ia: 1. Transcellular where transport occurs across the serosal and mucosal membranes of the cells in series; 2. Paracellular where transport occurs across the tissue by extracellular routes; and 3. Intercellular where molecules move from cell to cell via high conductance channels which remain patent as a result of metabolic activity. Microelectrode measurement of the cell membrane resistance and macroelectrode measurement of tissue resistance allows an assessment of the paracel lular contribution to transport across the tissue. When one microelectrode is used to send current into a cell and another is used to measure voltage changes in other cells as a function of distance from the current source, an estimate of route 2 is obtained (i.e. coupl ing).

With these techniques it was shown that: 1. The paracellular route is of minor significance in the gastric mucosa; 2. The different cell types, i.e. surface, parietal or oxyntic and chief cells act as a single electrically coupled system. The addition of aspirin, 10 mH, to the mucosal surface of the mucosa results in initially (a) an increase in tissue resistance accounted for by an increase in cell membrane resistance, and (b) a large decrease in intercellular conductance, which may be due to the mitochondrial effect of aspirin resulting in higher intracellular Ca++. This phase is followed by a decrease in tissue resistance due to an increase in paracellular conductance.

This approach thus allows a quantitative assessment of damage to the gastric "barrier", with precise definition of locus of action of various agents. (NIH, NSF).

Stephen Tint and Norma Deering. N.Y.U. Medical School, and New York V.A. Hospital, New York, New York. The neutral sterol composition was determined in gallstones removed surgically from 16 individuals (5 adult males and 11 young Southerwestern American Indian women). Sterols were separated by AgN03-thin-layer chromatography, and quantitated by gasliquid chromatography. The identification of each sterol was based upon the correspondence of its Rf and column retension times with values obtained for sterol reference standards. Sterols comprised about 80% of the dry weight of the stones (0.8at 0.16 grams/gram). Cholesterol (cholest-5-en-3~-01) was quantitatively the most important sterol and constituted 96 . 5t O.5% of~the total sterols. However, substantial quantities of cholestanol (SGt-cholestan-3r-ol), 1.8t0.3%, laJhosterol (5l(-cholest-7-en-3~ -01), l.stO.3%, and lanosterol (lanosta-5,24, diene-3p -01) plus other methyl sterols, 0.2% were found. The sterol fraction in gallbladder bile aspirated from 2 subjects with stones contained cholesterol, 97.0%, cholestanol, 2.1%, and lathosterol 0.9%, which indicated that the 5~-saturated sterols originated from the liver. In contrast, the sterol fraction in bile from 5 normal subjects contained over 99% cholesterol, and only traces of the 5al-saturated sterols. The results indlcat~ that substantial amounts of 5G(-dihydro saturated sterols are secreted into bile by the liver and participate in the formation of gallstones. These sterols are not detected by the usual colorimetric assays, because of their saturated configuration at carbon 5. Consequently, present reports may have underestimated total gallstone sterols. The secretion of these compounds in patients with cholelithiasis points to a defect in the hepatic biosynthesis and metabolism of sterols. 31 patients with an endoscopically proven duodenal ulcer were entered into a double-blind placebo controlled study of carbenoxolone sodium contained in positionrelease capsules 1. (Duogastrone). 11 of these patients had either negative or equivocal barium meals. The response to 6 weeks treatment was assessed symptomatically at 2, 4, 6 and 12 weeks and by duodenoscopy between 2 and 4 weeks, and at 6 and 12 weeks employing a standard procedure 2. Ulcers were recorded photographically and their size was assessed by reference to anatomical landmarks or by calibration with biopsy forceps. 13 of 16 (81%) patients treated with carbenoxolone showed unequivocal evidence of healing by 4 weeks, compared with 4 of 13 (J~) patients on placebo (P< 0.01). At 12 weeks however 7 of 12 (58%) patients on placebo showed endoscopic evidence of healing as compared with 10 of 14 (71%) on carbenoxolone (p~0.05). In those patients treated with carbenoxolone there was no relationship between symptomatic improvement and ulcer healing. Only 3 ulcers recurred at 12 weeks following early healing with carbenoxolone. The diagnosis and assessment of healing of duodenal ulcers has been shown to be more accurate by duodenoscopy than by either radiology or analysis of symptoms. Carbenoxolone sodium (Duogastrone) has been shown to promote early healing of duodenal ulcers in spite of the natural tendency for these ulcers to heal over a longer period.

I. Galloway, R. (Symposium on Carbenoxolone sodium, p. 203. 2. Salmon, P.R., Brown, P., Thein-Htut, Read, A.E. (Gut. 13: Butterworths 1968) In press 1972) The gastric electrical control activity (pacesetter activity) in the stomachs of anesthetized dogs (stimulated bipolarly through two silver wire electrodes, 1 cm long and 0.5 cm apart, usually with 100 msec pulses of 2 to 80 V) could be driven from anywhere in the electrically active region. The maximum driven frequency (MDF) in the intact stomach was 7.7 c/min. The control wave at the site of stimulation had phase lead over control waves at all other sites. The stomach was divided into three segments by circumferential cuts in the muscle layers. The distal (antral) segment could be driven up to a frequency of 9.1 c/min. The MDF of the proximal and middle segments was 6. 6 to 7.7 c/min. The MDF of the antral segment also depended on sympathetic or parasympathetic activities. 500 ~g of atropine (i.v.) raised the MDF in the antral segment to 9.1 c/min; 200-600 ~g of physostigmine (i.v.) lowered the MDF in the antral segment to about 6 c/min . The MDF of the proximal and middle segments was not much affected by either drug. In the intact stomach, single stimuli were applied at different sites and at different phases of the control wave cycle, to determine the refractory properties of oscillators. Results were similar to those reported earlier on i.a. injections of acetylcholine, and those predicted by the gastric computer model; 50-60% of the control wave cycle immediately after the occurrence of a control potential was absolutely refractory. The rest of the cycle was relatively refractory;

i . e., premature control potentials (PCP) could be produced in this period with a stimulus of adequate amplitude. The threshold value of the stimulus required to produce a PCP fell progressively during the relatively refractory period. Conclusions:

(1) Intrinsic frequencies of gastric segments, and abilities of upper and middle segments to follow electrical stimuli, are not determined by the refractory period; other characteristics of the oscillators and their mutual coupling are involved.

(2) The ability of the antrum to be driven may be limited by the refractory period under parasympathetic blockade, but not otherwise. This study suggests the possible use of electronic pacemakers to influence the gastric control activity, and hence the mechanical activity and gastric emptying. A previous prospective, randomized study to evaluate selective gastric vagotomy vs. truncal vagotomy showed that the selective gastric technic was superior in achieving vagal denervation of the stomach (2% incidence of incomplete vagotomy with selective gastric vagotom~ vs. 19% incidence incomplete vagotomy with the truncal technic), but the clinical benefits of preserving the extragastric vagal fibers were not impressive. Following the observation that selective gastric vagotomy will usually achieve complete vagal denervation of the stomach, another prospective, randomized study was initiated 5 years ago to compare the two adjuvant procedures (antrectomy or pyloroplasty) most widely utilized with vagotomy for the surgical treatment of duodenal ulcer. Follow-up studies have been obtained in 104 patients --53 with antrectomy and 51 with pyloroplasty. Evaluation after operation has included clinical follow-up at regular intervals, personal patient interviews, postoperative studies of gastrointestinal function with radiological examination, and gastric secretory tests. A preliminary report showed no difference between the two groups. Now this later report, with continuing followup observation to 5 years, indicates superior clinical results from selective vagotomy and pyloroplasty. Two recurrent ulcers have developed in patients undergoing antrectomy while no recurrent ulcers have occurred in patients with pyloroplasty. The incidence of dumping and diarrhea is no different in the two groups. . Acute hemorrhagic pancreatitis was induced in dogs by the technique of the closed duodenal loop. In the first series of experiments, survival times were studied in untreated dogs and in animals receiving injections of vasopressin. Ten untreated dogs had an average survival time of 33 hours with a range of 18 to 48 hours. Twelve dogs that were treated with 5 units of vasopressin/kg before surgery and with 2.5 units of vasopressin/kg every 6 hours until demise showed an increase in the average survival time to 72 hours with a range of 30 to 140 hours. In the second series of experiments, the gross and microscopic appearance of the pancreas and the closed duodenal loop were examined over a 24 hour period in 14 untreated dogs and compared with 11 dogs receiving 0.25 units of vasopressin/kg every hour. The pancreas of the vasopressin treated dogs showed some extravasation of RBC's and WBC's into the tissue spaces, some blood vessel engorgement and some edema but the acinar tissue disruption that was characteristically seen in the untreated dogs was not present. The closed duodenal loops of the treated dogs showed little or no necrosis of the villi and the fluid in the loops did not have the characteristic foul odor present in the untreated dogs. It is postulated that vasopressin reduced the intensity of experimentally-induced pancreatitis and increased the survival rate by increasing the permeability of the epithelial surfaces with the passage of fluid out of the duodenal loops and the pancreatic ducts, preventing the increased pressures that would rupture the pancreatic duct system. Intestinal growth. hexose absorption and disaccharidase activities are enhanced in the diabetic rat. Despite this stimulation of the gut. the diabetic animal fails 807 ro grow. Hence it is important to examine the effects of diabetes on transport of a "building block" nutrient such as calcium. We studied matched control (C) and alloxan diabetic (0) rats by in situ perfusion and balance studies. Ca balance was positive in C rats but was negiti~n D rats because of significantly increased fecal Ca excretion. Diabetics, however. showed significantly increased efficiency of fat absorption (excretion as % of intake: C. 18.5; O. 13.0; P < 0.05). In situ perfusion showed net absorption (NA) and lumen-to-plasma flux (lP) of Ca tooe-iignificantly decreased in the duodenum of 0 rats (NA: C. 44.0; D, 17.0; lP: C, 66.0; D, 39.7) . There was no difference between C and 0 rats in ileal Ca transport (Fed. Proc., 31: March-April,1972) . To examine the biochemical mechanism for the dif1'e!rence in duodenal Ca transport between 0 and C rats we studied calcium binding protein (CaBP) (Deluca, Biochem' J 10:2302 , 1971 . Mean specific activity of crude duodenal extracts (Chelex assay, % Q5Ca bound per mg/ml protein) was 0.92 in C and 0.58 in 0 rats.

Fractionation of the extracts with Biogel P 100 gave a sharp CaBP peak at 2.7 void volumes in C rats (sp. act. 21.6). This sharp peak was absent in 0 rats, but there was a broad peak of reduced activity (sp. act. 4.4) . Similarly, fractions from Biogel chromatography exhibited a peak of endogenous 40Ca concentration (X10-6M) at the CaBP peak in C rats (6.1) which was not present in 0 rats (2.5). Thus, the decreased duodenal Ca absorption in 0 rats was associated with decreased CaBP. Because transport is both stimul ated and suppressed in di abetes, the di abetic rat is a unique model for studying mechanisms and control of intestinal transport. Women with hepatitis B antigen (HBAg) positive icteric disease during pregnancy transmit the infection to their infants with high frequency. The previously described series at John Wesler Hospital (Schweitzer and Spears, New Eng. J. Med. 283:570, 1970) now includes 23 mothers with overt disease during pregnancy or within 2 months after delivery; HBAg positivity was found in 12 infants (52%). In contrast, our observations of women found to be asymptomatic carriers at the time of del ivery reveal a low frequency of transmission to infants. We have fol lowed 14 such mother-infant pairs; HBAg positivity was found in only one infant (7%). HBAg subtyping also demonstrated a definite difference between the two groups. Of the 23 mothers with overt hepatitis, 17 were y+ by direct testing, and 4 more with weak HBAg reactions were presumptively y+ on the basis of their infants' subtype. HBAg positivity of the y+ subtYPf was found in I I infants. Only 2 mothers with acute disease had d subtype; one of their 2 infants developed d antigenemia. Of the 14 mothers who were asymptomatic carriers, none had+the y+ subtype. The 14 d+ subtypes in this+group resulted in one d infection in an infant. Thus, there was y predominance among infected infants, which may have one of two explanations: (I) An acute infection is readi Iy transmitted, but a chronic infection is not. The frequency of either subtype among infants, therefore, is explained in terms of its relative frequency among acute cases. (2) The y strains of type B virus have biologic characteristics which faci litate transmission to the infant. Additional data from other centers where d+ infections are more frequent in acute hepatitis and y+ infections more frequent among carriers wi I I be needed to distinguish between these possibi lities. The high activity of the pentose phosphate shunt has been previously demonstrated for the isolated gastric mucosa of the rat (Fed. Proc. 30:477, 1971) . Insofar as it relates to gastric acid secretion, shunt activity might be expected to vary with changes in acid . secretion. For this reason, selected stimulants and inhibitors of acid secretion were added to the mucosal supernatant fraction that contains the shunt enzymes. When carbachol, histamine, gastrin, cyclic AMP, dinitrophenol, thiocyanate or ouabain was added to this mucosal supernatant fraction, however, the rate of shunt oxi dati on remai ned unchanged compared wi th that of the untreated, pai red control. Similarly, the ratio of labeled C02 evolved from glucose -1-14C to that evolved from glucose -6_ l4 C was not significantly increased by the addition of carbachol to rat mucosal slices that were incubated with these radioisotopes. Conversely, the rate of shunt oxidation in the rat mucosal supernatant fraction was increased nearly three-fold by the additi on of gl utathi one, an agen.t that di d not appreciably increase acid secretion by the isolated rat gastric mucosa. These results support the view that whereas basal acid secretion by the rat gastric mucosa may be supported through shunt activity, stimulated acid secretion may instead be maintained through some oxidative alternative like the citric acid cycle. Because hepatic bile obtained at surgery from patients with cholesterol gallstones is supersaturated with cholesterol, deranged hepatic function has been assigned a primary role in the formation of lithogenic bile (New Eng. J. of Med. 238 (1970) 53). However, control data is not available for patients without cholesterol gallstones, and it is not known whether the liver of gallstone patients continues to secrete lithogenic bile after recovery from cholecystectomy. We therefore examined the relative composition (J. Clin. Invest. 47( 1968) 1093) of hepatic bile obtained at surgery from 11 patients with cholesterol gallstones and from 17 control subjects: 9 with pigment stones and 8 without hepatobiliary disease. Indwelling T-tubes allowed daily postoperative collections of fasting and postprandial bile for 2-20 weeks in 9 cholesterol gallstone and 11 control patients. At surgery hepatic bile from the cholesterol gallstone group was supersaturated with cholesterol, while hepatkbile from control patients was not and contained less cholesterol (p < 0.01). Postoperatively, both fasting and postprandial biles became undersaturated in all patients within 3 days after clamping the T-tube to reestablish the enterohepatic circulation. In patients with cholesterol gallstones as well as in the control patients, fasting biles contained relatively more cholesterol (p ( .05) than did postprandial biles, but there were no significant differences between patient groups. Thus, patients with cholesterol stones have bile which is clearly different from controls at surgery. After cholecystectomy, their bile reverts to a pattern indistinguishable from either cholecystectized pigment stone or non-stone patients with functioning gallbladders. We conclude that in cholesterol gallstone disease, the gallbladder influences the liver to secrete lithogenic bile. After removal of the gallbladder, bile composition is restored to a ~orma 1 pattern. The hypothesis that the initial action of ethanol in the stomach is on active transport processes was investigated using electrophysiological and biochemical techniques. Transmucosal potential difference, short circuit current and electrolyte concentrations in an in situ rat stomach were determined. Activity of adenyl cyclase (the membrane-bound enzyme responsible for formation of cyclic AMP from ATP) was assayed by measuring the amount of radioactive 3' , S'-adenosine monophosphate formed from 14C-labeled adenosine triphosphate . (Krishna, Weiss and Brodie. J. Pharm. Exp. Therap. 163:379, 1968 ). Solutions of 10-20% ethanol in lSO mN NaCl or SO mN NaCl and 100 mN HCl were instilled on the mucosa of fasted fat stomachs for S-30 minutes. The threshold concentration for producing effects was 10% ethanol. 20% ethanol in 150 mN NaCl decreased transmucosal potential (-36+4 mV to -20+S mY) and increased tissue electrical resistance (18.:.4%). Changes were more pronounced when ethanol was present in the mixed NaCl-HCl solution than in the NaCl solution. Within 5 minutes after topical ethanol administration, luminal chloride concentration decreased from l50:!:.SmEq to 136.:.4 mEq (p<.OS). At this time there were no consistent changes in sodium or potassium concentrations or volume. 20% ethanol only slightly altered basal adenyl cyclase activity but inhibited sodium fluoride activated adenyl cyclase. The earliest action of alcohol appears to be inhibition of chloride transport. Since electrical resistance increased, the effect is on active transport of chloride rather ilian an increased permeability (back flux) of the ion. Inhibition of active chloride transport is postulated to be through inhibition of the membrane-bound enzyme, adenyl cyclase. (Supported by NIH grants #1 ROl MH21S22-01 and # AMll073-0S). The incidence of stress ulcer is increasing, yet mortality remains high and no agreement exists as to optimal treatment. Current experimental stress ulcers usually fail to produce the massive hemorrhage seen clinically. A simple, reproducible, exsanguinating gastric mucosal bleeding model is presented. One hundred ten cats were prepared under aseptic conditions with gastric cannulae and indwelling plastic femoral catheters. Ten days later, fasted cats were given histamine-beeswax (3 mg/kg) subcutaneously, and aspirin 100 roM (10 cc/kg intragastric). Two hours later the stomach was emptied and ethanol 20% (8 cc/kg) was instilled. After 18 hours doses were repeated. All cats were weighed daily, had hemoglobin determinations and received Ringers lactate 50-70 cc/kg. The first 40 cats were sacrificed for gross and histologic studies. Three groups were studied to evaluate treatments: (I) Controls, (II) Effect of fresh ACD blood replacement, (III) Effect of antacid (ge1usi1 70 cc BID intragastric). Results. Gross bleeding occurred in all 40 cats initially studied; diffuse erosions were present in 18, antral ulcers 6, cannula erosions 3, perforations 2 and combined lesions 11. Thirty-seven of 50 control animals exsanguinated; hemoglobin fall averaged 10.1 gr., (range 1.3-15) over 3.0 days (range 0.5-6.5), Five of 10 antacid treated animals exsanguinated; hemoglobin fall averaged 9.2 gr., (range 5,0-14), over 3.7 days (range 2,0-7,0), Two of ten transfused animals exsanguinated, receiving an average of 28.5 cc/kg (range 5-58 cc/ kg). Conclusions. (1) Gross hemorrhage occurred in 100% of animals studied; (2) Death from exsanguination occurred in 74% of controls but was preventable by trans- We have developed a mathematical model of circulatory control in the gut to determine whether a metabolic feedback mechanism could account for autoregulation of blood flow and autoregulatory escape. The model consists of four sets of equations describing hemodynamics, oxygen transport, and local and nervous effects on mesenteric arteriolar and precapillary sphincter tone. The local system controls interstitial p02 by I) regulating arteriolar resistance to change blood flow; and 2) by opening or closing precapillary sphincters, thereby determining capillary density (the number of open capillaries) and the diffusion parameters (surface area and diffusion distance). Sympathetic activation enhances arteriolar and precapillary sphincter tone. In computer simulations the model responded to step changes in arterial pressure by autoregulating blood flow to maintain oxygen delivery. The degree of autoregulation increased whenever the amount of extractable oxygen in the blood decreased. This has been observed experimentally in skeletal muscle but not yet in the intestine. In simulations of autoregulatory escape, sustained sympathetic activation caused a fall in mesenteric blood flow, capillary density and oxygen delivery. Capillary density remained depressed, but blood flow and oxygen delivery returned toward control. The end of sympathetic activation was followed by hyperemia and overshoot in capillary density. These results agree with those obtained experimentally. However, our model predicts that the degree of autoregulatory escape and post-stimulation hyperemia will increase as capillary density is depressed by the sympathetic input. This relationship has not yet been examined experimentally. The degree of autoregulation varied greatly with initial flow conditions, yet the "escape" index changed only slightly. Such a relationship could account for the failure of experimental attempts to correlate the degree of autoregulation with the degree of escape . We conclude that the metabolic feedback hypothesis can adequately explain blood flow autoregulation and autoregulatory escape; however, a description complete enough to account for intestinal hemodynamics in some experimental conditions (e.g. venous pressure elevation) must also include certain "myogenic" properties of the intestinal vasculature. (Supported by Grant Number HE 11678, a grant from Miss. Heart Assoc. and U.S. Army Contract Number DADA 17-69-C902S).

Recent reports that cyclic AMP (cAMP) may function intracellularly as a "second messenger" of beta adrenergic stimulation led us to postulate that within mesenteric vascular smooth muscle an increase in the intracellular cAMP concentration, regardless of its mode of production, would cause vasodilation. In anesthetized dogs we measured blood flow through the superior mesenteric artery with electromagnetic flowmetry while monitoring systemic arterial and portal venous pressures with transducers. Control measurements were compared with measurements during intra-arterial infusion of vasoactive agents which increase the intracellular level of cAMP. The agents which produced significant (p(O.OS) increases of 40 to 60% in superior mesenteric arterial flow were: prostaglandin El (0.01 ug kg-l min-l ), isoproterenol (O.04-0.0S uy kg-l min-l), papaverine (0.1-0.3 mg kg-l Tin-l), exogenous cAMP (0.3 mg kg-l min-), and dibutyryl cAMP (0.3-0.8 mg kg-l min-). Propranolol blocked the vasodilatory effect of isoproterenol but not that of the other agents. Our data are consistent with the hypothesis that an increase in the cAMP level of mesenteric vascular smooth muscle cells, whether by stimulating adenyl cyclase (isoproterenol and prostaglandin E l ), by inhibiting phosphodiesterase (papaverine), or by administering exogenous cAMP and its dibutyryl derivative, will cause vasodilation. Although non-adrenergic and beta adrenergic drugs may alter cAMP levels by different routes, these findings support the view that cAMP serves as an intracellular mediator of both beta adrenergic and other drug-jnduced vasodilation in the mesenteric vascular bed. Anti-Au was detected by incubation of 12SI-labeled Australia antigen (Au*) with test serum followed by precipitation of Anti-Au-bound Au* with rabbit antihuman IgG/ IgM plus polyethylene glycol. Although the precise threshold of Anti~Au positivity is difficult to determine, a cut-off point ( > 2S% Au* precipitation) was selected by inspection of the plotted data. By this criterion, 26 of 200 (13%) consecutive blood donor sera contained Anti-Au. (All donors had denied recognition of previous hepatitis). Among acute phase sera from 20 hepatitis patients found Au-positive by counterimmunoelectrophoresis (CIEP) and/or RIA, co-existent anti-Au was demonstr' ated in 3 (15%). It is anticipated that Au in sufficient excess would preclude the detection of co-existent Anti-Au and vice versa. Among sera from SO patients with Au-negative hepatitis (CIEP and RIA) 16 (33%) contained Anti-Au. This number includes 11 patients with fulminant hepatic necrosis, of whom 3 (27%) were Anti-Au positive. Possible explanations for the increased frequency of Anti-Au in this group include 1) These patients have hepatitis due to an agent unrelated to Au with a high prevalence of preexisting Anti-Au reflecting above average past exposure. The detection of Anti-Au in 9/27 (33%) sera from patients with Alcoholic liver disease supports this concept. 2) ~on-specific Au-binding factors develop in response to hepatic necrosis and/or cho1estasis. Against this is the "normal" prevalence of Anti-Au found among patients with halothane hepatitis (2/20; 10%) and primary biliary cirrhosis (2/13; lS%). 3) These patients have Au-associated hepatitis and the Au, if still present, is obscured by Anti-Au. This might be resolved by determination of Anti-Au titers in convalescent sera. Though results on such sera are not currently available for this group of patients, such a study is underway. Milk proteins have been implicated as etiologic factors in a wide variety of gastrointestinal problems of infants and children. A colitis like picture in infants below 3 months of age, occult blood loss and protein exudation in older infants, steatorrhea, and sensitization to gluten and other proteins, have all been attributed to intestinal intolerance of one or more cow's milk proteins. In many of these patients, of different ages and with mild symptamatology a common denominator is the finding of eosinophilic gastroenteropathy (EG), a syndrome characterized by (I) diarrhea or vomiting (2) peripheral eosinophilia and (3) eosinophilic infiltration in mucosal biopsies of affected gastrointestinal (GI) lesions. Fourteen infants under two years of age were diagnosed as EG, of whom nine were males. Seven inFants presented with bloody diarrhea and peripheral eosinophilia and were intolerant repeatedly of bovine milks and cereals, for 4 -28 months of Follow up. Seven otherwise thriving inFants, 8 -18 months of age presented with signs of allergic gastroenteropathy, i.e. iron deficiency anemia, accult blood in the stool and proteinlosing enteropathy, and relatively mild GI complaints. The anemia, excess stool protein losses and eosinophilia cleared with dietary manipulation -three with simple milk product withdrawal, three required the elimination of milk products plus beef and veal, one with removal of milk products and gluten, and the last cleared only after removal of milk products, gluten and all animal protein. Except For eosinophilic inFiltration, biopsies of small intestine and colon were normal -in all who were so studied. Steatorrhea occurred only with concurrent gluten-intolerance. Three of the older infants showed additional unexplained speciFic profound clinical intolerance with development of shock like states on ingestion of rice in two cases and chicken in one instance. Antibodies to bovine proteins in sera and stool were absent. No primary immunological abnormalities or changes in serum complement were noted. Protein intolerances have persisted for 8 -48 mOl'	hs pflf.ollow up, manifesting shock, bloody diarrhea, or eosinophilia with measured protein cha lenQes.

HYPERALlMENTATION SYNDROME. Stephen E. Silvis and Pablo D. Paragas, Jr. GI Section, Veterans Administration Hospital and Univ. of Minn., Minneapolis, Minn.

Three cases with an unusual complication of parenteral hyper alimentation have been observed in malnourished individuals. All of these cases developed marked hypophosphatemia (mean serum phosphate of 0.3 mg%) and muscular weakness. Two of the three patients had paresthesias, convulsions and coma. One patient died in coma.

These studies were done to produce a similar reaction in starved animals. Mongrel dogs with a 50% weight loss were infused with fibrin hydrolysate and/or glucose through a superior vena cava catheter at a rate of 140 cal/kilo/day. The starved dogs had a mean survival of 5.4 days. Control animals survived until sacrificed at 14 days. A profound fall in serum inorganic phosphate occurred in the starved animals. The mean serum phosphate in starved animals was 0.4 mg% on the fifth day while the serum phosphate in control animals was 3.4 mg% at the same time. SMA-12, serum electrolytes and osmolality were not remarkable. The same result was observed with glucose and amino acids or glucose alone in equivalent amounts. Maintenance of the serum phosphate at normal levels did not influence the mortality in starved dogs.

When starved rats (25% weight loss) were infused at 420 cal/kilo/day the mean survival was 1.4 days. Control rats were infused at the same rate the mean survival was 5.6 days. When this rate was reduced by 50% in starved rats, the survival increased to 4.0 days. When starved rats maintained a constant weight for 1 and 2 weeks were infused at 215 cal/kilo/day, the mortality was 0 to 20% respectively. A 60% mortality was observed in those infused at the same rate immediately after the weight loss. Starved rats infused at 165 cal/kilo/day, had a 10% mortality, when the calories were increased to 215 cal/kilo/day with either dextrose or amino acids the mortality increased to 40%. Totally fasted infused rats (25% weight loss) had decreased mortality compared to chronically starved infused animals.

These studies demonstrate fatal reactions in starved humans, dogs and rats by high caloric infusions. These fatal reactions can be produced with glucose alone or with glucose and amino acids. Chronic starvation with continuing weight loss appears to be important in the production of this syndrome. The etiology of this syndrome is undetermined. It is known that pancreatic digestive enzymes are transported from endoplasmic reticulum through Golgi complex into zymogen granules and thence secreted from the cell. By use of pulse-label techniques and subcellular fractionation, migration of C l4 -labeled proteins from microsomes to zymogen granules and secretion into medium may be examined. For these studies, rat pancreata were pulse labe led in vitro with L-phenylalanine-14 C and the influences of certain drugs on migration and secretion of pancreati c proteins investigated . Bethanechol chloride did not alter transport of Cl4-label from microsomes to zymogen granules but did increase secretion of label from granules into medium . Puromycin, a well known inhibitor of protein synthesis, did not alter rates of transport of label from microsomes to zymogen granules nor did it alter secretion of label into medium. Amphotericin-B and tetracycline impaired transport of Cl4-label from microsome into zymogen granules as well as secretion of label from zymogen granules into medium. Lidocaine impaired release of Cl4-label from zymogen granules into medium although it increased secretion of preformed amylase.

By use of these techniques, processes eventuating in synthesis, transport, and secretion of pancreatic digestive proteins can be examined. There is a renewed interest in ethanol metabolism because of demonstration of a microsomal ethanol oxidizing-system and differences found in the rate of ethanol metabolism in alcoholic SUbjects. The objective of this presentation is to summerize the results obtained with a standardized ethanol absorption test.

Studies were carried out in 60 volunteer subjects using 39.95% ethanol in water. On ml per kg body weight was administered in fasting state and blood samples taken at 0,15 30,60,120 and 180 minutes.

Ethanol blood concentration was determined by gas chromatography.

In 33 subjects (55%) the maximum blood ethanol concentration was reached in 30 minutes; in 9 subjects (15%) the peak was reached in 15 minutes; in 18 subjects (30%) a delayed absorption (peak at 60 minutes) was observed. According to maximum ethanol blood concentration the patients were divided into three groups: low maximum 0.15-0.40 giL (19 subjects); standard maximum 0.41-0 . 66 giL (36 subjects and high maximum 0.67-0.92 giL (5 subjects). The slope of ethanol oxidation curve varied and no relation was found with rate of absorption and Maximum Blood Ethanol Concentration.

It is concluded that considerable individual variation exists in ethanol absorption rate and that standardization of testing may be useful in further elucidation of various parameters of alcohol metabolism including gastric absorption, ethanol oxidation and effects of hepatic injury. Hepatic bile (HB) and gallbladder bile (GBB) of cholesterol stone formers have been shown to be saturated or supersaturated with cholesterol (lithogenic bile), while that from normal subjects is .uhsaturated. Since the population in this country has not previously been studied in detail, the composition of HB and GBB in stone formers was examined. Samples were obtained at operation from 65 stone formers and 15 controls (7 patients with pigment stones, 8 without gallstone disease). In 7 stone formers with biliary fistulas, multiple HB samples were obtained before and after ceasing external biliary drainage. The relative proportions of bile salt, phospholipid, and cholesterol were measured in all samples to determine their lithogenicity. Samples were examined microscopically for cholesterol microcrystals. The proportions' of individual bile salts , and the amounts of lecithin and lysolecithin, were determined . In the bile of stone formers, no changes were found in the relative amounts of individual bile salts; lysolecithin made up <.50/0 of the total phospholipid; and cholesterol crystals were found only in lithogenic bile. The previously accepted criteria for lithogenicity, therefore, appeared to be valid. However, only 600/0 of GBB samples from stone formers were lithogenic, and only 650/0 of HB samples. Although no controls produced lithogenic bile, there was marked overlap between stone formers and controls in the degree of cholesterol saturation of bile samples. HB in each bile fistula patient varied from unsaturated to supersaturated, even after reconstitution of the patient ' s bile salt pool. These results indicate that (l) contrary to earlier studies, it is not possible to separate stone formers from controls by studying random samples of hepatic or gallbladder bile, (2) the production of lithogenic bile by the liver is not a continuous process, but an intermittent one. In establishing 3'5' cyclic AMP as an intermediate in the stimulation of acid secretion by histamine, 4 criteria should be fulfilled~ 1. Inhibitors of phosphodiesterase should produce secretion; 2. Cyclic AMP should mimic the effects of histamine ; 3. Adenyl cyclase should be stimulated by histamine; and 4. Levels of cyclic AMP in the tissue or secretion should increase with addition of histamine. In the case of Necturus in vitro gastric mucosa, which is nonsecreting when mounted in the Ussing chamber, theophylline and cyclic AMP both stimulate acid secretion. Adenyl cyclase, present in the membrane fraction of the tissue, is stimulated by both histamine and pentagastrin. Cyclic AMP levels in the tissue are also increased by theophylline. In contrast, in the case of the dog, theophylline neither stimulates secretion, nor alters the histamine dose response. Injection of cyclic AMP does not stimulate acid secretion; and in fact may inhibit it. Cyclic AMP levels in the tissue are only slightly increased by theophylline, although the tissue contains a typical phosphodiesterase. In only one of five dogs did cyclic AMP levels in gastric juice increase with histamine stimulation. Adenyl cyclase, present in a purified membrane fraction as monitored by 5' AMPase or electron microscopy is only variably stimulated by histamine and is insensitive to fluoride. Based on these observations, there is considerable evidence for cyclic AMP as an intermediate in amphibian acid secretion. For dog, however, the situation appears considerably more complex with, at the moment, the evidence arguing against a second messenger role for this nucleotide. 

In 23 postoperative peptic ulcer patients (Vagotomy-Pyloroplasty -21, Vagotomy -81-1 and BI-I), insul in testing in 10 and tolbutamide testing in 13 resulted in significant and prolonged gastric acid inhibition. Basal acid outputs (meq/hr) were consistently higher than in any post-stimulation hour up to 4 hours. Maximal inhibition usually occurred within the first 2 hours and involved decreases in volume, acid concentration, and total acid output.

In 8 insul in and II tolbutamide-induced inhibitory responses repeat testing with the same or alternate agent resulted in conversion to a positive response. The most significant difference in acid response between initial and repeat tests occurred in the second hour after injection of hypoglycemic agents (t test:p = .01 to . 05). The use 6f KCL infusion in 2 instances was of doubtful significance in conversion to positive. In 4 cases of inhibition, 2 after insul in and 2 after tolbutamide, repeat or alternate tests failed to convert. Seven recurrent ulcers (4 duodenal, 2 gastric and I stomal) were associated with tests converted to clearly positive. The misinterpretation of acid inhibition after induced hypoglycemia as a negative response, has led to several erroneous reports of recurrent ulcer in the face of "complete vagotomy", The e" xact mechanisms and circumstances responsible for such acid inhibition after hypoglycemia are s till unc I ear, Prolonged remission, if not cure, can be accomplished with appropriate antibiotic therapy in Whipple's disease.

The clinical symptom complex and PAS positive material in biopsies at light microscopy are suggestive, but diagnostic confirmation must await electron micrographs revealing as yet unidentified bacilliform bodies in the small bowel mucosa.

Duodenoscopy in the present report disclosed a previously not described gross "buttermilk sky" appearance of the duodenal mucosa which if characteristic may prove very helpful in establishing early diagnosis.

The present finding correlates well with light and electron microscopic examinations and diminished and almost completely disappears as clinical and anatomic pathology improve at three and seven month follow-up examinations. The effect of synthetic C-terminal octapeptide (OP) of cholecystokinin (Squibb Institute for Medical Research, New Brunswick, N.J.) on gall bladder (GB) size was studied in 18 male volunteers. Three subjects were studied at each of the following doses: 2.5, 5, 10, 20, 40, and 80 ng/kg. OP was injected intravenously over one minute. Each subject underwent cholecystography twice. OP was injected on one day, and saline alone as a control on another day, in double-blind fashion. The order of injection was randomized. GB x-ray films were obtained before and at 1.5, 3, 5, 10, 15, 30, and 60 minutes after injection. Vital signs and serial blood chemistries were measured . Symptoms following injection were recorded. GB area was measured by planimetry. Percentage reduction in area was determined.

Dose 28 < 0.01 Maximum decrease in GB area occurred at 5 to 15 minutes after OP injection. With saline alone, mean area varied within 5% of the pre-injection area. One subject experienced nausea after 40 ng/kg, and one had nausea and abdominal cramps after 80 ng/kg. These symptoms lasted 3-5 minutes. There were no significant changes in vital signs or blood chemistries after OP. It is concluded that OP is a cholecystokinetic agent in man. 20 ng/kg was as effective as the higher doses tested. The aim of this project is to conduct a controlled therapeutic trial on the use of antacid, anticholinergic and antipeptic agents in active duodenal ulcer disease. The diagnosis was based upon clinical symptoms and roentgen demonstration of an ulcer crater. The design of this study was double-blind, with consecutive patients being randomly assigned to one of four treatment groups: (1) antacidaluminum hydroxide 225 mg. and magnesium oxide 75 mg. in a single capsule, (2) anticholinergic -propantheline 7.5 mg., (3) antipeptic agent -amylopectin sulfate 250 mg., and (4) a combination of 7.5 mg. of propantheline and 250 mg. of amylopectin sulfate. These capsules were identical in appearance and were coded. The patient was asked to take one capsule every two hours during the waking period. The therapeutic success was evaluated on the basis of rapid relief of symptoms and prompt healing of an ulcer. Persistence of ulcer distress beyond a 4-day therapy and/or the appearance of complications were considered a failure. A total of 293 patients with duodenal ulcer were admitted to the study; 15 were dropped. Of the 278 patients remaining in the study, the results are as follows: (1) Of the 71 patients on antacid therapy, good results were observed in 42, and poor results in 29 (41%);

(2) Of the 85 patients on propantheline therapy, good results were observed in 60, and poor results in 25 (2~1o); (3) Of the 70 patients on amylopectin sulfate therapy, good results were observed in 58 and poor results in 12 (lrlo); and (4) Of the 52 patients on combined propantheline and amylopectin sulfate, good results were observed in 43, and poor results in 9 (lrlo). Analysis of our data showed no evidence that the failure rate within each form of treatment is dependent upon the clinical parameters, i.e., duration of symptoms, intensity of abdominal pain and the occurrence of radiation and of nocturnal pain . Chi-square analysis revealed that the failure rate was significantly higher in the antacid and propantheline groups than in the amylopectin sulfate groups. Thus, amylopectin sulfate, an antipeptic agent, is a good anti-ulcer drug. Distinction has been made between the hepatitis associated with Australia antigen ~u(l~ and hepatitis which is not so associated. We have studied prospectively 24 patients with post-transfusion hepatitis at the Philadelphia General Hospital, 12 of whom had received at least one unit of Au(l)-containing blood, and 12 of whom had received only blood in which the antigen was not detected [Au(Oa . Both groups were similar with respect to age, sex, number of units transfused, and number of clinic follow-up visits. Seven of the 12 Au(l)-recipients developed detectable serum Au(l), and one developed antibody. Neither antigen nor antibody was found in sera of the Au(O)-recipients. Both groups had similar clinical signs and symptoms; three of the Au(l)-recipients and four of the Au(O)-recipients became jaundiced. Since not all were icteric, the incubation period was taken as the time from transfusion to first observed serum glutamic pyruvic transaminase (SGPT) elevation. Median incubation period in Au(l)-cases was 51 days (range 17 -132), and in Au(O)-cases 62.5 days (range 42 -94). This was not a significant difference, nor was there any difference in the time to reach maximum SGPT or the duration of SGPT elevation. The maximum SGPT level was similar in both groups, as was the maximum serum bilirubin and prothrombin time, and no histological distinction could be made. The median peak serum alkaline phosphatase in the Au(l)-recipients was 41 International Units (I.U.), and in the Au(O)-recipients, 113.5 I.U . This was the only significant difference between the two groups, suggesting an obstructive factor found more often in the recipients of Au(O)-blood. With this sole exception, the se findings indicate that, in hospital patients, there are no distinguishing features of hepatitis occurring in recipients of Au(l) vs . Au(O) blood with regard to incubation period, symptoms, physical findings, clinical course, laboratory findings, and severity of disease.

Supported by grants from the National Institutes of Health, Office of Naval Research, and an appropriation from the Commonwealth of Pennsylvania. The endogenous catecholamines, norepinephrine (NE) and epinephrine (E) are vasoconstrictors in the canine mesenteric circulation regardless of the dose or method of injection (Swan, Gastroenterology 61: 863, 1971 ). Since the response of this vascular bed to experimental shock is distinctly different between dog and monkey (Swan, Gastroenterology 61: 872, 1971 ), a question arises as to the effects of these catecholamines on the mesenteric circulation of the subhuman primate . Adult olive baboons (Papio anubis)were anesthetized with pentobarbital and superior mese. nteric arterial blood flow (MBF) was measured with an electromagnetic flowmeter. The first branch of the superior mesenteric artery was cannulated for intra-arterial (IA) drug infusions. Systemic arterial (AP) and portal venous (PP) pressures were measured. During the control period MBF was 218 + 38 (SE) ml/min. AP was 122 + 3 and PP, 6 + 2 mm Hg. IA infusion of ~, 0.05 pg/kg-mln, caused a progressive incre;se in MBF and-at 7 minutes the value, 435 + 23 ml/min, was Significantly greater than control (p< .001). MBF remained elevat;d at this level for the duration of the 10 minute infusion. Following cessation of the infusion, MBF returned to control. At the same concentration NE caused vasoconstriction characterized by autoregulatory escape. When the dose of catecholamine was increased to 0.5 pg/kg-min, E exerted a vasoconstrictor effect which reduced MBF from 147 + 39 to 107 + 37 ml/min within one minute (p < .05). MBF remained depressed with som; tendency towards autoregulatory escape at the end of the 10 minute infusion. NE exhibited a similar response at this dose. The findings reveal a vasodilator effect of epinephrine upon the mesenteric circulation of the subhuman primate. This effect is dose dependent in contrast to observations in the dog. Equating this dose with concentrations of circulating catecholamines measured during experimental shock in the monkey suggests that epinephrine would be more likely to effect an increase rather than a decrease in flow to the gut during shock. Cyclic AMP stimulates gastric acid secretion in vitro (frogs) and in anesthetized rats. This study on unanesthetized rats with permanent gastric fistulae shows the effects of Db-cAMP on basal gastric secretion in a more physiological setting. Male rats (mean wt lS6 Gm) fitted with stainless steel cannulae were trained for five or more days in restraining cages. After 48 hrs on a glucose diet, .09% saline (7.9~1/min) was infused continuously via a tail vein. Once the secretory rate stabilized, basal gastric secretion was collected for four IS-min periods by dependent drainage. Next, Db-cAMP (lOO.ug/min) was added to the infusion and gastric secretions collected for three IS-min periods. Db-cAMP was then removed from the infusion and the collection of gastric juice continued for two IS-min periods. Values shown are for 10 rats expressed as totals per IS min .

Periods Basal The role of the sympathetic nervous system in control of gastrointestinal function is uncertain . We have studied the sympathetic activity of the rat gut, using tritiated norepinephrine (3 HNE) . Tracer doses of 3HNE were injected into the tail vein of unanesthetized animals. The endogenous NE concentration, fractional turnover rate, half-life and turnover rate of NE were determined in the esophagus, non-glandular and glandular portions of the stomach, duodenum, jejunum, ileum and colon. The concentration of endogenous NE varied from .226±.010 to .340±.012 ug/g and was highest in the duodenum and colon (p<O . OOS) as compared to the other tissues studied. As a group, the 3HNE fractional turnover rates and half-lives for the duodenum (k= -.09S, tl/2 = 7.3 hr), jejunum (k= -.090, tl/2 = 7.7 hr), ileum (k= -.09S tl/2 = 7.S hr) and colon (k= -.102, tl/2 = 6.8 hr) were significantly greater than the esophagus (k= -.044, tl/2 = 15.8 hr), non-glandular stomach (k= -.OS8, t~/2 = 11 . 9 hr) and glandular stomach (k= -.064, tl/2 = 10.8 hr)(p<. OOS) . The NE turnover rate, which reflects the level of sympathetic activity (~g NE released/g of tissue/hour) was highest in the colon and duodenum. This study demonstrates the applicability of 3HNE tracer methods to the gut and suggests ways in which the effect of the sympathetic nervous system on normal gastrointestinal physiology and pathophysiology may be studied. Using silver/silver chloride electrodes placed above and below the phrenoesophageal ligament of the isolated perfused canine stomach and esophagus, electrical activity has been recorded similar to that in the intact dog. Low amplitude (50 mV) low frequency electrical recording was obtained from both electrodes, which usually had no definite pattern. In the distal sphincter the electrical activity was superimposed on a steady baseline which did not change even thou~h the resting gastro-esophageal sphincteric pressure altered phasically (2 c.p.m.). In the proximal part of the sphincter larger action potentials (200 mV; 7 c.p.s.) were superimposed, coinciding with the phasic increases of pressure in the gastro-esophageal sphincter. No change in the proximal or distal part of the gastro-esophageal sphincter was related to antral electrical phase or pressure. No electrical activity could ever be recorded from the body of the esophagus. Balloon distension of the esophagus caused the distal part of the sphincter to relax, accompanied by a reduction in the basic electrical activity distally. In the proximal part of the sphincter, relaxation was followed by contraction which was related to fast activity. Nor-adrenaline, adrenaline, large doses of acetylcholine and pentagastrin caused contraction of the sphincter with fast activity proximally. Isoprenaline produced relaxation with no change in electrical activity. None of the drugs altered the electrical activity of the distal part of the sphincter. Intra-arterial sodium thiopentone abolished electrical activity in both electrodes, associated with a fall in sphincteric pressure. 

There is considerable disagreement both about the normal levels of urinary excretion of indican and the value of this test in suspected stagnant loop syndr.omes. This study was undertaken to investigate these two problems. One hundred and forty one control patients were examined. Mean indican excretion was 18.3 mg/24 hrs (S.D. 10.5). There was a positive correlation between dietary tryptophane and indican excretion in this group (r = 0.46 p < 0.001). A dietary increase in tryptophane containing protein (an increase of 0.4 g of tryptophane per day) in a randomly selected group from the controls increased the urinary indican excretion Significantly to 36.0 mg/24 hrs (S.D. 16.4, t = 2.96 P < 0.05). One hundred and fifty post gastrectomy patients were examined. Raised indican levels were only found in patients who had had an afferent loop created at surgery (ardthereby liable to bacterial colonisation). The mean tryptophane intake of this group was significantly less than the controls and there was no correlation between dietary tryptophane and indican (p > 0.05). There was a significant association between weight loss and indican (X2 = 5.4 P < 0.05) and between faecal fat and indican (r = 0.61 p < 0.01).

In patients whose jejunal flora was examined, all with raised indican had E. coli above 106 organisms/mI. No patient with a normal indican had E. coli counts above 10 6 organisms/mI. It is concluded that the "normal" level of indican excretion depends on the dietary tryptophane intake, but this factor becomes insignificant when searching for evidence of small bowel bacterial overgrowth after gastric surgery. Gastrin is believed to playa crucial role in the control of rates of gastric acid secretion in man. Fasting serum gastrin levels in patients with duodenal ulcer have not been shown to be increased above normal. The present study was directed to examine and compare gastrin release in normal individuals and in patients with duodenal ulcer in response to a physiological stimulus . In addition, duodenal ulcer patients with prior pyloroplasty and vagotomy or antrectomy and vagotomy were studied . Studies were performed on 43 subjects. Fasting sera were obtained prior to ingestion of a protein-rich meal (180 gm beef) and at five 15 minute intervals thereafter. ~1easured by radi oimmunoassay, the mean fasti ng serum gastri n concentrati on for 22 patients with duodenal ulcer was 76 ± 7 (SEM) pg/ml which did not differ from that for 10 control subjects (71 ± 9.4 pg/ml). A relative plateau of serum gastrin concentrations was achieved in all groups during the interval from 30 to 75 minutes following feeding . In response to feeding the mean serum gastrin concentration in the control subjects increased by 38% (increment 27 pg/ml) whereas the mean serum gastrin concentration in the patients with duodenal ulcer disease increased by 71% (increment 53 pg/ml). The mean serum gastrin concentration in patients with prior antrectomy (47 pg/ml) did not increase following feeding: in contrast, patients with prior vagotomy and pyloroplasty (fasting level 95 ± pg/ml) had a 72% increase in serum gastrin concentration . This increase was significantly greater than for control subjects and did not differ from unoperated duodenal ulcer patients. In summary, results of thi s study i ndi cate increased gastri n release in response to feedi ng in duodenal ulcer patients and in such patients following vagotomy and pyloroplasty, but not following antrectomy and vagotomy . Thus these studies indicate, in spite of absence of fasting hypergastrinemia, the potential importance of gastrin in the hypersecretory state which frequently characterizes duodenal ulcer disease. We have previously identified increased concentrations of gastrin in fasting sera of patients with Zollinger-Ellison tumors, however to date there have been no reports of serum gastrin levels in patients with gastric carcinoma. In the present study in 22 patients with carcinoma of the stomach fasting serum gastrin concentrations were measured by radioimmunoassay and basal, maximum, and peak gastric acid output rates were determined. The patients were placed in three groups: those with carcinoma located (1) in the antrum, (2) in the body of the stomach, or (3) in the region of the antral-body junction. The mean fasting serum gastrin level for patients with antral carcinoma was 80 ± 18 (SEM) pg/ml which did not differ from that of a similar control population (92 ± 8 pg/ml). The mean fasting serum gastrin level for those with carcinoma of the antral-body juncture region and the gastric body were both elevated (P<O .OOl) and were 257 ± 32 pg/ml and 1028 ± 389 pg/ml respectively. For the three groups of patients marked differences in basal, maximal, and peak acid output gastric secretory rates were also found . Mean basal, maximal and peak acid outputs for patients with carcinoma of the antrum were 1.9, 8.0, and 10.7 mEq/hr respectively, whereas those with carcinoma involving the antral-body junction region were 0.2, 0.6, and 0.8 mEq/hr. Acid secretory rates were lowest for those with carcinoma of the gastric body with four of five patients achlorhydric. In these patients there was an inverse relationship between fasting serum gastrin concentrations and rates of gastric acid secretion. The marked fasting hypergastrinemia in these patients with carcinoma of the body of the stomach, almost all of which were achlorhydric, was of the same degree as that in patients with pernicious anemia. Negligible amounts of gastrin (less than 10 pg/mg) were found in tumor tissue, providing no evidence that the observed hypergastrinemia resulted from gastrin-releasing gastric carcinomas. This study reveals substantially elevated serum gastrin concentrations (P<O.OOl) in patients with carcinoma of the gastric body and the antral-body junction region, particularly in those with marked reductions in gastric acid secretory rates. GASTRIC Since 1959, when the Childhood Peptic Ulcer Registry was established, 772 peptic ulcers have been reported to the Registry. One hundred fifty nine of these children with ulcer have been studied and treated by me. Statistical studies show that there were 400 males and 315 females, 679 have had duodenal ulcers and 93 gastric ulcers. The family histories were positive for ulcer in 208 children, 20%. Fifty eight children had surgical procedures performed, which represents 7-1/2% of the total number with ulcer. Ulcer may occur with congenital disease of the gastrointestinal tract . When ulceration follows a burn, the usual time for hemorrhage is between the 8th and 10th day. The most frequent fatal p. ost-burn complication is sepsis. Ulcers may occur as complications of cystic fibrosis. Acute gastroduodenal ulcer may follow cardiovascular surgery. Ulcer is a rare complication of hiatal hernia. Hypoglycemia may be an etiologic factor in peptic ulcer or it may complicate gastric surgery. Gastric ulceration with pyloric obstruction is common after ferrous sulfate poisoning. Peptic ulcers may occur with liver disease, such as hepatitis or portal hypertension. Neonatal ulcers may occur, and usually hemorrhage or perforate before 14 days. Regional enteritis and ulcerative colitis may be complicated by a duodenal ulceration. Rokitansky-Cushing ulcers, occurring with intracranial disease, are acute ulcers which demand emergency therapy. Stress ulcers occur with hypoxia, fever, surgery, shock and in other life threatening situations. Ulcers may follow steroid therapy in children, occurring in such illnesses as polyarteritis, nephrosis, rheumatoid arthritis and ulcerative colitis. The world literature plus the Registry has yielded a total of 519 cases of peptic ulcers in children which were treated surgically . Trauma is an uncommon cause of ulcer in children. One case of the Zollinger-Ellison syndrome has been reported to the Registry, and 15 cases have recently been collected in a tumor registry and reported by Wilson.

CALCIUM TRANSPORT IN RAT CECUM. Ernest Urban and Thomas C. Smith. Dept. of Physiol. and Medicine, University of Texas Medical School at San Antonio, San Antonio, Texas.

In a previous study we reported that calcium (Ca) movement across the rat colon in vivo appeared to follow Ca concentration gradients between lumen and serum (Gastroenterology 60:728, 1971 ). In order to better define Ca transport in the colon, the effect of electrical forces require consideration. We have investigated the in vitro transmural Ca movement in cecum of normal rats using the short circuit current technique of Ussing and Zerahn (Acta Physiol. Scand. 23:110, 1951) . Cecum was clamped between two lucite half-cells, bathed with either sodium Ringer (NaR) or choline Ringer (ChR) usually containing 2 mM Ca and continuously aerated. Glucose (G: 100 mg /100 ml) was also present in some experiments. Short circuit current (SCC) was applied for 5.5 min . and the potential difference (PD) was measured for 0.5 min. during each 6 min. Unidirectional Ca fluxes were measured by adding radioactive 45Ca to either mucosal (M) Previous studies from this laboratory demonstrated that growth hormone partially pro t e c~ts against the production of restraint-stress ulcers in rats, and also promotes healing of these ulcers aftT they have been induced . The present study is a preliminary report of the use of human growth hormone (HGH)* in 6 patients with massive hemorrhage from stress ulcers. Stress ulcers were do cumented in each patient by gastroscopy. All patients bled massively requiring large volumes of blood over several days to weeks. Other potential bleeding lesions were ruled out and all patients continued to bleed massively on medical therapy, and after surgery in 3 of the patients. Neoplastic disease was present in 5/6 patients but was not the source of bleeding. HGH was administered subcutaneously to each patient in a dose of 10 mg. daily. Four of the six patients apparently stopped bleeding within 24 hours. A fifth patient stopped bleeding temporarily and then exsanguinated. A sixth patient's bleeding was slowing down and appeared to be stopping but the patient died suddenly of undetermined reasons. Repeat gastros copies in two patients showed healing of the gastri c erosions after administration of several doses of HGH. The mechanism of action of HGH in this beneficial effe ct on the bleeding from stress ulcers is not known. Our original hypothesis suggested that its a ction was based on a stimulation of DNA, RNA and protein synthesis to counteract the effe ct of stress on the gastri c mucosa. However, since some patients stopped bleeding in 24 hours after HGH it is possible that the benefi cial effect was mediated via hemostatic mechanisms. In view of the grave prognosis for patients with stress ulcers, this small experience with HGH seemed to warrant this report. In dogs, truncal vagotomy increases acid output from Heidenhain pouches in response to a meal. This increased acid output could be due either to increased serum gastrin release or to decreased release of an inhibitor which is under vagal control. Four dogs, each with Heidenhain pouch and gastric fistula, were fed 15 g/kg beef liver. Serum gastrin responses were measured by radioimmunoassay and Heidenhain pouch acid output was determined. Truncal vagotomy was performed and the studies were repeated 4 weeks later. Completeness of vagotomY was confirmed by failure of insulin to elicit an acid response from the gastric fistula. After vagotomY, Heidenhain pouch peak acid output in response to the meal increased from 0.58 to 0.90 mEq/10 min (p < 0.05). Basal serum gastrin after vagotomy (64 pg/ml) did not differ from prevagotomY (67 pg/ml). However, peak postprandial serum gastrin concentrations were lower following vagotomy (125 vs 177 pg/ml, p < 0.01) as were integrated 3 hr gastrin outputs above baseline (5.6 vs 11.2 ng-ml/3 hr, p < 0.05), indicating that vagotomy resulted in a moderate decrease in the serum gastrin response to feeding. Doseresponse studies revealed a significant increase (p < 0.05) in the calculated maximal response of the Heidenhain pouch to pentagastrin following vagotomy while the dose required to elicit a half-maximal response was unchanged. Responses to histamine were unchanged. These studies indicate that the increased Heidenhain pouch acid response to a meal following vagotomy is not due to increased antral gastrin release. Instead it appears that the pouch is more sensitive to gastrin because of the removal of a vagally-mediated noncompetitive inhibitor of gastrin. It is well known that in the carcinoid syndrome(CS) an abnormally large percentage of circulating tryptophan(T) is diverted from its normal metabolic pathway to be substrate for the increased synthesis of serotonin(SHT) and related indolic substances. A resultant decrease in blood T would be expected. This might playa part in the genesis of the carcinoid heart lesion since T deficiency is one of the prerequisites for the experimental production of a carcinoid-like heart lesion in the guinea pig. (Spatz, Lab.lnvest.13:288,1964) . In evaluation of this, measurements were made of plasma T and serum SHT levels in 11 CS patients and in a large number of normal individuals. Plasma T and serum SHT were each determined fluorimetrically after appropriate extractions. 7 CS patients had evidence of carcinoid heart disease and 4 had no evidence of any cardiac disease. Plasma T levels in 24 normal adults ranged from 10.0 to IS.S mcg./ ml.,mean 12.0 ± 1.3. In 11 CS patients plasma T levels ranged from 2.0 to 10.4 mcg.f ml.,mean 4.7 ±2.0. Our normal range for serum SHT is 0.08-0.32 mcg.fml.,mean 0.18±0.07, (previously established in S6 control subjects). Serum SilT levels in CS patients ranged from 0.42-4.00 mcg.fml.,mean 1.77 tl.04. The SilT levels did not correlate with the plasma T levels. The 7 CS patients with heart disease had T levels of 2.0-10.4 meg.fml.,mean 4.6 ± 2.0, and the 4 CS patients without heart disease had T levels of 3.2-9.8 mcg.fml.,mean S.4 ± 0.9. Serum SilT in the 7 CS patients with heart disease ranged from 0.42-4.00 meg. fml., and in the 4 CS patients without heart disease SilT ranged from 0.4S to 2.71 mcg.fml. These data support the previous reports by others that SilT levels in CS do not correlate with the presence or absence of carcinoid heart disease. In addition, it is concluded that the plasma T level is almost invariably reduced in the CS patient, but that the degree of reduction does not correlate with the presence or absence of carcinoid heart disease. This study was supported in part by The Carcinoid Tumor and Serotonin Research Foundation, Inc. Over 45 years have elapsed since Ivy clearly established that the intestine of dogs released a humoral stimulant of gastric acid secretion when appropriately stimulated. Little progress has been made since in defining the physiological characteristics of this substance. Three compounds that increase acid output are known to exist within gastrointestinal tissues: gastrin, cholecystokinin (CCK), and histamine. The following experiments were designed to test whether the intestinal stimulant of acid was physiologically similar to these substances. In the first experiment 3 dogs were prepared with chronic gallbladder fistulas, gastric fistulas, and Thiry-Vella loops of proximal jejunum. When the jejunal loops were perfused with 0.15 N HCl solution at 100 ml/hr the stomach secreted 4. 68±0. 47 mEq/15 min W but no gallbladder contraction occurred. Intestinal perfusion with 120 mM I-phenylalanine solution (to release CCK) contracted the gallbladder strongly but stimulated gastric acid weakly. In the next study serum gastrin measurements were made before and during gastric secretion stimulated by acid perfusion of the Thiry-Vella loops. Basal serum gastrin was 65±5 pg/ml while H+ output was zero; serum gastrin was 79±9 pg/ ml and W secretion was 4. 78±0. 85 mEq/15 min during stimulation. The rise is statistically insignificant and could not account for the observed secretion. In another study we found that when a maximal dose of histamine dihydrochloride 0.08 mg/kg-hr, was administered by continuous intravenous infusion, perfusion of the jejunal loop with acid increased the plateau of secretion from the gastric fistulas from 6.ll±0. 51 to 9.12±0. 72 mEq of W per 15 min. Additional histamine should not increase secretion further so histamine was probably not the acid stimulant coming from the small intestine. Therefore we conclude that the intestinal stimulant is a compound distinct from CCK, gastrin, or histamine. Radiologists have defined several constant and reproducihle landmarks in the distal esophagus. With experience and perseverance it is possihle to correlate roentgenological events in the distal esophagus with actual intralumenal contractions seen endoscopically. Movie films and repetitive sequential still pictures taken during esophagoscopy may he matched centimeter-hy-centimeter with the barium filled esophagram. The muscular "A-level" contraction at the entrance to the vestihule presents a dynamic, symmetric , contractile ring at a fixed level of the esophagus. The concentric closure of this ring gives the appearance of a rosette near the distal esophagus. The esophageal vestihule located distal to this ring, contracts by shortening and wrinkling its walls. A second rosette, previously descrihed as the distal esophageal rosette appears at the bottom of the vestibule as the walls converge to pass through the diaphragm. This rosette remains closed most of the time : distal to it, the submerged segment is visihle with the mucosal junction located within it.

A tongue of gastric mucosa is frequently seen on the posterior aspect of the esophageal side of the distal esophageal rosette.

Precise correlation of radiologic and fiber -optic endoscopic studies permit useful study of the functional anatomy of the distal esophageal closing mechanism. Movie and still photographs of these anatomical correlations will be shown. The relationship between stimulation of intrinsic nerves of the esophagus and contraction of its smooth muscle was studied in strips of opossum esophagus . Esophagi were taken from adult animals, opened lengthwise, and their mucosae removed. The distal 10-12 cm of each esophagus were cut into transverse strips 1/2 -1 cm wide. Each strip, which contained manily circular muscle, was placed in a tissue bath filled with Krebs-Ringer solution. One end of each strip was fixed and the other end attached to a force -displacement transducer. Contractions were elicited by transmural stimulation with 1 second trains of electrical pulses of 1 millisecond duration. The interval between delivery of the trains of pulses was varied from 0 . .8 to 4 seconds. The strips did not contract during the 1 second period of stimulation but did afterwards. When the interval between trains was too short, no contraction occurred until the last train was delivered . Two gradients in the responses of the strips to stimulation were observed. First, strips taken from orad portions of the esophagus responded to faster rates of stimulation than strips taken from more distal areas . Second, the time between the end of the stimulus and the beginning of the contraction increased progressively for strips taken from more distal areas of the esophagus. Measurements of membrane electrical activity and contractile responses to transmural stimulation were made with a sucrose-gap apparatus . During stimulation there was a hyperpolarization of the membrane. This was followed by a gradual depolarization to a level where spike potentials and contraction occurred. Conclusions: the peristaltic sequ~nce of contractions in the esophagus may be due partly to the gradient in delay between stimulation and contraction of the circular muscle. And the gradient in delay may be reflected in differences in the membrane electrical activity along the esophagus . (Supported by Research Grants AM 11242 and AM 08901). Using the triple lumen tube we have meas ured fat absorption rates during intraduodenal perfusion in man of a lipid mixed meal containing 6% corn oil , skimmed milk and glucose. Tracer amounts of glycerol-3H triether and 0. 5% polyethylene glycol were used as oil and aqueous dilution markers respectively. Each mea l was perfused into the mid-duodenum for 125 minutes with collections of jejunal contents made every 25 minutes from portals 15 and 55 cm distal to the site of infusion. Rates of total absorption at the distal portal during steady state conditions in normals were (± SD) 1.71 ± 0.11 mg/min/cm intestine. The absorption rate of fat correla ted positively with micellar fat concentra tion in normals, biliary depletion states and chronic pancreatitis. In the sprues micella r fatty acid was in the normal r ange but absorption rates were low (0.86 ± 0 .37). In one patient with chronic pancreatitis and steatorrhea the absorption rate and per cent lipolysis were diminished. Also, m icellar fatty acid concentration was low despite normal levels of bile salts, whereas in all other subjects studied the micellar fatty acid concentration was highly correlated with the bile salt concentra tion. These observations are consistent with t he idea that diminishing the rate of any of the steps of fat digestion may lead to l ower absorption rates. Furthermore, this lower absorptive rate i s a result of diminished micellar fatty acid secondary to decreased lipolysis or bile salt concentration. We have shown that one may examine a patient with malabsorption and determine which factors , lipolytic rate, micellar solubilization of fatty acid or cellular transfer, are responsbile for a measured diminished absorption rate in that indi vidual. It has been suggested on the basis of an increased occurrence of hepatitisassociated antigen (HAA) in some patients with hepatomas, that in various parts of the world this may reflect an etiologic factor. In Asia, HAA with hepatomas has been reported in only 3% of 114 patients in Singapore (Simons et al, Lancet 1:1149 , 1971 ) and as high as 80% of 55 patients in Taiwan (Tong et al, Ann. Intern. Med. 75:687, 1971) . No such data are available from South Vietnam. Since exposure to malaria, intestinal parasites and aspergillus products have been suggested as other possible etiologic factors in hepatomas, our preliminary studies were expanded to provide some data on these aspects. Twenty-six patients with hepatomas and 61 control subjects (14 hospitalized patients and 47 blood donors) seen in Saigon during a 4 month period form the basis for this report. The diagnosis of hepatoma was made on tissue examination in 11 cases and on clinical grounds in 15 (6/15 had alpha-fetoprotein). Of those with hepatoma, 17 were men and 9 women; 23 were Vietnamese, one Vietnamese-Chinese and 2 Chinese. HAA was determined by immunoelectro-osmophoresis, alpha-fetoprotein (AFP) by counter-immunoelectrophoresis, malaria and E. histolytica titers by indirect hemagglutination and the presence of antibodies to aspergillus by a double diffusion method. Of the 26 patients with hepatomas: 12/26 (46%) had AFP, 0/26 HAA, 12/21 (57%) positive malaria titers, 4/21 (19%) positive E. histolytica titers and 1/18 (6%) antibodies to aspergillus. Of the 61 control subjects: 0/61 had AFP, 8/61 (13%) HAA, 4/5 (80%) positive malaria titers, 12/51 (23%) positive E. histolytica titers and 6/53 (11%) antibodies to aspergillus. On the basis of these preliminary studies, patients with hepatomas in South Vietnam did not have a higher incidence of HAA, positive malaria or E. histolytica titers or antibodies to aspergillus than a control population.

A REVIEW OF 733 CASES OF PANCREATITIS FROM THREE SEATTLE HOSPITALS. T. T. White, and J. Murat, Department of Surgery, University of Washington, Seattle, Washington, 98195.

Seven hundred and thirty-three cases of acute and chronic pancreatitis were found in a search of records for 10 years from the University, Swedish, and Doctors Hospitals, Seattle. Of the 358 cases with acute pancreatitis, 192 were recorded to have gallstones. Only 24 of these patients were alcoholics, 177 had a cholecystectomy, 78 -common duct drainage for a dilated duct, 47 -sphincteroplasty, 23laparotomy only and 8 -splanchnicectomy. The remainder had medical therapy only. Of 179 patients with chronic pancreatitis, 106 were alcoholics, 54 had calcification. Primary treatment consisted of pancreaticojejunostomy for the patients with dilated ducts, splanchnicectomy for those with a small, hard pancreas without ductal dilatation, and resection where all other procedures failed. For the other 196 cases, 77 related to duodenal ulcer, 61 were postoperative, 15 to abdominal trauma, 46 related to renal failure, 34 to shock, 3 to hyperparathyroidism, 1 to hyperlipemia, and 5 to viral disease. The authors are impressed by the frequency of unrecognized gallstones and duodenal ulcer in patients with recurrent acute pancreatitis, by the part played by shock and coronary occlusions in setting off acute attacks, and by the fact that a good gastrointestinal work-up is usually not available in the average case. They also feel that any surgeon planning operation for this disease at an interval stage should envision anyone of a number of operations such as cholecystectomy, sphincteroplasty, choledochoduodenostomy, anastomosis of the pancreas to the intestine, or operation for duodenal ulcer, rather than a single one. In an effort to ascertain the effect of vagotomy on gallstone formation we have studied the bile acid pool as well as the other elements of bile in a series of 6 dogs before and after anesthesia alone, and before and after transthoracic truncal vagotomy. Chronic gallbladder fistulas were constructed first. There were no significant changes in concentrations of bile acids, cholesterol, or phospholipids or in the bile salt pool following nembutal anesthesia, only after vagotomy. The total bile acid pool increased from 1 . 9lg + 0.64 to 3.9lg + 0.62 (P< 0.001). In addition, the daily cholic acid production dropped from 0.49 +-0.28 to 0.29 + 0.05. The cholic acid rate constant decreased from 0.45 + 0.19 to 0.13 + 0.02, again a significant drop. In connection with this the cholic acid half life increased from 1.74 + 0.65 days to 5.46 + 0.82 (P< 0.01). The other interesting findings were that the ratio of cholic acid-to cholesterol decreased sharply from 95.8 + 37.22 to 46.35 + B.93 as did the ratio between cholic acid and phospholipids from-l.5l + 0.45 to 0~69 + 0.09. In addition the ratio between cholic acid and chenodeoxycholic acid was reduced from 23.98 + 7.34 to 9.83 + 3.03, a very significant change. The increase in pool size is probably secondary-to stasis and enlargement of the gallbladder. Therefore, it seems unlikely that cholesterol stones could be formed on the basis of vagotomy alone, unless stasis is the more important factor. Furthermore, the concentration of cholic acid increased at a greater rate than the volume of bile. The changes in ratios between cholic acid and the other two bile acids suggested that the latter were being held in the liver cells due to some change in membrane permeability or ability of the cells to extrude these acids. Perhaps, if dividing the vagus reduces secretin or other hormone output, the lack of choleretic effect of this hormone might account for these changes. During the past 20 years the author has operated personally on 86 patients with chronic pancreatitis (and 145 with acute or recurrent acute pancreatitis). A total of 39 sphincterotomies, 50 pancreaticojejunostomies, 37 cholecystectomies, 21 choledochoenterostomies, 22 splanchnicectomies, 5 95% left-to-right resections, and 4 pancreaticoduodenectomies were performed by him on these patients. Of the 85 patients, 66 were alcoholics, 1 developed chronic pancreatitis second to trauma, 2 were second to unrecognized retroperitoneal sarcomas, 3 due to congenital obstruction of the pancreatic duct, and 10 had no apparent connection to other disease. There was no apparent instance of hyperlipemia . The author feels that any patient with chronic pain which has warranted 2 or more admissions to hospital should be explored with the idea of performing a pancreaticojejunostomy if the patient has a duct dilated over 5 rom in diameter, decompression of the bile duct if necessary, a splanchnicectomy if the pancreatic ductal system is not dilated and the pancreas is small, hard, and painful, and resection if all else fails. Sphincterotomy has not been effective in treating patients with alcoholic or other chronic pancreatitis in the author's experience, but approximately 35% of patients undergoing splanchnicectomy are relieved of their pain for 5 years or more. FGrty-two of 50 patients with dilated duct on whom a 10 em long or lon~ er pancreaticojejunostomy was carried out were relieved of pain postoperatively for 2 to 16 years. Three required a subsequent splanchnicectomy and 3 required a subsequent 95% resection. Sphincterotomy was effective in only 1 of these patients, a patient who had a calcified cyst in the head of his pancreas adjacent to the sphincter of Oddi. The author feels that all patients undergoing surgery for chronic pancreatitis should re prepared for all of the above described operations, and the appropriate ones decided upon on the basis of clinical findings. No single operation appears to be appropriate for this problem, but one or more in combination. We prefer, however, to reserve pancreatic resection for cases where the splanchnicectomy and pancreaticojejunostomy have proven to be ineffective. Indications for specific operations will be discussed. Studies of the morphologi cal and fun ctional status of canine jejunal allografts have suggested that multiple sequential biopsies and histologi c evaluation is the best method of monitoring the reject ion process in such grafts. However, a more sensitive method would be useful if it provided earlier warning of re j e ction. Disaccharidase activities (U/gm wet wt) were assayed in biopsies obtained from canine jejunal allografts to determine whether changes in these enzymes preceded histological abnormality . Sequent ial assays were ' performed in 2 control dogs having Thiry-Vella loops constructed, 1 with and 1 without immunosuppression; and in 5 dogs with je,iunal allograft s, 4 with and 1 without immunosuppression . Sucrase, maltase and lactase activit ies were moderately stable in all control animals after surgery. Allografts showed a decrease in activities of all enzymes as re,ie ction proceeded. However, ,/ariability was seen among the specimens and temporally a single value was unreliable. The decreasing trend was most consistent for lactase. Changes in enzyme activities usually did not occur until some morphological abnormalities in the lamina propria were present. Changes in enzyme a ctivity usually occurred before villus epithelial cell abnormalities were seen . A representative study is: These studies demonstrate that disaccharidase activities decrease during rejection and usually follow morphological changes of early re ject ion but may pre cede the morphological indi cators of advanced rejection. Conclusion: Disaccharidase assays are less sensitive than morphologi cal observations in monitoring rejection but could be of value as a supplement to the morphological observat i ons. The mode of action of cholecystokinin (CCK), octapeptide fragment of CCK, caerulein and pentagastrin on small intestinal motor activity was investigated and their relative potency determined. Isolated segments of guinea pig ileum were bathed in aerated chambers containing Kreb's solution and maintained at 37°C. Isometric mechanical responses were recorded by force transducers connected to a Grass polygraph. The following results were obtained: 1) The muscle developed tolerance after the initial stimulus even when tested after 30-45 minutes with the same dose . 2) Single doseresponse studies indicated that the responses elicited by 1. 31 x 10-7M octa-CCK, 1.48 x 10-7M caerulein, 3.35 x 10-7M CCK and 7.80 x 10-5M pentagastrin were similar. 3) Cholinergic blockade (6 x 10-6M scopolamine) reduced comparable responses to both urecholine (3 x 10-6M) and the peptides (1.31 x 10-7M octa-CCK, 1.48 x 10-7M caerulein, 3.35 x 10-7M CCK and 7.80 x 10-5M pentagastrin). The inhibition was greater for the urecholine response (75%) than the peptide response (50%). The difference between these degrees of inhibition by scopolamine was significant (P<0.05). 4) Tissues aged in the cold for 24, 48 and 72 hours before experimentation were studied for their responses to the peptides. The data showed a progressive decrease in sensitivity to hormones with time. At the end of 72 hours, no response to the peptides could be obtained while the tissue was still capable of responding to urecholine. 5) Treatment of tissues with tetrodotoxin (3 x lO-7M) completely abolished the effect of all four peptides while the response to urecholine remained unchanged. 6) Alpha and beta blockers failed to block the hormonal effects on the intestine. We conclude that all four peptides were capable of contracting intestinal muscle. The fact that the effects of CCK and related analogues were altered by scopolamine suggests the possible operation of excitatory cholinergic comp~nents. Tetrodotoxin and cold ~t~rage d~ta further supported this view. It seems l1kely, therefore, that CCK part1c1~ates 1n feedback control over the intricate networks modulating the neural regulat10n of gut motility. Phenobarbital induces a rise in 7 n-hydroxylase activity in Wistar strain rats (E. H. Mosbach). Daily administration of phenobarbital to Rhesus monkeys increased the biliary secretion of bile salts (BS) and decreased that of cholesterol (C) indicating an increased conversion of hepatic cholesterol to bile acids (Redinger and Small). It has been suggested, therefore, that phenobarbital or some other inducers of 7 a.-hydroxylase may be useful in cholelithiasis. We have looked for similar effects of phenobarbital in mice. Standard methods for the determination of bile salts, cholesterol and lecithin have been adapted to permit analysis of gallbladder bile from a single mouse. Male mice treated with phenobarbital sodium, 30 or 75 mg/kg, i.p., for 3 to 14 days had increased BS/C ratios in gallbladder bile.

However, chronic administration of the drug in the diet (0.02%) for 3 weeks resulted in an increased level of cholesterol in bile and a corresponding decrease of BS/C (p < 0.05). The results provide indirect evidence that 7 n-hydroxylase can be induced by phenobarbital in the mouse. But evidently other factors can come into play during chronic drug administration to reverse the early effects, possibly a later induction of other enzymes involved in lipid metabolism (e.g., HMG-CoA reductase), and this raises some doubt as to the ultimate use fullness of general enzyme inducers in cholelithiasis. Hexose absorption is increased in the alloxan diabetic rat. Since hexose and disaccharide transport are interrelated, and most of the hexose absorbed is derived from dietary disaccharide and polysaccharide, we studied the effect of diabetes on intestinal disaccharidases. We compared matched control (C) and 8 day diabetic rats (D). The intestine was examined as mucosal and underlying tissue fractions of the proximal, mid and distal thirds of the small intestine. In both C and D rats more than three quarters of the disaccharidase activity was in the mucosal scrapings. Total intestinal activities in diabetes (pmoles hydrolyzed/min) were nearly doubled for lactase (C, 4.3~0.5; D, 7.0~0.5; p<O.Ol) and maltase (C, 85~4; D, 180~9; p<O.Ol) and tripled for sucrase (C, 17.5+1.4; D, 54.4+2.8; p<O.Ol). Since intestinal growth was only 17% greater in the diabetic than the control group, the increase in disaccharidases was chiefly in their specific activities (pmoles disaccharide hydrolyzed/min per g protein). Lactase specific activity was increased chiefly in the proximal segment. Maltase and sucrase specific activities were increased in all three segments in proportion to the activities originally present. Thus, stimulation by diabetes maintained the normal longitudinal gradient of high disaccharidase activity in proximal and mid intestine. This is also the normal longitudinal gradient for rate of hexose absorption. Thus, diabetes has parallel effects on hexose absorption and disaccharidases: the degree of stimulation is the same; the longitudinal pattern of disaccharidase stimulation follows the normal pattern for rate of hexose absorption along the length of the intestine.

Since the vitamin 0 (D) and calcium (Ca) status of the animal determines the pathways of 0 metabolism (Boyle, et al Proc Nat Acad Sci 62: 2131 , 1971 , this status would also be expected to influence the effects of vitamin on intestinal Ca transport. Most studies of the effects of 0 on Ca absorption have been in 0 depleted rats where the vitamin is converted chiefly to the 1 ,25-dihydroxy derivative, the metabolite active in the intestine (Frolik, et al, Arch Biochem Biophys 147:143, 1971 ). In depleted rats we found 0 increased net intestinal Ca absorption and its 1umen-to-plasma flux, but had no effect on flux of Ca into the lumen (Younoszai, et al Gastroenterol 60:810, 1971 ). The effects of vitamin 0 on Ca transport in normal rats, where the main metabolite is the 21,25-dihydroxy derivative, have not been examined. We studied intestinal plasma-to-lumen flux and net absorption of Ca in normal control (C) and in normal rats injected with 0 (+0). Forty-eight hr before study 10,000 U of 02 was given to 13 rats in a matched group of 26. Twenty-four hr before study all rats were loaded with 45Ca given parenterally. The duodenum (duo) and ileum (il) were perfused with 3.4 mM calcium. +0 rats showed significantly decreased flux of 45Ca and Ca (~moles in 0.5 hr/g mucosa wet wt) into the lumen: duo, C=0.60+0.03, +0 =0.51~0.02; p<0.05; il, C=l . 10~0.06, +0=0.86~0.05; p<0.05, but no si9nificant effect on net Ca absorption (~moles in 0.5 hr/g mucosa wet wt): duo, C=10.3+0.8, +0=12.0+1.0; il, C=7.5+0.8, +0=9.1+0.7. These studies demonstrate-effects of vTtamin 0 on intestinal Ca transport that appear to be specific for the metabolic state of the animal: the vitamin decreases flux of calcium into the lumen in normal rats but not in depleted rats where its major action is to increase net absorption, an effect not seen in normal animals. This may reflect differences in vitamin 0 metabolism. ON Esophageal manometry frequently employes perfused multi-catheter assemblies and pressure measurements are obtained, through side openings, at locations 5 cm apart.

Often there is great variability between pressures measured at one location by the different catheters. Previous investigators have shown that pressures measured simultaneously through a three lumen catheter with openings at the same location are not identical (Gastroenterology, 61: 213, 1971 ). As part of a continuing study on the mechanics of the esophagus and the methodology of esophageal manometry we recorded intraluminal pressures, in the body of the esophagus and in the high pressure zones, by means of a three catheter assembly (inner diameter=1.7 rom) with side openings at the same level. In the 20 subjects studied there was great variability between the three recording channels, such that the average difference between the highest and lowest pressure, at the same location and at the same time, was 30% of the mean of the three readings. This variability was noted both in normal subjects and in patients with altered esophageal manometry. Since the variability could be due to artifacts of pressure transmission along the catheter assembly, a systematic study of such factors was performed. We found that the pressure measurements depend only on the flow rate of the perfusing fluid; i.e. the perfusion rate must be high enough to record rapid pressure changes, and low enough not to alter the baseline through viscous effects. The perfusion rate can be calculated by a formula relating the deformability (D) of the system, the rate of flow (F) of the perfusing fluid, the maximal intraluminal pressure (p) and the time of contraction(t). This formula, F=DxP/ t, determines the proper rate of flow in any catheter system. The rise in baseline caused by viscous effects can be estimated by the Poiseuille equation, and is related to the length and diameter of the catheters. With the catheters of our system the pressure recorded by the transducers will be within 1 rom Hg of the pressure at the openings.

We conclude that differences in pressure measured in catheter assemblies with side openings ·at the same level, are not due to artifacts of pressure transmission along the assembly, but to actual differences of pressures occurring at the side openings. Proteolysis was estimated by the solid substrate digestion technic of Cammarata. Patients ingested two steel wire mesh capsules containing a weighted quantity of heat coagulated albumin attached to a small bore tube. Samples of gastric juice were aspirated and proteolytic activity was estimated by the hemoglobin assay. After three hours, the capsules were recovered , washed, lyophilized and reweighed to determined the quantity of substrate digested. Proteolysis was expressed as percent by weight of substrate digested In three hours. The second capsule was exposed to a 5% solution of sulfated amylopectin pepsin inhibitor (SN-263) and washed extensively prior to Ingestion. A small quantity of pepsin inhibitor, 6% by weight, remains tightly bound to the solid albumin substrate. The inhibitory effect of this substrate bound material was studied. Patients with endoscopically proven active duodenal or gastric ulcer were studied and compared to hospitalized patients without gastrointestinal tract disease and student volunteer controls. Duodenal ulcer patients (n=ll) had significantly greater gastric proteolysis 33~19% (mean~S.D.) than patient controls (n=ll) 21t19% (P<0.1). Proteolysis in gastric ulcer patients (n=5) 2J±22 was in an intermediate range between duodenal ulcer and control patients. Student volunteer controls had a greater proteolysis (n=IO) 41t25% than duodenal ulcer patients. These data suggest the presence of Increased gastric proteolysis In active duodenal and gastric ulcer patients, but emphasize the need for matched control groups . In vivo substrate Inhibition of proteolysis by sulfated amylopectin was demonstrated by the marked inhibition of proteolysis In ulcer patients (53~19% inhibition) and in patient controls (76t 22% Inhibition) . Two pernicious anemia patients with achlorhydria demonstrated 15% gastric proteolysis not inhibited by SN-263 that presumably is non-peptic In origin. Comparison of solid substrate and hemoglobin methods of measuring proteolysis reveals a marked underestimation of proteolytic activity by the hemoglobin method. This appear to result from the production of pepsin Inhibitors during in vitro proteolysis.

AUTOMATED ASCITES ULTRAFILTRATION-REINFUSION. D. S. Zimmon. J. Beckwith and F. Iber. V.A. Hospital, New York, New York, and Lemuel Shattuck Hospital, Boston Massachusetts.

The response of 11 cirrhotic patients with diuretic resistant ascites to continuous ascites ultraflltratlon-relnfusion was studied using a new automated fall-safe device. During constant monitoring of body weight. urine output, central venous pressure and abdominal girth, ascites was continuously aspirated at controlled rates up to 1200 ml/hr . An ultraflltering module removed crystalloids . The concentrated proteins were reinfused intravenously. The volume of ascites removed ranged from 4 to 31 liters per procedure. Protein concentration of the relnfusate could be Increased 2 to 3 fold by varying the back pressure In the ultraflltering module. Similarly, ultraflltrate/relnfuslon volume ratios can be varied from III to 2/1. Volumes reinfused ranged from 2 to 15 liters at rates from 300 to 600 ml/hr . These intravascular volume loads generated spontaneous or diuretic assisted diuresis . Example : In a cirrhotic patient (BUN 54, creatinine 3.1) with an initial ascites volume measured wi th I-131 albumin of 24 liters 29 hou r ultraflltration-reinfuslon removed 31 liters of ascites with 16 liters ultrafiltered and 15 liters reinfused. In response to furosamlde urine volume Increased to 10 liters with a net weight loss of 56 lbs. This initial experience demonstrates the mechanical safety and efficiency of the device. It is clinical useful in obviating the compartmental izatlon of ascites and may have particular value in the management of tense ascites, ascites following abdominal surgery, ascites with high protein concentrations or diuretic resistant ascites .

Department of Medicine

AlcohOl (A), tolbutamide (T), and meprobamate (M) were given to alcoholic patients to PHYSIOLOGICAL EFFECTS OF AN ELEMENTAL DIET IN

Program in Physiology, The University of Texas Medical School at Houston

Fecal excretion of l4c-cholate was studied 32 times (16-day studies) in 22 normal paid subjects by the method of Stanley and Nemchausky (J. Lab. Clin. Med. 70: 627, 1967) . Each subject ate a constant diet (2-day rotation) closely resembling his usual fare. During third and fourth (4-day) periods effects of medications, randomly assigned, administered double-blind, were compared to those of methyl cellulose administered during the second period. Methyl cellulose does not increase bile salt excretion in fasting subjects (Stanley, Metabolism 19: 865, 1970 ) but increases stool mass as do the comparison medications. The absolute excretion rate during period 2 (=0.10 ± 0.007 SE pools / day) is designated as 1.0; absolute excretion rates during periods 3 and 4 are expressed as multiples of this control rate:(A) Methyl cellulose (N =32) 1.0 ± 0.07 S.E. (B) Cholestyramine (N =11) 6.8 ± 0.6; t = 9.43, P<O.OOI (C) Metamucil (N =11) 1.7 ± 0. Severe diarrhea associated with acute shigellosis has considerable importance for international health. The pathophysiology of the fluid loss, however, is poorly understood. Recently, a potent exotoxin has been isolated from a strain of Shigella dysenteriae associated with epidemic diarrhea in Central America and found to have both cytotoxic and enterotoxic properties. Using Shigella enterotoxin (S.E.) the secretory response of in vivo rabbit jejunal loops has been studied by a recirculatory perfusion technique with P.S . P. as a volume marker. S.E. in concentrations (I-SOD ug/ml) were included in the isotonic equilibrium perfusion solution. Serial samples (0.5 ml) from the reservoir (20 ml) were taken at 30-60 minute intervals for 3-9 hr periods. A control loop in each animal was perfused with isotonic solution alone . Another group of animals were perfused in a similar manner with cholera toxin (C.T.). At levels 200 ug/m! S. E. induced consistent fluid secretion which was similar in composition to C.T. induced fluid. S.E. induced rates of secretion (+0.194 ml/cm/hr) were similar to C.T. during the period of linear response. However, S.E. secretion differed from C.T. in the greater latency of response and shorter duration of effect. No response to S.E. was observed until I hr and significant fluid secretion only developed after 2 1/2 hrs exposure. This study was done to determine the effect of surgical procedures on fasting and postprandial serum gastrin concentrations in patients with duodenal ulcer disease. The following groups of patients were fed a standard meal of eggs, toast, and orange juice: nonu1cer; active duodenal ulcer (DU); previous vagotomy and pyloroplasty (V&P); previous subtotal gastrectomy and vagotomy with either Bi11roth I gastroduodenostomy (BI) or Bi11roth II gastrojejunostomy (BII). Each postoperative subject was tested at least 6 weeks after surgery. The size of the gastric remnant and absence of gastric obstruction were determined by X-ray. Venous blood was drawn during a 30 min basal period and at graduated intervals for 2 hr after feeding. Serum gastrin was measured by radioimmunoassay. Basal and peak stimulated serum gastrin values and integrated 2 hr gastrin output above baseline were: (Mean ± SE) When compared with nonu1cer or with DU patients, BII patients had lower fasting (p < 0.01) and peak stimulated (p < 0.01) gastrin. BII patients also had lower fasting (p < 0.05) and peak (p < 0.01) gastrin values than BI patients. There was no significant difference in basal and peak values between BI and DU or nonu1cer patients. V&P patients had higher fasting and postprandial gastrin than patients in any other category. Conclusions: Vagotomy did not diminish and may have enhanced the gastri n response to feedi ng inman ;gastri n responses to feedi ng after antrectomy were retained when food entered the duodenum but were markedly depressed when food passed directly into the jejunum. The release of duodenal gastrin in response to feeding may be of importance in man. The finding of significant levels of circulating carcinoembryonic antigen (CEA) in patients with Crohn's disease and ulcerative colitis, along with recent data suggesting organ-specific cellular reactivity in ulcerative colitis (Bendixen, Gut 10:631-636 1969) , prompted this study of in vitro leucocyte response to meconium antigen. The reconium antigen used here produced lines of identity with purified CEA obtained from Dr. Phil Gold, and with saline extracts of colonic cancer when precipitated in Ochterlony gell double diffusion plates with mono-specific anti-CEA antibody also obtained from Dr. Gold. The antigen was used in protein concentration of 500 ugm/ml and contained 9 ng/ml of CEA as measured by radioimmunoassay. This antigen also contained small amounts of albumin, gamm~ globulin, blood group substances M and N, and normal colon antigens. The leucocyte migration technique was employed to test the response of peripheral leucocytes from ten normal individuals, five patients with cancer of the colon, nine patients with Crohn's disease, and seven patients with ulcerative colitis. fhe migration of leucocytes from normal controls and from patients with carcinoma of the colon was not inhibited. Consistent inhibition of migration was found in all patients with Crohn's disease, while inhibition of migration was found in only two patients with ulcerative colitis. The component of meconium responsible for this consistent difference in pattern of response is not yet known. These studies, however, suggest a method of separating Crohn' s disease and ulcerative coli tis by immunologic means.

LIVER. Edward C. Wilson, Dept. Med . , Univ. of Va. School of Med., Charlottesville, Va.One of the more striking effects of ethanol on lipid metabolism is its suppression of the release of triglycerides from the perfused liver of the rat. Little is known of the effect of ethanol on the hepatic release of other constituents of lipoproteins. This study was conducted to define the acute effects of ethanol at various concentrations on the secretion of phospholipids and triglycerides from the isolated, perfused liver of the rat. Female rats, fed ad libitum, served as the liver donors. The livers were perfused with a mixture of synthetic plasma and washed rat red blood cells to which had been added palmitate-l-C14, as the free fatty acid, which served as the substrate for hepatic lipid synthesis. The rates of release of triglycerides and phospholipids from the livers were defined by determining the concentrations of these lipids in the perfusate after 60, 120 and 180 minutes of perfusion. In the experimental perfusions ethanol was added to the perfusate in amounts sufficient to establish initial concentrations of either 285 or 395 mg. ethanol per 100 ml perfusate. The control perfusions were without additions. At the lower concentration of ethanol only the release of triglycerides from the liver was depressed. At the higher concentration ethanol inhibited the release of phospholipids and triglycerides from the perfused liver. These data suggest that ethanol adversely effects hepatic lipoprotein synthesis and secretion but in different ways depending on the concentration of the drug existent in the perfusate. It appears likely that at the lower concentration the synthesis and/or secretion of triglyceride-rich very low density lipoproteins is mainly suppressed, while at higher concontrations the secretion of high-density lipoproteins is also impaired.MANOMETRIC ASYMMETRY OF THE LOWER ESOPHAGEAL HIGH PRESSURE ZONE. C.S. Winans. Department of Medicine, The University of Chicago, Chicago, Illinois.To assess the relation of orifice orientation to recorded pressure within the lower esophageal high pressure zone (LEHPZ) and to determine the structures responsible for this zone, studies were performed on 8 normal subjects, 8 patients with hiatus hernia and one patient with advanced progressive systemic sclerosis with free gastroesophageal reflux and an aperista1tic distal esophagus. Pressures were detected with a special 8-1umen polyvinyl recording catheter, the lateral orifices of which were uniformly spaced at 45-intervals about the circumference of the catheter at the same axial level. Resting pressures were recorded at each station as the catheter was withdrawn by 1 cm increments from stomach to esophagus. End-inspiratory and endexpiratory pressures recorded from each orifice were tabulated relative to the corresponding intragastric pressure. Similar pressures were recorded from all orifices when they were located within stomach or esophageal body. However, within the LEHPZ of normal subjects recorded pressure was significantly greater from orifices directed to the patients' left (36 mm Hg) and left posterior (34 mm Hg) than from those directed in the other 6 directions (14.5 to 23.5 mm Hg). Mean maximal end-inspiratory and end-expiratory pressures for each orifice position were nearly identical, but maximal pressures were recorded 1 cm more distal during inspiration. For the hernia patients greatest pressures were also recorded from the orifices directed in the left (19.5 mm Hg) and left posterior (18 mm Hg) directions, but these pressures did not differ significantly from those recorded from the other 6 orifice positions (8 to 13.5 mm Hg~ Study of the scleroderma patient revealed pressure elevations from the left and left posterior orifices similar in magnitude to those of normal subjects, but no detectable pressure elevation from other orifice positions. It is concluded that pressures recorded from within the LEHPZ depend upon the radial orientation of the recording orifice within the lumen. The predominance of pressure recorded from the left posterolateral direction, even in a scleroderma patient without intrinsic sphincter activity, suggests that in normals an extrinsic force from this direction, possibly due to the lateral margin of the diaphragmatic hiatus, augments the pressure resulting from the contraction of an intrinsic lower esophageal sphincter. A three dimensional rec·onstruction of the pancreatic ducts has been made from the secondary branches of the main duct to the terminal acini. This has been carried out by cutting serial sections from surgical specimens of pancreas from 10 patients with alcoholic calcifYing chronic pancreatitis and specimens from two patients with a small carcinoma obstructing the ~in pancreatic duct. A mean of 800 sections of be~ ween 7.5 and 12 jU were cut from each specimen. The lesions of calcifying pancreatitis demonstrated by this method were characteristic but poorly demonstrated on single sections. The main features of the lesions were : 1 -The spotty distribution of the lesions which were related to the ducts; 2 -The frequent finding in the ducts of plugs of precipi tated protein, which tended to evolve into calculi by deposition of calcium on thei r surface. These plugs rarely contained muc opolysaccharide ; 3 -All sizes of duct were involved and showed irregular dilatation, loss of lining epithelium, encroachment of periductalcOOf\l:C.bvc. tissue to form stenosis, cystic distension or complete disappearance ~ acini and ducts in the territory drained by a blocked duct. Consideration of these lesions suggests that the initiating lesion in this condition is protein precipitation which forms plugs which,in turn, calcify and obstruct the ducts and. lead to the other lesions by increasing the pressure in the ducts above the obstruction. In chronic pancreatitis secondary to an obstruction in the main pancreatic duct, the lesions are different in almost every respect to those described above.