Robert Werkman

A fistula from the portal vein to the bile duct: an unusual complication of transjugular intrahepatic portosystemic shunt

We describe a case of a portal vein bile duct fistula as a complication of transjugular intrahepatic portosystemic shunt (TIPS) placement. The patients course was complicated by endocarditis, hemobilia, recurrent episodes of fever, and bacteremia, followed by liver transplant. The findings of fever, bacteremia (especially with Gram-negative organisms), and a decreased hematocrit after shunt placement should raise the suspicion of an infected shunt with a possible fistula.

Assessment of gastric electrical activity and autonomic function among diabetic and nondiabetic patients with symptoms of gastroesophageal reflux.

Gastroesophageal reflux disease (GERD) may present differently in patients with diabetes mellitus (DM) than in nondiabetics (NDM). We compared three tests in two patient groups with GERD symptoms: a DM group (n = 10) and a NDM group (n = 13). The tests were 24-hr esophageal pH, autonomic function testing (AFT), and electrogastrography (EGG). Analysis of the 23 patients revealed the DM group had normal 24-hr pH values (9 of 10 patients, mean pH 3.1 ± 1.7), while NDM displayed abnormal pH values (9 of 13 patients, mean pH 21.2 ± 5.9). AFT results were abnormal in DM (demonstrating cholinergic/adrenergic dysfunction), but normal in NDM. EGG values were abnormal in both groups (mean 3.31 ± 0.1 in each). We conclude that in GERD-symptomatic patients, those with DM frequently have normal 24-hr pH, but abnormal autonomic functioning, in contrast to NDM, who have abnormal 24-hr pH but normal autonomic function. Both groups had identically abnormal mean EGG values.

Sequential Group Trial to Determine Gastrointestinal Site of Absorption and Systemic Exposure of Azathioprine

Azathioprine (AZA) is used in the treatment of patients with refractory inflammatory bowel disease; however, its use is limited because of systemic toxicity associated with long-term use. Ileocecal delivery of AZA might be advantageous if local intestinal therapeutic effects could be provided with decreased systemic side effects. Decreased cecal systemic absorption would allow higher dosages of AZA to be administered. A two-phase study was performed to compare the systemic exposure of AZA and 6-mercaptopurine (6-MP) following administration of AZA into the stomach, jejunum, and cecum and to compare the systemic exposure to AZA and 6-MP following administration of three different dosages of AZA into the cecum. In phase I, six healthy male volunteers received three 50 mg sequential doses of AZA via an oral tube directly placed into the stomach, jejunum, and cecum, respectively. In phase II, six healthy male volunteers received three different dosages (50, 300, 600 mg of AZA) into the cecum. Plasma concentrations of AZA and 6-MP at various times were quantified and area under the plasma concentration-time curve (AUC) and mean residence time (MRT) were determined. No significant differences in the AUC of AZA were seen at the different sites. The AUC of 6-MP following administration of AZA into the jejunum (67.0 ± 30.1 ng×hr/ml) was higher compared to the stomach (39.9 ± 38.1 ng/hr/ml) and cecum (29.2 ± 10.9 ng×hr/ml). Jejunal absorption was 68% higher than absorption from the stomach and 129% higher than that of the cecum. Gastric absorption was 27% higher than that of the cecum. Increased dosages given into the cecum resulted in increased AUCs of AZA and 6-MP. The AUCs of AZA following 50, 300, and 600 mg dosages were 16.9 ± 7.4, 52.3 ± 67.2, and 132 ± 151 ng×hr/ml, respectively, and the AUCs of 6-MP were 22.2 ± 14.9, 63.4 ± 50.6, and 104 ± 115 ng×hr/ml, respectively. Systemic exposure to 6-MP is reduced following administration of AZA into the cecum, most likely secondary to reduced absorption of 6-MP from the colon. Higher dosages of AZA presented to the cecum do result in increased systemic absorption, but may still allow more drug to be administered with less toxicity than the same dose received orally.

Biliary, Pancreatic, and Sphincter of Oddi Electrical and Mechanical Signals Recorded During ERCP

Measurements of biliary tract motility havefocused on radiologic and pressure measurements toquantify biliary motility rather than measurements ofelectrical activity of the biliary tract. We previously reported the recording of biliary electricalsignals during ERCP and now report on the continueddevelopment and validation of a system to measurebiliary tract electrical activity as well as biliarymechanical activity. In 26 patients presenting with avariety of clinical indications, we recordedmeasurements of electrical activity from the common bileduct sphincter (16 patients), pancreatic duct sphincter(eight patients), and/or sphincter of Oddi (eightpatients). Electrical recordings were performed with aspecially modified ERCP catheter, using two circularelectrodes as well as a custom catheter that measured both electrical and mechanical activity.Electrical activity of the biliary tract wassuccessfully recorded in 25 of 26 patients (96%),including the common bile duct sphincter (16 patients,62%), pancreatic duct sphincter (eight patients, 31%) andsphincter of Oddi (eight patients, 31%). Along with theelectrical recordings, common bile duct sphinctermechanical activity was recorded in 12 patients (67%), pancreatic duct sphincter mechanical activityin six patients (33%), and sphincter of Oddi mechanicalactivity in six patients (33%). Frequency analysis ofelectrical signals revealed a mean frequency(cycles/min) of 4.7 ± 0.5 in the common bile ductsphincter, 4.1 ± 0.6 in the pancreatic ductsphincter, and 4.9 ± 0.7 in the sphincter ofOddi. Phasic mechanical frequency in cycles per minutewas recorded at a frequency of 4.8 ± 0.5 in common bileduct sphincter, 4.0 ± 0.6 in pancreatic ductsphincter, and 5.3 ± 0.9 in sphincter of Oddi.Tonic pressure (averaged 12.1 ± 1.5 mm Hg) incommon bile duct sphincter, 12.4 ± 1.4 mm Hg inpancreatic duct sphincter, and 15.0 ± 5.1 mm Hgin sphincter of Oddi. Analysis of wave form propagations(noted as percentage antegrade, retrograde, orindeterminant) revealed 50% antegrade, 23% retrograde, and 27%indeterminant). One patient was recorded on twooccasions via ERCP; the same patient had anintraoperative recording. All three recordings showedsimilarities. We conclude that measurements of biliary,pancreatic, and sphincter of Oddi electrical andmechanical activity are feasible and can be done as partof ERCP. There was good correlation between biliarytract electrical and mechanical events and differentwave form characteristics were noted for different partsof the biliary tree. Further studies are warranted toevaluate the potential usefulness of measurement of biliary tract electrical activity, and toconfirm its correlation with mechanical events in thepancreato-biliary tree.

Comparison of azathioprine (AZA) and 6-mercaptopurine (6-MP) plasma concentrations after administration of aza solution into the stomach jejunum, or cecum

AZA is used in the treatment of patients with refractory inflammatory bowel disease. Its use, however, has been limited by its delayed onset of action and toxicity associated with long term use. Van Os et ai (Gut 1996) reported reduced bioavailability of 6-MP after colonic administration of azathioprine, suggesting the possibility that if a local therapeutic effect occurs, this may result in reduced systemic side effects. The purpose of this study was to compare the systemic exposure to 6-MP following administration of AZA to different absorption sites. Six healthy male volunteers who had normal thiopurine methyltransferase activity, received three doses of Imuran (50mg injectable) via an oral-enteric tube directly into the stomach (S), jejunum (J), or cecum (C), separated by 24 hours. Plasma concentrations of AZA and 6-MP at various times were quantified by HPLC; area under the plasma concentration, time curve (AUC) and mean residence time (MRT) were determined by statistical moment theory. Mean -+ SD results were: