Isosorbide dinitrate, a long-acting organic nitrate, has been shown to decrease portal pressure in the experimental animal and humans. We conducted a double-blind randomized hemodynamic evaluation of the effects of placebo and 10 mg and 20 mg isosorbide dinitrate in stable individuals with alcoholic cirrhosis. Baseline values for all three groups were similar. Isosorbide dinitrate resulted in a peak reduction of the hepatic venous gradient of 24.7% +/- 3.0%, with significantly decreased values 4 h after the administration of the 20-mg dose. A reduction of arterial pressure and cardiac index (peak decrease of 25.7% +/- 1.5%) was well tolerated by 13 of 15 patients. Changes in mean arterial pressure were not predictive of modifications in the hepatic vein wedge pressure. There was no relation between the area under the plasma isosorbide dinitrate concentration curve and hemodynamic changes. Levels of isosorbide-5-mononitrate, a vasoactive metabolite, were detectable for an 8-h period. Isosorbide dinitrate significantly reduced portal pressure in stable cirrhotics, in association with systemic hemodynamic changes. Thus, titration of isosorbide dinitrate is required to maximize hemodynamic benefits in individual patients. As the decrease in portal pressure is more predictable than the effect of previously tested pharmacologic agents, isosorbide dinitrate should be evaluated for its efficacy in the management of portal hypertension.

To investigate whether paracentesis could be an alternative therapy for ascites, 117 cirrhotics with tense ascites were randomly allocated into two groups. Fifty-eight patients (group 1) were treated with paracentesis (4-6 L/day until disappearance of ascites) and intravenous albumin infusion (40 g after each tap). Fifty-nine patients (group 2) were treated with spironolactone (200-400 mg/day) plus furosemide (40-240 mg/day). Patients from group 2 not responding to diuretics were treated with a LeVeen shunt. After disappearance of ascites, patients from both groups were discharged from hospital and were instructed to take diuretics. Patients developing tense ascites during follow-up were readmitted to hospital and treated according to their initial schedule. Paracentesis was effective in eliminating the ascites in 56 patients from group 1 (96.5%) and did not induce significant changes in renal and hepatic function, plasma volume, cardiac index, peripheral resistance, plasma renin activity, plasma norepinephrine and antidiuretic hormone concentration, and urinary excretion of prostaglandin E2 and 6-keto-prostaglandin F1 alpha. Diuretics were effective in eliminating the ascites in 43 patients from group 2 (72.8%) (p less than 0.05). Ten patients in group 1 and 36 in group 2 developed complications during their first hospital stay (p less than 0.001). This difference was due to the significantly higher incidence of hepatic encephalopathy, renal impairment, and electrolyte disturbances occurring in patients treated with diuretics. The duration of hospital stay was 11.7 +/- 1.5 days for patients from group 1 and 31 +/- 2.8 days for patients from group 2 (p less than 0.001). The two groups did not differ significantly with respect to the probability of requiring readmission to hospital during follow-up, reasons for readmission, survival probability after entry into the study, and causes of death. These results indicate that paracentesis associated with intravenous albumin infusion is a fast, effective, and safe therapy for ascites in patients with cirrhosis.

The effects of cyclooxygenase inhibition by indomethacin on gastric acid secretion were studied in 8 healthy men. Oral doses of indomethacin (200 mg), administered 15 and 2 h before testing, were known to inhibit prostaglandin synthesis by 90% in 3 of the subjects as determined by prostaglandin E2 generation assay on endoscopically obtained gastric mucosal biopsy specimens. Acid-induced inhibition of gastric acid secretion was evaluated in a randomized and blinded study in which acid output was measured for 2 h during basal conditions by aspiration, for the next 2 h by intragastric titration during distention with isotonic glucose, and for the following 2 h by intragastric titration during meal stimulation with peptone. The studies were done on separate days, and intragastric pH was maintained at either 2.5 or 5.5 after administration of indomethacin or placebo. Basal acid output was not altered by indomethacin treatment. Distention of the stomach stimulated acid output significantly to a similar degree in all groups, without affecting plasma gastrin. Meal stimulation increased plasma gastrin and acid output significantly more at pH 5.5 (47 +/- 12 pM, 13 +/- 2 mmol/30 min) than at pH 2.5 (30 +/- 8 pM, 6 +/- 2 mmol/30 min). No effect of indomethacin treatment was observed. It is concluded that the participation of cyclooxygenase products in the mechanisms by which acid inhibits the gastric phase of acid secretion in humans is likely to be minor. These results also cast doubt on an important physiologic role for cyclooxygenase products in the regulation of basal acid secretion or of acid secretion stimulated by distention or a peptone meal.

The efficacy and safety of 4-g 5-aminosalicylic acid enemas were assessed in 153 patients with ulcerative colitis involving up to 50 cm of distal colon. Seventy-six patients received active medication and 77 received a placebo. There were 20 dropouts (6 in the active group and 14 in the placebo group) during the study because of insufficient efficacy. After 6 wk of therapy, 48 of the 76 patients (63%) receiving 5-aminosalicylic acid were considered to be "much improved" by the study physician compared to 22 of the 77 patients (29%) on placebo (p = 0.001). A disease activity index based on patient symptoms and sigmoidoscopic appearance was used to assess efficacy. Mean disease activity index declined 55% for patients on 5-aminosalicylic acid and 24% for patients on placebo (p = 0.0001). Analysis of subgroups indicated that patients most likely to respond were those with disease confined to the 20-40 cm from the anus. Response was not affected by concurrent sulfasalazine, but patients requiring concurrent oral steroids had a diminished response. Rapid onset of efficacy was shown by a significant reduction in rectal bleeding within 3 days of treatment initiation. 5-Aminosalicylic acid enemas are well tolerated and are of benefit in the treatment of ulcerative colitis confined to the distal colon.

A unicenter, single-blind, randomized study was conducted on 283 patients with active duodenal ulcer to compare possible factors that may affect healing and relapse in patients treated with a potent antisecretory agent, cimetidine, or a site-protective and cytoprotective agent, sucralfate. The endoscopic healing rates at 4 wk were 76% and 79%, respectively, and cross-over treatment of the failures for a further 4 wk resulted in 68% healing with cimetidine and 69% healing with sucralfate, both differences being not statistically different. Unlike cimetidine, healing by sucralfate was unaffected by cigarette smoking, reluctance to give up smoking, habitual use of alcohol, high maximal acid output, and large ulcer diameter. In particular, the healing rate of smokers treated with sucralfate (82%) was significantly greater than that of smokers treated with cimetidine (63%). Duodenal bulb deformity significantly affected healing in both groups, and was the only offsetting factor identifiable for sucralfate out of 46 factors examined. Of the patients with healed ulcers, 238 participated in a 24-mo follow-up study consisting of interviews at 2-mo intervals and endoscopy at 4-mo intervals or whenever symptoms recurred. The cumulative relapse rate was significantly (p less than 0.007) greater in patients healed with cimetidine than with sucralfate, 50% relapse occurring at 6 and 12 mo, respectively. In both, the cumulative relapse rate was significantly greater in cigarette smokers than in nonsmokers, but smokers and nonsmokers treated with cimetidine relapsed (50% at 4 and 8 mo, respectively) faster than the corresponding smokers and nonsmokers treated with sucralfate (50% at 8 and 18 mo, respectively). Furthermore, in cimetidine- but not sucralfate-healed patients, early ulcer relapse (within 6 mo) was associated with short duration of illness, short remission period, long symptomatic spell, and reluctance to give up smoking. We conclude that smoking adversely affects duodenal ulcer healing by cimetidine and hastens subsequent relapse, and that sucralfate overcomes the adverse effect of smoking on healing as encountered with cimetidine, and results in a subsequent remission period double that of cimetidine.

Delayed gastric emptying is common in primary anorexia nervosa. We investigated in 12 patients whether gastric emptying could be accelerated by the prokinetic drug cisapride. Patients were studied on two occasions 1 wk apart and received, under random double-blind conditions, 8 mg of cisapride and placebo intravenously. Gastric emptying of an isotopically labeled semisolid meal and antral motor activity were measured using a dual-headed gamma-camera for 50 min. Emptying was significantly slower (half-emptying time, 50-191 min; median, 121 min) than in 24 healthy volunteers (half-emptying times, 21-119 min; median, 47 min). Cisapride accelerated emptying significantly (p less than 0.001; half-emptying time after cisapride, 22-80 min; median, 42 min). Antral contraction amplitude increased and contraction frequency decreased significantly (p less than 0.001), whereas the propagation velocity of contractions remained unchanged. We concluded that intravenous cisapride accelerates gastric emptying and increases antral contraction amplitude in patients with anorexia nervosa. Whether or not these effects can prove beneficial in diminishing the patients' symptoms and in helping them to gain weight can only be answered from studies involving long-term treatment with cisapride.

Twenty-nine patients with esophageal symptoms and contraction abnormalities of the esophageal body completed a 6-wk, double-blind, placebo-controlled trial of trazodone (100-150 mg/day). Measures of esophageal and psychologic symptoms were completed at entry and at each follow-up visit. Esophageal manometry was repeated at the termination of the trial. Upon completion of the treatment, patients receiving trazodone (n = 15) reported a significantly greater global improvement than those receiving placebo (n = 14; p = 0.02). Although a variable clinical response was observed, the trazodone group had less residual distress over esophageal symptoms compared with the placebo group (59% +/- 9% vs. 108% +/- 19%, p = 0.03). Manometric changes observed during the course of the trial were not influenced by treatment nor by clinical response. Remarkable reductions in ratings of chest pain were reported by both treatment groups, emphasizing the importance of controlled trials when studying this patient population. We conclude that low-dose trazodone therapy can be of benefit in the management of symptomatic patients with esophageal contraction abnormalities. In addition, our findings support recent observations that manometric abnormalities characterizing this patient group may not be solely responsible for symptoms.

Previous therapeutic trials with prostaglandins have shown them to be effective in healing duodenal ulcers when used at doses that are highly effective suppressors of gastric acid secretion. We undertook this trial to determine if a cytoprotective dose of arbaprostil (10 micrograms q.i.d. for 4 wk) would also be efficacious in this disease state. Eighty-two patients between the ages of 19 and 72 yr with endoscopically documented duodenal ulcers were entered into this randomized double-blind placebo-controlled trial. The patients were monitored with biweekly endoscopies and laboratory examinations, weekly interviews during the period when drug was administered, and a follow-up interview plus laboratory examinations 1 wk after drug administration was completed. No statistically significant differences between the arbaprostil and placebo treatment groups were found for ulcer healing rates, pain relief, antacid consumption, side effects, or laboratory examinations. It is presumed that this prostaglandin may not have sufficient duodenal cytoprotective capacity to effectively heal duodenal ulcers, or that some suppression of gastric acid secretion may be required to achieve significant clinical efficacy.

Eighteen colorectal cancer patients with liver metastases receiving hepatic arterial infusion of chemotherapeutic agents (mitomycin-C + 5-FUDR) were randomized to receive misoprostol (a prostaglandin E1 analogue) or placebo as a mode of cytoprotection against inadvertent gastroduodenal mucosal injury. Four of 10 patients on misoprostol and 3 of 8 patients on placebo developed gastroduodenal mucosal injury (p greater than 0.1), which was confirmed by endoscopy. Significant tumor responses to chemotherapy were equally distributed between the two groups. Although our sample size was small and type II error (false-negative trial) cannot be excluded, these data strongly suggest that prophylaxis with misoprostol does not prevent the acute gastroduodenal mucosal injury associated with hepatic arterial infusion of chemotherapeutic agents.

Percutaneous liver biopsies were performed in 40 adult patients with acute or chronic liver disease. Real-time sector scan ultrasound of the right upper quadrant was obtained just before biopsy, 1 day after biopsy, and in 10 randomly selected cases 7 days after biopsy. Twenty patients were randomized, by sealed envelope, to 6-h and 20 patients to 24-h postbiopsy bed rest. Nine patients (23%) had ultrasound-detected hematomas 1 day after liver biopsy (7 intrahepatic, 2 subcapsular). Five of these patients had been randomized to 6-h and 4 to 24-h bed rest. Follow-up ultrasound at 7 days in the 10 randomly selected patients failed to reveal any hematomas not seen on day 1 (n = 6) and showed complete or partial resolution of those that had been detected previously (n = 4). Significant drops in systolic and diastolic blood pressure and increases in pulse rate were common postbiopsy findings, and were unrelated to hematoma formation. The results of this study indicate that intrahepatic and subcapsular hematomas are common after percutaneous liver biopsy. The length of postbiopsy bed rest (6 vs. 24 h) does not appear to influence the frequency of this complication.

Cisapride is a prokinetic agent believed to facilitate acetylcholine release from the myenteric plexus of the gut. The effect of cisapride on gastric emptying of solids was studied in 9 diabetic patients, all of whom had delayed gastric emptying of indigestible solids (gastroparesis). Six patients had chronic nausea and vomiting, and 3 had no symptoms. Cisapride (5 mg) was given intravenously 15 min before ingestion of a 400-kcal test meal and 10 indigestible solid radiopaque markers. On separate days and in random order each patient also received intravenous metoclopramide (10 mg) or placebo 15 min before ingestion of the meal and markers. Mean gastric emptying of radiopaque markers, assessed by serial radiographs of the gastric region, was accelerated by metoclopramide and cisapride, but the difference reached significance only with cisapride (p less than 0.05). There was considerable intersubject variability in gastric emptying responses to cisapride and metoclopramide. No side effects occurred with either drug. This study indicates that acute, intravenous administration of cisapride accelerates gastric emptying of indigestible solids in patients with diabetic gastroparesis.

The efficacy and safety of famotidine, a potent new long-acting H2-receptor antagonist, was compared with placebo in a multicenter, double-blind, randomized, placebo-controlled study in the United States. A total of 384 patients with endoscopically proven acute duodenal ulcer disease were enrolled. Patients received either famotidine or a placebo. The patients receiving famotidine were treated with one of three dose regimens, 40 mg h.s., 40 mg b.i.d., or 20 mg b.i.d. Patients were reassessed by endoscopy at 2, 4, and 8 wk if ulcer healing had not occurred sooner. A diary was kept to record the duration and intensity of the day and night pain and the amount of Gelusil antacid (Parke-Davis, Morris Plains, N.J.) ingested. Three hundred sixty-three patients met the evaluation criteria. The results revealed a 4-wk healing rate of 70%, 75%, 67%, and 31% for the famotidine 40 mg h.s., 40 mg b.i.d., 20 mg b.i.d., and placebo groups, respectively. The 8-wk healing rates for the same respective groups were 83%, 82%, 82%, 45%. Ulcer pain and antacid consumption occurred less often in the famotidine groups. The clinical and laboratory safety profile of the famotidine groups was similar to that of the placebo group. Famotidine appears to be an effective and safe once-a-day therapy for the treatment of acute duodenal ulcer disease. The recommended dosage is 40 mg h.s.

Twenty-four patients with primary biliary cirrhosis were entered into a prospective, randomized trial of chlorambucil therapy. Thirteen patients received chlorambucil (0.5-4 mg/day) and 11 patients received no therapy; all have been followed for 2-6 yr (mean, 4.1 yr). Two control but no treated patients died. Average serum bilirubin, serum aspartate aminotransferase activities, and albumin levels improved or remained unchanged in treated patients but worsened in controls. Serum alkaline phosphatase levels did not change in either group. Immunoglobulin M levels decreased and became normal in all treated patients but in only 3 control patients. Liver biopsy histology revealed an improvement in inflammatory cell infiltrate in treated patients in comparison with controls, but no significant change in degree of fibrosis or the histologic stage of disease. Side effects of therapy included bone marrow suppression necessitating discontinuation of the drug in 4 patients. These findings indicate that chlorambucil therapy may retard the progression of primary biliary cirrhosis. Whether such therapy will ultimately decrease morbidity and improve survival in this disease can only be demonstrated by large-scale, placebo-controlled trials.

In 108 patients the healing and relapse of reflux esophagitis, defined endoscopically by the presence of epithelial defects (erosions and ulcerations) of the esophageal mucosa, were studied. In the first study, with open treatment of ranitidine, the healing rate after 6 wk was 50%. The most important factor that negatively influenced healing was the extent of esophageal erosions. Patients with isolated erosions had a 6-wk healing rate of 78%; the healing rate was 38% in patients with longitudinally confluent lesions and 23% in those with circumferential erosions of the distal esophagus. Smoking also had an unfavorable effect. Age, sex, duration of history, body weight, and alcohol consumption were not related to outcome. Symptoms improved during treatment with ranitidine, but the correlation between symptoms and endoscopic findings at 6 wk was weak. In the second study, relapse was investigated in 61 patients with healed esophagitis in a randomized, double-blind trial comparing placebo and ranitidine (150 mg at bedtime for 6 mo). In both groups, relapse occurred in more than one-third of the patients, with no significant difference between ranitidine and placebo treatment. Patients with worse daytime symptoms at the time of previous healing had a higher relapse rate. The initial severity of esophagitis and smoking did not influence recurrence. Thus, the initial endoscopic findings are of prognostic value in reflux esophagitis. Smoking retards healing. Low-dose maintenance treatment with ranitidine does not prevent relapse.

The natural history of chronic antral gastritis in relation to the healing of duodenal ulcer and its response to treatment, if any, are unknown. We performed a double-blind controlled trial using an oral prostaglandin E1, misoprostol, in 229 patients with active duodenal ulcer randomized to receive placebo (n = 76), misoprostol 200 micrograms (n = 77), or misoprostol 300 micrograms (n = 76), q.i.d. orally. Healing of duodenal ulcer was assessed biweekly up to 12 wk by endoscopy, during which procedures at least two antral and two fundal biopsy specimens were taken. The activity and the degree of chronic inflammation of gastritis, as assessed histologically by the infiltration of polymorphs and chronic inflammatory cells, respectively, was graded blindly by two pathologists as nil, mild, moderate, or severe. Before treatment, 99% of patients had chronic antral gastritis and 1.5% had chronic fundal gastritis. In the placebo group, healed duodenal ulcer was associated with significantly (p less than 0.01, life table analysis) higher incidence of improvement of the activity of the antral gastritis (nil or mild as endpoint) than unhealed ulcer (30% vs. 4% at week 8). Irrespective of whether duodenal ulcer was healed or unhealed, significantly (p less than 0.01) more patients on misoprostol (50% at week 8) showed improvement (nil or mild as endpoint) than the placebo group. The degree of chronic inflammation of the antral gastritis showed similar significant changes in favor of misoprostol. Smoking and alcohol intake had no significant effect on the improvement of chronic antral gastritis. In conclusion, healing of duodenal ulcer was associated with improvement of the activity of chronic antral gastritis, which, as shown for the first time, could be further enhanced by a therapeutic agent--prostaglandin E1.

Since 1976, we have compared the end-to-side portacaval shunt (PCS) with the distal splenorenal shunt (DSRS) in patients with alcoholic liver disease and recurrent variceal bleeding. Fifty-four patients were randomly assigned to receive either shunt procedure. There were 27 patients in each group and both groups were highly comparable in clinical and laboratory characteristics. Median follow-up was 31 mo in each group. Postoperative complications and operative mortality (7% after PCS, 12% after DSRS) were comparable. Spontaneous portasystemic encephalopathy developed in 32% of the patients at risk after PCS and in 39% after DSRS. Rebleeding from varices occurred in 4% of the patients after PCS and in 27% after DSRS. Cumulative survival was not significantly different between groups (5-yr survival: 31% after PCS, 43% after DSRS). We have failed to demonstrate superiority of DSRS in our patients with alcoholic liver disease with respect to postoperative encephalopathy or survival, and have experienced an unusually high rate of variceal rebleeding after DSRS.