Prevalence of short segments of specialized columnar epithelium in distal esophagus: association with gastroesophageal reflux.

BACKGROUND AND AIMS: Recent studies have reported high prevalence rates of short segments of specialized columnar epithelium (SCE) in the distal esophagus. The association of SCE with gastroesophageal reflux disease is not well established. We studied the prevalence and associations of short segments of SCE in the distal esophagus amongst Indians. METHODS: 271 patients (mean age 36 [14] y; 160 men) undergoing diagnostic upper gastrointestinal endoscopy were interviewed regarding symptoms of gastroesophageal reflux, and history of medications, smoking or chewing tobacco and alcohol ingestion. At endoscopy, presence and grade of esophagitis and hiatus hernia were recorded. One biopsy each was taken from the squamocolumnar junction and 2 cm proximal to it. Biopsies were stained with hematoxylin/eosin and alcian blue/periodic acid-Schiff. The pathologist was blinded to the clinical and endoscopic data. RESULTS: Short segments of SCE in the distal esophagus were present in 16/271 (6%; CI 5.03-6.97) patients. Increasing age (p<0.01), and endoscopic (p<0.01) and histologic (p<0.001) esophagitis were associated with its presence, whereas symptoms of gastroesophageal reflux, smoking, tobacco chewing, use of alcohol or non-steroidal anti-inflammatory drugs, and hiatus hernia were not. One patient with SCE had dysplasia. CONCLUSION: Prevalence of short segments of SCE in the distal esophagus amongst Indians is low and is usually associated with inflammation in the esophagus.

High frequency of bacteremia with endoscopic treatment of esophageal varices in advanced cirrhosis.

OBJECTIVES: The fregency of bacteremia after endoscopic variceal band ligation (EVL) is reported to be lower when compared to that after endoscopic variceal sclerotherapy (EVS). However, there are conflicting reports on the infectious sequelae after EVL. AIM: To compare the frequency on bacteremia and infectious sequelae after EVL and EVS in patients with cirrhosis of liver. METHODS: Bacteremia and infectious sequelae were studied in 32 sessions of EVL in 18 cirrhotic patients (Child-Pugh class A-6, B-5, C-7), 30 sessions of EVS in 22 cirrhotic patients (Child-Pugh class A-2, B-5, C-15) and 14 diagnostic upper gastrointestinal endoscopies. Blood cultures were collected before, during and 30 minutes after the procedure. Patients were observed for infectious sequelae during subsequent hospitalization. RESULTS: Before the procedure, bacteremia was present in 7/62 (11%) sessions. Significant bacteremia during and 30 min post-procedure developed in 8/32 (25%) and 12/30 (40%) of EVL and EVS sessions, respectively (p = ns), and in 1/14 (7%) upper gastrointestinal endoscopy sessions. There was more frequent bacteremia with severe liver disease (Child-Pugh class A-0/6, B-1/5. C-7/21; p = 0.09) in the EVL but not in the EVS group (Child-Pugh class A-1/3, B-2/5, C-9/22; p = ns). The incidence was higher with emergency sclerotherapy compared to elective sclerotherapy (6/8 [75%] vs 6/22 [27%]; p <0.01). One patient in the EVS group developed spontaneous bacterial peritonitis. CONCLUSIONS: Bacteremia occurs frequently following EVL and EVS in patients with advanced liver diseases. In the EVS group it is more common after emergency sclerotherapy. This bacteremia is rarely associated with significant infectious sequelae.

Bilioma secondary to choledocholithiasis.

Bilioma secondary to choledocholithiasis is rare. We report a patient in whom a large common bile duct stone was responsible for leak from the infraduodenal segment of the bile duct. Choledochotomy with extraction of stone followed by T-tube drainage of the bile duct and evacuation of the bilioma resulted in complete recovery.

Long-term anti-inflammatory drug use and Helicobacter pylori infection: a clinical, endoscopic and histological study.

BACKGROUND: Non-steroidal anti-inflammatory drugs (NSAIDs) and Helicobacter pylori are independent risk factors for gastroduodenal damage and peptic ulcer. OBJECTIVE: To study the frequency and effect of H pylori infection on gastroduodenal mucosa in patients on long-term NSAID use. METHODS: A total of 125 subjects were studied: 65 patients (Group 1) on NSAID therapy (> 6 months), 30 patients (Group 2) with arthritic disorders prior to starting NSAID therapy, and 30 healthy volunteers (Group 3). Dyspeptic symptoms were evaluated using a questionnaire. All patients underwent endoscopy and antral and duodenal biopsies were obtained to assess the extent of gastroduodenal damage and H pylori status. RESULTS: H pylori infection was less frequent in Group 1 (37%) compared to Group 2 (57%, p = ns) and 3 (60%, p < 0.05). Among Group 1 patients, H pylori infection did not increase the risk of gastroduodenal damage (52% vs 45%) or ulceration (32% vs 27%). Group 1 patients with H pylori infection were more likely to be symptomatic (48% vs 27%) and have chronic active gastritis (76% vs 12%) and chronic active duodenitis (68% vs 5%). Gastric metaplasia was seen only in patients with H pylori infection, chronic active gastritis and duodenitis. Chemical gastritis was observed more commonly in Group 1 (34% vs 3%) compared to Group 2; its was not seen in Group 3. H pylori infection was less commonly observed in patients with chemical gastritis (8% vs 50%). CONCLUSION: Patients on long-term NSAIDs are not at increased risk of H pylori infection. Presence H pylori infection is not associated with increased risk of gastroduodenal damage in these patients. H pylori infection correlated with presence of chronic active gastritis, and NSAID with presence of chemical gastritis.

Gastric metaplasia and Helicobacter pylori infection in intestinal tuberculosis.

BACKGROUND: Gastric metaplasia occurs in intestinal epithelium of patients with Helicobacter pylori infection as well as intestinal tuberculosis, but the relationship between the three is not clear. SETTING: Teaching hospital. METHODS: Staining of intestinal sections from patients with ileocecal tuberculosis for H pylori and for metaplasia. RESULTS: Twenty seven of 92 (29.3%) patients demonstrated gastric metaplasia (superficial epithelial and pyloric gland metaplasia) in resected specimens of ileum. Gastric metaplasia was of antral type in all 27 patients. Helicobacter pylori was present in 3 of 27 (11%) patients with gastric metaplasia and was absent in all 65 patients without metaplasia. CONCLUSION: Gastric metaplasia (both superficial epithelial and pyloric gland) occurs in one-third of patients with ileocecal tuberculosis and such mucosa is poorly colonized by H pylori.

Heterotopic gastric mucosa and Helicobacter pylori infection in Meckel's diverticulum in Indian subjects.

Two of twenty nine (6.8%) patients who had undergone resection of Meckel's diverticulum demonstrated heterotopic gastric mucosa. In both these patients gastric mucosa was of antral type. Remaining 27 of 29 (93.2%) patients with Meckel's diverticulum had intestinal mucosa. 9 of 29 (31%) patients with Meckel's resected were symptomatic (8 bleeding, 1 perforation) and all 9 were negative for heterotopic gastric mucosa. Helicobacter pylori (H. pylori) was absent in both patients with intestinal mucosa. Antral biopsy from 13 partial gastrectomy specimens (surgery done for carcinoma stomach) were used as methodological controls and 7 of 13 (54%) showed H. pylori colonization. We conclude that (i) incidence of gastric mucosa is 6.8% in Meckel's diverticulum in our population, (ii) presence of heterotopic gastric mucosa is unlikely to be associated with complications of Meckel's diverticulum and (iii) heterotopic mucosa in Meckel's diverticulum is not commonly colonized by H. pylori.

New H2-receptor antagonist--roxatidine acetate--in treatment of duodenal ulcer.

A double blind randomised trial, comparing a new H2-receptor antagonist, Roxatidine acetate, with Cimetidine was carried out in 47 patients of uncomplicated, endoscopically proven duodenal ulcer. Twenty seven patients were treated with Roxatidine 75 mg twice daily and 20 patients were treated with Cimetidine 200 mg 3 times a day and 400 mg at bed time for 4 weeks. At the end of 4 weeks, total pain relief was obtained in 74% and 70% patients receiving Roxatidine and Cimetidine respectively. Complete endoscopic healing at the end of 4 weeks was observed in 92.3% patients receiving Roxatidine and 85% patients receiving Cimetidine. These differences were statistically not significant. No significant side effects were observed in either group. We conclude that Roxatidine acetate is comparable to cimetidine in relieving pain and endoscopic healing of duodenal ulcer and has an excellent safety profile.

Hepatitis C virus infection in chronic liver disease in Bombay.

To find out the prevalence of antibody of hepatitis C virus (anti-HCV) in patients with chronic liver disease in Bombay, sera from 126 patients (93 men, 33 women; aged 9-70 years, mean 39.7) with chronic liver disease (cirrhosis 103, cirrhosis with hepatocellular carcinoma 3, chronic active hepatitis 20) were tested for HBsAg and anti-HCV antibody. HBsAg positive sera were tested for anti-delta antibody and IgM anti-HBc. All the tests were carried out by ELISA. Of 126 patients, 51 (40.5%) were HBsAg positive, 49 (38.8%) alcoholic and 21 (16.6%) anti-HCV positive. The prevalence of anti-HCV in HBsAg positive, alcoholic and cryptogenic (HBV negative and no alcohol) liver disease patients was 13.7%, 14.7% and 20.5% respectively. Of 21 anti-HCV antibody positive patients, 8 (38%) had received blood transfusions previously. HCV is present in 15-20% of patients with chronic liver disease in Bombay.

Prevalence of spontaneous bacterial peritonitis.

To assess the prevalence of spontaneous bacterial peritonitis (SBP), ascitic fluid cell count, and ascitic fluid culture by conventional method and by bedside inoculation in blood culture bottles were performed in 31 consecutive patients of liver cirrhosis. Seven (22.58%) patients had ascitic fluid polymorphonuclear count (PMN) more than 500/mm. Ascitic fluid culture by conventional method was negative in all the patients, while in 4 patients culture was positive by bedside inoculation method. Six of 7 patients with SBP or its variant were in Child class C. Clinical features in these patients were abdominal pain (5 patients), fever (4) and encephalopathy (2); serum bilirubin level was 6.8 +/- 5.5 mg/dl, serum albumin 1.98 +/- 0.2 g/dl, prothrombin index 59.8 +/- 12.2%, ascitic fluid protein 0.78 +/- 0.24 g/dl. Three of 7 patients with SBP or its variant expired during hospital stay; the other 4 patients recovered after appropriate antibiotic therapy. We conclude that SBP is a serious complication in patients of liver cirrhosis with ascites. Ascitic fluid PMN count and bedside inoculation of blood culture bottles with ascitic fluid are sensitive indicators of SBP. Hence they should be performed routinely for early detection of SBP.

Effect of acute ingestion of tobacco on gastric mucosa. An endoscopic study.

One hundred patients with non ulcer dyspepsia with history of chronic tobacco chewing were examined endoscopically to assess the effect of tobacco ingestion on the gastric mucosa. Gastric erosions were seen in 20 patients in the fasting state. The remaining 80 patients in whom gastroscopy did not reveal erosions were subjected to repeat gastroscopy after tobacco ingestion. In 40 patients, endoscopy was repeated 30 minutes after 200 mg of tobacco ingestion (Group I) and in another 40 patients endoscopy was repeated 1 hour after 400 mg of tobacco ingestion (Group II). Eleven patients (27.5%) in Group I and 19 (47.5%) in Group II developed gastric erosions. Erosions were observed mainly along the lesser curvature, and in the fundus and the body of stomach. Gastric pH, determined after tobacco ingestion, was 2.4 +/- 0.43 in patients with erosions and 3.0 +/- 0.67 in patients without erosions. It is concluded that tobacco ingestion produces dose-dependent damage to the gastric mucosa as seen on endoscopy. Hence, history of tobacco ingestion should always be asked for in patients with gastric erosions.

Ascitic fluid protein and cellular changes during diuretic therapy in cirrhosis of liver.

Thirty four patients with peritoneoscopy and/or biopsy proven uncomplicated cirrhosis of liver with ascites were studied for the effect of diuretic therapy on ascitic fluid protein and cell count. Renal function tests, liver function tests, ascitic fluid protein concentration and cell count were estimated prior to diuretic therapy and once every week till the end of therapy. There was no change in mean total serum protein (5.71 +/- 0.58 g/dl to 5.72 +/- 0.63 g/dl). The rise in mean ascitic fluid protein from 1.55 +/- 0.77 g/dl to 1.76 +/- 0.79 g/dl was not significant (P greater than 0.05). Twenty one patients (Group I) showed a rise in ascitic fluid protein concentration while 13 patients (Group II) did not show a rise or showed a fall in protein concentration during diuretic therapy. The rise in ascitic fluid protein concentration in Group I from 1.62 +/- 0.69 g/dl to 2.05 +/- 0.67 g/dl was statistically significant (P less than 0.05). Group I patients had a mean weight loss of 6.21 +/- 3.66 kg as compared to 3.15 +/- 2.00 kg in Group II patients (p less than 0.05). There was no other difference between Group I and Group II patients. Only 5 patients showed a rise in ascitic fluid cell count (more than 50 cell/mm3). It is concluded that diuretic therapy alters ascitic fluid protein concentration in a majority of patients but has no significant effect on cell count.