Gastric carcinoma after surgical treatment of peptic ulcer: an analysis of morphologic features and a comparison with cancer in the nonoperated stomach.

BACKGROUND: Surgery for peptic ulcer is associated with an increased risk of later development of gastric carcinoma. This risk applies to cancer occurring in the distal stomach, not at the cardia. Mucosal alterations occurring in the nonneoplastic mucosa, adjacent to postgastrectomy carcinomas, are currently poorly defined. METHODS: Between 1975 and 1995, the authors collected records of 76 patients with gastric carcinoma developing after previous ulcer surgery. Thirty-three gastrectomy specimens were available for study and were compared with a control series of gastric carcinomas occurring in the intact stomach. Morphologic features studied were macroscopic findings, tumor type, extent of inflammation, atrophy, intestinal metaplasia, reactive gastropathy, and presence of Helicobacter pylori. When possible, a semiquantitative grading system (Sydney system) was used. RESULTS: The 33 patients with resected carcinoma after ulcer surgery were representative of the total 76 patients. There were no differences between the seven postulcer surgery cardia carcinomas and the control cardia carcinomas. The 18 distal carcinomas occurring after prior gastrectomy had significantly less intestinal metaplasia and Helicobacter pylori in the nonneoplastic mucosa than did the controls. Eight distal carcinomas occurring after a vagotomy and pyloroplasty had adjacent mucosa with findings intermediate between the gastrectomy carcinomas and the controls. CONCLUSIONS: The differences shown suggest that for distally located postgastrectomy carcinomas Helicobacter pylori infection with extensive intestinal metaplasia may be relatively less important in pathogenesis and that intestinal reflux with reactive gastropathy more important. There was no evidence to support a different mechanism of pathogenesis for postsurgical carcinomas occurring at the cardia from that of control cardia carcinomas.

Comparison of esophagocardiac and more distal gastric cancer in patients with prior ulcer surgery.

BACKGROUND: Adenocarcinoma of the gastric cardia and/or lower esophagus differs from cancer of the more distal stomach. METHODS: The authors compared the proportion of patients with these two types of cancer who underwent surgery for peptic ulcers more than 5 years before the diagnosis of cancer during the periods 1977-82 and 1983-88. RESULTS: The total number of patients with esophagocardiac cancer increased from 99-241 between the two periods, but the number with previous ulcer surgery remained the same (n = 7) in each period. The total number of more distal gastric cancers decreased from 262 to 237 between the two periods, but the number with previous ulcer surgery increased from 9-26 (P = 0.002). Only 4 of 14 patients with esophagocardiac cancer and a previous ulcer had a partial gastrectomy compared with 27 of 35 patients with more distal gastric cancers (P = 0.005). CONCLUSION: These data indicate a relationship between partial gastrectomy and the late development of gastric cancer, but this finding is confined to cancers not involving the cardia.

Cigarette and alcohol use in patients with adenocarcinoma of the gastric cardia or lower esophagus.

The authors studied the cigarette and alcohol use of 243 patients with adenocarcinoma of the cardia or lower esophagus, including 66 with Barrett's metaplasia (esophagocardia group). They compared results of that investigation with the cigarette and alcohol use of 303 patients with squamous carcinoma of the esophagus, a cancer in which an etiologic association is proven, and of 338 patients with stomach cancer not involving the cardia, a cancer in which there is little or no association with cigarettes or alcohol. Controlled for other variables, patients with squamous cancer used more cigarettes and alcohol than the other two groups. There was no significant difference in cigarette or alcohol use between patients with esophagocardia and stomach cancer or between those in the esophagocardia group with or without Barrett's metaplasia. Because cigarette and alcohol use was not greater in patients with esophagocardia than in those with stomach cancer, the authors do not think that such factors explain the increasing incidence of adenocarcinoma of the cardia or lower esophagus.

Carcinoid tumour of the gastrointestinal tract: prognostic factors and disease outcome.

This study represents retrospective analysis of 87 patients with a carcinoid tumour of the gastrointestinal tract seen and followed in the British Columbia Cancer Agency (BCCA) from 1960 to 1986. In 49 cases, the primary site was the small bowel. The rest of the cases were distributed as follows: 11 appendix, 10 rectum, 5 stomach, and 7 undetermined. We extrapolated the Dukes' and modified Astler-Coller surgicopathological classifications used for colorectal cancer for use in our cases of carcinoid tumour of the gastrointestinal tract. A strong correlation was found, using this staging, with disease-specific survival. Other prognostic factors included histologic differentiation, the presence of macroscopic residual disease after initial surgery, and level of 5-hydroxyindoleacetic acid (5-HIAA) in urine. Among 51 patients with surgically grossly removed disease, there was a tendency for the development of distant and distant/locoregional recurrence more often than locoregional recurrence alone. The liver was the commonest site of distant recurrence. Analysis of the effect of radiotherapy or chemotherapy on carcinoid tumour of the gastrointestinal tract proved unsuccessful because only a small portion of the patients had this treatment, and it was used mainly for palliation.

Fecal occult blood screening for colorectal cancer.

Fecal occult blood testing with guaiac gum is a simple, safe, and cheap method for screening for colorectal cancer. In the office setting, compliance is good. Asymptomatic colorectal cancers are detected at a more favourable pathologic stage than those detected in symptomatic patients. Because of the low positive predictive value of the test, many persons testing positive undergo unnecessary bowel examinations. As yet, randomized controlled studies have not shown that fecal occult blood screening reduces mortality from colorectal cancer. At present, screening with this test is left to the judgement of the individual physician.

Adenocarcinoma of the esophagus and/or gastric cardia.

One hundred twenty-nine adenocarcinomas involving the esophagus and/or gastric cardia differed significantly from 212 cancers of the rest of the stomach as follows: male-female ratio, 6:1 versus 2:1, birth outside Canada, US or UK, 12% versus 34%; parent or sibling with gastric cancer, 5% versus 13%; previous duodenal ulcer, 23% versus 9%; chronic reflux symptoms, 25% versus 3%; hiatal hernia, 51% versus 11%. Of the 129 esophagocardia cancers, 24 involved the esophagus alone, 48 the cardia and esophagus, 33 the cardia alone or cardia and fundus, and 24 the upper stomach and lower esophagus extensively. Thirty-four were associated with Barrett's esophagus. The 72 patients with involvement of both the upper stomach and lower esophagus (48 cardia and esophagus, 24 extensive) were identical with the esophagocardia group as a whole. The 24 patients with esophageal cancer and the 34 with Barrett's epithelium were the same clinically as the whole esophagocardia group except more had chronic reflux and hiatal hernia. The 33 patients with cancer confined to the cardia or cardia and fundus resembled the whole esophagocardia group but did not have Barrett's esophagus. Adenocarcinoma of the esophagocardia region is probably a different disease from cancer of the rest of the stomach.

Chemotherapy of advanced gastric cancer with 72-hour continuous intravenous 5-fluorouracil infusion at 2-week intervals.

Twenty-nine patients with advanced gastric adenocarcinoma were treated with 72-hour continuous intravenous infusion of 5-fluorouracil (30 mg/kg/24 hr) at 2-week intervals. Six patients (21%) achieved partial response and four (14%) had stabilization of their disease. The median duration of partial response was 8 months, with a range of 5-19 months. The median survival time for all patients from the start of chemotherapy was 8 months. Various prognostic factors important in determining response and survival are analyzed.

Mechanism of aspirin injury to human gastroduodenal mucosa.

The effect of oral plain aspirin tablets (ASA) on gastric mucosal endoscopic appearance and histology and on the prostaglandin E (PGE) content of the gastric juice and gastroduodenal mucosa was determined in ten healthy young adults. In all subjects the initial endoscopic appearances and microscopy of biopsies from antrum and duodenum were entirely normal. After ASA (2.6 g/day for 5 days) all subjects showed obvious endoscopic damage and 6 showed a histological gastric lesion consisting of loss of mucus and an alteration in nuclear appearance. In addition there was a significant reduction (p less than 0.01) in the PGE content of the gastric juice and of the antral and duodenal mucosa. There was no correlation between the severity of the mucosal damage and the degree of reduction in PGE. It is concluded that aspirin damages the human gastric mucosa by at least two distinct mechanisms one of which is inhibition of protective prostaglandins.

Long-term effect of shoyu (Japanese soy sauce) on the gastric mucosa of the rat.

The long-term effect of shoyu (Japanese soy sauce) on the gastric mucosa was assessed in intact rats and in those with a fundusectomy. One hundred female 4-month-old Wistar rats were divided into 4 groups:intact and operated rats on a control diet and intact and operated rats on the same diet saturated with shoyu. The animals given shoyu were smaller than the controls; 15 intact rats receiving the sauce were healthy at 33 months compared with 7 controls. Only 9 of 50 operated rats lived 33 months. Breast tumors developed in 10 control rats but in none given shoyu. The distribution of other tumors was the same. Two malignant neoplasms appeared to originate in the perigastric tissues of operated rats: a reticulum cell sarcoma in the shoyu group and a lymphosarcoma in the control group. None involved the gastric mucosa. Most of the operated rats (with or without shoyu) developed small benign gastric tumors. Mucosal histology was well preserved in 22 intact rats on shoyu: In 15 it was normal, in 7 there was mucus loss and nuclear changes in the cells of the gastric pits; 2 of the 7 had mild gastritis. One intact control showed patchy mucus loss. Shoyu did not appear to be a carcinogen in the rat; its prolonged use impaired neither health nor longevity. Long-term use was associated with gastric mucus loss and nuclear changes in a few rats, occasionally with a mild gastritis.

Diagnosis of superficial or very small gastric cancers.

Thirteen superficial or very small gastric cancers are described, 10 of which were diagnosed preoperatively. Five patients presented with gastric ulcer, five with radiological deformities of uncertain nature and two with polypoid lesions; one cancer was an incidental finding at emergency operation. Two of the patients had a family history of gastric cancer, two had pernicious anemia and three were born in countries where gastric cancer is common. Gastric cytology was positive in eight of the nine patients examined. A follow-up study of patients investigated for stomach cancer three to five years previously suggests that suspected gastric cancer is seldom missed for long periods if the diagnostic workup is thorough.

Histological effect of certain pickles on the human gastric mucosa. A preliminary report.

The histological effects of several types of pickles and control substances were assessed by repeated suction (per os) biopsies of the gastric antrum in three healthy men. The ingestion of 30 ounces of fukujinzuke (assorted vegetables pickled in soy sauce) or of vinegared gherkins over a three-day period caused marked changes in the surface epithelium and gastric pits in all three subjects. The abnormalities consisted of loss of cellular mucus, nuclear enlargement, prominent nuclear chromatin and an increased number of mitotic figures. All control biopsies were normal, as were biopsies after the ingestion of 30 ounces of raw carrots or cucumbers over a three-day period. This study shows that eating a sufficient quantity of certain types of pickles causes marked changes in the human stomach. Peroral suction biopsy of the gastric antrum appears to be a practical method for testing the effect of various foods on the human gastric mucosa.