PATHOPHYSIOLOGICAL EFFECTS OF LONG-TERM ACID SUPPRESSION IN MAN

A critical evaluation has been made of the available evidence in man of the effects of prolonged low acid states on the structure and function of the stomach. Various human models have been examined. 1. Ageing does not affect acid output from the normal male stomach, and there may be an increase in women. With progressive atrophy of the corpus mucosa, which is more frequent and rapid in patients with gastric ulcer, there is an associated loss of secretory function. Chronic gastritis and atrophy are the most important age-related changes, which in many cultures are hypothesized to develop via a prior Helicobacter pylori-related gastritis. However, H. pylori colonization of the mucosa decreases with increasing grades of gastric atrophy probably because intestinal metaplasia provides a hostile environment. Atrophy and intestinal metaplasia are associated with precancerous lesions and gastric cancer. Apparent hyperplasia of the gastric argyrophil endocrine cells is a common and spontaneous phenomenon in patients with atrophic gastritis, which in part may be related to the preferential loss of nonendocrine cells. 2. Pernicious anemia is associated with a complete lack of acid production, marked hypergastrinemia, and endocrine cell hyperplasia in the majority of patients. ECL-cell carcinoids and gastric cancer occur with a prevalence of 3-7%, and endoscopic surveillance in routine clinical practice is not warranted. 3. Gastric ECL-cell carcinoids are rare events that have been described in association with two diseases in man, pernicious anemia and Zollinger-Ellison syndrome as part of multiple endocrine neoplasia syndrome type I, and usually relate to marked hypergastrinemia and the presence of chronic atrophic gastritis with gastric antibodies or a genetic defect rather than the presence or absence of acid. Regression or disappearance of ECL-cell carcinoids, either spontaneously or after removal of the gastrin drive, has been recorded. Lymph node, and rarely hepatic, metastases are documented but death in these cases has been anecdotal. 4. Therapy with H-2 antagonists may result in up to a twofold rise in serum gastrin levels but in man no endocrine cell hyperplasia has been recorded. However, the data for H-2 antagonists on these aspects are very limited. There is no drug-related risk of gastric or esophageal cancer, although the incidence of the latter may be raised. Long-term treatment with omeprazole is associated with a two- to fourfold increase in gastrin levels over baseline values in one third of patients and apparent endocrine cell hyperplasia in 7% of cases overall. The endocrine cell hyperplasia is correlated with both levels of hypergastrinemia and the changes of progressive atrophic gastritis. No metaplastic, dysplastic, or neoplastic changes have been reported to date on long-term therapy with omeprazole. Monitoring patients on any form of long-term antisecretory therapy, for changes in serum gastrin or endoscopy with biopsy, is not recommended as part of routine clinical practice. Bacterial overgrowth in patients on any of the antisecretory drugs has not proven to be a problem clinically. 5. Gastric surgery may have profound effects on gastric function, depending on the type of operation. Hypergastrinemia, generally higher than that reported in patients on H-2 antagonists or omeprazole, has been reported following all types of vagotomy. Endocrine cell changes have not been adequately studied. The issue of nitrosation and cancer risk remains hypothetical, dogged by methodological problems and conflicting results. Overall, the risk of gastric cancer after gastric resection does not become significant until 20-25 years later, and even then endoscopic screening is not justified. 6. The nutritional consequences of diseases and therapies in which there is a low acid state cannot easily be predicted but are only likely to occur over a very long time course, over 20 years in many reports. 7. The evidence for any increase in the occurrence of cancer at extragastric sites, such as pancreas or colon, in patients with prolonged low acid states is limited and conflicting. Overall, the risks of significant changes in gastric structure or function as a result of long-term low acid status in man have been over-stated and analogies with animal data have not been supported by the currently available evidence in humans.