A topographic analysis of atrophic gastritis and stomach cancer risk. A clinicoepidemiological study in Poland.

The purpose of this cross-sectional study was to assess the relation between intestinal and diffuse stomach cancer at its various locations with topography of atrophic gastritis. The study population consisted of 3435 patients who reported over the period of 1991-1994 for the first time to gastroenterological outpatient clinics of 7 university medical centers in Poland. Among these subjects there were 131 histologically proved consecutive cases of gastric carcinoma. The reference group consisted of 1540 patients among whom endoscopic examination did not reveal peptic ulcers, polyps, deformations of antrum or bulbus duodenum and mucosa erosions. Gastroscopy on gastric cancer patients and the reference group was performed and biopsy specimens were obtained from the tumour and from the antrum and stomach corpus distant from the tumour. Among the gastric cancer cases there was a higher prevalence of atrophic gastritis in the intestinal than in the diffuse type. The highest prevalence of atrophic gastritis irrespective of its degree and stomach area was observed in the tumour-area of intestinal cancer located distally (78.9%), and the lowest in the tumour-free area in diffuse proximal cancer (18.5%). Prevalence of atrophic pangastritis (atrophic gastritis present both in the corpus and antrum) was also highest in intestinal distal cancer (69.2%) and lowest in diffuse proximal cancer (21.7%). The age-adjusted correlation coefficients between gastritis score in the tumour-area and tumour-free area were highly significant. It was shown that only atrophic pangastritis was significantly associated with gastric cancer irrespective of its histology and location (OR = 3.8, 95% CI:2.4-6.0), however, it was much stronger related to the intestinal gastric cancer (OR = 5.9, 95% CI: 3.1-11.0), than to the diffuse carcinoma (OR = 2.2, 95% CI: 1.1-4.3).