The aim of this work was to investigate the extension of intestinal metaplasia (IM), as well as to quantitate various components of IM (namely sialomucins, sulfomucins and Paneth cells), in entire gastrectomy specimens from Swedish and Japanese patients. The length of the gastric mucosa was assessed by morphometry. The percent of sections with IM was regarded as the extension of IM in the specimens. Histochemically labeled sialomucins, sulfomucins and Paneth cells (the 3 main findings in gastric IM) were quantified in separate sections with the aid of an image analyzer. In total, 1,321 sections corresponding to 6 gastrectomy specimens were quantified. Sialomucins and sulfomucins were more extensively distributed in the 4 specimens with carcinoma than in the 2 without carcinoma (one having a peptic ulcer and the other, hereditary gastric cancer syndrome (HGCS) without carcinoma). On the other hand, quantitative analysis in Swedish specimens indicated that the highest values for sialomucins, sulfomucins and Paneth cells were present in HGCS. When Swedish and Japanese specimens with adenocarcinoma were compared, only sulfomucins (denoting Types II and III IM) were significantly higher in those carrying an intestinal-type carcinoma (ITC) than in those with diffuse-type carcinoma (DTC). The results substantiate those obtained with gastric biopsies by other authors. On the other hand, the mucosal extension and the amount of sulfomucins are not comparable parameters (since that mucin was not equally distributed, but "concentrated" in certain areas in the mucosa). One possible conclusion is that the focal distribution of acidic mucins and of Paneth cells in the gastric mucosa may strongly influence their detection rate in gastric biopsies. Thus, haphazard biopsy of the gastric mucosa may fail to sample areas with sulfomucins in population studies aiming to detect individuals at risk. Such sampling errors in gastric biopsies may explain the conflicting results on this subject appearing in the literature.