Risk factors for intestinal metaplasia in adult residents of Matzu: a cross-sectional study

• Published in: Journal of the Formosan Medical Association Vol. 101; no. 12; pp. 835 - 840
• Main Authors: Chang, Chia-Hsuin, Wu, Ming-Shiang, Chang, Yu-Ting, Chang, Ming-Chu, Shun, Chia-Tong, Liu, Cheng-Ying, Lin, Jaw-Town
• Format: Journal Article
• Language: English
• Published: Singapore 01.12.2002
• Subjects: Adult; Aged; Antibodies, Bacterial - blood; Cross-Sectional Studies; Female; Helicobacter pylori - immunology; Humans; Immunoglobulin G - blood; Intestines - pathology; Male; Metaplasia - diagnosis; Middle Aged; Pepsinogen A - blood; Precancerous Conditions - diagnosis; Risk Factors; Stomach Neoplasms - diagnosis; Taiwan - epidemiology; Taiwan
• ISSN: 0929-6646

People living on Matzu, a group of small islets, have the highest mortality rate from gastric cancer (GC) in Taiwan. Intestinal metaplasia (IM) is a precursor lesion of GC. We therefore conducted a gastroscopic screening program in this district to evaluate the risk factors associated with IM. A total of 274 residents of Matzu, aged 30 years or older, underwent testing for anti-Helicobacter pylori IgG antibody and serum pepsinogen (PG) I/PG II levels, and gastroscopic examination. Univariate and multivariate logistic regression models were used to identify risk factors associated with IM in terms of odds ratios (ORs) and 95% confidence intervals (CIs). Among the 274 subjects, 118 (43%) were confirmed to have IM by histologic examination. Individuals aged 60 years or older had a 2.66 times higher risk (95% CI 1.38-5.13) of having IM, as compared with those less than 60 years of age. Subjects with duodenal ulcer had a 60% lower risk of coexisting IM (OR, 0.40; 95% CI, 0.19-0.83). Meanwhile, subjects with gastric ulcer were not at increased risk of IM (OR, 0.84; 95% CI, 0.34-2.08). Low serum PG I level, low PG I/PG II ratio, and positive anti-H. pylori IgG antibody, were associated with IM in the univariate analysis. While age and duodenal ulcer maintained their independent effects in the multivariate analysis, only a low PG I/PG II ratio and H. pylori positivity were significantly associated with IM. This study of residents of Matzu found that age and H. pylori infection are risk factors for IM, while subjects with duodenal ulcer have a lower incidence of coexisting IM. Low PG I/PG II ratio is a potential noninvasive biomarker of IM.