The association between Helicobacter pylori infection and early gastric cancer:: A meta-analysis

Objective: Helicobacter pylori (H. pylori) is recognized as a definite carcinogen for gastric cancer. The prevalence of H. pylori infection in patients with gastric cancer varies widely among studies and no meta-analysis on the prevalence of H. pylori infection in early gastric cancer (EGC) has been performed. We aimed to review systematically the relationship between H. pylori infection and EGC, and different types of EGC. Methods: Observational studies reporting raw data on the prevalence of H. pylori infection in EGC and controls, or comparing different types of EGC, conducted in adult populations, and published in the English language were identified through MEDLINE and EMBASE up to June 2006. Results: Of 87 relevant studies, 19 case-control studies met inclusion criteria. Of these, 15 studies compared EGC (N = 2,722) and non-neoplasm controls (N = 13,976) or advanced gastric cancer (AGC) (N = 1,130), 9 studies compared the intestinal-type (I-EGC) and diffuse-type (D-EGC) of EGC, and 2 studies compared the differentiated-type (DF-EGC) and undifferentiated-type (UDF-EGC) of EGC and were included in the meta-analysis. The prevalence of H. pylori infection was significantly higher in patients with EGC (87.3%) than in non-neoplasm controls (61.4%) (OR 3.38, 95% CI 2.15-5.33, P < 0.00001). However, significant heterogeneity was seen (P < 0.00001). Four large sample (N >= 100) studies (N = 2,060) may result in the heterogeneity, but the conclusion remained unchanged when sensitivity analysis was made with the other 11 homogeneous small sample studies alone, in which the prevalence of H. pylori infection was significantly higher in EGC (N = 662) than that in controls (N = 5,898) (87.8% vs 68.6%, P < 0.00001), and the odds ratio (OR 3.28, 95% CI 2.34-4.61) was similar to the large sample studies alone (OR 3.40, 95% CI 1.14-10.12). The prevalence of H. pylori infection in EGC was significantly higher than in AGC (6 studies) (OR 2.13, 95% CI 1.75-2.59) and 16-fold higher in patients with DF-EGC than in those with UDF-EGC (OR 16.53, 95% CI 2.64-103.43). No significant difference in the prevalence of H. pylori infection was seen between the patients with I-EGC and D-EGC types (OR 0.75, 95% CI 0.26-2.18). Conclusions: This study indicates that H. pylori infection is strongly associated with early gastric cancer when compared with non-neoplasm controls or advanced gastric cancer. To determine more accurately the effect size of H. pylori in EGC, age-matched normal controls or adjusting for age in the analysis should be considered in H. pylori-related gastric cancer case-control studies.