Helicobacter pylori infection and early onset myocardial infarction: case-control and sibling pairs study.
Danesh J., Youngman L., Clark S., Parish S., Peto R., Collins R.
OBJECTIVES: To examine the association between coronary heart disease and chronic Helicobacter pylori infection. DESIGN: Case-control study of myocardial infarction at young ages and study of sibling pairs with one member affected and the other not. SETTING: United Kingdom. PARTICIPANTS: 1122 survivors of suspected acute myocardial infarction at ages 30-49 (mean age 44 years) and 1122 age and sex matched controls with no history of coronary heart disease; 510 age and sex matched pairs of siblings (mean age 59 years) in which one sibling had survived myocardial infarction and one had no history of coronary heart disease. MAIN OUTCOME MEASURES: Serological evidence of chronic infection with H pylori. RESULTS: 472 (42%) of the 1122 cases with early onset myocardial infarction were seropositive for H pylori antibodies compared with 272 (24%) of the 1122 age and sex matched controls, giving an odds ratio of 2.28 (99% confidence interval 1.80 to 2.90). This odds ratio fell to 1.87 (1.42 to 2.47; P<0.0001) after smoking and indicators of socioeconomic status were adjusted for and to 1.75 (1.29 to 2.36) after additional adjustment for blood lipid concentrations and obesity. Only 158 of the 510 pairs of siblings were discordant for H pylori status; among these, 91 cases and 67 controls were seropositive (odds ratio 1.33 (0.86 to 2.05)). No strong correlations were observed between H pylori seropositivity and measurements of other risk factors for coronary heart disease (plasma lipids, fibrinogen, C reactive protein, albumin, etc). CONCLUSION: In the context of results from other relevant studies, these two studies suggest a moderate association between coronary heart disease and H pylori seropositivity that cannot be fully accounted for by other risk factors. But even if this association is causal and largely reversible by eradication of chronic infection, very large randomised trials would be needed to show this.