Objective: We tested whether two single-nucleotide polymorphisms (SNPs) (rs2383207 and rs10757278), previously strongly associated with myocardial infarction, are independently associated with stroke and coronary events in patients with hypertension.
Methods: The Nordic Diltiazem study compared the effects of calcium antagonist and beta-blocker or diuretic-based antihypertensive treatment on cardiovascular events in 10 881 patients with hypertension, of whom 5262 patients provided DNA for the present study. We related SNPs rs2383207 and rs10757278 to stroke and to myocardial infarction and coronary revascularizations (coronary events) using crude and multivariate adjusted Cox proportional hazards models.
Results: The G-allele of both SNPs predicted coronary events in crude recessive models [hazard ratios = 1.36, 95% confidence interval (CI) = 1.04-1.79, P = 0.02 for rs10757278 and hazard ratios = 1.40, 95% CI = 1.08-1.81, P = 0.01 for rs2383207] as well as after adjustment for classical cardiovascular risk factors. The G-allele of both SNPs predicted incident stroke in crude additive models [rs2383207 hazard ratios = 1.25 (95% CI = 1.02-1.53), P = 0.04 and rs10757278 hazard ratios = 1.34 (95% CI = 1.09-1.65), P = 0.006], as well as after adjustment for classical cardiovascular risk factors and after additional adjustment for prevalent and incident coronary events, atrial fibrillation, ischemic heart disease and congestive heart failure. As was the case for coronary events, the excess genetic risk of stroke was driven by subjects homozygous for the risk allele.
Conclusion: Genetic variation at the CDKN2A/CDKN2B locus predicts stroke in hypertensive patients. The genetic association with stroke is independent of classical cardiovascular risk factors and of all prevalent and incident coronary events, suggesting that gene variation at this locus promotes either atherosclerosis or another disease mechanism that is common to both coronary and cerebrovascular disease.