Angiotensinogen gene and essential hypertension in the Japanese: extensive association study and meta-analysis on six reported studies

Background Accumulating evidence has supported the pathophysiological role of angiotensinogen in essential hypertension. However, some studies of molecular genetics have implicated that there may be an ethnic variation concerning the disease susceptibility of the AGT gene. Objectives and methods To evaluate the importance of this candidate gene for hypertension, we undertook an extensive association study in the Japanese. This case-control study was conducted in a total of 1232 individuals consecutively enrolled in a single institution, divided into two subgroups: one subgroup comprised 254 hypertensive and 224 normotensive subjects and the other comprised 463 hypertensive and 291 normotensive subjects. A meta-analysis was subsequently performed on six Japanese studies including the present study. Results No significant association was observed between a molecular variant of AGT, Thr235, and hypertension status in our case-control study. Moreover, this finding was extendible to another AGT polymorphism, G-6A, one of the potential functional polymorphisms in the promoter region, because these two polymorphisms proved to be in complete linkage disequilibrium in the studied population. The mete-analysis revealed that the pooled estimate of the odds ratio across the studies was 1.22 (95% CI 1.05-1.42), and that there was significant evidence against homogeneity of the odds ratios among the studies included (chi(2) = 19.8, df = 5, P = 0.0014), In particular, a large range of variation (60-83%) was found for the allele frequency of Thr235 among control subjects of the six Japanese case-control studies. Conclusions Although the meta-analysis appears in favour of association between the AGT variant and essential hypertension in the Japanese, there is considerable heterogeneity among the studies and the evidence is also rather borderline. Further comprehensive approaches are needed to resolve this debatable issue. J Hypertens 1999, 17:757-763 (C) Lippincott Williams & Wilkins.