摘要

Whether hypertension is associated with -572 C>G or -174 G>C polymorphism in interleukin (IL)-6 genes still remains hazy and ambiguous. We conducted a meta-analysis to offer a more reliable and clearer evaluation about the association. Electronic literature databases including PubMed, Web of Science, EMBASE, Google Scholar, Chinese National Knowledge Infrastructure and Wanfang database were searched. The study included the following: evaluating associations between -572 C>G or -174 G>C polymorphism in IL-6 gene and hypertension; case-control design; essential information must be offered; precise diagnostic criteria of hypertension; and no language restriction. Patients who met the diagnostic criteria and controls without a history of hypertension were included. Interventions were not available. A quality assessment was conducted using Newcastle-Ottawa scale. Combined odds ratios with 95% confidence intervals were calculated in 5 genetic models. Sources of heterogeneity were explored by subgroup analysis, meta-regression, and Galbraith plots. Finally, test for publication bias was performed to prove the stabilization. Fifteen studies were finally included. Eleven articles were judged high-quality reports. Overall, the -572 C>G polymorphism was proved to be significantly associated with hypertension in 4 genetic models. Subgroup analysis based on ethnicity revealed significant associations in Asian population in recessive model and homozygote comparison. The association in Europeans and Mid-East required further confirmation. No significant association was observed between the -174 G>C polymorphism and hypertension under all of the genetic models. The limitations of the study were the following: restrictive number of eligible studies limited the extrapolation range in subgroup analysis; gene-environment factors could not be described due to lack of data; some relevant studies could not be included because of various reasons. Current researches supported the association between the development of hypertension and the -572 C>G rather than -174 G>C polymorphism. Future well designed epidemiological studies may evaluate the possible gene-environment interactions.