Fasting plasma glucose and incident heart failure risk: a population-based cohort study and new meta-analysis

Background: There remains uncertainty regarding the association between fasting plasma glucose (FPG) and risk of heart failure (HF) in individuals without a history of diabetes.

Methods and results: We assessed the association between FPG and HF risk in a population-based cohort of 1,740 men aged 42-61 years who were free from HF or diabetes at baseline. During a mean follow-up of 20.4 years, 146 participants developed HF. In age-adjusted analysis, the hazard ratio (HR) for HF per 1 mmol/L increase in FPG was 1.34 (95% confidence interval 1.22-1.48). This association persisted after adjusting for established HF risk factors: HR 1.27, 95% confidence interval 1.14-1.42. The findings remained consistent across several clinical subgroups and in analyses excluding incident coronary heart disease or diabetes during follow-up. In a meta-analysis of 10 prospective studies involving 4,213 incident HF cases, the HR for HF per 1 mmol/L increase in FPG level was 1.11 (95% confidence interval 1.04-1.17), with evidence of heterogeneity between studies (I(2) = 79%; 95% confidence interval 63%-89%; P < .001). The corresponding HR was 1.12 (95% confidence interval 1.08-1.18) on exclusion of the single study that accounted for the heterogeneity.

Conclusions: There exists a positive, continuous, and independent association between FPG and risk for HF. Studies are warranted to evaluate the causal relevance of these findings.