Multiple Dietary Indexes Associated With Lower Risk of Heart Failure and Its Subtypes in the Health Professionals Follow-Up Study

Abstract

ObjectivesDue to the increasing disease burden and the limited effectiveness of pharmacological therapies, strategies to predict and prevent heart failure (HF) are urgently needed. Healthy eating has been shown associated with lower risks of multiple cardiovascular diseases, while studies of dietary quality and HF incidence are still sparse. The Prime Diet Quality Score (PDQS) was developed to overcome the limitations of applying the well-established Alternative Healthy Eating Index (AHEI) in clinical settings. In this study, we aimed to investigate whether the AHEI, as well as the PDQS, is associated with the risk of HF and its subtypes, heart failure with preserved ejection fraction (HFpEF) and heart failure with reduced ejection fraction (HFrEF). MethodsOur study included 41,520 men who were free from cardiovascular disease and cancer at baseline from the Health Professionals Follow-up Study (HPFS). The AHEI and PDQS were computed based on dietary data that were collected every 4 years beginning in 1986 using semi-quantitative food frequency questionnaires (FFQs). Heart failure, as well as HFpEF and HFrEF, were first self-reported and then adjudicated based on the review of medical records. Data were analyzed from 1986 through 2008. Associations between the exposures and outcomes were estimated with multivariate-adjusted Cox proportional hazards models. ResultsDuring 930,597 person-years of follow up, 803 HF cases were documented. Among those with ejection fraction data, 184 participants were identified as HFpEF and 181 were identified as HFrEF. A higher AHEI was significantly associated with a lower risk of HFrEF after adjusting for potential confounders (HRQ5 vs. Q1 = 0.60, 95% CI: 0.38–0.93, P-linear = 0.01). Similar inverse associations with incident HFrEF were also observed for the PDQS (HRQ5 vs Q1 = 0.54, 95% CI: 0.34–0.85, P-linear = 0.009). We did not observe significant associations of the dietary indexes with HFpEF, while the two dietary indexes were inversely, but non-significantly associated with total HF. ConclusionsA healthy overall diet was associated with lower risk of HFrEF. Both AHEI and PDQS showed similar predictability of the HF outcomes. Moreover, the food-based PDQS can be readily computed with wider clinical applicability. Funding SourcesThe HPFS is funded by the National Heart, Lung, and Blood Institute and National Cancer Institute.