Course and trajectories of insulin resistance, incident heart failure and all-cause mortality in nondiabetic people.

Background: In nondiabetic people, the long-term effects of insulin resistance (IR) on heart failure (HF) and all-cause mortality have not been studied.

Objectives: To examine the association between IR trajectories and incident HF and all-cause mortality in a nondiabetic population.

Methods: We studied 7835 nondiabetic participants from the Atherosclerosis Risk in Communities (ARIC) Study. We estimated IR with several methods: Homeostatic Model Assessment-Insulin Resistance (HOMA-IR), triglyceride to high-density lipoprotein cholesterol ratio (TG/HDL-C), triglyceride glucose Index (TyG Index), and metabolic score for insulin resistance (METS-IR). The latent class analysis identified two trajectories for HOMA-IR ('low level' and 'high level'), and three trajectories for TG/HDL-C, TyG index, and METS-IR ('low level', 'moderate level', and 'high level'). Cox proportional hazard models were employed to examine the association.

Results: Participants in the 'high level' group of HOMA-IR trajectory patterns were more likely to have incident HF and all-cause mortality with HRs (95% CIs) of 1.29 (1.11-1.50) and 1.31(1.19-1.44), respectively, compared to the 'low level' group. Similarly, participants in the 'moderate level' and 'high level' groups of TG/HDL-C, TyG index, and METS-IR trajectories had elevated risks of incident HF and all-cause mortality. However, no increased risk was found for all-cause mortality for men in the 'moderate level' and 'high level' group of TG/HDL-C, TyG index, and METS-IR relative to the 'low level' group.

Conclusions: Long-term moderate and high IR levels were positively associated with increased risks of incident HF for both males and females. For all-cause mortality, however, consistent associations were found only in women.