Glycohemoglobin Variations and Long-term Risk of Incident Heart Failure in Patients With Type 2 Diabetes

Mayo Clin Proc. 2025 Jul 1:S0025-6196(25)00002-3. doi: 10.1016/j.mayocp.2024.12.022. Online ahead of print.

Abstract

Objective: To investigate the association between long-term glycohemoglobin variations and the onset of incident heart failure (HF) in individuals with diabetes, addressing the phenotype concerning left ventricular ejection fraction and etiology.

Patients and methods: This retrospective study identified patients with type 2 diabetes from a multicenter database between January 2007 and December 2010 who had multiple hemoglobin A1c (HbA1c) measurements during the 4 years after the initial diagnosis of diabetes, calculating variations using average real variability. Patients were followed from the end of run-in period, with HF hospitalization as the primary outcome. Cox proportional hazard models, adjusted for mean HbA1c and baseline characteristics, were used to evaluate the relationship between HbA1c variations and outcomes.

Results: We identified 53,748 patients with an average of 12.7 HbA1c measurements. The average follow-up time was 6.2 years, and the cumulative incidence of HF hospitalization was 6.0 per 1000 person-years. Patients with an average visit-to-visit change in HbA1c exceeding 0.778% were independently associated an elevated risk of hospitalization (HR, 1.28; 95% CI, 1.09 to 1.52). This association persisted regardless of reduced (HR, 1.31; 95% CI, 1.02 to 1.68) or preserved ejection fraction (HR, 1.26; 95% CI, 1.01 to 1.58) and ischemic (HR, 1.30; 95% CI, 1.04 to 1.61) or nonischemic HF (HR, 1.71; 95% CI 1.34 to 2.18).

Conclusion: Hemoglobin A1c variations independently correlated with increased risk of HF hospitalization, encompassing both reduced and preserved ejection fraction, as well as ischemic and nonischemic HF. An average visit-to-visit change in HbA1c exceeding 0.778% can serve as a simple indicator for the risk of HF in patients with type 2 diabetes.