Circulating ketone bodies and risk of incident atrial fibrillation: insights from the MESA and UK Biobank cohorts

Abstract Aims Circulating ketone bodies (KB) are an important source of metabolic fuel for the myocardium under physiological and pathological conditions. Prior studies have linked KB levels with adverse cardiovascular outcomes. However, their relationship with atrial fibrillation (AF) risk remains unknown. The objective is to evaluate the association of KB levels with the risk of incident AF. Methods and results This prospective, multicentre cohort study recruited patients without baseline AF from the Multi-Ethnic Study of Atherosclerosis (MESA) and the UK Biobank (UKB), respectively. Total KB levels were measured by nuclear magnetic resonance spectroscopy in both studies. The associations between total and individual KB and incident AF were evaluated using multivariable-adjusted Cox proportional hazard models. Prespecified interaction analyses were performed for smoking status, history of diabetes mellitus, history of hypertension, body mass index, and self-reported race/ethnicity (to represent a social construct). A total of 6783 participants [mean (standard deviation, SD) age 62 (10) years, 52% women, 38% White, 27.5% Black, 21.8% Hispanic, and 11.6% Chinese American] from MESA and 116 480 participants [mean (SD) age 56.5 (8) years, 54% women, 94% White] from the UKB were included. Over a median follow-up of 16.5 (10.3–17.4) years in MESA and 13.6 (12.8–14.4) years in the UKB, higher levels of total KB were associated with a higher risk of incident AF. In MESA, each doubling of baseline total KB levels was associated with a 1.08-fold increased hazard of incident AF on follow-up [95% confidence interval (CI): 1.00–1.16] after adjustment for confounding factors. Consistent results were obtained in a subgroup analysis stratified by different types of KB (β-hydroxybutyrate, acetoacetate, and acetone). There was a significant interaction with current smoking, such that per doubling of total KB levels was associated with a 1.32-fold increased hazard for incident AF in current smokers (95% CI: 1.08–1.61), but this association was not seen among former/never smokers [hazard ratio (95% CI): 1.05 (0.95–1.13)]. These results were replicated in the UKB. Conclusion Higher concentrations of KB are associated with an increased risk of AF in predominantly healthy, community-based cohorts, acknowledging that many participants also had comorbidities such as hypertension and diabetes, particularly among smokers. Ketone bodies could serve as a potential biomarker for identifying populations at risk for AF.