Integrating ketone bodies in multi-marker risk prediction of incident heart failure in the multi-ethnic study of atherosclerosis

Abstract Aims Circulating ketone bodies (KB) have emerged as a potential adjunctive biomarker for incident heart failure (HF) risk and might provide incremental information beyond established biomarkers. A multi-marker risk score may improve risk stratification of incident HF in the community. The authors aim to develop a risk score using N-terminal proB-type natriuretic peptide (NT-proBNP) and high-sensitivity troponin (hs-cTnT) and a unique systemic biomarker of KB to predict HF among participants without cardiovascular disease. Methods and results A multi-marker score was developed incorporating NT-proBNP ≥ 125 pg/mL, hs-cTnT ≥ 14 ng/L, and total KB ≥ 75th percentile (316 μmol/L), with one point allocated for each abnormal marker among Multi-Ethnic Study of Atherosclerosis participants. Multivariable Cox model was used to assess the association between multi-marker risk score and the risk of incident HF. Among 6748 participants, there were 383 incident HF events over a median follow-up of 15.7 years. The three biomarkers exhibited poor correlation with one another (r < 0.06 for all). The addition of KB to NT-proBNP and hs-cTnT to identify incident HF improved 5- and 10-year risk prediction (C-statistic 0.74 vs. 0.77, P = 0.02 and 0.70 vs. 0.73, P = 0.02, respectively). There was no evidence of miscalibration using the multi-marker score for predicting 5- and 10-year HF risk (P > 0.05). A graded association was observed between the multi-marker score and risk of HF independent of established clinical factors. Conclusion The addition of plasma KB to a clinical risk score using biomarkers of cardiac injury and stress may further improve the prediction of incident HF.