Abstract Introduction/Purpose Recent studies suggest that the ketone bodies (KB) beta-hydroxybutyrate (HBA) and acetoacetate (ACA) are relevant substrates for myocardial energy metabolism in heart failure (HF). However, the amount of KB-dependent mitochondrial respiration in HF has not yet been quantified. High-resolution respirometry (HRR) is the gold-standard method for quantifying substrate-dependent mitochondrial oxidative capacity. Therefore, we aimed to evaluate the association between myocardial KB-dependent oxidative capacity and human HF using HRR. Methods The primary endpoint of this single-center prospective cohort study was KB-dependent and overall mitochondrial oxidative capacity in permeabilized myocardium. We used two different protocols for our measurements. In the first HRR protocol, conventional respirometry substrates including fatty acids, complex I- and complex II substrates were used to quantify the maximum coupled oxidative phosphorylation capacity (OXPHOS). Leak respiration was quantified using oligomycin to calculate the respiratory control ratio (RCR, state 3/state 4o), and leak control ratio (LCR, state 4o/state u). In the second protocol which was recently established by our working group, HBA and ACA were used as respirometry substrates to quantify KB-dependent respiration down to the enzyme level. The relative contribution of the KBs was obtained by comparing HBA- and ACA-derived respiration to OXPHOS. We applied both protocols to myocardial samples from individuals with advanced HF (HF group, catheter-acquired endomyocardial biopsies from non-ischemic HF patients or samples of explanted hearts collected during orthotopic heart transplantation surgery in end-stage heart failure patients) as well as previously heart transplanted individuals without current heart failure (Control group). Results Controls and HF had similar demographic characteristics, with comparable age distribution (mean±SD, 56.3±10.3 vs. 54.2±10.7 years, p=0.36), sex (68.75% vs. 66.67% male, p=0.87), or body mass index (25.0±5.8 vs. 29.0±10.6 kg/m2, p=0.08) and differed significantly in cardiac index (3.06±0.67 vs. 1.93±0.44) and ejection fraction (55.8±10.2 vs. 26.6±9.1) (both p<0.0001). The HF group exhibited lower OXPHOS and respiration for all substrate combinations tested (Fig. 1A). In contrast, KB-dependent respiration did not differ between HF and controls (Fig. 1B). The relative HBA-dependent respiration (Fig. 1C) and the relative and absolute ACA-dependent respiration (Fig. 1C and D) were higher in the HF group compared to controls. Mitochondrial uncoupling (LCR) (0.47±0.29 vs. 0.44±0.09 [AU], p=0.79) and coupling efficiency (RCR) (2.75±1.38 vs. 2.11±0.39 [AU], p=0.15) were not different between the groups. Conclusion These findings suggest that mitochondria in the failing heart can utilize KBs for OXPHOS to a greater extent and further hint towards KB metabolism as a potential therapeutic target for HF.