AimsPerturbations of myocardial metabolism and energy depletion are well-established hallmarks of heart failure (HF), yet methods for their systematic assessment remain limited in humans. This study aimed to determine the ability of computational modelling of patient-specific myocardial metabolism to assess individual bioenergetic phenotypes and their clinical implications in HF.Methods and resultsBased on proteomics-derived enzyme quantities in 136 cardiac biopsies, personalised computational models of myocardial metabolism were generated in two independent cohorts of advanced HF patients together with sex- and body mass index-matched non-failing controls. The bioenergetic impact of dynamic changes in substrate availability and myocardial workload were simulated, and the models' ability to predict the myocardial response following left ventricular assist device (LVAD) implantation was assessed. Compared to controls, HF patients had a reduced ATP production capacity (p < 0.01), although there was remarkable interindividual variance. Utilisation of glucose relative to fatty acids was generally higher in HF patients, depending on substrate availability and myocardial workload. The ratio of fatty acid to glucose utilisation was associated with reverse cardiac remodelling after LVAD implantation and highly predictive of an improvement in left ventricular ejection fraction ≥10% (C-index 0.94 [0.81-1.00], p < 0.01). System-level simulations identified fatty acid administration and carnitine supplementation in those with low mitochondrial carnitine content as potential pharmacological interventions to restore myocardial substrate utilisation.ConclusionsComputational modelling identified a subset of advanced HF patients with preserved myocardial metabolism despite a similar degree of systolic dysfunction. Substrate preference was associated with the myocardial response after LVAD implantation, which suggests a role for substrate manipulation as a therapeutic approach. Computational assessment of myocardial metabolism in HF may improve understanding of disease heterogeneity, individual risk stratification, and guidance of personalised clinical decision-making in the future.