We thank Nabben and colleagues for their constructive Correspondence (Metabolic remodelling in heart failure revisited. Nat. Rev. Cardiol. https://doi.org/10.1038/s41569-018-0115-8; 2018)1 on our Review (Metabolic remodelling in heart failure. Nat. Rev. Cardiol. 15, 457–470; 2018)2. We agree with their comments and acknowledge that most of these critical points result from the necessity of giving readers a comprehensive overview of a broad area of cardiovascular research within the limited space of a Review article. Here are our responses to their comments.

We decided to discuss diabetic cardiomyopathy separately from pressure overload-induced and ischaemic heart failure (HF) because they differ completely from a metabolic perspective2. Consequently, we refer in the text to the pressure-overloaded or ischaemic heart as the “failing heart”, whereas we refer to ventricular dysfunction associated with diabetes mellitus in the absence of hypertension or coronary artery disease as the “diabetic heart”2,3. This distinction is, of course, a simplification, which was intended to make the topic more understandable to readers, and we apologize if it had the opposite effect.

We completely agree with Nabben and colleagues when they state that the role of lipotoxicity and insulin resistance in the development and progression of HF is not fully understood1. However, several animal models in which lipotoxicity was induced with genetic manipulation have clearly shown that this process detrimentally affects cardiac function (reviewed previously4), and intramyocardial accumulation of toxic lipid intermediates was observed in patients with HF with or without diabetes5. On these grounds, we propose that metabolic derangements in patients with HF might contribute to the progression of cardiac dysfunction; accordingly, throughout our Review, we never present this hypothesis as an established model.

With regard to fatty acid inhibition as a therapeutic option in HF, we again agree with the comment made by Nabben and co-workers1. However, we think that this point was adequately discussed in our Review. Indeed, the section on ‘Diabetic cardiomyopathy’ should provide readers with sufficient insight to exclude the concept that inhibition of fatty acid oxidation could be applied to patients with diabetic cardiomyopathy, and we did cite all the relevant studies in which this therapeutic strategy was tested in humans, including the randomized study showing the lack of benefit of giving trimetazidine treatment in addition to optimal medical therapy in patients with non-ischaemic HF6.

Given that substrate metabolism in the normal heart has been the subject of other excellent reviews7, we decided to refer to these works and dedicate more space to other, unresolved issues in the field. Although we omitted to mention the important role of glucose and fatty acid uptake in regulating substrate utilization, we did refer to studies in which the expression of glucose and fatty acid transporters was genetically manipulated (see Figure 3 in our Review2).

Finally, it is, of course, true that metabolic derangements vary during the progression of HF, and we referred to the necessity of distinguishing between compensated hypertrophy and end-stage HF in the section on ‘Altered substrate metabolism in heart failure’ with regard to both fatty acid and glucose utilization.