Metabolic dysfunction in arrhythmogenic cardiomyopathy: a cardiopulmonary exercise testing and metabolomics study
G De Zan, K Taha, J Ciapaite, J Jans, N Van Der Wilt, M Harakalova, P Van Der Harst, M J Cramer, M Guglielmo, T Takken, A S J M Te RieleAbstract
Background
Arrhythmogenic cardiomyopathy (ACM) is an inherited myocardial disease associated with ventricular arrhythmias (VAs) and progressive right ventricular systolic dysfunction. Exercise triggers most VAs and increases phenotypic penetrance among ACM-associated pathogenic variant carriers. However, evidence on the interaction between exercise and ACM remains limited. Moreover, while metabolic dysfunction has been demonstrated in other cardiomyopathies and heart failure, data in ACM are scarce(1,2).
Purpose
To investigate the metabolic response of ACM-associated pathogenic variant carriers to cardiopulmonary exercise testing (CPET) using indirect calorimetry and untargeted metabolomics.
Methods
This cross-sectional case-control study included 20 plakophilin-2 (PKP2) pathogenic variant carriers and 10 age- and sex-matched healthy controls (35 ± 11 vs 30 ± 3 years, respectively, p=0.20; 60% males in both groups). The PKP2 group included 10 patients with a definite ACM diagnosis as per 2010 Task Force Criteria. Participants performed a CPET with incremental workload until volitional exhaustion. Fat and carbohydrate (CHO) oxidation were calculated using indirect calorimetry(3). CPET data were collected at rest, low-intensity exercise (i.e. increase in heart rate of 30 bpm from baseline), moderate-intensity exercise (i.e. reaching the ventilatory anaerobic threshold; VT1) and peak-intensity exercise. Untargeted metabolomics by mass spectrometry on dried blood spots was performed from samples collected 3 minutes after peak exercise. Metabolites significantly different in both univariate analysis (fold-change analysis with an FC >2 cut-off) and multivariate analysis (s-PLSDA) were considered of interest.
Results
Maximal oxygen uptake (VO2max), peak heart rate and peak power output were significantly lower in the PKP2 group than in controls, despite no difference in peak respiratory exchange ratio (Figure 1). PKP2 carriers and controls did not show differences in fat and CHO oxidation at rest, while at low intensity exercise (i.e. before VT1) fat oxidation was significantly lower (0.18 ± 0.10 g/min vs 0.32 ± 0.14 g/min, p=0.002) and CHO oxidation higher (0.80 ± 0.34 g/min vs 0.30 ± 0.26 g/min, p<0.001) in the PKP2 group than in controls, even when correcting for VO2max and body mass index (BMI). Out of 1792 detected peaks, levels of 31 metabolites were significantly different in both statistical analyses (Figure 2). Among those, metabolites involved in fatty acid metabolism (dodecenoylcarnitine; tetradecadiencarnitine; oxo-methylthiobutanoic acid) and branched-chain amino acid metabolism ([iso]leucine; methyl-ketovaleric acid) were significantly lower in PKP2 carriers after exercise.
Conclusion
Our results support metabolic dysfunction upon exercise in PKP2 carriers. CPET results showed impaired fat oxidation and untargeted metabolomics further suggests a shift towards a more CHO-based metabolism.Fig 1.CPET dataFor image description, please refer to the figure legend and surrounding text.Fig 2.Heatmap of metabolites of interestFor image description, please refer to the figure legend and surrounding text.