Abstract
Pediatric cardiomyopathy is a major cause of diastolic heart failure; yet, its mechanisms remain unclear. Global myocardial transcriptomic and blood lipidomic profiling revealed a distinct metabolic signature of diastolic dysfunction marked by dysregulated lipid signaling. A machine learning model using these gene markers accurately classified diastolic dysfunction across cardiomyopathy subtypes. Lipidomic changes included excess saturated lipids and impaired oxidation that correlated with myocardial gene expression. Induced pluripotent stem cell cardiomyocytes from patients with diastolic dysfunction exhibited lipid accumulation, and mitochondrial dysfunction that was rescued by semaglutide. These findings define a new molecular phenotype of diastolic dysfunction and point to abnormal lipid signaling as a promising therapeutic target in childhood cardiomyopathy.