Identification of immunometabolism-associated genes and immune infiltration in sepsis-induced cardiomyopathy using integrated bioinformatics and machine learning approaches

ObjectiveSepsis-induced cardiomyopathy, a life-threatening complication of sepsis, carries a significant mortality risk. Despite its clinical severity, the immunometabolic mechanisms driving its pathogenesis remain poorly understood. This study aimed to identify immunometabolism-associated genes in sepsis-induced cardiomyopathy and analyze immune infiltration.MethodsWe integrated single-cell and bulk RNA sequencing data from septic cohorts, using a multistep bioinformatic approach involving machine learning and weighted gene co-expression network analysis to identify key immunometabolism-associated genes and explore their role in sepsis-induced cardiomyopathy. Key findings were validated using in vitro and in vivo experiments.ResultsWe identified two distinct macrophage subsets (functional and senescent) and prioritized four hub genes, including ABLIM1, BCL9L, SMAD3, and DHRS3, which formed a robust prognostic model (area under the curve = 0.724). In vivo and in vitro validation confirmed significant downregulation of these hub genes in sepsis-induced cardiomyopathy. Immune infiltration analysis demonstrated significant increases in monocytes, M0/M1 macrophages, and neutrophils in sepsis, with hub genes showing significant correlations with specific immune cell subsets.ConclusionsThis study identifies ABLIM1, BCL9L, SMAD3, and DHRS3 genes as pivotal biomarkers and potential therapeutic targets in sepsis-induced cardiomyopathy, linked to immune-metabolic dysregulation and myocardial injury. These findings provide new insights for early diagnosis and precision immunotherapy of sepsis-induced cardiomyopathy.