Mitral valve repair for pediatric dilated cardiomyopathy with severe functional mitral regurgitation: a mechanism-anchored, proof-of-concept preoperative selection framework with supportive cohort and pooled outcomes

Functional mitral regurgitation (FMR) develops in 40–65% of children with dilated cardiomyopathy (DCM) and independently predicts death or transplantation. Published pediatric mitral valve repair (MVR) series from 2003 to 2024 report clinical stabilization rates of 0% to 100% and perioperative mortality of 0% to 50%. This heterogeneity is multifactorial, reflecting the surgical era, technique, institutional experience, patient acuity, and reporting standards, with patient selection being a contributing factor that has not yet been systematically examined. No preoperative framework currently exists to identify which children are most likely to benefit. A mechanism-anchored, three-category proof-of-concept clinical framework was defined a priori. Patients were classified as Non-MR-dominant if any of four affirmative features were documented (N1: toxic, storage, infiltrative, or myopathic etiology; N2: primary structural mitral valve disease; N3: preoperative irreversible end-organ failure or ECMO with multi-organ dysfunction; N4: severe RV dysfunction, fixed pulmonary hypertension, or severe tricuspid regurgitation from RV/annular dilatation). Patients without any N-feature who met three volume-overload criteria (idiopathic, familial, or sarcomeric DCM; functional MR ≥ 3+; LVEDd Pettersen Z-score ≥ + 5) were classified as MR-dominant. A Borderline MR-dominant category captured patients who did not meet all MR-dominant criteria but satisfied one of three predefined conditions (infancy with intermediate fractional shortening; post-myocarditis with documented regional scar or fibrotic substrate; or an LVEDd Pettersen Z-score of + 3.5 to < + 5 with severe functional MR). The framework was applied to a single-center cohort of six consecutive children and, masked to outcome, to all 22 previously published pediatric MVR-receiving patients. In the index cohort, three MR-dominant children underwent MVR and sustained remodeling, with transplant-free survival at a median 30-month follow-up and NT-proBNP reductions of 86–99%. Three children not selected for MVR (an MR-dominant child whose family declined surgery, and two Non-MR-dominant children) all died. In the pooled MVR-receiving cohort ( n = 25), MR-dominant plus Borderline patients achieved clinical stabilization and/or remodeling in 16/20 (80%) versus 1/5 (20%) Non-MR-dominant patients (Fisher exact p = 0.024; odds ratio 11.0, 95% CI 1.3–91.8). The association is statistically significant in the primary specification but is fragile to single-patient reclassification, an unavoidable consequence of reporting only five Non-MR-dominant patients across three decades of global pediatric literature. A fully preoperative, mechanism-anchored framework that places the burden of proof on Non-MR-dominant assignment offers a coherent, biologically grounded interpretation of three decades of heterogeneous pediatric MVR outcomes. In MR-dominant disease, MVR may support clinical stabilization with the potential for remodeling and may be a viable bridging or alternative strategy where transplantation and durable ventricular assist device support are inaccessible. The framework should be regarded as a proof-of-concept clinical hypothesis rather than a validated decision algorithm; prospective multicenter validation is required before clinical adoption.