Prognostic Value of a Cardiac Magnetic Resonance‐Derived Left Atrioventricular Coupling Index From Routine Cine Images in Dilated Cardiomyopathy: A Multicenter Study

BACKGROUND: Accurate risk stratification in dilated cardiomyopathy (DCM) is challenging. Conventional markers like left ventricular ejection fraction (LVEF) and late gadolinium enhancement (LGE) fail to fully capture the pathophysiological interplay between left atrial (LA) and left ventricular (LV) remodeling. PURPOSE: To determine the prognostic value of a cardiac MRI-derived left atrioventricular coupling index (LACI) to predict adverse outcomes in DCM and assess its incremental utility beyond conventional markers. STUDY TYPE: Retrospective. POPULATION: 403 patients with DCM from two centers, randomly split into a training set (n = 282, 70%) for model development and an internal validation set (n = 121, 30%). FIELD STRENGTH/SEQUENCE: Balanced steady-state free precession cine and inversion-recovery prepared fast gradient-echo (LGE) sequences at 1.5T and 3.0T. ASSESSMENT: LACI was quantified as (LA minimal volume/LV end-diastolic volume) × 100% from routine cine images. LA/LV volumes, function, LGE, and global longitudinal strain (GLS) were assessed by blinded radiologists. The primary endpoint was a composite of all-cause mortality or heart failure hospitalization. STATISTICAL TESTS: Student's t-test, Mann-Whitney U test, chi-square test, Spearman's correlation, Cox regression, Kaplan-Meier analysis with log-rank test, receiver operating characteristic analysis, C-index, net reclassification improvement (NRI). A two-sided p value < 0.05 was significant. RESULTS: Over a median 43.6-month follow-up, 97 patients (24.1%) reached the primary endpoint. In the training set, elevated LACI (≥ 45.75%) was the strongest independent predictor (adjusted hazard ratio HR = 3.485, 95% CI 1.821-6.667). The LACI-enhanced model (including NYHA class, LGE, GLS) achieved a C-index of 0.804 (95% CI 0.774-0.835) in the training set, which remained robust in the validation set (C-index = 0.824, 95% CI 0.751-0.897). Adding LACI to the baseline model significantly improved reclassification (continuous NRI = 0.102, 95% CI 0.043-0.161). DATA CONCLUSION: LACI derived from routine cine images is strongly and independently associated with adverse outcomes in DCM and provides significant incremental prognostic value beyond established markers, supporting its potential as a practical risk stratification tool. EVIDENCE LEVEL: 3. TECHNICAL EFFICACY: Stage 5.