M yocardial fibrosis is implicated in late right ventricular (RV) dysfunction in patients who have undergone atrial redirection surgery (Mustard/Senning operation) for transposition of the great arteries.1,2 The RV in these patients functions as the systemic ventricle which predisposes to arrhythmia, premature systemic RV failure, and sudden death.3-13 Atrial tachyarrhythmia has been reported as the most common early clinical complication, 5,7,11,13,14 is associated with significant clinical morbidity, has been suggested to reflect systemic RV dysfunction, and is a recognized marker of elevated atrial pressure and increased mortality. 5,7,13,15 Thereafter, RV dysfunction precedes the onset of symptoms, clinical heart Background-We hypothesized that fibrosis detected by late gadolinium enhancement (LGE) cardiovascular magnetic resonance predicts outcomes in patients with transposition of the great arteries post atrial redirection surgery. These patients have a systemic right ventricle (RV) and are at risk of arrhythmia, premature RV failure, and sudden death. Methods and Results-Fifty-five patients (aged 27±7 years) underwent LGE cardiovascular magnetic resonance and were followed for a median 7.8 (interquartile range, 3.8-9.6) years in a prospective single-center cohort study. RVLGE was present in 31 (56%) patients. The prespecified composite clinical end point comprised new-onset sustained tachyarrhythmia (atrial/ventricular) or decompensated heart failure admission/transplantation/death. Univariate predictors of the composite end point (n=22 patients; 19 atrial/2 ventricular tachyarrhythmia, 1 death) included RV LGE presence and extent, RV volumes/mass/ejection fraction, right atrial area, peak Vo 2 , and age at repair. In bivariate analysis, RVLGE presence was independently associated with the composite end point (hazard ratio, 4.95 [95% confidence interval, 1.60-15.28]; P=0.005), and only percent predicted peak Vo 2 remained significantly associated with cardiac events after controlling for RV LGE (hazard ratio, 0.80 [95% confidence interval, 0.68-0.95]; P=0.009/5%). In 8 of 9 patients with >1 event, atrial tachyarrhythmia, itself a known risk factor for mortality, occurred first. There was agreement between location and extent of RV LGE at in vivo cardiovascular magnetic resonance and histologically documented focal RV fibrosis in an explanted heart. There was RV LGE progression in a different case restudied for clinical indications. Conclusions-Systemic RV LGE is strongly associated with adverse clinical outcome especially arrhythmia in transposition of the great arteries, thus LGE cardiovascular magnetic resonance should be incorporated in risk stratification of these patients. (Circ Cardiovasc Imaging. 2015;8:e002628.