Cardiac Magnetic Resonance Scar Burden and Appropriate Defibrillator Therapy in Cardiac Sarcoidosis

Background Shock-only outcomes may underestimate device-treated ventricular arrhythmias in cardiac sarcoidosis (CS). Objective To characterize implantable cardioverter-defibrillator (ICD)-treated arrhythmias and factors associated with appropriate shock and mortality. Methods We retrospectively studied 148 patients with CS who underwent ICD implantation. Appropriate shock was primary; appropriate antitachycardia pacing (ATP) and any appropriate ICD therapy (ATP or shock) were secondary. Quantitative left ventricular (LV) late gadolinium enhancement (LGE) analyses used cardiac magnetic resonance (CMR) obtained before events or censoring. Firth logistic and Cox regression were used. Results During a median 2.81-year follow-up, 22 patients (14.9%) received appropriate shock, 38 of 144 (26.4%) received appropriate ATP, and 53 of 146 (36.3%) received any appropriate ICD therapy; 31 received ATP without shock. In the exploratory pre-outcome quantitative CMR cohort (n=64; 8 shocks), LV LGE burden per 10 percentage points was associated with appropriate shock (odds ratio, 1.59; 95% confidence interval, 1.04–2.42; P=.032) and time to first shock (hazard ratio, 1.44; 95% confidence interval, 1.05–1.98; P=.025). Absolute LV LGE mass yielded directionally concordant but nonsignificant estimates. Neither LGE measure was associated with any appropriate ICD therapy (n=64; 21 events). Twenty-seven patients died; age and paced rhythm or heart block were associated with mortality. Conclusion ATP-inclusive assessment identified arrhythmic burden beyond shock alone. Proportional LV LGE showed an exploratory association with appropriate shock but not the ATP-or-shock endpoint. These findings do not establish ICD candidacy or an LGE threshold.