Propensity score matched comparison of subcutaneous and transvenous implantable cardioverter-defibrillator therapy in the SIMPLE and EFFORTLESS studies

Aims: Comparison of outcomes between subcutaneous and transvenous implantable cardioverter-defibrillator (S-ICD and TV-ICD) therapy is hampered by varying patient characteristics and complication definitions. The aim of this analysis is to compare clinical outcomes of S-ICD and TV-ICD therapy in a matched cohort. Methods and results: Patients implanted with de novo implantable cardioverter-defibrillators without need for pacing were selected from two studies: SIMPLE (n = 1091 single and n = 553 dual chamber TV-ICDs) and EFFORTLESS (n = 798 S-ICDs). Subcutaneous implantable cardioverter-defibrillator patients were 1:1 matched on propensity score to TV-ICD patients. Propensity scores were calculated using 15 baseline characteristics including diagnosis. The Kaplan-Meier estimates for complications requiring invasive intervention, appropriate shocks, and inappropriate shocks were calculated at 3 years follow-up. The primary analysis yielded 391 patients pairs with balanced baseline characteristics, with mean age 55 ± 14 years, 49% ischaemic cardiomyopathy, mean left ventricular ejection fraction 40%, 71% primary prevention, and 89% of TV-ICDs were single chamber. Follow-up was mean 2.9 years in the S-ICD arm vs. 3.3 in the TV-ICD arm. All-cause complications occurred in 9.0% of S-ICD vs. 6.5% of TV-ICD patients, P = 0.29. Appropriate shocks occurred in 9.9% of S-ICD vs. 13.8% in TV-ICD patients, P = 0.03 and inappropriate shocks in 11.9% in S-ICD vs. 8.9% in TV-ICD patients (P = 0.07). Total shock burden (20 vs. 31, P = 0.05) and appropriate shock burden per 100 patients years (9 vs. 18, P = 0.02) were lower for S-ICD patients, while inappropriate shock burden was equal (11 vs. 13, P = 0.56). Conclusion: The earliest experience of the S-ICD demonstrates similar outcomes as contemporary TV-ICD therapy in a matched comparison with predominately single-chamber devices at 3 years follow-up.