Clinical outcomes of subcutaneous vs. transvenous implantable defibrillator therapy in a polymorbid patient cohort

Abstract

Background: The subcutaneous implantable cardioverter-defibrillator (S-ICD) has been designed to overcome lead-related complications and device endocarditis. Lacking the ability for pacing or resynchronization therapy its usage is limited to selected patients at risk for sudden cardiac death (SCD).

Objective: The aim of this single-center study was to assess clinical outcomes of S-ICD and single-chamber transvenous (TV)-ICD in an all-comers population.

Methods: The study cohort comprised a total of 119 ICD patients who underwent either S-ICD (n = 35) or TV-ICD (n = 84) implantation at the University Hospital Frankfurt from 2009 to 2017. By applying an inverse probability-weighting (IPW) analysis based on the propensity score including the Charlson Comorbidity Index (CCI) to adjust for potential extracardiac comorbidities, we aimed for head-to-head comparison on the study composite endpoint: overall survival, hospitalization, and device-associated events (including appropriate and inappropriate shocks or system-related complications).

Results: The median age of the study population was 66.0 years, 22.7% of the patients were female. The underlying heart disease was ischemic cardiomyopathy (61.4%) with a median LVEF of 30%. Only 52.9% had received an ICD for primary prevention, most of the patients (67.3%) had advanced heart failure (NYHA class II-III) and 16.8% were in atrial fibrillation. CCI was 5 points in TV-ICD patients vs. 4 points for patients with S-ICD (p = 0.209) indicating increased morbidity. The composite endpoint occurred in 38 patients (31.9 %), revealing no significant difference between patients implanted with an S-ICD or TV-ICD (unweighted HR 1.50, 95 % confidence interval (CI) 0.78-2.90; p = 0.229, weighted HR 0.94, 95% CI, 0.61-1.50, p = 0.777). Furthermore, we observed no difference in any single clinical endpoint or device-associated outcome, neither in the unweighted cohort nor following inverse probability-weighting.

Conclusion: Clinical outcomes of the S-ICD and TV-ICD revealed no differences in the composite endpoint including survival, freedom of hospitalization and device-associated events, even after careful adjustment for potential confounders. Moreover, the CCI was evaluated in a S-ICD cohort demonstrating higher survival rates than predicted by the CCI in young, polymorbid (S-)ICD patients.

Keywords: S-ICD; TV-ICD; implantable cardioverter-defibrillator (ICD); subcutaneous ICD; sudden cardiac death; transvenous ICD.

Figure 1

Forrest plot for standardized mean differences of covariates included in the IPW model.

Figure 2

Adjusted Kaplan-Meier curves in patients with subcutaneous and transvenous ICD following inverse probability weighting for (A) the composite endpoint, (B) overall survival, (C) freedom of hospitalization, and (D) freedom of the device-related events at 1.5 year.