Device-Related Complications and Inappropriate Therapies Among Subcutaneous vs. Transvenous Implantable Defibrillator Recipients: Insight Monaldi Rhythm Registry

INTRODUCTION: In the context of randomized clinical trials, subcutaneous implantable cardiac defibrillators (S-ICDs) are non-inferior to transvenous ICDs (T-ICDs) concerning device-related complications or inappropriate shocks in patients with an indication for defibrillator therapy and not in need of pacing. We aimed at describing the clinical features of patients who underwent S-ICD implantation in our clinical practice, as well as the ICD-related complications and the inappropriate therapies among S-ICD vs. T-ICD recipients during a long-term follow-up. MATERIALS AND METHODS: All patients undergoing ICD, both S-ICD and TV-ICD, at Monaldi Hospital from January 1, 2015 to January 1, 2019 and followed up at our institution were included in the present analysis. The clinical variables associated with S-ICD implantation were evaluated by logistic regression analyses. We collected the ICD inappropriate therapies, ICD-related complications (including both pulse generator and lead-related complications), ICD-related infections, appropriate ICD therapies, and overall mortality. Kaplan-Meier (KM) analyses were performed to assess the risk of clinical outcome events between the two subgroups. A time-dependent Cox regression analysis was performed to adjust the results. RESULTS: Total 607 consecutive patients (mean age 53.8 ± 16.8, male 77.8%) with both TV-ICD (n: 290, 47.8%) and S-ICD (n: 317, 52.2%), implanted and followed at our center for a mean follow-up of 1614 ± 1018 days, were included in the study. At multivariate logistic regression analysis, an independent association between S-ICD implantation and ionic channel disease [OR: 6.01 (2.26–15.87); p < 0.0001] and ischemic cardiomyopathy [OR: 0.20 (0.12–0.35); p < 0.0001] was shown. The KM analysis did not show a significantly different risk of the inappropriate ICD therapies (log rank p = 0.64) between the two subgroups; conversely, a significant increase in the risk of ICD-related complications (log rank p = 0.02) and infections (log rank p = 0.02) in TV-ICD group was shown. The adjusted risk for ICD-related infections [OR: 0.07 (0.009–0.55), p = 0.01] and complications [0.31 (0.12–0.81), p = 0.01] was significantly lower among patients with S-ICD. CONCLUSIONS: The choice to implant S-ICD was mainly driven by younger age and the presence of ionic channel disease; conversely ischemic cardiomyopathy reduces the probability to use this technology. No significant differences in inappropriate ICD therapies were shown among S-ICD vs. TV-ICD group; moreover, S-ICD is characterized by a lower rate of infectious and non-infectious complications leading to surgical revision or extraction.