Single-chamber transvenous and subcutaneous defibrillators: clinical outcomes

Abstract

Introduction/Background By using an entirely extra-thoracic lead placement, subcutaneous implantable cardioverter–defibrillators (S-ICD) were designed to avoid lead-related complications of single-chamber transvenous implantable cardioverter-defibrillators (VVI ICD). Purpose Our objective was to assess and compare outcomes following first VVI ICD or S-ICD implantation in an exhaustive nationwide matched cohort. Methods This French longitudinal cohort study was based on the national hospitalization database covering hospital care from for the entire population. All adults (age ≥18 years) hospitalized in French hospitals From January 1, 2010 to September 1, 2020, who underwent a VVI ICD or S-ICD implantation were included. Patients with a previous pacemaker or ICD or with a history of infective endocarditis were excluded. Results 21,667 patients were included in the cohort, 19,493 patients had a transvenous VVI ICD and 2,174 had a subcutaneous ICD. Mean follow-up was 28.8±31.8 months. S-ICD patients was associated with higher rate of all-cause death (HR: 1.684, 95% CI: 1.309–2.165, p<0.001). There were no significant differences in cardiovascular death (HR: 1.092, 95% CI: 0.697–1.711, p=0.70) and infective endocarditis (HR: 0.354, 95% CI: 0.067–1.433, p=0.15) between the two groups Using propensity score, 1,582 patients with VVI ICD were matched 1:1 with S-ICD patients. Mean follow-up was 4.5±7.2 months. In the matched analysis, there were no significant differences in all-cause death (HR: 1.090, 95% CI: 0.728–1.633, p=0.68) and cardiovascular death (HR: 1.167, 95% CI: 0.603–2.260, p=0.65) between the two groups. A trend toward a lower risk of infective endocarditis in the S-ICD group was also observed without reaching significance (HR: 0.219, 95% CI: 0.047–1.017, p=0.053). A sensitivty analysis in patients with coronary artery disease in the matched cohort was performed. 1,024 patients had a VVI ICD and 977 had a S-ICD. Same trends were observed without significant differences in all-cause death (HR: 0.966, 95% CI: 0.605–1.543, p=0.88) and cardiovascular death (HR: 1.307, 95% CI: 0.610–2.799, p=0.49). Conclusion Our nationwide study highlighted a higher risk of all-cause death in patients treated with subcutaneous which however was not statistically significant after propensity score matching. No differences regarding cardiovascular mortality was found. An interesting trend toward diminution of infective endocarditis was also observed without reaching significancy. Funding Acknowledgement Type of funding sources: None. Baseline characteristicsCardiovascular death