Longevity of implantable cardioverter-defibrillators in a single-center population

Abstract

Real-life comparative data of implantable cardioverter-defibrillator (ICD) longevity are needed to identify the best possible device longevity for optimal patient comfort and to minimize risk for multiple replacements. We retrospectively studied ICD longevity in a large single-center population. At our institution, 1272 consecutive patients were implanted with 1665 ICD or cardiac resynchronization therapy with defibrillator (CRT-D) devices between 1998 and 2010. Device data and follow-up were retrieved from the continuous ICD clinic documentation, and generator replacements were verified from the hospital charts. Kaplan-Meier event probabilities for the time to generator replacement were calculated according to device type, pacing percentage (sum of atrial, right and left ventricular pacing, if applicable) and right ventricular pacing threshold, incidence of ICD shocks, and manufacturer. Multivariate Cox proportional hazards regression was performed in addition. A total of 470 devices were replaced for elective replacement indicator. These occurred after 5.4 years (95 % confidence interval, 5.1 to 5.6 years) in 175 of 625 implanted Boston Scientific or Guidant devices, after 5.7 years (5.5 to 5.9 years) in 266 of 883 Medtronic devices, and 5.2 years (5.0 to 5.4 years) in 29 of 157 Biotronik devices, respectively. Differences between manufacturers remained significant upon multivariate analysis (Medtronic vs. Boston Scientific: P = 0.01; both vs. Biotronik: P < 0.01) as did differences of device type (single- and dual-chamber ICD vs. CRT-D, P < 0.01) and pacing percentage (P < 0.01). Pacing output and ICD shocks did not influence battery longevity. ICD longevity differed significantly between manufacturers, independent of device type or pacing parameters.