Incidence of Ventricular Arrhythmias and 1‐Year Predictors of Mortality in Patients Treated With Implantable Cardioverter‐Defibrillator Undergoing Generator Replacement

Abstract

Background When implantable cardioverter defibrillator (ICD) battery is depleted most patients undergo generator replacement (GR) even in the absence of persistent ICD indication. The aim of this study was to assess the incidence of ventricular arrhythmias and the overall prognosis of patients with and without persistent ICD indication undergoing GR. Predictors of 1‐year mortality were also analyzed. Methods and Results Patients with structural heart disease implanted with primary prevention ICD undergoing GR were included. Patients were stratified based on the presence/absence of persistent ICD indication (left ventricular ejection fraction ≤35% at the time of GR and/or history of appropriate ICD therapies during the first generator's life). The study included 371 patients (82% male, 40% with ischemic heart disease). One third of patients (n=121) no longer met ICD indication at the time of GR. During a median follow‐up of 34 months after GR patients without persistent ICD indication showed a significantly lower incidence of appropriate ICD shocks (1.9% versus 16.2%, P <0.001) and ICD therapies. 1‐year mortality was also significantly lower in patients without persistent ICD indication (1% versus 8.3%, P =0.009). At multivariable analysis permanent atrial fibrillation, chronic advanced renal impairment, age >80, and persistent ICD indication were found to be significant predictors of 1‐year mortality. Conclusions Patients without persistent ICD indication at the time of GR show a low incidence of appropriate ICD therapies after GR. Persistent ICD indication, atrial fibrillation, advanced chronic renal disease, and age >80 are significant predictors of 1‐year mortality. Our findings enlighten the need of performing a comprehensive clinical reevaluation of ICD patients at the time of GR.

Impact of the timing of first appropriate shock on outcomes in patients with an implantable cardioverter‐defibrillator: Early versus late

“…In the SCD‐HeFT trial, the HF death rate among ICD patients after an appropriate shock was 43%, with a median time from shock to death of 168 days, which was similar to our result 4 . However, recent studies reported that among ICD patients who received generator exchange, advanced age and non‐cardiac comorbidities were associated with all‐cause mortality after generator exchange (mean time from initial implantation of 5 years), suggesting that the mode of death transitions from cardiac causes to non‐cardiac causes over time 9,10 . Interestingly, our data showed that in the late group, HF death was still a major cause of death after an appropriate shock.…”

“…suggesting that the mode of death transitions from cardiac causes to non-cardiac causes over time. 9,10 Interestingly, our data showed that in the late group, HF death was still a major cause of death after an appropriate shock. There are several explanations for this result.…”

“…8 Furthermore, several studies demonstrated that patients without an appropriate shock during the 1 st generator life period of 5−10 years experienced the first appropriate shock after generator exchange. 9 , 10 However, it is unclear whether the timing of the first appropriate shock is related to prognosis. In addition, no study has explored whether late occurrence of the first appropriate shock is associated with worse outcomes.…”

“…Approximately 20% of ICD patients receive an appropriate shock during the 5 years after implantation 8 . Furthermore, several studies demonstrated that patients without an appropriate shock during the 1 st generator life period of 5−10 years experienced the first appropriate shock after generator exchange 9 , 10 . However, it is unclear whether the timing of the first appropriate shock is related to prognosis.…”

Incidence and Predictors of Implantable Cardioverter-defibrillator Therapies After Generator Replacement—A Pooled Analysis of 31,640 Patients’ Data

“…A total of 31,640 patients from 20 studies (14 retrospective and 6 prospective registries) were included in our meta-analysis (Figure 1) with a mean ± standard deviation age of 65.6 ± 4.7 years (mean age range, 57-76 years) and follow-up period of 3.1 ± 1.5 years. 2,6,10,11,[16][17][18][19][20][21][22][23][24][25][26][27][28][29][30][31] The critical appraisal performed using the Newcastle-Ottawa scale revealed that the quality of studies was good (Table S1).…”