Implantable cardioverter defibrillators (ICD) and cardiac resynchronization therapy (CRT) reduce mortality in many patients with heart failure (HF), but the current use and effectiveness of ICD/CRT in patients with chronic kidney disease (CKD) are uncertain. We examined associations between kidney function and guideline-recommended prescription of ICD/CRT in the Get With The Guidelines-Heart Failure registry, a performance improvement program for hospitalized patients with HF. We compared differences in ICD and CRT prescription between the following categories of estimated glomerular filtration rate (eGFR; mL/min/1.73 m²): ≥60, 59 to 30, <30, and dialysis dependent. From 2008 through 2014, 26,286 patients were eligible for ICD or CRT, and 16,123(61%) had an eGFR <60. De novo ICD and CRT prescription in this group was low at 45% and 30.5%, respectively. Compared to patients with eGFR ≥60, patients with eGFR 30 to 59 were more likely to receive an ICD (adjusted odds ratio [aOR] 1.08, 95% confidence intervals [CI] 1.01 to 1.14), whereas dialysis patients were less likely (aOR 0.61, 95% CI 0.5 to 0.76). Worse kidney function was associated with a decreased likelihood of CRT prescription (aOR 0.97 per 10 ml/min eGFR decrease, p = 0.03). During the study period, the likelihood of both ICD and CRT prescription increased over time among patients with CKD (ICD aOR 1.12, 95% CI 1.07 to 1.18; CRT aOR 1.14, 95% CI 1.06 to 1.23, per year). Prescription of an ICT/CRT was associated with greater 1-year survival in all eGFR groups. In conclusion, there are significant CKD-based differences in prescription of ICD and CRT in HF. However, given the current state of evidence, it is unclear whether improved prescription of ICD and CRT in the CKD population will result in improvement in outcomes.

Background

Sudden cardiac death is the leading cause of death among end-stage kidney disease patients (ESKD) on dialysis, but the benefit of primary prevention implantable cardioverter defibrillators (ICDs) in this population is uncertain. We conducted this investigation to compare the mortality of dialysis patients receiving a primary prevention ICD with matched controls.

Methods

We used data from the National Cardiovascular Data Registry's ICD Registry to select dialysis patients who received a primary prevention ICD, and the Get with the Guidelines-Heart Failure Registry to select a comparator cohort. We matched ICD recipients and no-ICD patients using propensity score techniques to reduce confounding, and overall survival was compared between groups.

Results

We identified 108 dialysis patients receiving primary prevention ICDs and 195 comparable dialysis patients without ICDs. One year (3-year) mortality was 42.2% (68.8%) in the ICD registry cohort compared with 38.1% (75.7%) in the control cohort. There was no significant survival advantage associated with ICD [hazard ratio (HR) 0.87, 95% confidence interval (CI) 0.66-1.13, log-rank P = 0.29]. After propensity matching, our analysis included 86 ICD patients and 86 matched controls. Comparing the propensity-matched cohorts, 1 year (3 years) mortality was 43.4% (74.0%) in the ICD cohort and 39.7% (76.6%) in the control cohort; there was no significant difference in mortality outcome between groups (HR = 0.94, 95% CI: 0.67-1.31, log-rank P = 0.71).

Conclusions

We did not observe a significant association between primary prevention ICDs and reduced mortality among ESKD patients receiving dialysis. Consideration of the potential risks and benefits of ICD implantation in these patients should be undertaken while awaiting the results of definitive clinical trials.