Impact of Renal Function on Survival in Patients with Implantable Cardioverter-Defibrillators

Authors


  • This study is supported in part by a postdoctoral fellowship merit-based grant from the American College of Cardiology and Merck Foundation (MPT) and by a Veterans Administration Health Services Research and Development Research Career Development Award (PDV).

Address for reprints: Jeffrey Olgin, M.D., University of California San Francisco, Box 1354, San Francisco, CA 94143-1354. Fax: (415) 476-6260; e-mail: olgin@medicine.ucsf.edu

Abstract

Background:Although chronic renal insufficiency (CRI) is associated with increased cardiac and noncardiac mortality, there is limited data on the relationship between CRI and survival in patients with implantable cardioverter-defibrillators (ICDs), particularly across a wide range of renal function.

Methods:We studied 507 consecutive patients receiving first-time ICDs from 1993–2003 at a single center. Preimplant serum creatinine measurements were used to determine glomerular filtration rate (GFR) and stage of chronic kidney disease (CKD). The primary outcome was time to death.

Results:During a mean follow-up of 4 years, all-cause mortality through completion of follow-up increased stepwise by GFR stage (I: 16%, II: 20%, III: 35%; IV: 40%; V: 50%; P < 0.001 for trend). After multivariate adjustment, CRI was independently associated with death (HR = 1.7, P = 0.02), as were a serum creatinine ≥2.0 mg/dL (HR 2.5, P = 0.003) and the presence of end-stage renal disease (HR 6.8; P < 0.001). For every 10-unit decrease in GFR, the adjusted hazard of death increased 12% (P = 0.04).

Conclusion:The presence of CRI prior to implant is independently associated with increased mortality in patients receiving ICDs. The risk is proportional to the degree of renal dysfunction and is apparent even when GFR is only mildly reduced. Differences in mortality are observed within the first year of implant, and patients on dialysis are at highest risk. Because randomized trials of ICDs have not included large numbers of patients with moderate or severe renal disease, our findings may have important implications in prognosis and case selection of patients who otherwise meet current indications for ICD implantation.

Ancillary