Singh SM, Wang X, Austin PC, Parekh RS, Lee DS, for the Ontario ICD Database Investigators. Prophylactic Defibrillators in Patients With Severe Chronic Kidney Disease. JAMA Intern Med. 2014;174(6):995-996. doi:10.1001/jamainternmed.2014.1208
Heart failure (HF) and chronic kidney disease (CKD) are increasing in prevalence individually and in combination.1 It is unclear whether the survival benefits associated with prophylactic implantable cardioverter defibrillator (ICD) implantation in traditional populations extends to individuals with severe CKD (stage ≥4 or estimated glomerular filtration rate [eGFR] <30 mL/min/1.73 m2) given their underrepresentation in clinical trials2 and elevated risk of non–cardiac-related death.3 The absence of data and presence of competing risks complicates decision-making in this population. This study assessed the association between prophylactic ICD implantation and survival in individuals with severe CKD.
This study was approved by the institutional review board of Sunnybrook Health Sciences Centre. Patients with HF and non–dialysis-dependent severe CKD undergoing prophylactic ICD implantation in the Ontario ICD database from February 2007 through November 2011 were identified.4 Patients with HF, defined using the Framingham criteria, and severe CKD not receiving an ICD were identified in the Enhanced Feedback for Effective Cardiac Treatment (EFFECT Phase II) study5 from 2004 through 2005. Only patients with a documented ejection fraction of less than 35%, surviving at least 40 days postdischarge, and assessed by a cardiologist were included.
A propensity-score matched cohort of patients with and without ICD (hereinafter, ICD group and non-ICD group) was created using a greedy, nearest-neighbor matching algorithm. Variables in the propensity score included sex, eGFR, QRS duration, atrial fibrillation, prior revascularization, medications, comorbidities, and number of hospitalizations for HF or myocardial infarction (MI) in the preceding 5 years. In addition, patients were matched from the time of their last hospitalization for HF or MI.
Kaplan-Meier estimates of survival were determined for each group and compared using the stratified log-rank test. A robust variance estimator was used to account for the matched nature of the sample. The Cox model was adjusted for baseline covariates whose standardized difference exceeded 0.1 in the matched sample. All analyses were performed using SAS statistical software (version 9.3; SAS institute Inc), with a 2-tailed P < .05 indicating statistical significance.
A total of 108 unique pairs of patients (87% of eligible patients in ICD group) were matched (Table). Imbalances in age, QRS duration, hypertension, prior MI, additional comorbidities, and medication use were distributed in a pattern that potentially disadvantaged the non-ICD group. The mortality rates were 16.7 deaths per 100 person-years in the ICD group vs 17.1 per 100 person-years in the non-ICD group (P = .92) (Figure). The unadjusted hazard ratio (HR) for death in the ICD group was 1.05 (95% CI, 0.63-1.74; P = .86). After adjusting for variables with imbalance, the HR was 2.21 (95% CI, 0.45-10.8; P = .33).
Our data suggest that prophylactic ICD implantation may not confer a survival advantage in patients with HF and severe CKD. The primary limitation of our work was that ICD implantation was not randomized. Despite rigorous methods to adjust for the propensity of ICD receipt, including additional analyses adjusting for variables with imbalance, residual confounding cannot be excluded. The strengths of our work include the sample size (which to our knowledge is the largest matched cohort of patients with advanced CKD receiving ICDs), rich clinical data, and follow-up available in the population-based databases employed.
The results of our study are not unexpected. It is well known that patients with severe CKD are at increased risk for device-related infections, increased defibrillation thresholds, refractoriness to electrical therapy, and non–cardiac-related death, which may erode the survival benefit afforded by ICD implantation.3,6 Despite this, current device guidelines place no restrictions on prophylactic ICD implantation based on stage of CKD. Our findings, in combination with the increasing prevalence of CKD and HF, should serve as a call for prospective randomized investigation to rigorously assess the true benefits of prophylactic ICD implantation in this group of patients. Such work is critical to avoid the twin traps of overtreatment and therapeutic nihilism.
Corresponding Author: Douglas S. Lee, MD, PhD, Institute for Clinical Evaluative Sciences, 2075 Bayview Ave, Room G-106, Toronto, ON M4N 3M5, Canada (firstname.lastname@example.org).
Published Online: April 28, 2014. doi:10.1001/jamainternmed.2014.1208.
Author Contributions: Dr Lee had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Singh, Parekh, Lee.
Acquisition, analysis, or interpretation of data: Singh, Wang, Austin, Lee.
Drafting of the manuscript: Singh, Wang, Lee.
Critical revision of the manuscript for important intellectual content: Singh, Austin, Parekh, Lee.
Statistical analysis: Wang, Lee.
Obtained funding: Lee.
Administrative, technical, or material support: Singh, Lee.
Study supervision: Austin, Lee.
Conflict of Interest Disclosures: Dr Austin is a career investigator of the Heart and Stroke Foundation of Ontario, and Dr Lee is a clinician-scientist of the Canadian Institutes of Health Research. No other disclosures are reported.
Funding/Support: This research was supported by an operating grant from the Canadian Institutes of Health Research (CIHR MOP 111150) and the Ontario Ministry of Health and Long-Term Care (MOHLTC). The Institute for Clinical Evaluative Sciences (ICES) is supported in part by a grant from the Ontario Ministry of Health and Long-Term Care.
Role of the Sponsors: The funding sources had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Group Members: The Ontario ICD Database Investigators members are Jeffrey S. Healey, MD, MSc, Hamilton Health Sciences Centre; Christopher S. Simpson, MD, Kingston General Hospital; Andrew D. Krahn, MD, and Raymond Yee, MD, London Health Sciences Centre; Amir Janmohammed, MD, Rouge Valley Health System; Iqwal Mangat, MD, and Paul Dorian, MD, St. Michael's Hospital; Yaariv Khaykin, MD, Southlake Regional Health Centre; Eugene Crystal, MD, Sunnybrook Health Sciences Centre; Douglas Cameron, MD, Toronto General Hospital; Catherine LeFeuvre, MD, Trillium Health Centre; and David Birnie, MD, University of Ottawa Heart Institute.
Disclaimer: The opinions, results and conclusions are those of the authors and no endorsement by the Ministry of Health and Long-Term Care or by the Institute for Clinical Evaluative Sciences is intended or should be inferred.