Predictors of Mortality Amongst Recipients of Implantable Cardioverter Defibrillators for Secondary Prevention of Cardiac Arrest

• Main Authors: Sunil K. Agarwal, Ish Singla, Haitham Hreybe, Samir Saba
• Format: Article
• Language: English
• Published: Elsevier 2007-10-01
• Series: Indian Pacing and Electrophysiology Journal
• Subjects: Implantable Cardioverter Defibrillators; Cardiac Arrest; Mortality; Predictors
• Online Access: http://www.ipej.org/0704/saba2.htm

Background: Implantable Cardioverter-defibrillators (ICD) reduce mortality in survivors of cardiac arrest (CA). We investigated the predictors of mortality after ICD implantation in survivors of CA. Methods: Retrospective review of clinical records and social security death index of all patients who received an ICD in a preexisting database of survivors of CA at the University of Pittsburgh Medical Center was performed. Multivariate analyses using the Cox proportional hazard model were performed with backward elimination to identify independent predictors of the time to death, and Kaplan-Meier curves were plotted.Results: Eighty patients (64 men) with a mean age of 64.4±12.5 years were followed for 4.7±2.3 years after ICD implantation. Survival rates were 93.8%, 65% and 50% at 1, 5, and 10 years, respectively. Independent predictors of time to death were determined to include age (hazard ratio (HR) = 1.91 per 10-year increase, p = 0.003), serum creatinine ≥ 1.3 mg/dL (HR = 2.56, p = 0.004), and QRS width >120 ms (HR = 5.14, p = 0.012). Conclusions: In this sample of ICD recipients secondary to CA, older age, elevated serum creatinine, and wider QRS duration were independent predictors of mortality. The presence of more than one risk factor in the same patient was associated with higher mortality rates. Whether interventions such as biventricular pacing can offset this increase risk of death warrants further investigation.