Athough clinical trials have expanded implantable defibrillator (ICD) indications, octogenarians have been poorly represented in these studies. Overall, survival in this subgroup is ill-defined.

Consecutive patients ≥80 years of age at ICD implant between July 1995 and September 2003 were retrospectively analyzed. Kaplan-Meier survival analysis was performed, and mortality predictors were identified. Consecutive nonelderly patients aged 60 to 70 years (60-70 group) who received ICDs over the same period were analyzed as a reference. Mortality predictors in the ≥80 group were compared to the 60-to-70 group.

A total of 348 patients (age, ≥80 years [n = 107]; age, 60-70 years [n = 241]) were included. Mean follow-up time for the entire cohort was 3.3 ± 2.2 years. Other than the estimated glomerular filtration rate (eGFR) (58 ± 22 vs 66 ± 22 mL/min) in the ≥80 group versus the 60-to-70 group, no other differences in baseline characteristics were observed. Median survival was 4.2 years after implant in the ≥80 group versus 7 years in the 60-to-70 group (P < .01). Mortality predictors in the ≥80 group included ejection fraction (EF) ≤30% (hazard ratio [HR], 2.2; 95% confidence interval [CI], 1.3-4.0) and eGFR <60 mL/min (HR, 2.2; 95% CI, 1.3-3.7). In the 60-to-70 group, EF ≤30% (HR, 2.7; 95% CI, 1.6-4.5), eGFR <60 mL/min (HR, 3.4; 95% CI, 2.2-5.3), diabetes (HR, 1.8; 95% CI, 1.9-2.9), and QRS width >120 ms (HR, 2.1; 95% CI, 1.4-3.3) predicted mortality. QRS >120 ms and diabetes were not predictors in octogenarians (HR, 1.1 and 1.5, respectively; 95% CI, 0.7-1.9 and 0.8-2.7, respectively). Analysis of octogenarians subgrouped by EF ≤30% and eGFR <60 mL/min identified patients whose median survival was 6.1 years (neither predictor present; n = 28), 4.7 years (either predictor present; n = 46), and 19 months (both predictors present; n = 33) (P < .01 between groups). Survival analysis in the ≥80-year-old cohort grouped by eGFR quartile identified groups with median survival of 5.6, 4.7, 3.5 years, and 18 months, respectively, in the highest to the lowest eGFR quartile (>75, 61-75, 41-60, and <41 mL/min).

Median survival in octogenarian ICD recipients is greater than 4 years. In addition to baseline EF, eGFR is a strong predictor of mortality in elderly ICD candidates. These easily identifiable clinical variables may assist clinical decision making and help to provide appropriate post-ICD implant survival expectations in this elderly patient group.