Abstract 2427: Risk Stratification for Primary Implantation of an ICD in the Multicenter Automatic Defibrillator Implantation Trial - II (MADIT-II)
Background: ICD therapy reduced mortality in the MADIT-II trial (post MI patients with an EF≤30%). We hypothesized that certain low-risk patients may not benefit by virtue of being low-risk for sudden death and at the other extreme very high-risk (VHR) patients may not benefit due to advanced disease. We sought to better identify these subgroups.
Methods and Results: Best-subset proportional-hazards regression analysis was used to develop a simple clinical risk score for the end point of all-cause mortality in patients allocated to the conventional therapy arm (CONV) (n = 468) of MADIT-II. The selected risk score model comprised of 5 clinical factors: New York Heart Association functional class > II, age > 70 years, BUN > 26 mg/dL, QRS duration > 0.12 sec, and atrial fibrillation (all hazard ratios between 1.5 and 2.0, all p-values < 0.05). The benefit of ICD therapy was assessed within risk categories, and within a prespecified separate subgroup of VHR patients with BUN 3 50 mg/dL and/or serum creatinine 3 2.5 mg/dL. Crude mortality rates in CONV arm were 8% and 28% in patients with 0 (low-risk) and 3 1 risk factors (intermediate-risk), respectively, and 43% in VHR patients. ICD therapy was associated with a 49% reduction in the risk of death among patients with 3 1 risk factors [HR 0.51, p < 0.001 (n = 296 in CONV arm and 490 in ICD arm)], whereas no ICD benefit was identified in patients with 0 risk factors [HR 0.96; p = 0.91 (n = 149 in CONV arm and 196 in ICD arm)] and in VHR patients [HR 1.00; p > 0.99 (n = 23 in CONV arm and 37 in ICD arm). Using data on utilization, cost and survival within 3.5 years of the study (n = 1095 U.S. patients), the intermediate-risk group had an increase in survival of 124 days (0.3397 years) and an additional cost of $40,700, with an incremental cost-effectiveness ratio of $120,000/ years-of-life saved (YOLS) (95% CI $69,000 to $267,000). As there were no YOLS in the low-risk and very high-risk groups, the cost per life year saved was virtually infinite in these subsets.
Conclusions: Our data suggest a U-shaped pattern for ICD efficacy in the post MI low EF population, with pronounced benefit in intermediate risk patients and attenuated efficacy in lower- and higher-risk subsets. These preliminary findings need to be validated in similar independent populations.