Comparison of mortality prediction scores in elderly patients with ICD for heart failure with reduced ejection fraction.

Background: This investigation aimed to examine and compare the predictive value of MADIT-II, FADES, PACE and SHOCKED scores in predicting one-year and long-term all-cause mortality in implantable cardioverter-defibrillator (ICD) implanted patients, 75 years old and older, since there has been an area of uncertainty about the utility and usefulness of these available risk scores in such cases.

Methods: In this observational, retrospective study, 189 ICD implanted geriatric patients were divided into two groups according to the presence of long-term mortality in follow-up. The baseline characteristics and laboratory variables were compared between the groups. MADIT-II, FADES, PACE and SHOCKED scores were calculated at the time of ICD implantation. One-year and long-term predictive values of these scores were compared by a receiver-operating curve (ROC) analysis.

Results: A ROC analysis showed that the best cutoff value of the MADIT-II score to predict one-year mortality was ≥ 3 with 87% sensitivity and 74% specificity (AUC 0.83; 95% CI 0.73-0.94; p < 0.001) and that for long-term mortality was ≥ 2 with 83% sensitivity and 43% specificity (AUC 0.68; 95% CI 0.60-0.76; p < 0.001). The predictive value of MADIT-II was superior to FADES, PACE and SHOCKED scores in ICD implanted patients who are 75 years and older.

Conclusion: MADIT-II score has a significant prognostic value as compared to FADES, PACE and SHOCKED scores for the prediction of one-year and long-term follow-up in geriatric patients with implanted ICDs for heart failure with reduced ejection fraction.