Usefullness of Different Mortality Risk Scores in Patients with Heart Failure. A Retrospective Observational Study

pp 322-328

Abstract

Background: The Cardiac and Comorbid Conditions - Heart Failure (3C-HF) and the Meta-Analysis Global Group in Chronic Heart Failure (MAGGIC) are two score models developed to predict mortality in patients with heart failure (HF). The performance of these scores has been little studied in our setting.

Objective: The aim of this study was to assess the performance of the 3C-HF and the MAGGIC scores to predict one-year mortality in a population of patients with HF.

Methods: Ambulatory HF patients discharged after hospitalization due to acute HF in two centers were included in the study. The 3C-HF and MAGGIC scores were calculated and one-year mortality was the study endpoint. The discrimination ability of the scores was analyzed from the calculated area under the ROC curve and their calibration quality was assessed applying the Hosmer-Lemeshow test. Both areas under the ROC curve were compared using the Hanley-Mc Neil test.

Results: A total of 704 patients with mean age of 73±11 years and 39.6% women were included in the study. One-year mortality was 12.4% (n=87). Both scores were independent predictors of mortality, with HR of 1.03 (95% CI 1.008-1.06; p=0.02) and 1.08 (95% CI 1.02-1.13; p=0.004) for the 3C-HF and MAGGIC scores, respectively. The area under the ROC curve for the 3C-HF score was 0.70 (95% CI 0.64-0.75) and for the MAGGIC score 0.67 (95% CI 0.61-0.73), without significant differences between them (p=0.41). Both
scores presented adequate calibration (p=0.06 and p=0.32, respectively).

Conclusions: The 3C-HF and MAGGIC scores were predictors of one-year mortality, with a moderate ability to discriminate events and adequate calibration. The discrimination ability between both scores was not significant.