Predictive value of CHA2DS and CHA2DS2VASC scores on mortality in different groups of heart failure patients

Background: Atrial fibrillation (AF) and Heart Failure (HF) share many risk factors. CHA2DS and CHA2DS2Vasc scores predict stroke occurrence and mortality in AF patients. Aim of this study was to evaluate whether these scores can also predict mortality in different subsets of HF patients. Methods: ESC-HF Pilot survey was a prospective multicentre observational study run in 136 Cardiology Centres in 12 European Countries. From October 2009 to May 2010, 5118 patiens were included and followed up for one year. CHADS and CHADVASC scores were calculated for each patient. Patients were classified in three groups according to their CHADS score (Group A CHADS 1, Group B CHADS 2-3, Group C CHADS >3) and their CHADSVasc score (Group D CHADSVasc 1, Group E CHADSVasc 2-3, Group F CHADSVasc >3). All-cause 1-year mortality rate was calculated in three subsets of interest: Patients with or without AF, patients with preserved vs reduced left ventricular function, patients admitted for acute heart failure vs outpatients with chronic heart failure Results: Table shows one year all-cause mortality rate in these three subsets of patient split according to their CHADS and CHADVASC score, showing a clear relation between increasing scores and one year mortality. Multivariate anaysis showed that compared to Group A (CHADS score of 1), the odds of mortality were 1.78 (95% CI 1.3-2.4) in Group B (CHADS score 2-3) and 3.04 (95% CI 2.2-4.3) in Group C (CHADS score >3). Compared to Group D (CHADSVasc score 1), the odds of mortality were 2.83 (95% CI 1.5-5.5) in Group E (CHADSVasc score 2-3) and 5.00 (95% CI 2.6-9.5) in Group F (CHADSVasc score >3). Mortality by CHADS and CHADSVasc Score Conclusions: Both CHA2DS and CHA2DS2Vasc scores are independently associated with one year all-cause mortality in a wide variety of heart failure patients.