Influence of weight variation on long-term mortality of patients with heart failure

Abstract

BackgroundIn heart failure, weight loss predicts dismal prognosis. Weight variations have not been addressed in obese patients with heart failure. AimTo study the impact of weight variation on heart failure mortality according to body mass index strata. MethodsRetrospective study of patients with chronic heart failure with left ventricular ejection fraction<50%. Only patients with ≥1 year of follow-up were included. Patients with missing data for body mass index at the index and 1-year appointments were excluded. Patients were classified into three groups according to weight variation: weight gain>5%; weight loss>5%; and weight stability. Follow-up was set from the 1-year appointment. Cox-regression analysis was used to assess the prognostic impact of weight variation. ResultsWe studied 589 patients: 69.8% male; mean age, 69 years. Over 1 year, 148 patients (25.1%) gained>5% weight, 97 (16.5%) lost>5% weight and the remaining 344 were weight-stable. During 49 months of median follow-up, 248 patients died. Patients who lost>5% of their weight presented a higher death risk than the others (hazard ratio 1.61, 95% confidence interval 1.18–2.19). After multivariable adjustment, the hazard ratio for death for low/normal-weight patients who lost>5% of their weight was 1.81 (95% confidence interval 1.02–3.21; P=0.04) compared with the others. Among the overweight, those who lost>5% of their weight had a hazard ratio of 2.34 (95% confidence interval 1.32–4.12). In the initially obese subgroup, weight loss>5% was not associated with prognosis (hazard ratio 1.08, 95% confidence interval 0.53–2.19). ConclusionsWeight loss predicted mortality in low/normal-weight and overweight patients with heart failure. However, in obese patients, significant weight loss did not predict poorer survival. Weight loss should not be discouraged in obese patients with heart failure.