Body mass index, gender, and clinical outcome among hypertensive and diabetic patients with stage A/B heart failure


  • Conflicts of Interest Statement: Competing interests: the authors have no competing interests.

Cardiology Unit, S. Maria Annunziata Hospital, Florence, Italy. E-mail:



The existence of an “obesity paradox” in asymptomatic patients with preclinical heart failure (HF) has not been investigated. The prognostic value of BMI in a cohort of hypertensive and diabetic patients with stage A/B HF enrolled in the PROBE-HF study was explored.

Design and Methods

BMI was measured in 1003 asymptomatic subjects (age 66.4 ± 7.8 years, 48% males) with hypertension and/or type 2 diabetes and no clinical evidence of HF. Predefined endpoints were all-cause mortality and a composite of death and hospitalization for cardiac causes.


During a follow-up of 38.5 ± 4.1 months, 33 deaths were observed. Mortality in the normal BMI group (1.6 deaths per 100 patient-years) did not differ to that in the overweight group (1.1 per 100 patient-years, p = 0.31), but was higher than that in the obese group (0.4 per 100 patient-years, p = 0.0089). In multivariable analysis, obesity (hazard ratio [HR] 0.27 [0.09-0.85], p = 0.025) but not overweight (HR 0.68 [0.32-1.45], p = 0.32) was associated with lower risk of death. Obesity was also independently associated with reduced risk of the composite endpoint (HR 0.54 [0.28-0.99], p = 0.047).


In asymptomatic hypertensive and diabetic patients with preclinical HF, obesity is associated with better survival and reduced risk of events.