Background:
Important sex-related differences in body composition exist and may influence the predictive utility of body mass index (BMI) in heart failure (HF) populations. However, whether sex modifies the association between BMI and other adiposity-related anthropometrics and outcomes in HF with mildly reduced or preserved ejection fraction (HFmrEF/HFpEF) has not been rigorously explored.
Methods:
In this participant-level pooled analysis of five international, multicenter, randomized-controlled trials enrolling adults HFmrEF/HFpEF, the association between sex and adiposity-related anthropometrics (BMI, waist circumference [WC], and waist-to-height ratio [WHtR]) and the primary composite outcome of time-to-first HF hospitalization or cardiovascular (CV) death was evaluated using multivariable adjusted Cox proportional hazards models (stratified by age, treatment, and trial) and restricted cubic splines.
Results:
Overall, BMI was available in 21,606 participants (mean age, 71±9 years; 83% White; mean BMI: 30±6 kg/m
2
), of whom 10,681 (49%) were female and 10,925 (51%) were male. WC was available in 7,854 participants. Female participants were more likely to be older, have higher BMI, WHtR, and left ventricular ejection fraction. Males had higher rates of the primary outcome compared with females across the BMI, WC, and WHtR spectra (
Pcomparison
<0.01 for all), but higher BMI, WC, and WHtR were similarly associated with a higher rate of the primary outcome irrespective of sex (
Figure
). However, among those with BMI 18.5 to <25 kg/m
2
, WC was elevated in 8% of males, but 34% of females. Higher WC was associated with a higher rate of primary events (adjusted hazard ratio per 10 cm increase, 1.13, 95% CI 1.08-1.19;
P
<0.001) independent of BMI, with consistent findings by sex (
Pinteraction
=0.25).
Conclusions:
In this participant-level pooled analysis of HFmrEF/HFpEF trials, higher BMI, WC, and WHtR were similarly associated with HF hospitalization or CV death in females and males. However, use of BMI alone disproportionately underestimated abdominal adiposity among females, stressing the importance of implementing additional anthropometrics to mitigate sex-based differences in risk stratification in the HF population.