Disentangling the Impact of Adiposity From Insulin Resistance in Heart Failure With Preserved Ejection Fraction.

BACKGROUND

Obesity, insulin resistance (IR), and diabetes are common in heart failure with preserved ejection fraction (HFpEF) and are associated with worsening heart failure, but their independent contributions remain unknown.

OBJECTIVES

In this study, we sought to determine the contribution of diabetes vs obesity to left heart abnormalities in HFpEF

METHODS

Indices of adiposity (body mass index [BMI], bioimpedance fat mass, waist circumference) and IR (homeostasis-model assessment [HOMA]) were measured among PVDOMICS study participants with HFpEF. Rest and exercise pulmonary capillary wedge pressure (PCWP) responses were compared, stratified by obesity (BMI ≥30 kg/m2), IR status (HOMA-IR ≥2.6), and diabetes diagnosis. Findings were also tested in an independent HFpEF cohort.

RESULTS

Of 276 patients with HFpEF, 246 (89%) had increased waist/height ratio, and 166 (60%) had BMI ≥30 kg/m2, with 114 (69%) of the latter having IR and 75 (45%) having diabetes. Of 110 (40%) with HFpEF and BMI <30 kg/m2, 44 (40%) had IR and 27 (25%) had diabetes (both P < 0.0001 vs obesity phenotype). The presence of IR was not associated with worse left heart remodeling or PCWP. In contrast, obesity (regardless of IR status) was associated with greater biventricular enlargement, worse exercise performance, poorer quality of life, and higher rest and exercise PCWP (P < 0.01 for all). Obesity was associated with higher rest and dynamic PCWP responses (+4.4 mm Hg; 95% CI: +2.5 to +6.4 mm Hg; P < 0.0001), even after adjustment for HOMA-IR (+4.7 mm Hg; 95% CI: +2.7 to +6.7 mm Hg; P < 0.0001). Greater fat mass, BMI, and waist circumference were associated with higher PCWP at rest and exercise (P < 0.0009 for all), but HOMA-IR was not (+0.01 mm Hg; 95% CI: -0.13 to +0.16 mm Hg; P = 0.84). Findings were similar evaluating diabetes in place of IR, and were replicated in the independent HFpEF cohort (n = 254), where BMI remained independently associated with higher rest and exercise PCWP (+0.19 mm Hg [95% CI: +0.11 to +0.27 mm Hg] per kg/m2; P < 0.0001), but diabetes was not.

CONCLUSIONS

Excess adiposity is present in most patients with HFpEF, even among those not considered obese according to BMI, calling for further study of cardiometabolic therapies among patients with HFpEF and excess adiposity with BMI <30 kg/m2. Although excess body fat is associated with IR and diabetes, cardiac remodeling, hemodynamics, and functional impairment are independently correlated with body fat, but not IR. These findings suggest that diabetes is primarily a marker of greater adiposity in HFpEF, with less direct impact on heart failure severity. (Pulmonary Vascular Disease Phenomics Program [PVDOMICS]; NCT02980887).