Nonlinear association of body mass index with 1-year outcomes and consistent treatment effects across BMI in HFpEF: insights from a nationwide chinese registry

Background/Introduction Obesity is highly prevalent in heart failure with preserved ejection fraction (HFpEF), yet the relationship between body mass index (BMI) and prognosis remains controversial, with prior studies suggesting an “obesity paradox.” Moreover, whether BMI modifies the effectiveness of guideline-directed medical therapy (GDMT) in real-world HFpEF populations remains uncertain. Purpose To evaluate the association between BMI and 1-year clinical outcomes and to determine whether treatment effects of contemporary GDMT vary across the continuous BMI spectrum in HFpEF. Methods We analyzed 87,397 hospitalized HFpEF patients (LVEF >50%) from a nationwide Chinese heart failure registry. BMI was assessed at discharge and analyzed as both categorical and continuous variables. The primary outcome was 1-year all-cause death; secondary outcomes included cardiovascular (CV) death, rehospitalization, and a composite of CV death and HF rehospitalization. Multivariable Cox proportional hazards models were adjusted for demographics, comorbidities, hemodynamics, renal function, NT-proBNP (log-transformed with missing indicator), and discharge therapies. Restricted cubic spline analyses evaluated nonlinear associations. Treatment effect modification across continuous BMI was assessed for ARNI vs ACEI/ARB, β-blockers, MRAs, and SGLT2 inhibitors. Results Underweight patients (BMI <18.5 kg/m²) had significantly higher risk of all-cause death compared with normal weight, whereas overweight and class I obesity were associated with lower mortality. Restricted cubic spline analysis demonstrated a significant nonlinear association between BMI and all-cause mortality (P_overall<0.001; P_nonlinear<0.001), with excess risk at low BMI and nadir risk within the overweight to class I obesity range. Similar patterns were observed for CV death and composite outcomes. Across the continuous BMI range, treatment-associated risk reductions with ARNI, β-blockers, and SGLT2 inhibitors were broadly consistent, with no strong evidence of effect modification (all P for interaction >0.05). Associations were consistent across sex and other prespecified subgroups. Conclusion(s) In a large nationwide real-world HFpEF cohort, BMI demonstrated a nonlinear association with 1-year mortality, with increased risk among underweight patients and lower risk in overweight to mildly obese individuals. Importantly, therapeutic associations of contemporary GDMT were largely preserved across BMI levels, supporting guideline-based treatment irrespective of body size.Figure 1Figure 2