Obesity and Change in NT-proBNP with Angiotensin-Neprilysin Inhibition in Patients with Mildly Reduced or Preserved Ejection Fraction and a Worsening Heart Failure Event: PARAGLIDE-HF.

Abstract

Objective: To determine the relationship between obesity status and the effect of sacubitril/valsartan (sac/val) vs valsartan (val) on change in NT-proBNP and a hierarchical outcome of clinical benefit among patients with heart failure with mildly reduced or preserved ejection fraction (HFpEF).

Background: In the PARAGLIDE-HF trial, patients with HFpEF and a recent worsening HF event randomized to sac/val vs. val had a greater decrease in NT-proBNP over 8 weeks and showed a trend towards clinical benefit. Patients with HFpEF and obesity represent a unique population with high symptom burden and elevated risk for HF hospitalization. The efficacy and safety of sac/val in this high-risk population following a recent worsening HF event is not well characterized.

Methods: To assess the interaction between obesity status and sac/val vs val on NT-proBNP, the time-averaged proportional change in NT-proBNP from baseline to Weeks 4 and 8 was assessed with an analysis of covariance model that included the interaction of treatment-by-BMI group (obese with BMI ≥30 kg/m² vs not obese with BMI 18.5 to <30 kg/m²). A win ratio consisting of a hierarchy of cardiovascular death, total HF hospitalizations, urgent HF visits, and change in NT-proBNP was calculated for both BMI groups.

Results: Overall, 65% (n=298) of participants had obesity and 35% (n=164) of participants did not have obesity. When the interaction between BMI category and the effect of sac/val vs val on the time-averaged proportional change from baseline NT-proBNP was assessed, participants with obesity had a numerically larger decrease in NT-proBNP (ratio of change 0.82 (95% CI 0.68 to 0.99) with obesity vs 0.96 (95% CI 0.71 to 1.29) without obesity that was not statistically significant, p interaction 0.37. Participants with obesity were significantly more likely to derive benefit overall from sac/val vs val as compared to participants without obesity (win ratio 1.38, 95% CI 1.01-1.90 vs 0.90, 95% CI 0.62-1.32, p-interaction = 0.09).

Conclusion: In this prespecified secondary analysis of PARAGLIDE-HF, patients with HFpEF and obesity had a similar magnitude of decrease in NT-proBNP with randomization to sac/val vs val, but may derive greater clinical benefit as assessed by a win ratio. This hypothesis-generating evidence suggests that patients that patients with obesity and HFpEF may particularly benefit from sac/val.