Natriuretic peptides, body mass index and heart failure risk: Pooled analyses of SAVOR-TIMI 53, DECLARE-TIMI 58 and CAMELLIA-TIMI 61

Siddharth M Patel,David A Morrow,Andrea Bellavia,David D Berg,Deepak L Bhatt,Petr Jarolim,Lawrence A Leiter,Darren K McGuire,Itamar Raz,P Gabriel Steg,John P H Wilding,Marc S Sabatine,Stephen D Wiviott,Eugene Braunwald,Benjamin M Scirica,Erin A Bohula
DOI: https://doi.org/10.1002/ejhf.3118
Abstract:Aim: N-terminal pro-B-type natriuretic peptide (NT-proBNP) concentrations are lower in patients with obesity. The interaction between body mass index (BMI) and NT-proBNP with respect to heart failure risk remains incompletely defined. Methods and results: Data were pooled across three randomized clinical trials enrolling predominantly patients who were overweight or obese with established cardiometabolic disease: SAVOR-TIMI 53, DECLARE-TIMI 58 and CAMELLIA-TIMI 61. Hospitalization for heart failure (HHF) was examined across strata of baseline BMI and NT-proBNP. The effect of dapagliflozin versus placebo was assessed for a treatment interaction across BMI categories in patients with or without an elevated baseline NT-proBNP (≥125 pg/ml). Among 24 455 patients, the median NT-proBNP was 96 (interquartile range [IQR]: 43-225) pg/ml and the median BMI was 33 (IQR 29-37) kg/m2, with 68% of patients having a BMI ≥30 kg/m2. There was a significant inverse association between NT-proBNP and BMI which persisted after adjustment for all clinical variables (p < 0.001). Within any range of NT-proBNP, those at higher BMI had higher risk of HHF at 2 years (comparing BMI <30 vs. ≥40 kg/m2 for NT-proBNP ranges of <125, 125-<450 and ≥450 pg/ml: 0.0% vs. 0.6%, 1.3% vs. 4.0%, and 8.1% vs. 13.8%, respectively), which persisted after multivariable adjustment (adjusted hazard ratio [HRadj] 7.47, 95% confidence interval [CI] 3.16-17.66, HRadj 3.22 [95% CI 2.13-4.86], and HRadj 1.87 [95% CI 1.35-2.60], respectively). In DECLARE-TIMI 58, dapagliflozin versus placebo consistently reduced HHF across BMI categories in those with an elevated NT-proBNP (p-trend for HR across BMI = 0.60), with a pattern of greater absolute risk reduction (ARR) at higher BMI (ARR for BMI <30 to ≥40 kg/m2: 2.2% to 4.7%; p-trend = 0.059). Conclusions: The risk of HHF varies across BMI categories for any given range of circulating NT-proBNP. These findings showcase the importance of considering BMI when applying NT-proBNP for heart failure risk stratification, particularly for patients with low-level elevations in NT-proBNP (125-<450 pg/ml) where there appears to be a clinically meaningful absolute and relative risk gradient.
What problem does this paper attempt to address?