DOI: 10.1093/ejhf/xuag193.945 ISSN: 1388-9842

Factors associated with NT-proBNP reduction in acute heart failure: a secondary analysis of the PREMIER trial

Y Nabeshima, A Tanaka, T Imai, Y Matsue, K Kida, K Node

Abstract

Background/Introduction

N-terminal pro-B-type natriuretic peptide (NT-proBNP) reduction is a key surrogate for heart failure (HF) outcomes. The PREMIER trial showed that in-hospital sacubitril/valsartan (Sac/Val) initiation yielded a greater 8-week NT-proBNP reduction than angiotensin-converting enzyme inhibitor or angiotensin II receptor blocker (ACEi/ARB) in Japanese patients with acute heart failure (AHF). However, clinical characteristics associated a clinically meaningful NT-proBNP reduction remain undefined. Notably, atrial fibrillation (AF) can elevate the natriuretic peptide levels via atrial structural remodelling, complicating the interpretation of NT-proBNP changes.

Purpose

To identify independent predictors of ≥ 50% NT-proBNP reduction by Week 8 in patients with AHFand to further explore if they differed by baseline AF status (non-AF vs AF).

Methods

This post-hoc analysis of the multicentre, randomised PREMIER trial included 376 (Sac/Val n=183, ACEi/ARB n=193) patients haemodynamically stabilised after an AHF event. Independent predictors of achieving the reduction in NT-proBNP were identified using multivariable logistic regression in the AF-stratified cohorts. Interaction tests within the AF cohort tested treatment effect modification by baseline characteristics.

Results

Overall, 39.7% of patients achieved a ≥ 50% NT-proBNP reduction by Week 8. Its independent predictors included Sac/Val use (adjusted odds ratio [aOR] 2.21, 95% confidence interval [CI] 1.32 to 3.67), younger age (aOR 1.98 per 10-year decrease, 95% CI 1.29 to 3.04), de novo HF (aOR 2.40, 95% CI 1.41 to 4.10), higher baseline NT-proBNP (aOR 3.06 per 2-fold increase), and higher systolic blood pressure (aOR 1.44 per 10 mm Hg increase, 95% CI 1.03 to 2.02) (Figure 1). Stratified by AF status (non-AF n=214, AF n=162), 43.4% of non-AF and 34.6 % of AF cohorts achieved the reduction in NT-proBNP. In the non-AF cohort, Sac/Val use remained an independent predictor of response (aOR 3.12, 95% CI 1.57 to 6.21). Conversely, in the AF cohort, Sac/Val use was not a significant predictor (aOR 1.33, 95% CI 0.53 to 3.33). Instead, higher baseline NT-proBNP (aOR 4.07, 95% CI 2.09 to 7.93) and lower left ventricular ejection fraction (LVEF) (aOR 3.38 per 10% decrease, 95% CI 1.23 to 9.29) were significant predictors (Figures 1 and 2). Interaction analysis in the AF cohort showed that HF chronicity modified treatment effect (P=0.026): Sac/Val tended to favour NT-proBNP change (−25% [95% CI −43 to 10%]) in patients with de novo HF, but not in those with a history of HF (+36% [95% CI −3 to 91%]).

Conclusions

Predictors of NT-proBNP response in stabilised patients after an AHF event differed by AF status. In non-AF, Sac/Val was the primary driver of NT-proBNP reduction, whereas in AF, NT-proBNP response depended on some background HF status. Our findings highlight a need for a patient-centric therapeutic approach for this population.For image description, please refer to the figure legend and surrounding text.For image description, please refer to the figure legend and surrounding text.

More from our Archive