Hospital Variability in the Use of Vasoactive Agents in Patients Hospitalized for Acute Decompensated Heart Failure for Clinical Phenotypes.

Abstract

Background: The absence of practice standards in vasoactive agent usage for acute decompensated heart failure has resulted in significant treatment variability across hospitals, potentially affecting patient outcomes. This study aimed to assess temporal trends and institutional differences in vasodilator and inotrope/vasopressor utilization among patients with acute decompensated heart failure, considering their clinical phenotypes.

Methods: Data were extracted from a government-funded multicenter registry covering the Tokyo metropolitan area, comprising consecutive patients hospitalized in intensive/cardiovascular care units with a primary diagnosis of acute decompensated heart failure between January 2013 and December 2021. Clinical phenotypes, that is, pulmonary congestion or tissue hypoperfusion, were defined through a comprehensive assessment of clinical signs and symptoms, vital signs, and laboratory findings. We assessed the frequency and temporal trends in phenotype-based drug utilization of vasoactive agents and investigated institutional characteristics associated with adopting the phenotype-based approach using generalized linear mixed-effects models, with random intercepts to account for hospital-level variability.

Results: Among 37 293 patients (median age, 80 years; 43.7% female), 88.6% and 21.2% had pulmonary congestion and tissue hypoperfusion status, respectively. Throughout the study period, both overall and phenotype-based vasodilator utilizations showed significant declines, with overall usage dropping from 61.4% in 2013 to 48.6% in 2021 (P_trend<0.001). Conversely, no temporal changes were observed in overall inotrope/vasopressor utilization from 24.6% in 2013 to 25.8% in 2021 or the proportion of phenotype-based utilization. Notably, there was considerable variability in phenotype-based drug utilization among hospitals, with a median ranging from 48.3% to 77.8%. In multivariable-adjusted models, a higher number of board-certified cardiologists were significantly associated with lower rates of phenotype-based vasodilator utilization and reduced inappropriate inotrope/vasopressor utilization, while tertiary care hospitals were linked to more appropriate inotrope/vasopressor utilization.

Conclusions: Substantial variability existed among hospitals in phenotype-based drug utilization of vasoactive agents for patients with acute decompensated heart failure, highlighting the need for standardized treatment protocols.

Registration: URL: https://www.umin.ac.jp/ctr/index.htm; Unique identifier: UMIN000013128.