Abstract 4339562: High-output Heart Failure in Acute Respiratory Distress Syndrome: A Secondary Analysis of the FACTT Trial

Abstract

Introduction: Acute respiratory distress syndrome (ARDS) is characterized by pulmonary edema, hypoxia and systemic inflammation. The presence and clinical implications of a high cardiac output state in patients with ARDS remain unknown given the lack of invasive hemodynamic monitoring in most patients with ARDS. Hypothesis: A subset of patients with ARDS will have high output heart failure with high cardiac output and elevated pulmonary capillary wedge pressure (PAWP) representing a unique subphenotype of ARDS. Methods: Patients from the Comparison of Two Fluid-Management Strategies in Acute Lung Injury (FACTT) Trial were stratified according to baseline cardiac output states and filling pressures: HOLHF (high-output with left heart failure) was defined as cardiac index (CI) ≥ 4 L/min/m 2 and PAWP ≥ 15 mmHg; HOnoLHF (high-output without left heart failure) as CI ≥ 4 L/min/m 2 and PAWP < 15 mmHg; and NOHO (non-high-output) as CI < 4 L/min/m 2 . Hemodynamics, oxygenation, metabolic demand (VO 2 ) and clinical outcomes were compared across groups. Results: Of the 399 patients included, high-output states were present in 52% of all patients (Figure 1). High-output cohorts demonstrated higher myocardial workload, lower systemic vascular resistance (SVR), lower pulsatile arterial load, higher VO 2 and more impaired systemic O 2 utilization despite the highest peripheral O 2 delivery compared to NOHO (Table). Among the high-output cohorts, the HOLHF cohort demonstrated higher right ventricular stroke work (25.2 ± 12.1 vs. 20.3 ± 12.3, P<0.001), worse A-a gradient (326.0 ± 136.5 vs. 276.1 ± 114.7, P=0.048) and higher VO 2 (382.3 ± 234.2 vs. 289.8 ± 171.7, P<0.001) compared to HOnoLHF. Increasing cardiac output was individually related to decreasing SVR (r=-0.742, P<0.001) and increasing VO 2 (r=0.565, P<0.001) (Figure 2). When adjusted, both variables remained strongly related to increasing cardiac output (P<0.001). There were no significant differences among the cohorts with respect to refractory hypoxia and mortality. Conclusions: High-output states in ARDS are prevalent and underrecognized. Such states appear to be driven by arterial vasodilatation and heightened VO 2 with additional manifestations of increased myocardial workload and inefficient peripheral utilization with impaired systemic O 2 utilization. Whether specific strategies to mitigate the high output state might improve gas exchange, hypoxia and outcomes in patients with HOLHF and ARDS requires further study.