Novel subgroups in acute respiratory failure based on the trajectories of three endotheliopathy biomarkers: A cohort study

Abstract

Baseline levels of endotheliopathy are associated with worse respiratory outcomes and mortality in undifferentiated acute respiratory failure (ARF), but knowledge is lacking on the development of endotheliopathy over time in ARF. We, therefore, aimed to evaluate the prognostic significance of trajectories of endotheliopathy during the first days of ARF. We performed a secondary, exploratory analysis of a single-center prospective cohort including 459 patients requiring mechanical ventilation. Based on Days 1–3 Syndecan-1, soluble Thrombomodulin (sTM), and Platelet Endothelial Cell Adhesion Molecule-1 (PECAM-1), we divided patients into subgroups using latent class mixed modeling and correlated subgroups with clinical outcomes using Cox regression. Based on Syndecan-1 and sTM, respectively, we identified two subgroups. Based on PECAM-1, we identified three subgroups. Subgroups based on Syndecan-1 and sTM were identifiable from the baseline levels, but subgroups based on PECAM-1 were not. Patients with persistently high levels of both sTM and PECAM-1 were liberated from mechanical ventilation more slowly (Group high vs. Group low, sTM: hazard ratio [HR]: 0.66, 95% confidence interval [CI]: 0.50–0.88, p =.01, PECAM-1: HR: 0.59, 95% CI: 0.37–0.93, p =.02) and had higher 30-day mortality (sTM: HR: 1.90, 95% CI: 1.20–3.01, p =.01, PECAM-1: HR: 4.25, 95% CI: 1.99–9.07, p