Biological subphenotypes of acute respiratory distress syndrome show prognostic enrichment in mechanically ventilated patients without acute respiratory distress syndrome

Abstract

Rationale: Recent studies showed that biological subphenotypes in acute respiratory distress syndrome (ARDS) provide prognostic enrichment and show potential for predictive enrichment.

Objectives: To determine whether these subphenotypes and their prognostic and potential for predictive enrichment could be extended to other patients in the ICU, irrespective of fulfilling the definition of ARDS.

Methods: This is a secondary analysis of a prospective observational study of adult patients admitted to the ICU. We tested the prognostic enrichment of both cluster-derived and latent-class analysis (LCA)-derived biological ARDS subphenotypes by evaluating the association with clinical outcome (ICU-day, 30-day mortality, and ventilator-free days) using logistic regression and Cox regression analysis. We performed a principal component analysis to compare blood leukocyte gene expression profiles between subphenotypes and the presence of ARDS.

Measurements and Main Results: We included 2,499 mechanically ventilated patients (674 with and 1,825 without ARDS). The cluster-derived "reactive" subphenotype was, independently of ARDS, significantly associated with a higher probability of ICU mortality, higher 30-day mortality, and a lower probability of successful extubation while alive compared with the "uninflamed" subphenotype. The blood leukocyte gene expression profiles of individual subphenotypes were similar for patients with and without ARDS. LCA-derived subphenotypes also showed similar profiles.

Conclusions: The prognostic and potential for predictive enrichment of biological ARDS subphenotypes may be extended to mechanically ventilated critically ill patients without ARDS. Using the concept of biological subphenotypes for splitting cohorts of critically ill patients could add to improving future precision-based trial strategies and lead to identifying treatable traits for all critically ill patients.