Description

Rationale

Even before the pandemic of the new coronavirus disease (COVID-19), acute respiratory distress syndrome (ARDS), the most severe form of acute hypoxemic respiratory failure (AHRF), constituted a public health challenge. Despite the intense research on identifying targeted pharmacological therapies for ARDS, there is no available treatment for the syndrome.

The failure of clinical trials exploring pharmacological therapies for ARDS has been attributed to incomplete understanding of the pathogenesis and heterogeneity of the syndrome. As examples of the heterogeneity of ARDS, our recent work has identified differential outcomes of patients with ARDS depending on whether hypoxemia is rapidly improving or persistent severe or associated with non-identifiable risk factors or associated with neutropenia. It is advocated that the heterogeneity of ARDS can be tangled by a precision approach, which identifies endotypes of ARDS; i.e., subtypes characterized by a distinct biological profile that might share mortality risk, clinical course, or treatment responsiveness.

Notwithstanding their contributions, current research efforts on endotyping ARDS might be limited by the fact that they are based on the current conceptual framework of the syndrome, which has been widely questioned. Indeed, for reasons such as high interobserver variability of radiological criteria of ARDS and exclusion of patients requiring high-flow nasal oxygen, influential experts have even suggested to completely abandon the term.

Objective

Accordingly, in the current study, we attempt to perform endotyping of critically ill patients presenting in the emergency department with de novo AHRF, which is a simpler and more reliable phenotype than ARDS. The approach for endotyping will be dynamic rather than static; i.e, two blood samples with a 24-hour interval will be used for endotyping to trace trajectories of biomarkers (over 1500 unique human proteins). In addition, we attempt to identify what clinical, radiological, physiological and biological variables collected early in the course of AHRF correlate with subsequent mortality and/or persistent severe hypoxemia.

Thus, the research protocol is organized as 3 aims.

Aim#1 will organize a registry and biobank of critically ill patients presenting in the emergency department with de novo AHRF.

Aim#2 will build a predictive model to identify what variables among those collected in Aim#1 are associated with subsequent mortality and/or persistent severe hypoxemia.

Aim#3 will use an agnostic discovery approach to explore novel proteomic biomarkers-based dynamic endotypes in critically ill patients presenting in the emergency department with de novo AHRF. Also, Aim#3 will create a multiprotein model of biomarkers associated with subsequent mortality and/or persistent severe hypoxemia and will determine whether inclusion of this multiprotein panel in the predictive score developed in Aim#2 improves risk prediction.