Introduction
Acute respiratory distress syndrome (ARDS) requiring mechanical
ventilation is the most severe manifestation of coronavirus disease 2019
(COVID-19) caused by the Severe Acute Respiratory Syndrome –
Coronavirus 2 (SARS-CoV-2). Mortality has varied since initial reports
but remains consistently high, above 40%, even in the most recent
publications (1). A significant debate concerning the existence and
effects of hyperinflammation in severe COVID-19 is ongoing (2, 3) and
most of the therapeutic trials have focused on dampening the systemic
immune response. Immune modulation has become the therapeutic norm even
after non-corticosteroid immunosuppressive trials registered only small
effect sizes. Clinical and treatment effect heterogeneity persists and
only recent reports have overlapped the recognized ARDS phenotypes –
hyper- and hypo-inflammatory – onto COVID-19 (4, 5). Both these
phenotypes seem to respond to corticosteroids, suggesting treatment
heterogeneity within the phenotypes.
There is also a signal that the SARS-CoV-2 viral load may play a role in
the therapeutic response and ultimately the survival of critically ill
patients with severe hypoinflammatory COVID-19 (5). Autopsy studies have
not definitively determined the direct cause of demise in patients with
severe COVID-19 ARDS and confirm that the abnormal pulmonary response in
fatal COVID-19 can be triggered by direct viral cytopathic damage, an
exaggerated immune infiltrate and even ventilator-induced lung injury.
(6–8). It remains unclear how to best differentiate and treat patients
who succumb to the uncontrolled viral replication vs an exaggerated,
virus-independent immune response.
The main objective for this study was to identify groups of patients
with severe COVID-19 with similar viral loads and immune response
intensities using prespecified measures of viral load and immune markers
in bronchoalveolar lavage (BAL) and plasma. We hypothesized that using
surrogates of semi-quantitative SARS-CoV-2 RT-PCR in BAL would better
define COVID-19 ARDS phenotypes, and that lower airways samples would
provide a more accurate window into viral infection and the triggered
host immune response when compared with matched plasma samples.