Early versus late awake prone positioning in non-intubated patients with COVID-19

Background Awake prone positioning (APP) is widely used in the management of patients with coronavirus disease (COVID-19). The primary objective of this study was to compare the outcome of COVID-19 patients who received early versus late APP. Methods Post hoc analysis of data collected for a randomized controlled trial (ClinicalTrials.gov NCT04325906). Adult patients with acute hypoxemic respiratory failure secondary to COVID-19 who received APP for at least one hour were included. Early prone positioning was defined as APP initiated within 24 h of high-flow nasal cannula (HFNC) start. Primary outcomes were 28-day mortality and intubation rate. Results We included 125 patients (79 male) with a mean age of 62 years. Of them, 92 (73.6%) received early APP and 33 (26.4%) received late APP. Median time from HFNC initiation to APP was 2.25 (0.8-12.82) vs 36.35 (30.2-75.23) hours in the early and late APP group (p < 0.0001), respectively. Average APP duration was 5.07 (2.0-9.05) and 3.0 (1.09-5.64) hours per day in early and late APP group (p < 0.0001), respectively. The early APP group had lower mortality compared to the late APP group (26% vs 45%, p = 0.039), but no difference was found in intubation rate. Advanced age (OR 1.12 [95% CI 1.0-1.95], p = 0.001), intubation (OR 10.65 [95% CI 2.77-40.91], p = 0.001), longer time to initiate APP (OR 1.02 [95% CI 1.0-1.04], p = 0.047) and hydrocortisone use (OR 6.2 [95% CI 1.23-31.1], p = 0.027) were associated with increased mortality. Conclusions Early initiation (< 24 h of HFNC use) of APP in acute hypoxemic respiratory failure secondary to COVID-19 improves 28-day survival. Trial registration ClinicalTrials.gov NCT04325906.