Lung ultrasound response to awake prone positioning predicts the need for intubation in patients with COVID-19 induced acute hypoxemic respiratory failure: an observational study

Background: Awake prone positioning (APP) reduces the intubation rate in COVID-19 patients treated by high-flow nasal cannula (HFNC). However, the lung aeration response to APP has not been addressed. We aimed to explore the lung aeration response to APP by lung ultrasound (LUS). Methods: This two-center, prospective, observational study enrolled patients with COVID-19-induced acute hypoxemic respiratory failure treated by HFNC and APP. LUS score was recorded 5–10 min before, 1 h after APP, and 5–10 min after supine in the first APP session within the first three days. The primary outcome was LUS score changes in the first three days. Secondary outcomes included changes in SpO₂/FiO₂ ratio, respiratory rate and ROX index (SpO₂/FiO₂/respiratory rate) related to APP, and the rate of treatment success (patients who avoided intubation). Results: Seventy-one patients were enrolled. LUS score decreased from 20 (interquartile range [IQR] 19–24) to 19 (18–21) (p < 0.001) after the first APP session, and to 19 (18–21) (p < 0.001) after three days. Compared to patients with treatment failure (n = 20, 28%), LUS score reduction after the first three days in patients with treatment success (n = 51) was greater (− 2.6 [95% confidence intervals − 3.1 to − 2.0] vs 0 [− 1.2 to 1.2], p = 0.001). A decrease in dorsal LUS score > 1 after the first APP session was associated with decreased risk for intubation (Relative risk 0.25 [0.09–0.69]). APP daily duration was correlated with LUS score reduction in patients with treatment success, especially in dorsal lung zones (r = − 0.76; p < 0.001). Conclusions: In patients with acute hypoxemic respiratory failure due to COVID-19 and treated by HFNC, APP reduced LUS score. The reduction in dorsal LUS scores after APP was associated with treatment success. The longer duration on APP was correlated with greater lung aeration. Trial registration This study was prospectively registered on clinicaltrials.gov on April 22, 2021. Identification number NCT04855162.