Dose-response relationship between awake prone-positioning duration and PaO2/FiO2 changes and risk of disease aggravation in patients with severe COVID-19: A prospective cohort study.

ABSTRACT

BACKGROUND: Patients not mechanically ventilated often fail to achieve the recommended duration of awake prone positioning due to treatment interruption and discomfort. Few studies have investigated the link between treatment outcome and prone-positioning duration, the inability to accurately guide patients to perform awake prone positioning. OBJECTIVES: The aim of this study was to characterise and explore the relationship between awake prone-positioning duration with the ratio of the partial pressure of arterial oxygen to the fraction of inspired oxygen (PaO2/FiO2 [P/F]) changes and the risk of disease aggravation. METHODS: A prospective cohort study; dose-response relationship was used. Awake prone positioning was performed on patients with severe Corona Virus Disease 2019 (COVID-19) for 5 consecutive days from 1 February to 21 March 2023. Linear and logistic regression models were utilised to assess the association between prone-positioning duration with P/F changes and risk of disease aggravation, respectively. Meanwhile, the restricted cubic spline was used to evaluate the dose-response relationships. RESULTS: A total of 408 patients with severe COVID-19 were analysed. The daily prone positioning duration was 4.57 ± 2.74 h/d, and the changes in P/F were 67.63 ± 69.17 mmHg. On the sixth day of hospitalisation, the condition of 52 (12.8%) patients deteriorated. There was a positive, nonlinear dose-response relationship (Poverall < 0.001, Pnonlinearity = 0.041) and a strong, significant positive correlation (ß = 29.286, t = 4.302, P < 0.001) between the prone-positioning duration and P/F changes. The risk of disease aggravation gradually decreases with the increase of prone-positioning duration. Nonetheless, the prone-positioning duration was not statistically associated with disease aggravation (odds ratio = 0.986, 95% confidence interval 0.514-1.895). CONCLUSIONS: Awake prone positioning for ≥4 h/d is effective on oxygenation (not mortality/intubation) and is achievable for patients with severe COVID-19. Prolonged prone positioning is promising in improving patients' oxygenation but does not alleviate their risk of disease aggravation.