Prone Positioning in Patients with COVID-19: Analysis of Multicenter Registry Data and Meta-analysis of Aggregate Data

Background/Aim: Evidence suggests a beneficial effect of prone positioning (PP) in COVID-19. Materials and Methods: Meta-analysis of individual (7 investigators’ groups) and aggregate data (PubMed/EMBASE) regarding the impact of PP on the ratio of arterial partial pressure of oxygen to fraction of inspired oxygen (PO₂/FiO₂) in patients with COVID-19. Results: Among 121 patients (mean age±SD 59.1±10.7 years, 55% males, 57% intubated) the mean post-versus pre- PP PO₂/FiO₂ difference was: (i) 50.4±64.3 mmHg, p<0.01, (ii) similar in awake (58.7±72.1 mmHg) versus intubated patients (44.1±57.5 mmHg, p=NS), (iii) inversely correlated with body mass index (r=–0.43, p<0.01). Meta-analysis of 23 studies (n=547, weighted age 58.3±4.1, 73% males, 59% intubated) showed a pooled PO₂/FiO₂ difference of 61.8 [95% confidence intervals=49.9-73.6] mmHg. Meta-regression analysis revealed no associations with baseline demographics, the time in PP before assessment, and the risk of bias of the studies. Conclusion: PP seems to improve oxygenation of patients with COVID-19.