Prone positioning to improve oxygenation in COVID-19 patients outside critical care (prone-COVID study)

Introduction: Prone positioning has potential efficacy in improving oxygenation in patients with coronavirus disease (COVID-19). The UK Intensive Care Society has published recommendations for prone positioning in awake patients with COVID-19 but it remains unknown whether it is beneficial in those patients prior to requiring respiratory support, and whether there is applicability for patients with non-COVID-19 pneumonia. Published studies are limited by their retrospective nature, inclusion of minimal time-points for physiological assessment and lack of information on tolerability. This prospective study aimed to assess the tolerability and physiological effects of prone positioning in non-ventilated patients with or without COVID-19 pneumonia. Methods: This interventional case-control study (ClinicalTrials.gov Identifier: NCT04589936) is currently being conducted at a tertiary hospital, with the aim of recruiting 56 patients with pneumonia. Inclusion criteria include those able to provide informed consent and rotate independently through a cycle of supine, lateral (for a duration of 15 minutes) and prone position (for as long as tolerable). The tolerability of each position was qualitatively assessed using a questionnaire and visual analogue scores (VAS). Continuous assessment of oxygenation, respiratory rate, end tidal carbon dioxide and pulse rate will be performed throughout the cycle of position changes. Thirtytwo of these patients will be proned for a longer duration, using a non-invasive positional sensor which will correlate body positions with the tolerability and physiological effects. Statistical analysis of ordinal VAS data was performed using a non-parametric Freidman test and demographic data presented as median (range). Results: The progress of participant recruitment is summarized in figure 1. To date, 73 patients with pneumonia were identified, 22 of whom were approached, of which eleven patients (eight COVID-19 and three non-COVID-19 pneumonia) underwent proning (6 female, 5 male;age, 67 [25-88] years;body mass index, 27.3 [22.8-32.0] kg/m2). There were no significant differences between different positions in the VAS for breathlessness (p=0.41), although the VAS for discomfort worsened between the supine (median score 2) and prone (score 5) position with a trend towards significance (p=0.100). Conclusion: Whilst awake prone positioning is recommended in national guidance, our prospective study to date highlights the challenge in recruiting patients who are suitable and are successfully able to self-prone. Our qualitative data suggests some patients experience discomfort in the prone position. Further detailed characterisation of physiological variables with ongoing recruitment will help inform the feasibility of performing prone positioning in hospitalised patients with pneumonia.