Prone Position to Improve Oxygenation in COVID-19 Patients Outside Critical Care

COVID-19 (coronavirus disease 2019) caused by the infectious agent, SARS-CoV-2, was first
recognised in Wuhan, Hubei Province, China in December 2019. Whilst a large proportion of
COVID-19 illness within the community is mild, patients who are admitted to hospital with
COVID-19 often have more severe disease and have a higher risk of developing COVID-19
complications including ARDS. Patients with ARDS are often managed on intensive care units
that have evidence-based protocols for managing patients with ARDS. Prone positioning for
such patients is a well-recognised component of ARDS critical care management. It is known to
improve oxygenation and reduce oxygen requirements, as well as reduce mortality in patients
with moderate to severe ARDS managed in critical care. The benefits of prone positioning as a
technique to improve oxygenation in awake, non-intubated patients is less well-established. A
few studies have showed prone positioning to be well-tolerated and improved oxygenation
amongst non-intubated patients without respiratory support, as well as those requiring
high-flow nasal oxygen or non-invasive ventilation support.

It remains unknown whether prone positioning is beneficial in patients with COVID-19 earlier
in their disease, prior to requirement of non-invasive or invasive ventilation. This study
therefore aims to better understand the physiological effects of prone positioning in
patients with pneumonia with and without COVID-19.

This will be a single-centred interventional case-control study, comparing physiological
effect of prone positioning in COVID-19 cases versus pneumonia unrelated to COVID-19. A
sub-study of a smaller number of enrolled patients, with approximately even spread of
COVID-19 and non-COVID cases, will allow an additional exploratory descriptive report on
prone positioning over a 24-hour period.

Participant will be fitted with a Masimo monitoring device that enables continuous
monitoring, and subsequent data storage and download of SpO2, heart rate, end tidal CO2 and
respiratory rate. The sub-study investigating the effects of a longer duration of proning
will involve applying a non-invasive positional sensor to automatically detect the
participant's position, and correlate it with the patient's physiological parameters as well
as tolerability. Qualitative data on patient's tolerability of prone positioning will also be
collected.

It is anticipated that each participant will be their own control, enabling comparison of
SpO2 in supine, lateral, prone and again supine positions. Summary statistics of mean,
median, range, interquartile range, and range for each position for participants will be
summarised. The data from the two arms will initially be analysed separately, then pooled to
determine if a larger sample size impacts results. A multilevel regression model with average
SpO2 as outcome will be fitted to assess the effect of lying position on SpO2. Additional
models such as mixed models incorporating multiple repeated measurements/endpoints may be
considered/fitted for exploratory analyses. Data from questionnaires will be reported by
summary statistics where possible. Free text responses may be grouped into themes, and
parametric or non-parametric tests will be used to analyse the VAS data further.

All data will be transferred into a Case Report Form (CRF) which will be anonymised. The
investigator will obtain written informed consent from each participant before any
study-specific activity is performed. Before the start of the study, or implementation of any
amendment, we will obtain approval of the protocol, protocol amendments, informed consent
forms and other relevant documents from the REC.