Modelling oxygen-haemoglobin dissociation curves in COVID-19 [COVID-19]
Research type
Research Study
Full title
Modelling oxygen-haemoglobin dissociation curves in COVID-19 to predict outcomes: a multicentre cohort study
IRAS ID
294295
Contact name
David Russell-Jones
Contact email
Sponsor organisation
Royal Surrey NHS Foundation Trust
Duration of Study in the UK
0 years, 1 months, 1 days
Research summary
The COVID-19 pandemic has shown no evidence of resolution to date, and still presents a significant risk to population health worldwide. A proportion of patients present to hospital with extremely low blood oxygen levels (hypoxia) and require supplementary oxygen. Some treatments, for example steroid therapy, have been shown to reduce the risk of death but mortality rates remain high. The underlying cause of hypoxia is yet to be fully determined. In order for the lungs to effectively oxygenate blood, the amount of oxygen travelling to different areas of lung needs to be proportionate to the blood supply, a phenomenon known as ventilation-perfusion matching. If blood travels to areas that do not receive any oxygen at all, this blood effectively bypasses the lungs and is not oxygenated (a phenomenon known as intrapulmonary shunting). Our group recently published a paper demonstrating that intrapulmonary shunting may contribute to mortality in the sickest patients. We used a novel mathematical model which uses fingertip pulse oximetry to plot oxygen-haemoglobin curves [ODCs] (which plots the amount of oxygen in the blood against the amount of oxygen in the inspired air) and compared the curves to a reference curve to quantify shunt. Our current study aims to validate our preliminary retrospective findings in a larger, multicentre cohort. Our findings have the potential to further define the underlying mechanisms responsible for hypoxia in COVID-19 pneumonia which may ultimately impact management decisions. Invasive and non-invasive ventilation is known to improve intrapulmonary shunting and may improve outcomes if used earlier in the disease course in this subset of patients. Furthermore, we aim to validate our model for use as a rapid, inexpensive, and non-invasive triage tool to identify patients at high risk of death on admission to hospital which may have far reaching implications.
REC name
N/A
REC reference
N/A