Estimated Sp02/Fio2 ratio to predict mortality in patients with suspected COVID-19 in the Emergency Department: a prospective cohort study

BackgroundThis study examined whether the presence and severity of Type 1 Respiratory Failure (T1RF), as measured by the ratio of pulse oximetry to estimated fraction of inspired oxygen (SpO2/eFiO2 ratio), is a predictor of in-hospital mortality in patients presenting to the ED with suspected COVID-19 infection. MethodsWe undertook a prospective observational cohort study of patients admitted to hospital with suspected COVID-19 in a single ED in England. We used univariate and multiple logistic regression to examine whether the presence and severity of T1RF in the ED was independently associated with inhospital mortality. Results180 patients with suspected COVID-19 infection met the inclusion criteria for this study, of which 39 (22%) died. Severity of T1RF was associated with increased mortality with odds ratios (OR) and 95% confidence intervals of 1.58 (0.49 - 5.14), 3.60 (1.23 - 10.6) and 18.5 (5.65 - 60.8) for mild, moderate and severe T1RF, respectively. After adjusting for age, gender, pre-existing cardiovascular disease, neutrophil-lymphocyte ration (NLR) and estimated glomerular filtration rate (eGFR), the association remained, with ORs of 0.63 (0.13 - 3.03), 3.95 (0.94 - 16.6) and 45.8 (7.25 - 290). The results were consistent across a number of sensitivity analyses. ConclusionsSeverity of T1RF in the ED is an important prognostic factor of mortality in patients admitted with suspected COVID-19 infection. Current prediction models frequently do not include this factor and should be applied with caution. Further large scale research on predictors of mortality in COVID-19 infection should include SpO2/eFiO2 ratios or a similar measure of respiratory dysfunction. KEY MESSAGESO_ST_ABSWhat is already known on this topicC_ST_ABSA number of studies have identified potential variables to predict mortality in patients with COVID-19 but have mainly focused on laboratory measures. The primary pathophysiology and cause of death seems to be lung injury leading to an acute respiratory distress syndrome (ARDS) like illness, clinically presenting as type 1 respiratory failure (T1RF). Despite this, only a very small number of studies have included measurements of respiratory dysfunction as predictor of mortality. What this study addsTo our knowledge, this is the first study to examine the association between T1RF as measured by the ratio of pulse oximetry (SpO2) and estimated fraction of inspired oxygen (FiO2) and mortality in patients admitted with suspected COVID-19. The presence and severity of T1RF are strongly associated with mortality. At the same time, other factors previously shown to be associated with mortality are potentially less important than currently assumed, once adjusted for the severity of T1RF.