Pre-hospital lowest recorded oxygen saturation independently predicts death in patients with COVID-19.

Background: The coronavirus disease 2019 (COVID-19) results in hypoxia in around a fifth of adult patients. Severe hypoxia in the absence of visible respiratory distress ('silent hypoxia') is increasingly recognised in these patients. There are no published data evaluating lowest recorded pre-hospital oxygen saturation or pre-hospital National Early Warning Score 2 (NEWS2) as a predictor of outcome in patients with COVID-19.

Methods: In this retrospective service evaluation, we included adult inpatients with laboratory confirmed COVID-19 who were discharged from hospital or who died in hospital between 12 March and 28 April 2020 (n = 143). Pre-hospital and in-hospital data were extracted and analysed to explore risk factors associated with in-hospital mortality to inform local triage and emergency management.

Results: The lowest recorded pre-hospital oxygen saturation was an independent predictor of mortality when controlling for age, gender and history of COPD. A 1% reduction in pre-hospital oxygen saturation increased the odds of death by 13% (OR 1.13, p < 0.001). Lower pre-hospital oxygen saturation predicted mortality after adjusting for the pre-hospital NEWS2 (OR for a 1% reduction in pre-hospital oxygen saturation 1.09, p = 0.02). The pre-hospital NEWS2 was higher in those who died (Median 9; IQR 7-10; n = 24) than in those who survived to discharge (Median 6; IQR 5-8; n = 63).

Conclusion: This service evaluation suggests that the lowest recorded pre-hospital oxygen saturation may be an independent predictor of mortality in COVID-19 patients. Lowest pre-hospital oxygen saturation should be recorded and used in the assessment of patients with suspected COVID-19 in pre-hospital and emergency department triage settings.