Association Between Hypoxemia and Mortality in Patients With COVID-19

OBJECTIVE: To identify markers associated with in-hospital death in patients with coronavirus disease 2019 (COVID-19)–associated pneumonia. PATIENTS AND METHODS: A retrospective cohort study was conducted of 140 patients with moderate to critical COVID-19–associated pneumonia requiring oxygen supplementation admitted to the hospital from January 28, 2020, through February 28, 2020, and followed up through March 13, 2020, in Union Hospital, Wuhan, China. Oxygen saturation (SpO(2)) and other measures were tested as predictors of in-hospital mortality in survival analysis. RESULTS: Of 140 patients with COVID-19–associated pneumonia, 72 (51.4%) were men, with a median age of 60 years. Patients with SpO(2) values of 90% or less were older and were more likely to be men, to have hypertension, and to present with dyspnea than those with SpO(2) values greater than 90%. Overall, 36 patients (25.7%) died during hospitalization after median 14-day follow-up. Higher SpO(2) levels after oxygen supplementation were associated with reduced mortality independently of age and sex (hazard ratio per 1-U SpO(2), 0.93; 95% CI, 0.91 to 0.95; P<.001). The SpO(2) cutoff value of 90.5% yielded 84.6% sensitivity and 97.2% specificity for prediction of survival. Dyspnea was also independently associated with death in multivariable analysis (hazard ratio, 2.60; 95% CI, 1.24 to 5.43; P=.01). CONCLUSION: In this cohort of patients with COVID-19, hypoxemia was independently associated with in-hospital mortality. These results may help guide the clinical management of patients with severe COVID-19, particularly in settings requiring strategic allocation of limited critical care resources. TRIAL REGISTRATION: Chictr.org.cn Identifier: ChiCTR2000030852