Abstract
BACKGROUND: We aimed to investigate risk factors predicting oxygen demand in COVID-19 patients. METHODS: Patients admitted to Shizuoka General Hospital with COVID-19 from August 2020 to August 2021 were included. First, we divided patients into groups with and without oxygen demand. Then, we compared patients' clinical characteristics and laboratory and radiological findings to determine factors predicting oxygen demand. RESULTS: One hundred seventy patients with COVID-19 (aged 58±15 years, 57 females) were enrolled. Common comorbidities were cardiovascular diseases (47.6%), diabetes mellitus (28.8%), and dyslipidemia (26.5%). Elder age, higher body mass index, cardiovascular diseases, diabetes mellitus, lower lymphocyte count, albumin, hepatic attenuation value, and the liver-to-spleen ratio (L/S), higher D-dimer, aspartate aminotransferase, lactate dehydrogenase, troponin-T, C-reactive protein, KL-6, chest and abdominal circumference, and visceral fat were found in patients with oxygen demand. According to the multivariate logistic regression analysis, L/S, lymphocyte count, D-dimer, and abdominal circumference under the diaphragm were independent risk factors predicting oxygen demand in COVID-19 patients. CONCLUSIONS: On admission, L/S, lymphocyte count, D-dimer, and abdominal circumference were predictive factors for oxygen demand. These factors may help in the appropriate triage of COVID-19 patients in the decision to admit them to the hospital.