The use of laboratory parameters and computed tomography score to determine intensive care unit requirement in COVID-19

Objectives: It is vital to determine the intensive care unit (ICU) requirement at an early stage to reduce the mortality rate in COVID-19 patients. The aim of the study was to find reliable predictive markers to determine the ICU requirement. Methods: We retrospectively reviewed the clinical and laboratory records of 151 COVID-19 patients. The predictive abilities of biochemical parameters and computed tomography (CT) score were evaluated to determine of ICU requirement. Results: The area under curve (AUC) values for procalcitonin, D-Dimer, C reactive protein (CRP), glucose, lactate dehydrogenase (LDH) and CT score were found higher than those for other parameters in the prediction of ICU requirement. The negative predictive values of these markers were higher than their positive predictive values. CT score was found to be highly correlated with fibrinogen and CRP. The glucose levels [odd ratio (OR): 95% CI; 1.07, p-value: 0.014] and CT score [OR: 95% CI; 1.05, p-value: 0.022] were associated with ICU requirement in COVID-19 patients. Conclusions: CT score, procalcitonin, D-Dimer, CRP, glucose, and LDH are potential predictors to rule out ICU requirement on hospital admission. Fibrinogen and CRP can be used to follow up and predict lung damages in patients with COVID-19.