The aim of this study was to evaluate the association between baseline clinical and CT characteristics and to identify factors associated with intensive care unit (ICU) admission in hospitalized COVID-19 patients. This retrospective study included 176 adult patients with laboratory-confirmed SARS-CoV-2 infection hospitalized at the COVID Hospital of the Clinical Hospital Center Kosovska Mitrovica during 2021–2022 (Delta and Omicron variants). Patients were divided into two groups according to intensive care unit requirement: those treated in a general inpatient ward (No ICU) and those requiring ICU admission (ICU group). Demographic and clinical characteristics, lifestyle factors, CT findings, CT severity score (CTSS) values, and therapeutic interventions were compared between groups. Of the total cohort, 113 patients (64%) were hospitalized in a general inpatient ward, while 63 (36%) required intensive care unit admission. Independent predictors of ICU admission identified in the multivariate logistic regression analysis were obesity (B = 2.96, p < 0.001), dyspnea (B = 1.51, p = 0.041), higher CT severity score (B = 0.68, p < 0.001), and lower glucose levels (B = −0.27, p = 0.014). Furthermore, for each one-point increase in the CTSS, the odds of ICU admission nearly doubled (OR = 1.97). Total CT score values above the cut-off point (15.0) demonstrated significant reliability in discriminating the need for ICU transfer in patients with COVID-19. These findings suggest that combined clinical and radiological assessment at hospital admission may facilitate early identification of patients at high risk of requiring ICU care, with the CT severity score representing the strongest radiological predictor.
Milenkovic et al. (2026) studied this question.