Evaluation of the Prognostic Value of Chest Computed Tomography (CT) Scan in COVID-19 Patients

Aminzadeh, B.; Layegh, P.; Foroughian, M.; Tavassoli, A.; Emadzadeh, M.; Teimouri, A.; Maftouh, M.

Objectives: To evaluate the prognostic value of chest computed tomography (CT) imaging features in patients with coronavirus disease 2019 (COVID-19) pneumonia. Methods: In this cross-sectional study, 201 patients with COVID-19 were enrolled consecutively. The patients' chest CT scans were analyzed, and the disease severity was rated using two methods: (1) total lung involvement (TLI) in which each lobe is scored from 0 to 4 based on the percentage of involvement;and (2) modified TLI in which each lobe involvement score is multiplied by the number of its segments, and the sum is recorded as the modified TLI. The patients were categorized into four groups depending on their prognosis (patients admitted to hospital wards, patients admitted to intensive care units (ICUs), patients with intubation during hospitalization, and expired patients). The relationship between both scoring methods and the clinical outcomes of patients was examined in the four groups. Results: The receiver operating characteristic (ROC) curve analysis showed no significant difference between the two scoring methods (TLI and modified TLI) in predicting the patients' prognosis. The average disease severity based on the two scoring methods was significantly different between the four groups. Patients who were intubated during hospitalization and patients who expired had significantly higher scores than patients admitted to the ICUs and hospital wards (P = 0.001). The area under the ROC curve for the prediction of mortality was 0.81 (95% CI: 0.72 - 0.90;P < 0.001);the TLI score of 18.5 could predict mortality with specificity of > 95%. Conclusion: The TLI scoring system can be used for predicting in-hospital mortality and ICU admission in COVID-19 patients. This scoring method can help us devise a better strategic healthcare plan during the COVID-19 pandemic.