Radiographic Findings of Lung CT Scans of COVID-19 Patients: Delta and Omicron Variants

The present study was designed with the aim of comparing the radiological findings of the CT scan of the lungs of patients with Delta and Omicron variants of COVID-19 hospitalized in the intensive care unit. This was a retrospective comparative study conducted during the fifth peak of COVID-19 in Iran, August to November 2021 (the predominance of Delta variant) and March to April 2022 (the predominance of Omicron variant) in Qom, Iran. Radiologic findings of lung CT scans reported by a single radiologist were compared between these two cohorts. To analyze the collected data, SPSS version 22 statistical software is used through chi-square or fisher exact test, independent t-test, and Mann-Whitney test. A p-value of lower than 0.05 was considered as statistically significant. The total number of 50 patients in Delta (mean age of 58.68±14.77 years, 56% male) and 46 patients in Omicron (mean age of 70.45±18.02 years, 47% male) periods were included in study. Lesions of Right middle lobe (RML), left upper lobe (LUL), and Lingual lobes were more common in Delta period than Omicron (P<0.05). GGOs were more common in Delta period than in the Omicron ‎ (P=0.013).  There were no Unilateral GGOs during the Delta period, while 9 of 32 (28.1%) patients having GGO lesions in Omicron era had unilateral ones (P<0.001). Peripheral lesion was not differentiating Omicron from Delta. Crazy Pavings, Parenchymal bands, Halo signs, and Pleural effusions were more common in omicron period than in the Delta (P<0.05). Total Chest CT severity score was statistically higher in Delta period (22.75±5.24 units) than Omicron period (13.97±9 units). There were the significant differences in the radiologic characteristics between the two variants. This study suggests that with evolving the virus variants, while severity of lung involvement has decreased, GGOs of COVID-19 might have tended to manifest as unilateral lesions with more specific radiographic features.