Radiological imaging of viral pneumonia cases identified before the COVİD-19 pandemic period and COVİD-19 pneumonia cases comparison of characteristics

OBJECTIVE: Computed tomography of the thorax is used as a common diagnosis method in the diagnosis of Coronavirus Disease 2019 (COVID-19) which causes pandemic. As in the pre-pandemic period, during the COVID-19 pandemic period, the radiological differential diagnosis of other viral agents that cause pneumonia in patients with normal immunity or in immunosuppressed patients with seasonal epidemics has gained importance in early diagnosis and isolation. Therefore, it was aimed to investigate the difference between CT imaging findings defined as characteristic in COVID-19 pneumonia and CT findings detected in pneumonia due to other viral agents previously encountered. MATERIALS AND METHODS: 249 COVID-19 patients aged 18 and over who applied to Istanbul Faculty of Medicine between March 15, 2020 and May 30, 2020, who were positive for Reverse Transcriptase Polymerase Chain Reaction (RT-PCR) in nasopharyngeal swab samples taken at the application, and who were found to have pneumonia in the thoracic CT examination at admission with; resulting negative bacterial and fungal agents in other sputum and blood cultures with thoracic CT findings consistent with viral pneumonia between January 2015 and December 2019 and positive results of viral respiratory tract panel or bronchoalveolar lavage / blood viral PCR within an average of 5.67 ± 7.95 days, 94 patients with detected non-COVID viral pneumonia (Viral panel results; Influenza AB n = 26, Adenovirus n = 5, CMV n = 28, RSV n = 8, Parainfluenza n = 10, HMPV n = 3, Endemic Coronaviruses (HCoV- NL63, HCoV-HKU, HCoV-229E, HCoV-OC43) n = 16, Rhinovirus n = 7, Bocavirus (HBoV) n = 1) were included in the study. Thorax CT examination protocol: image slice thickness is 1 mm-5 mm, CT images obtained in supine position in full inspiratory in all patients are -600 to +1600 HU for lung parenchyma, +50 to +350 HU for mediastinum using window width; it was retrospectively analyzed using the PACS System. CT findings were evaluated by two radiologists who did not know to which group the patient belonged, and it was decided by consensus. Age, gender, known chronic disease demographic data of the patients; leukocyte, lymphocyte, C reactive protein (CRP) laboratory data; CT findings include the percentage of involvement, number of lesions, transverse distribution preference, antero-posterior distribution preference, dominant pattern, if there is ground glass opacity (GGO), its distribution pattern, nodule, tree in bud sign, interstitial changes, crazy paving sign, reverse halo sign, vacuolar sign, halo sign, vascular prominence, fibrosis-fibrotic linear opacities, traction bronchiectasis, peribronchial wall thickness increase, air trapping, pleural retraction, pleural effusion, pericardial effusion, cavitation, mediastinal/hilar lymphadenopathy, dominant lesion size, consolidation, subplevral curvilinear opacities, air bronchogram, pleural thickening were examined. CT findings were also evaluated with the RSNA consensus guideline and the CORADS scoring system, data obtained were divided into two main groups as non-COVID-19 pneumonia and COVID-19 pneumonia; statistically compared with chi-square tests and multiple regression analysis of independent variables. RESULTS: While all of the non-COVID-19 patient group had chronic diseases, 33% of the COVID-19 patient group had no chronic disease. When the patient group with non-COVID-19 viral pneumonia was compared with the patient group with COVID-19 pneumonia; RSNA class showing atypical and indeterminate features, CORADS 2 and 3 scores, parenchymal involvement more than 75%, multiple lesions, coexistence of both peripheral and central distribution, dominant nodular pattern, absence of BCO, diffuse BCO, non-round perihilar BCO distribution, presence of a non-dominant nodule, tree in bud sign, septal thickening, fibrosis-fibrotic linear opacities, peribronchial wall thickness increase, air trapping, pleural retraction, pleural effusion, pericardial effusion, cavitation, presence of significant lymph nodes with a short axis > 1 cm, increased or decreased leukocyte count, presence of lymphopenia, increased or decreased neutrophil count were different in favor of non-COVID-19 pneumonia (p <0.01). Class showing typical features of RSNA, CORADS 5 score, single lesion, peripheral distribution, BCO dominant pattern, peripheral-bilateral BCO distribution pattern, thin reticular opacity, crazy paving sign, reverse halo sign, vascular prominence sign, presence of traction bronchiectasis, normal leukocyte, normal lymphocyte and normal neutrophil counts were found to be different in favor of COVID-19 pneumonia (p <0.01). In the multiple linear regression analysis performed to determine the effect of independent variables on COVID-19 pneumonia; when the regression coefficients were examined, it was found that those with inverted halo sign (β = 0.097, p <0.05) and those with pleural effusion (β = 10.631, p <0.05) had a significant effect on COVID-19 pneumonia; it was found that COVID-19 pneumonia was more common in patients with reverse halo sign compared to those without pleural effusion. CONCLUSİON: In the diagnosis of viral pneumonia, radiological imaging, which is evaluated together with laboratory examinations, especially clinical and gold standard RT-PCR test, has an important role in diagnosis and patient management. RSNA classification and CORADS scoring system can be used to distinguish COVID-19 pneumonia from non-COVID-19 pneumonia. The presence of reverse halo sign and absence of pleural effusion was found to be efficient in the diagnosis of COVID-19 pneumonia.