Methods
Study Design and Inclusion Patients
This study was approved by the research ethics committee Mazandaran University of Medical Sciences, Sari, Iran (IR.MAZUMS.REC.1399.7418). All patients who were admitted to Emam Khomeini hospital, Sari, Mazandaran from February 26, 2020, to September 21, 2020, were included. The written informed consent was given up from all participants with COVID-19 infection. Simple random sampling was used for including patients. The diagnosis, hospitalization, and treatment of patients was performed according to Iran’s Ministry of Health and Medical Education guidelines. Iranian guidelines mentioned the laboratory and radiological examinations as a first line of COVID-19 confirmation along with clinical symptoms such as fever, cough, and shortness of breath. The cases without chest CT and laboratory test confirmation were excluded. Assuming 95% confidence level and p=0.7 (lymphopenia 70%), d= 0.07, and Cochran’s formula the sample size was calculated as 596. Clinical outcomes including CBC, CRP, and ESR were evaluated within a 7-days monitoring period. All medical data including epidemiological, demographic, laboratory data, and patient’s discharge and/or death were extracted via the hospital information system (HIS).
CT scans
All CT examinations were performed using a multi‐detector computed tomography scanner (CT-Scan) Somatom Emotion eco (16‐slice configuration, Siemens, Erlangen, Germany). The supine CT scan was performed and no contrast medium had been administrated. The CT images were observed with both lungs and in parenchymal and mediastinal window settings. Two radiologists appraised the lung scans by the picture archiving and communication system (PACS, INFINITT Healthcare co., Ltd., Seoul, Korea). The features of CT imaging were typically focused on unilateral and bilateral Ground Glass Opacity (GGO), consolidation, and pleural effusion. The pulmonary involvement of COVID-19 patients was categorized into two groups: less than 50% (Figure 1-A) and more than 50 % involvement (Figure 1-B). Clinical data were screened on the date of symptom onset and 7th day and abstracted from the medical records of Imam Khomeini hospital, Sari, Iran. All data containing age, sex, underlying medical conditions, CT scan results, lymphocyte count, CRP, ESR, Platelet, length of stay (LOS), and mortality was recorded according to the checklist. Data analysis was performed using SPSS, and Mann-Whitney and Kruskal-Wallis tests. The logistic regression model interprets the odds ratio and P <0.05 was considered as statistically significant.