Patients with COVID-19 pneumonia have rapid progression of CT abnormalities after symptom onset, and the great majority still have CT abnormalities at hospital discharge, new findings show. "The temporal changes of the diverse CT manifestations followed a specific pattern, which might indicate the progression and recovery of the illness," Dr. Min Zhou of Huazong University of Science and Technology in Wuhan, China, and colleagues write in Radiology. To better understand how CT can be used in diagnosing and managing COVID-19 pneumonia patients, Dr. Zhou and colleagues analyzed 366 CT scans from 90 patients admitted to the isolation wards of their hospital with a positive SARS-CoV-2 rRT-PCR result from January 16 to February 17, 2020. Each patient had a median of four scans, with a median interval of 6 days between scans. Ten of the patients were scanned for the first time before they were symptomatic, and six of these patients had normal scans, two had ground-glass opacity and two had consolidation; these patients developed symptoms from 2 to 6 days after their abnormal CT scan. Ground-glass opacity was the most frequent type of abnormality, appearing in 62% of patients on illness days 0-5 and 45% of scans from illness days 12-17. During illness days 0-5 and 6-11, consolidation was the second most frequent finding, in 23% and 24% of patients, respectively. On the following days, the mixed pattern was the second most common finding, ranging from 38% of patients on illness days 12-17 to 22% of patients at day 24 or later. Patients spent a median of 16 days in the hospital, and their median illness duration was 24 days. Seventy patients were discharged from the hospital and only four had no residual disease on their final CT scan. Ground glass opacity was the most common type of abnormality in the discharged patients. The authors estimate sensitivities for CT scans of 84% for 0-5 days and 99% for days 6-11. One limitation of the study, Dr. Zhou and colleagues note, is that it does not include subgroup analyses of patients with mild and severe disease, which could help identify prognostic factors. "Larger sample size and longer follow-up were needed to better predict the development of the illness," they add. —Reuters Staff Source
