ABSTRACT
Background The long-term pulmonary sequelae of COVID-19 is not well known. Purpose To characterize patterns and rates of improvement of chest CT abnormalities 1 year after COVID-19 pneumonia. Materials and Methods This was a secondary analysis of a prospective, multicenter observational cohort study conducted from April 29 to August 12, 2020, to assess pulmonary abnormalities at chest CT approximately 2, 3, and 6 months and 1 year after onset of COVID-19 symptoms. Pulmonary findings were graded for each lung lobe using a qualitative CT severity score (CTSS) ranging from 0 (normal) to 25 (all lobes involved). The association of demographic and clinical factors with CT abnormalities after 1 year was assessed with logistic regression. The rate of change of the CTSS at follow-up CT was investigated by using the Friedmann test. Results Of 142 enrolled participants, 91 underwent a 1-year follow-up CT examination and were included in the analysis (mean age, 59 years ± 13 [SD]; 35 women [38%]). In 49 of 91 (54%) participants, CT abnormalities were observed: 31 of 91 (34%) participants showed subtle subpleural reticulation, ground-glass opacities, or both, and 18 of 91 (20%) participants had extensive ground-glass opacities, reticulations, bronchial dilation, microcystic changes, or a combination thereof. At multivariable analysis, age of more than 60 years (odds ratio [OR], 5.8; 95% CI: 1.7, 24; P = .009), critical COVID-19 severity (OR, 29; 95% CI: 4.8, 280; P < .001), and male sex (OR, 8.9; 95% CI: 2.6, 36; P < .001) were associated with persistent CT abnormalities at 1-year follow-up. Reduction of CTSS was observed in participants at subsequent follow-up CT (P < .001); during the study period, 49% (69 of 142) of participants had complete resolution of CT abnormalities. Thirty-one of 49 (63%) participants with CT abnormalities showed no further improvement after 6 months. Conclusion Long-term CT abnormalities were common 1 year after COVID-19 pneumonia. © RSNA, 2022 Online supplemental material is available for this article. See also the editorial by Leung in this issue.