Changes on computed tomography in post-acute COVID-19 syndrome: systematic review and meta-analysis

The objective of this systematic review and meta-analysis of observational studies was to estimate the prevalence of residual alterations in the lung parenchyma on computed tomography (CT) after coronavirus disease 2019 (COVID-19), correlating those alterations with the severity of the acute phase of the disease. We reviewed data related to adult patients evaluated at 3, 6, and 12 months after the diagnosis of moderate-to-critical COVID-19. We performed structured searches of 14 databases, encompassing works published between January 2020 and January 2024. Thus, 44 primary studies were selected. Data on mild cases of COVID-19 were excluded, as were those related to assessment of the acute phase of the disease. The results were analyzed descriptively, and meta-analyses were conducted to estimate prevalence. The estimated prevalence of altered CT scans at post-diagnosis months 3, 6, and 12 was 69.0% (95% CI: 60.0-77.6%; I ² = 86%; p < 0.001), 62.0% (95% CI: 52.0-71.5%; I ² = 77%; p < 0.001), and 54.0% (95% CI: 40.0-67.5%; I ² = 88%; p < 0.001), respectively. There was no correlation between severity of the acute phase and the persistence of alterations on CT in general. Among the CT scans acquired at post-diagnosis month 3, alterations indicative of fibrosis were observed in 22% (95% CI: 13-30%; I ² = 85%; p < 0.001), and no reduction in that prevalence was observed at the subsequent time points (rho-s = 0.952; p < 0.000). The severity of the acute phase showed a positive correlation with the presence of lesions indicative of pulmonary fibrosis on CT scans acquired at 3 months after the diagnosis of COVID-19.