Thin-section CT in patients with severe acute respiratory syndrome following hospital discharge: Preliminary experience

Abstract

PURPOSE: To report the initial experience regarding thin-section computed tomographic (CT) findings in patients with severe acute respiratory syndrome (SARS) who improved clinically after treatment. MATERIALS AND METHODS: Twenty-four patients (10 men, 14 women; mean age, 39 years; age range, 23-70 years) with confirmed SARS underwent follow-up thin-section CT of the thorax. The scans were obtained on average 36.5 days after hospital admission and were analyzed for parenchymal abnormality (ground-glass opacification, consolidation, or interstitial thickening) and evidence of fibrosis (parenchymal band, traction bronchiectasis, irregular interfaces). Patients were assigned to group 1 (with CT evidence of fibrosis) and group 2 (without CT evidence of fibrosis) for analysis. Patient demographics, length of hospital stay, rate of intensive care unit admission, peak lactate dehydrogenase level, pulsed intravenous methylprednisolone therapy, and peak opacification on chest radiographs were compared between the two groups. RESULTS: Parenchymal abnormality was found in 96% (23 of 24) of patients and ranged from residual ground-glass opacification and interstitial thickening in group 2 (nine of 24, 38%) to fibrosis in group 1 (15 of 24, 62%). Patients in group 1 were older (mean age, 45 vs 30.3 years), had a higher rate of intensive care unit admission (27% [four of 15] vs 11% [one of nine]), more requirement for pulsed intravenous methylprednisolone (87%, [13 of 15] vs 67% [six of nine]), higher peak lactate dehydrogenase level (438.9 vs 355.6 U/L), and higher peak opacification on chest radiographs (estimated area, 14% vs 11%) than patients in group 2. CONCLUSION: Pulmonary fibrosis may develop early in patients with SARS who have been discharged after treatment. Patients who are older and have more severe disease during treatment are more likely to develop thin-section CT findings of fibrosis. © RSNA, 2003.