Background Organizing pneumonia (OP) is an interstitial lung disease linked to infections and other conditions. Cryptogenic OP (COP) is a rarer, idiopathic form. OP has become one of the most frequent respiratory complications following SARS-CoV-2 pneumonia. Standard treatment involves the use of glucocorticoids. The objective of the study is to determine the optimal dosage of oral corticosteroids for the treatment of this disease. Methods Randomized, open-label, assessor-blinded, parallel-group, single-center, non-inferiority clinical trial with an active control group comparing two oral prednisone regimens. The control group received prednisone 0.75 mg·Kg −1 ·day −1 for 4 weeks; 0.5 mg·Kg −1 ·day −1 4 weeks; 20 mg·day −1 4 weeks; 10 mg·day −1 6 weeks; 5 mg·day −1 6 weeks (6 months) and the experimental group prednisone 0.5 mg·Kg −1 ·d −1 3 weeks, 20 mg·day −1 3 weeks; 15 mg·day −1 2 weeks; 10 mg·day −1 2 weeks, 5 mg·day −1 2 weeks (3 months). Visits were carried out at baseline and at 3, 6 and 12 months, recording clinical history, physical examination, pulmonary function tests and chest CT. The primary outcome was non-inferiority with a margin of 10% in DLCO at 6 months. Results Seventy-nine patients were randomized according to intention to treat, 40 to the experimental group and 39 to the control group. Mean baseline DLCO was 59.3% ( sd 16.7) in the experimental group and 55.5% ( sd 16.5) in the control group, with mean improvements at 6 months of 11.7% (95%CI 7.18–16.17) and 12.8% (95%CI 8.57–17.11) respectively. The difference between groups at 6 months was 1.23% (95%CI −5.13–7.59, the experimental regimen being non-inferior to the control group with the margin of 10%. Patients in the experimental group had fewer than half as many adverse effects as controls: 9 (23%) versus 22 (56%) (p=0.0027). Conclusions Treatment with oral corticosteroids lasting three months (initial dose of prednisone 0.5 mg·Kg −1 ·day −1 ) may be recommended in patients with post-COVID OP.
Espejo-Castellanos et al. (Thu,) studied this question.