Corticosteroid dose escalation in non-ICU COVID-19 patients with worsening lung lesions reduces lesion severity without improving clinical outcomes.

The effect of increasing corticosteroid doses on clinical outcomes and chest findings in patients with coronavirus disease (COVID-19) pneumonia and lung disease remains unknown. We aimed to investigate the effects of increasing steroid dosage on chest lesion area and clinical outcomes in patients with moderate or severe COVID-19 and progressive lung involvement on chest computed tomography (CT). A total of 105 patients with radiological progression during methylprednisolone (MP) therapy either received an increased MP dose (n = 79) or were maintained on the same MP dose (n = 26). These patients were divided into dose-increment and no-change groups according to the MP dose adjustment strategy. Clinical features, changes in CT severity scores within 7 days after steroid adjustment, and outcomes were compared between the groups. Six (7.6%) and one (3.8%) patients in the dose-increment and no-change groups, respectively, had increasing World Health Organization outcome scores 96 h after MP adjustment (P = 0.678). Length of stay [15 days (IQR: 10-24) vs. 14 days (IQR: 10-25); P = 0.994] and in-hospital death rate (7.6% vs. 3.8%; P = 0.678) showed no significant differences between the groups. Logistic regression analyses revealed that an increased MP dose was significantly associated with improvement in CT lesion area compared with no change in MP dose, but the CT lesions deteriorated subsequently (79.7% vs. 53.8%, P = 0.044). In conclusion, increasing the MP dose in patients with worsening CT findings ameliorates CT lesions but fails to prevent serious adverse outcomes.