Skip to main content

Corticosteroid treatment of patients with coronavirus disease 2019 (COVID-19)

Our take —

Unclear inclusion/exclusion criteria and non-randomized retrospective design limit inferences about whether corticosteroids are beneficial or harmful in SARS-CoV-2 infection in this small observational study. Future randomized control trials are needed.

Study design

Retrospective cohort

Study population and setting

The study took place at two tertiary hospitals in Wuhu, China, and enrolled 31 patients with COVID-19 who were hospitalized between January 24 and February 24, 2020. All participants had laboratory-confirmed SARS-CoV-2 infection. The primary outcome was time to virologic clearance, defined as two negative nasopharygeal swabs 24 hours apart. Secondary outcomes were time to recovery and length of hospital stay. The length of follow-up time was not specified.

Summary of Main Findings

All patients had mild symptoms upon presentation, and none developed acute respiratory distress syndrome; 29/31 had CT scan findings of pneumonia. All patients received lopinavir-ritonavir and inhaled interferon alpha, n=5 received umenifovir, and n=14 received moxifloxacin. Eleven of 31 patients received methylprednisolone within 24 hours of presentation. People who received steroids had higher temperatures, higher C-reactive protein, lower lymphocyte counts, and were more likely to have bilateral findings on CT scan upon presentation than those who did not receive steroids. At study end, 26/31 were discharged and 5/31 were still hospitalized. No differences in virologic or clinical outcomes were found between those who received corticosteroids and those who did not.

Study Strengths

This is one of the first studies to consider methylprednisolone use in SARS-CoV-2 infection.


This is a small study of 31 patients, and it is unclear how they selected the patients included in this study. For example, while all patients had mild symptoms upon presentation, it is unknown whether this was deliberate, because inclusion and exclusion criteria are not clearly specified. The frequency of methylprednisolone varied between 40 mg by mouth 4 times/day or 2 times/day, and no breakdown was given of how many received which dosing. The inclusion of only patients with mild disease limits our ability to generalize these results to people with severe or critical disease who clinicians would be more likely to consider treating with corticosteroids.

Value added

Lays the groundwork for future randomized controlled trials to study the role of corticosteroids in SARS-CoV-2 infection.