Corticosteroid treatment and 2019-nCoV lung injury
Russell CD, Millar JE, Baillie JK. Clinical evidence does not support corticosteroid treatment for 2019-nCoV lung injury. Lancet. 2020;395(10223):473–475. doi:10.1016/S0140-6736(20)30317-2
Shang L, Zhao J, Hu Y, Du R, Cao B. On the use of corticosteroids for 2019-nCoV pneumonia. Lancet. 2020;395(10225):683–684. doi:10.1016/S0140-6736(20)30361-5
Zhou W, Liu Y, Tian D, et al. Potential benefits of precise corticosteroids therapy for severe 2019-nCoV pneumonia. Signal Transduct Target Ther. 2020;5:18. Published 2020 Feb 21. doi:10.1038/s41392-020-0127-9
Review written by:
Clare Howard, edited by Naomi Ali
Russell et al: Yellow – The letter was published in the Lancet as a high-level review of many studies. However, the evidence is primarily observational, mostly without controls, and un-randomized.
Shang et al: Yellow – The letter was published in the Lancet and summarizes the Chinese Thoracic Society expert consensus statement. However, the evidence is primarily observational, mostly without controls, and un-randomized.
Zhou et al: Red – The letter was published in a peer-reviewed Nature Group Journal, but the comparisons due to small sample size and lack of a control group.
The literature about the use of corticosteroids in viral pneumonias primarily comes from MERS, SARS, Influenza, and RSV. The role of steroids in this setting remains highly controversial, as is illustrated by two Lancet articles, published 2 weeks apart, presenting opposing points of view: one literature review published by Russel, et al on Feb. 15, 2020 draws the conclusion that clinical evidence does not support a role for steroid use in COVID-19, while a rebuttal letter published Feb 29th (Shang, et al) disputes Russel’s assessment and presents the Chinese Thoracic Society’s expert consensus of a potential role for limited corticosteroid use.
Russel’s basis for skepticism regarding steroids in SARS-CoV2 infection (outside of a clinical trial) derives from a literature review of corticosteroid use in SARS, MERS, Influenza, and RSV.
Russel cites a retrospective observational study of 309 adults with MERS, ~50% of whom received steroids and showed no improvement in 90-day mortality, but did see increased likelihood of mechanical ventilation, vasopressors, renal replacement therapy, and delayed clearance of viral RNA from respiratory secretions.
Russel then cites an earlier review of four studies in patients with SARS highlighting harmful effects of steroids.
One case-control study (45 patient total) found an association between higher corticosteroid doses and SARS-related psychosis.
One randomized control trial (16 non-critically-ill patients) showed longer duration of viremia in patients given corticosteroids.
The remaining two studies showed moresteroid-related complications, including diabetes and avascular necrosis.
For influenza, Russel cites a meta-analysis of 10 observational studies (6548 patients) that found increased mortality, longer ICU stays, and an increased rate of secondary bacterial and fungal infection in patients given corticosteroids.
Finally, Russel cites an observational study of 50 adults hospitalized with RSV infection suggesting that those given corticosteroids had impaired antibody responses at 28 days.
Shang’s letter follows Russel’s by 2 weeks, and cautions that nearly all studies cited by Russel were observational. There may be a mortality bias as physicians use corticosteroids in patients that are already critically ill. While Russel cites four studies demonstrating evidence of possible harm from corticosteroid treatment for SARS, Shang points out that 25 other studies in the same literature review were inconclusive. Shang cites to support a possible role for corticosteroids:
A 401-patient retrospective study of SARS patients in which corticosteroids were found to reduce mortality and shorten hospital stay for critically ill patients. (Median of the mean daily doses was reported at 163 mg methylprednisolone equivalent).
A prospective cohort study of 2141 patients with H1N1 influenza, showing that low-moderate doses (25-150 mg/day methylprednisolone-equivalent) can reduce mortality in patients with oxygen index lower than 300 mm Hg.
The Chinese Thoracic Society consensus statement supports the use of low to moderate dosages (<0.5-1 mg/kg per day methylprednisolone or equivalent) and short courses (<7 days) of corticosteroids in critically ill COVID 19 patients.
One last observational study, also published by a Chinese group at the end of February, reports a favorable effect of low-moderate dose steroid use with the first 15 COVID-19 cases in the ICU in January, 2020. All patients (average age 61.7 y) had bilateral pneumonia, hypoxemia, and moderate to severe ARDS. 14 out of 15 patients had infections; eight had shock; nine had multiple organ injuries. Prior to ICU admission, all were initially treated with non-invasive O2 and antibiotics or antivirals. Upon ICU admission, they were all started on corticosteroids (400 hydrocortisone 400 mg/day) for 9.5 days and O2 saturation, WBC count, and inflammatory markers such as CRP were recorded every two days. Mortality in these 15 patients was 46.7%. Lacking a control group, authors compared mortality prevalence to the reported mortality from MERS treated without steroids (study from 2018) concluding that steroid use in their study did not confer a mortality benefit. In their analysis of oxygenation and inflammation measures, the authors do cite small improvements in O2 sat, PaO2/FiO2 fraction, and inflammatory markers over time with steroid treatment. However, with the study’s limitations, it is impossible to make conclusions about treatment efficacy from these results.