Three main reasons have been put forth for using corticosteroids in critically ill patients with COVID-19, but only one – the hope of preventing lung fibrosis in patients with unresolved acute respiratory distress syndrome – is reasonable to employ now outside of formal randomized trials, Peter Pickkers, MD, PhD, asserted at a webinar on COVID-19 sponsored by the European Society of Intensive Care Medicine.
The most commonly invoked rationale for giving steroids in patients with severe COVID-19 is to modulate the destructive inflammatory immune response that occurs with advancing disease. Another justification cited for giving steroids is to treat suspected adrenal insufficiency in those with refractory shock. Of note, both practices are endorsed in the recent Surviving Sepsis Campaignon management of critically ill patients with COVID-19 ( ).
But those recommendations – numbers 22 and 42 out of a total of 50 recommendations included in the guidelines – should never have been made, according to
The Surviving Sepsis Campaign guidelines, which were developed by a panel comprising 36 experts in 12 countries, are quite frank in conceding that the guidance in favor of corticosteroids are weak recommendations based on low-quality evidence.
The guidelines recommend against using corticosteroids to try to modulate the immune system in mechanically ventilated COVID-19 patients without acute respiratory distress syndrome (ARDS), but do recommend steroids in those with COVID-19 and ARDS. However, the guidelines also note that, because of the very low quality of the evidence, some experts on the panel preferred not to issue a pro-steroids recommendation at all until higher-quality evidence becomes available. Dr. Pickkers said he believes that the minority view should have prevailed. Moreover, current COVID-19 guidance from the World Health Organization is at odds with the Surviving Sepsis Campaign recommendations; the WHO advises against corticosteroids unless the treatment is indicated for a reason other than immunomodulation, he noted.
The evidence in favor of steroids in an effort to blunt the immune response in COVID patients with ARDS is based largely upon a single small, retrospective, non–peer-reviewedthat 5-7 days of treatment with 1-2 mg/kg per day of methylprednisolone was associated with shortened fever duration and need for supplemental oxygen.
The evidence against steroids for immunomodulation comes mainly from earlier studies of the SARS and MERS novel coronaviruses. For example, in a multicenter study of 309 patients with the MERS (Middle East respiratory syndrome) virus, those who received corticosteroids received no benefit and experienced delayed viral clearance ().
“The thing is, virtually all COVID-19 patients in the ICU fulfill the criteria for ARDS, so following the Surviving Sepsis Campaign guidelines would have far-reaching consequences,” Dr. Pickkers said.
Those consequences include a theoretic potential for serious harm arising from dampening the immune response at a point in the course of COVID-19 when the virus is still present, which could result in slowed viral clearance and prolonged viral shedding. Moreover so far no one has been able to identify a sweet spot in the disease course where the viral load has waned and the immune response is sufficiently early that intervention with corticosteroids might have an optimal benefit/risk ratio, he continued.
“My opinion is that at this moment there is no benefit at all for corticosteroids for immunomodulation in patients with COVID-19,” Dr. Pickkers said. “My personal recommendation, in contrast to the Surviving Sepsis Campaign recommendation, is not to use this therapy outside of a study.”
He added that randomized, controlled trials of corticosteroid therapy in critically ill patients with COVID-19 are ongoing in Europe and the United States, including the large RECOVERY study of dexamethasone in the United Kingdom.
As for the Surviving Sepsis Campaign recommendation to use corticosteroids to treat refractory shock in COVID-19, Dr. Pickkers dismissed this guidance as largely irrelevant. That’s because few patients with COVID-19 who need mechanical ventilation have refractory shock as evidenced by the need for a high infusion rate of norepinephrine. Anyway, he noted, that Surviving Sepsis recommendation is based upon extrapolation from evidence of benefit in bacterial septic shock patients, which he deemed to be of questionable relevance to the COVID-19 pandemic.
Attacking the fibroproliferative phase of ARDS
First off, Dr. Pickkers conceded, there is no evidence that treatment with corticosteroids to prevent lung fibrosis in COVID-19 patients with nonresolving ARDS is an effective strategy; the pandemic is simply too new at this point for the appropriate studies to have been done. But this much is known: Postmortem pathologic studies show fibroplastic proliferation is present in the lungs of COVID-19 patients, as in those who die of ARDS of other causes. Also, COVID-19 patients typically aren’t admitted to the ICU until day 11 or 12 after developing their first symptoms, so by the time they display indications that their ARDS is not resolving, the virus has typically left the scene; thus, there is little risk at that point that corticosteroids will promote viral proliferation. Additionally, studies in critically ill patients with nonresolving ARDS of other causes show clinically meaningful benefits for corticosteroid therapy.
Dr. Pickkers cited as “must reading” an analysis of five randomized trials of corticosteroid therapy in a total of 518 patients with acute lung injury ARDS of non–COVID-19 origin. The analysis by investigators at the University of Tennessee, Memphis, concluded that treatment resulted in clinically meaningful reductions in duration of mechanical ventilation and ICU length of stay. Moreover, in the 400 patients whose steroid therapy commenced before day 14 of ARDS, there was a statistically significant 22% reduction in risk of death, compared with patients in whom corticosteroids were started later ().
Session cochair, struck a cautious note, remarking, “It’s my perception that we’re using the evidence that we’ve gathered in other conditions and are now trying to apply it in COVID-19. But quality data on patients with COVID-19 itself is pretty scare, and it’s really hard to say whether this disease behaves similarly to bacterial septic shock or to other viral infections.”
“We need more information about the use of corticosteroids in COVID-19, although I think a lot of people are using it at this moment,” added Dr. De Waele, a surgical intensivist at Ghent (Belgium) University.
That being said, he asked Dr. Pickkers when he considers using corticosteroids to prevent pulmonary fibrosis.
Dr. Pickkers said that when he notices that a COVID-19 patient’s lung compliance is worsening, that stiff lung is a clue that fibrosis is occurring and is having clinical consequences. “We also measure blood procollagen, a not very sensitive but moderately specific marker of fibroproliferation. If we see an increase in this biomarker and the lung mechanics are changing, then we do treat these patients with corticosteroids,” Dr. Pickkers replied.
He and his colleagues try to start steroids before day 14 of ARDS, and they continue treatment for longer than 7 days in order to prevent a rebound inflammatory response upon treatment discontinuation. They also avoid using neuromuscular agents and engage in meticulous infection surveillance in order to minimize potential complications of corticosteroid therapy in the ICU.
Dr. Pickkers reported having no financial conflicts regarding his presentation.