The basis for the protocol
The protocol is based on common sense, an understanding of scientific literature, and an understanding of COVID-19, Marik says. The website includes links to past research trials and observational studies examining ascorbic acid and thiamine in critically ill patients and early looks at anticoagulants in COVID-19 patients.
They chose methylprednisolone as their corticosteroid based on the expertise of group member G. Umberto Meduri, MD, professor of medicine at the University of Tennessee Health Science Center in Memphis, Tennessee, who had found the steroid effective in treating acute respiratory distress syndrome. On the MATH+ website, the physicians link to multiple observational studies posted on preprint servers in April and May that suggest methylprednisolone helped COVID-19 patients.
“What’s happened with time is all the elements have been validated by scientific studies, which makes this so cool,” says Marik. The RECOVERY Trial results in particular validated the push to use corticosteroids in COVID-19 patients, he says. But that study used a different steroid, dexamethasone, in much smaller doses than what MATH+ recommends. Revised guidance from the Infectious Diseases Society of America recommends dexamethasone for severely ill patients, but says methylprednisolone and prednisone can be used as substitutes at equivalent doses.
Marik and Kory say that mortality rates for COVID-19 patients at their respective hospitals decreased after they began using the protocol. The physicians have been collecting observational data on their patients, but have not yet published any, and do not plan to conduct a randomized trial.
Several physicians who were not involved in the creation of the protocol say the evidence the physicians cite is not robust enough to warrant the promotion of MATH+ and call for randomized controlled trials. Coming up with a protocol is fine, says Kaul, but “you have to do the hard work of doing a randomized control trial to determine if those drugs given in those combinations work or not.”
“When I looked at it, I thought it was actually not very evidence based,” says Michelle Gong, MD, chief of the Division of Critical Care Medicine at Montefiore Health System in New York City. “It is not something I would recommend for my doctors to do outside of a clinical trial.”
The protocol authors push back against the necessity and feasibility of randomized control trials.
There is no time for a randomized control trial right now, says Jose Iglesias, DO, associate professor at Hackensack Meridian School of Medicine at Seton Hall and critical care specialist at Community Medical Center and Jersey Shore University Medical Center in New Jersey. “Time is limited. We’re busy bedside clinicians taking care of patients, and patients who are dying.”
Marik argues there is not equipoise: It wouldn’t be ethical to randomize patients in a placebo group when the physicians are confident the steroids will help. And the protocol is personalized for each patient, making the standardization required for a randomized control trial incredibly difficult, he says. He also cites “the people who are unwilling to accept our results and just think it’s too good to be true.”
Hugh Cassiere, MD, director of critical care medicine at Northwell Health’s North Shore University Hospital in Manhasset, New York, said he finds it “very disturbing that this is being propagated.” In the context of a pandemic in which physicians from other specialties are helping out colleagues in ICUs and might follow the protocol uncritically, he worries, “this could potentially lead to harm.”
“I understand the intention; everybody wants to do something, these patients are so sick and the crisis so sharp that we all want to do something to make patients better,” Gong said. “But as physicians taking care of patients we need to make sure we separate the noise from the evidence.”