Sepsis is the LaBrea Tar Pit of medicine. Many investigators go in; few come out, and those few are covered in ooze.
As one review [1] notes: “More than 100 randomized clinical trials have tested the hypothesis that modulating the septic response to infection can improve survival. With one short-lived exception, none of these has resulted in new treatments.”
This story in Stat News outlines the objections of the watchdog group Public Citizen to the CLOVERS trial, claiming that it is “akin to an experiment that would be conducted on laboratory animals”.
Sepsis is the over-response of the immune system to infection. It is a leading cause of death worldwide. The CDC estimates 1.5M new cases per year in the US, with 250K deaths [2] . That puts it right behind cancer and heart disease as the leading causes of death – yet it often doesn’t show up in lists because it is usually a complication of an existing disease.
CMS (the agency that runs Medicare and Medicaid) has instituted a “bundle” of sepsis interventions, and includes adherence to this bundle as one of its measures of hospital care quality. Since payment rates to hospitals are based on quality scores, failure to comply has significant financial downside and hospitals are falling in line, albeit slowly.
The objections of Public Citizen to the CLOVERS sepsis trial are that they deviate from standard practice guidelines – essentially, the bundle – and thus compromise patient care. That would be a damning accusation if the bundle were proven to be effective.
But it is not. The only element of the sepsis care bundle that is clearly critical is the prompt administration of appropriate antibiotics [3] [4] . The CLOVERS trial does not alter antibiotic administration. Rather it varies the timing and amount of fluid administration and anti-hypotensive therapy. A recent meta-analysis of 20 studies of the hemodynamic interventions in the sepsis bundle shows no evidence that they are effective [5] . Thus it is not clear that changing these interventions is risky.
More to the point, there is little evidence that the current approach is reducing sepsis mortality rates:
From Incidence and Trends of Sepsis in US Hospitals, 2009-2014
It wouldn’t be fair to say that what we are doing in sepsis isn’t working. Doing nothing would undoubtedly be much worse. But there is a great deal of room for improvement. The basic premise of Public Citizen’s objections are that patients are being denied standard care that is known to be effective. This objection is understandable – after all, why would it be standard if it were not effective? – but false.
Guidelines, particularly for complex diseases like sepsis, are not immutable truth. Instead, they are often little more than best guesses by the leading experts in the field. They are almost always helpful but are far from infallible, and are commonly revised as knowledge advances. Running trials is a key way to advance knowledge and improve treatment.
Given that the standard of care results in mortality rates of 20% or higher, there is good reason to look for improvements. The trial protocol has been reviewed by Institutional Review Boards whose job it is to make sure that ethical guidelines are followed. I see no reason to believe that this trial – although it undoubtedly contains an element of risk – fails to meet our current standards of ethical trial design.
UPDATE 9/25/18: The NYT has now covered this story. It’s a perfect example of “he said, she said” journalism – superficially balanced, yet fails to ask the critical question of whether the standard of care has been proven effective. In other words, they covered the controversy but missed the real story. The result is an article that contains no errors and yet is uninformative if not downright misleading. Another sci journo fail.
Disclosure: I consult for MBio Diagnostics, which is developing a rapid test for sepsis patient stratification, but is not involved in this trial.
Footnotes
[1] Why have clinical trials in sepsis failed?
[3] Physician Variation in Time to Antimicrobial Treatment for Septic Patients Presenting to the Emergency Department.
[4] The Timing of Early Antibiotics and Hospital Mortality in Sepsis.
1 thought on “Is the CLOVERS sepsis trial unethical?”