New sepsis trial results reveal how much we have to learn.
A massive study looked at sepsis-induced low blood pressure and found that two very different approaches had similar outcomes.
A few years ago, a research assistant approached me to ask whether my patient in Room 19 would be appropriate to include in a sepsis study that was being conducted. If I said yes, the patient would be randomized either to receive a lot of intravenous fluids to raise their blood pressure or medications called vasopressors to achieve that goal.
“Sure,” I said. The patient met the criteria to be included in the study. Plus, I thought it was an important idea. Patients with sepsis—a condition in which the immune system’s fight against an infection causes organ damage and even death—causing low blood pressure need help in raising it to safer values. But the best way to accomplish that was a “known unknown.”
I would have accepted any outcome of the study. But I might have had a slight hope that the vasopressor group might fare as well as (if not better than) the fluids-first group. My reasons for this were two-fold. The first was patient-centered. Vasopressors work quickly. If we could show that skipping the administration of liters and liters of IV fluids over hours was safe, their condition might improve sooner (also, for some patients, high volumes of fluids can be unsafe). The second was selfish—a doctor-centered concern. If anything was shown to be as good or better than the IV fluid approach, then maybe the HHS federal regulation on sepsis care (called SEP-1) would go away, ideally replaced by another protocol that has a better chance to help patients while being less burdensome on patients and clinicians alike.
It turned out that this sepsis trial, called the CLOVERS trial, had gotten into the crosshairs of a consumer advocacy organization. That group was convinced that this trial was not ethical. It lobbied the NIH to halt the trial. That made some news (though I tend to agree with those who say the trial probably should not have been targeted).
The central argument the consumer advocates advanced was the protocol meant that neither group was getting “usual care” (or the “standard of care”). In their view, that meant that this study could not show whether fluids-first or early vasopressors really helped patients, because the results were not being compared to what doctors would otherwise have done, were the study not being conducted.
The NIH looked into it. (In fact, some researchers from the NIH were the ones who pulled the alarm on this.) The end result was that the trial continued, with a couple of alterations.
This month, the results were published in The New England Journal of Medicine. The upshot is that neither approach was any worse or better. The mortality rate in the fluids-first group was 14.9% versus 14% in the early vasopressors group (For clinicians wanting the nitty-gritty, Dr. Lauren Westafer and I covered this in our podcast/website, which can be found here).
This is yet another blow to the HHS sepsis regulation (SEP-1), which has to be one of the least popular and least effective efforts in the history of the government-driven healthcare and hospital quality measurement.
But to me, the more important lesson is that while well-conducted trials (which this study certainly was) are crucial, government quality metrics clearly do not always define “usual care” or the “standard of care.” They just reflect who happened to win a scientific political battle. For example, the “right” amount of IV fluids for patients with sepsis is simply unknown, and may not be monolithic. (Precision medicine, anyone?) For NIH scientists to mistake what was written in the federal government’s sepsis regulation as anything resembling the evidence-based standard of care tells us more about how little agreement there is on the topic of sepsis care than anything roguish about this trial’s design.
In any case, the patient safety advocates pulled the wrong alarm. Instead of fretting that a research protocol (that was approved by dozens of research ethics boards in the participating hospitals all over the country) was off-base, they should have instead spent their energy opposing the federal government’s strange unearned certainty about treating sepsis-induced low blood pressure in the first place.
Old saying, attributed to Mark Twain: “It ain’t what you don’t know that gets you into trouble. It’s what you know for sure that just ain’t so.”
We know precious little about managing low pressure in sepsis. We need to continue to study this problem accordingly. Pretending we know more than we do is the problem, not the solution.
New sepsis trial results reveal how much we have to learn.
Thank you for helping me understand more about sepsis and the resulting low blood pressure it causes. At this very moment my brother is hospitalized with pneumonia and sepsis. The doctors are trying to raise his blood pressure with the IV fluids.
Thank you Dr Faust for the update on SEPSIS treatment research. My family uses standard Medical Power of Attorney ( POA's ) which are updated as needed & on file with family physicians under the relevant patient identifier.