Acetaminophen Fails to Improve Mortality, Need for Organ Support in Sepsis

—
However, treated patients were less likely to develop ARDS, respiratory dysfunction

by
Katherine Kahn
,
Staff Writer, MedPage Today

May 20, 2024

SAN DIEGO — Intravenous acetaminophen for critically ill patients with sepsis did not improve mortality or reduce the need for organ support, according to the phase IIb ASTER trial.

In the randomized study, patients treated with acetaminophen had a mean 20.2 days alive and free of organ support, statistically the same as the mean 19.6 days observed among those who received a placebo (P=0.56), Michael Matthay, MD, of the University of California San Francisco, reported at the American Thoracic Society annual meeting.

“For clinicians, I would not change therapy for patients with sepsis at this time,” Matthay told attendees. “There’s not enough evidence here for us to change therapy.”

Results from the ASTER trial were simultaneously published in JAMA.

All-cause mortality at 28 days was 18% for patients receiving acetaminophen versus 23% in the placebo group (P=0.19). At 90 days, all-cause mortality was 26% in the acetaminophen group and 32% in the placebo group (P=0.19). And, at 28 days, there were no significant differences in days free of assisted ventilation, new renal replacement therapy, or vasopressors.

“Unfortunately, over the past several decades, virtually all drug trials in sepsis have been negative, so I’m never surprised by another negative one,” Robert Dickson, MD, of the University of Michigan Medical School in Ann Arbor, told MedPage Today. “The investigators had a promising hypothesis, and the study was well-executed, but unfortunately it just didn’t translate into the differences in their primary outcomes.” Dickson was not associated with the study.

On a more positive note, in a safety analysis of the study, no differences were found in markers of hepatic injury, systemic hypotension, or other adverse events between the treatment arm and the placebo group.

“The study further confirms the safety of acetaminophen in patients with sepsis, so we should feel comfortable using it to help with patients’ fever and pain,” Dickson said. “But we can’t say that it keeps people alive and gets them off of life support any faster than placebo.”

Several observational and small phase IIb studies have suggested a mortality benefit of acetaminophen in critically ill patients, perhaps due to its effects of blocking the damaging effects of cell-free hemoglobin, according to the researchers.

Some of the trial’s secondary outcomes did reach statistical significance. For example, researchers found that only 2.2% of patients in the acetaminophen arm developed acute respiratory distress syndrome (ARDS) compared with 8.5% in the placebo arm (P=0.01). Also, patients receiving acetaminophen had significantly greater improvements in the overall daily Sequential Organ Failure Assessment (SOFA) scores on days 2 (P<0.05) through 4 (P<0.01), as well as significant improvements in the SOFA respiratory scores on days 1 through 4.

Researchers also conducted several additional exploratory analyses to see whether patient responses varied by plasma cell-free hemoglobin levels, to see if it could serve as a potential biomarker to identify patients who might benefit from acetaminophen. However, among the patients with cell-free hemoglobin levels >10 mg/dL (a little less than half of the patients), there was also no significant difference in 28-day mortality for those treated with acetaminophen (12%) versus placebo (21%, P=0.10).

“There are actually some really provocative mortality differences across the treatment groups, especially among patients who had elevated cell-free hemoglobin levels,” Dickson said, commenting on this finding. “But unfortunately the study wasn’t large enough to rule out the possibility that this difference was due to chance. I’d love to see a larger study exclusively studying septic patients with elevated cell-free hemoglobin.”

Of note, among patients with a baseline cell-free hemoglobin higher than 10 mg/dL, just 8% required assisted ventilation versus 23% of those who received placebo (P=0.02 for interaction between study treatment and hemoglobin level).

“One problem in critical care is the patients get sick so fast, that we do not normally have time to figure out which biomarkers help predict which therapy could give the best outcome,” Matthay commented in a press release. “We hope that these findings will underscore the potential therapeutic value of using a biomarker to help successfully find a treatment that will work when patients need it the most.”

Patients were eligible for the ASTER trial study if they had hypotension with a need for vasopressors or rehydration, or assisted ventilation or supplemental oxygen. Enrolled patients had to have been admitted to the ICU within 36 hours of presentation to the emergency department or acute care.

The trial enrolled 447 adult patients with a mean age of 64 years. About three-quarters of participants were white and about half were female. Pneumonia was the most common infection, occurring in approximately 44% of patients, followed by urinary tract infections, intra-abdominal infections, and infections of unknown origin. In the 3 weeks prior to enrollment, 9% had tested positive for SARS-CoV-2. At baseline, 76% were receiving vasopressors and 42% were receiving assisted ventilation. The mean SOFA score was 5.4.

Patients were randomized to receive IV acetaminophen at a dose of 1g in 100 mL diluent or a placebo of 5% dextrose IV infusion every 6 hours for 5 days. Patients in the acetaminophen arm received about 12 doses and those in the placebo arm received about 13 doses.

In addition to the primary outcome of days alive and free from organ support and 28-day all-cause mortality, researchers also analyzed 15 secondary outcomes that included 90-day hospital mortality, 90-day all-cause mortality, changes in SOFA scores, changes in serum creatinine, Major Adverse Kidney Events at 28 days, ICU days, and the development of ARDS.