Antibiotic for Treating Sepsis Linked to Increased Mortality

— "Physicians should think carefully" about using anti-anaerobic antibiotics in sepsis

A photo of a hand holding a vial of piperacillin-tazobactam for injection.

In adults hospitalized with suspected sepsis, piperacillin-tazobactam was associated with a higher mortality rate and increased duration in organ dysfunction compared with cefepime, according to a retrospective cohort study.

In an instrumental variable analysis, empirical treatment with vancomycin and piperacillin-tazobactam resulted in a 5% absolute increase in 90-day mortality compared with empirical treatment with vancomycin and cefepime (22.5% vs 17.5%, P=0.002), Rishi Chanderraj, MD, MSc, of the Ann Arbor Veterans Affairs Hospital in Michigan, and colleagues reported.

In addition, piperacillin-tazobactam was also associated with 2.1 (95% CI 1.4-2.7) fewer organ failure-free days, 1.1 (95% CI 0.57-1.62) fewer ventilator-free days, and 1.5 (95% CI 1.01-2.01) fewer vasopressor-free days when compared with cefepime, the findings in showed.

"When it comes to early treatment in sepsis, all broad-spectrum antibiotics are not created equal," Chanderraj told app. "Unless the patient has a specific indication for anti-anaerobic antibiotics, physicians should think carefully before using them and exposing their patients to unnecessary harm."

"A 2% to 5% mortality difference may not initially sound like much, but given how common and deadly sepsis is, the routine use of anti-anaerobic antibiotics in sepsis may lead to thousands of deaths per year," he added.

More than 65% of patients with sepsis receive broad-spectrum antibiotics targeting methicillin-resistant Staphylococcus aureus and Pseudomonas aeruginosa, the authors pointed out. The combination of vancomycin and piperacillin-tazobactam is the most frequently prescribed empirical regimen for sepsis. Given the results of their study, the combination may contribute to one additional death per every 20 patients with sepsis, the authors wrote.

In a secondary analysis, the researchers looked at the association between receipt of any anti-anaerobic antibiotic and patient outcomes when compared with no receipt of anti-anaerobic antibiotics. Anti-anaerobic antibiotic treatment was associated with a 12% increase in 90-day mortality using a two-stage regression model (95% CI 3-21, P <0.001) and with a 14% increase in mortality using a non-linear two-stage residual inclusion model (95% CI 6-22, P<0.001). Treatment with anti-anaerobic antibiotics was again associated with fewer ventilator-free days, vasopressor free-days, and organ failure-free days.

"Most patients with sepsis do not have a specific indication for anti-anaerobic antibiotics, and our study suggests they could cause real harm," Chanderraj stressed. "Making the right antibiotic choice is not just a matter of preventing future antibiotic resistance -- these choices directly impact clinical outcomes for the patients in front of us."

Prior observational studies also found a mortality difference for piperacillin-tazobactam, Chanderraj noted. On the other hand, the recent randomized, controlled ACORN trial, comparing piperacillin-tazobactam with cefepime, found that piperacillin-tazobactam did not increase the incidence of death among hospitalized adults, he pointed out.

The ACORN trial, however, only looked at 14-day outcomes and did not consider other anti-anaerobic antibiotics. "It was really designed to answer a very different question than our study," he explained. In a post-hoc analysis, Chanderraj and colleagues did look at 14-day outcomes and found no significant difference in mortality between piperacillin-tazobactam and cefepime -- similar to the ACORN trial.

The primary analysis of the current study included a final sample of 7,569 adult patients who presented to the emergency department at the University of Michigan from July 2014 through December 2018 and received empirical treatment with sepsis. (Patients with repeated admissions, who did not receive vancomycin, or who did not receive either piperacillin-tazobactam or cefepime treatment were excluded.) Of these, 4,523 were treated with vancomycin and piperacillin-tazobactam and 3,046 were treated with vancomycin and cefepime. The median age was 63 and 55% were men.

Importantly, a piperacillin-tazobactam shortage occurred during part of the study period, from June 12, 2015 to Sept. 18, 2016. The shortage allowed the researchers to conduct the retrospective study similar to a randomized trial because it affected which of the two antibiotics patients received at the study site, Chanderraj explained. As a result of the shortage, nearly all patients (97%) treated with piperacillin-tazobactam were admitted outside the shortage period, "enabling a direct comparison of the association between these antibiotics and 90-day mortality in a clinical practice setting," the authors wrote.

The study had some limitations. It relied on data from electronic medical records, but the authors emphasized that it had a good balance of covariates across treatment groups. Also, data was from a single center and may not be generalizable to other clinical settings or populations.

  • author['full_name']

    Katherine Kahn is a staff writer at app, covering the infectious diseases beat. She has been a medical writer for over 15 years.


The study was funded by the National Heart, Lung, and Blood Institute, the National Institute on Aging, and the Agency for Healthcare Research and Quality.

Chanderraj reported no conflicts of interest; one study author reported receiving personal fees from Charles River Laboratories and Shionogi Pharmaceuticals.

Primary Source

JAMA Internal Medicine

Chanderraj R, et al "Mortality of patients with sepsis administered piperacillin-tazobactam vs cefepime" JAMA Intern Med 2024; DOI: 10.1001/jamainternmed.2024.0581.