If you work in the emergency department, you’re out there in the trenches “suspecting sepsis” on a daily basis. Adhering to guideline recommendations, considered by some hand in hand with “quality,” requires prudent empiric antibiotic coverage. Vancomycin remains the typical straightforward choice for patients in whom methicillin-resistant Staphylococcus aureus coverage is indicated. However, the choice for antipseudomonal coverage is muddier; the split between cefepime and piperacillin-tazobactam is more or less equal in current practice.¹

Surely, these two differing antibiotic formulations cannot possibly be interchangeable?There must be a difference favoring one over the other? These are among the questions at the forefront in the exciting world of infectious disease. The opening volley in the debate came from the Antibiotic Choice On ReNal outcomes (ACORN) trial, which tested these two antibiotics against a primary outcome of acute kidney injury (AKI) or death by day 14 of treatment.² A secondary prespecified outcome included measures of days alive and free of delirium and coma, a nod to the observed increase in neurotoxicity suspected with cefepime.

Nearly 95 percent of the 2,511 patients included in this trial were enrolled in the emergency department, with the most common sources of infection suspected to be intra-abdominal or pulmonary. Most patients received the antibiotic as per randomization, though the number of patients receiving the study antibiotic dropped steadily through the first three days. By study day four, fewer than 40 percent of the original cohorts were still hospitalized and receiving the antibiotic exposures of interest.

Grossly speaking, the results were a wash. Small differences, not reaching statistical significance, favor piperacillin-tazobactam, including renal outcomes, neurotoxicity, and death. It is probably not reliable to focus on these small differences in renal outcomes and mortality, as there are differences in the baseline characteristics of patients enrolled that potentially contributed to the skew. Most prominently, there was a two percent absolute excess of intensive care unit patients and mechanically ventilated patients randomized into the cefepime. Without further parsing through baseline characteristics, this is enough of an indicator that the non-significant excess mortality should not be emphasized. By any stretch, however, there are no clear signals of harm associated with piperacillin-tazobactam.

Unfortunately, the suspected association of cefepime with neurotoxicity was likely confirmed. While the measure of days “alive and free of delirium and coma” was skewed by the increased mortality seen in the cefepime cohort, further analyses still demonstrate an excess of delirium and coma. These various post-hoc analyses required accounting for receipt of sedation and varying alternative definitions of delirium and coma, but do little to refute the prevailing notion of increased neurotoxicity relating to cefepime.

Upending these results entirely comes a subsequent study whose conclusions are completely different.³ In this alternative study, the authors took advantage of a sort of “natural experiment,” a 15-month national piperacillin-tazobactam shortage. Preceding and following the period of shortage, piperacillin-tazobactam was the clear primary option at the authors’ institution. During the shortage period, the prevailing anti-pseudomonal antibiotic coverage for their institution switched, by necessity, to cefepime. The authors make an argument that the shortage exists as a random exogenous factor creating the conditions necessary to interpret their results as a sort of emulated clinical trial.

The results of this emulated clinical trial are starkly different. In their study, comprising over 7,500 patients, these authors demonstrated an opposite, and larger, absolute effect on mortality. Rather than the non-significant difference observed at 14 days in ACORN, here the mortality outcome at 90 days demonstrated a 5.0 percent absolute advantage to cefepime. The authors ultimately conclude, when not otherwise indicated, the use of piperacillin-tazobactam results in potentially “thousands of additional fatalities each year.”

Any conclusion providing evidence of “thousands of additional fatalities” ought give any clinician pause. Clearly, it is always prudent to narrow the spectrum of antibiotic coverage as clinically appropriate. However, each of these antibiotics has historically been used interchangeably for empiric coverage of gram-negative organisms.

There are, of course, concerns regarding the magnitude of this observed effect size. Firstly, and most prominently, this remains a retrospective analysis prone to the effects of uncontrolled biases and confounding. Even a so-called “natural experiment” such as this remains limited in its ability to isolate a causal effect. Of note, despite the proposed disadvantage of piperacillin-tazobactam, crude mortality was unchanged, rather than mproved, during its absence. Most of the observed mortality advantage favoring cefepime emerged following statistical adjustments.

In particular, one specific adjustment, for metronidazole exposure, is suspected to have been problematic. Patients who received metronidazole as extended anaerobic coverage were generally more unwell, leading to commentary suggesting the observations from this analysis are subject to collider bias.⁴ Collider bias is a type of confounding in which an unobserved feature is associated with both an outcome and a covariate used for adjustment. In this case, the concern voiced is high-risk clinical factors independently influence both mortality and metronidazole use, and an adjustment solely for metronidazole neglects this “collider.” The net effect of this adjustment generates an over-correction, after a fashion, bestowing an exaggerated advantage upon cefepime.

All controversy aside, however, these observations favoring cefepime must be respected because they are consistent with other observational data. The unavoidable fact remains that these data have the fundamental limitations of all retroactive or observational studies. The only true method to settle the question remains a prospective, randomized trial designed specifically to address medium-term mortality, rather than short-term renal outcomes. Considering the wide usage of these antibiotics in the current approach to sepsis, an urgent answer ought be sought.

Dr. Radecki (@emlitofnote) is an emergency physician and informatician with Christchurch Hospital in Christchurch, New Zealand. He is the Annals of Emergency Medicine podcast co-host and Journal Club editor.

References

1. Rhee C, Chen T, Kadri SS, et al. Trends in empiric broad-spectrum antibiotic use for suspected community-onset sepsis in US hospitals. JAMA Netw Open. 2024;7(6):e2418923.
2. Qian ET, Casey JD, Wright A, et al. Cefepime vs piperacillin-tazobactam in adults hospitalized with acute infection: the acorn randomized clinical trial. JAMA. 2023;330(16):1557-1567.
3. Chanderraj R, Admon AJ, He Y, et al. Mortality of patients with sepsis administered piperacillin-tazobactam vs cefepime. JAMA Intern Med. 2024;184(7):769-777.
4. Hamilton F, Lee TC, Butler-Laporte G. Instrumental variable analysis: choice of control variables is critical and can lead to biased results. medRxiv 2024.07.11.24310262 [Preprint]. July 11, 2024 [cited Oct. 29,2024].