ACORN

Published

Qian E, et al. "Cefepime vs. piperacillin-tazobactam in adults hospitalized with acute infection". JAMA. 2023. 330(16):1557-1567.
PubMed • Full text

Clinical Question

In adults presenting to hospital for an acute infection requiring anti-pseudomonas therapy, does piperacillin-tazobactam as compared to cefepime increase the risk of acute kidney injury or mortality?

Bottom Line

Among acutely ill patients requiring antipseudomonal antibiotics, piperazillin-tazobactam and cefepime yielded similar rates of kidney injury or death, while neurologic dysfunction was more common with cefepime.

Major Points

Cefepime and piperacillin-tazobactam are two common antipseudomonal antibiotics used for empiric coverage in hospitalized adults with suspected infection. However, observational studies suggested that piperacillin-tazobactam may increase the risk of acute kidney injury.

The ACORN trial, enrolling over 2,500 patients, compared piperacillin-tazobactam (n=1,297) and cefepime (n=1,214) for empiric antibiotic coverage in acutely ill adults at risk of acute kidney injury or death. The primary outcome was a measure of acute kidney injury or death at 14 days. The measure of acute kidney injury or death was not significantly different between groups. Moreover, the incidence of major renal adverse events was similar between groups (8-10%). There was more neurologic dysfunction in patients receiving cefepime.

The authors point out several limitations of the study, including its limited generalizability because of single-center design. There were also concerns that the study's design violated the proportional odds assumption. The study was also notable for its design which relied on best practice alerts embedded within the electronic health record, and for its use of the EHR for data collection. There was significant crossover during the study, with close to 20% of patients receiving a dose of the alternate antibiotic and 77% of patients were also exposed to vancomycin. Overall, ACORN suggests that rates of acute kidney injury are similar between cefepime and piperacillin-tazobactam. Overall, an individual's risk for acute kidney injury and neurotoxicity should be considered when choosing between agents.

Guidelines

IDSA 2023 Guidance on the Treatment of Antimicrobial Resistant Gram-Negative Infections, adapted

Question 4.1: What are preferred antibiotics for the treatment of infections caused by MDR P. aeruginosa?

• If P. aeruginosa isolates are susceptible to both, choose non-carbapenems over carbapenems.
• If P. aeruginosa isolates not susceptible to carbapenems but susceptible to traditional beta-lactams, suggest adminstering as high-dose extended-infusion
• For critically ill or poor source control with P. aeruginosa isolates resistant to carbapenems but susceptible to traditional β-lactams, use of novel beta-lactam agent that tests susceptible is suggested

Design

• Single center, open label, parallel-group, randomized, pragmatic trial
• N=2,511
  • Piperacillin-tazobactam (n=1,297)
  • Cefepime (n=1,214)
• Setting: Single US center
• Enrollment: November 2021 to October 2022
• Mean follow-up: 28 days
• Analysis: Intention-to-treat
• Primary outcome: Acute kidney injury or death by day 14

Population

Inclusion Criteria

• Age ≥18 years
• ED or medical ICU setting
• Antibiotics ordered within 12 hours of presentation
• Antibiotic order placed for antipseudomonal cephalosporin or penicillin

Exclusion Criteria

• Allergy to cephalosporins or penicillins
• Receipt of one dose of an antipseudomonal cephalosporin or penicillin within prior 7 days (other antipseudomonal agents acceptable)
• Incarcerated
• If the treating clinician determined that 1 of the 2 study drugs represented a better treatment option for that patient

Baseline Characteristics

Cefepime group displayed

• Demographics: median age 57 years, 43% female, 5% hispanic, 16% black, 77% white, 2% other race
• Facility: median hours from presentation to enrollment 1.3 hours, 93.5% enrolled in ED, 6.5% enrolled in ICU
• Physiologic parameters: median SOFA 2, 9% mechanical ventilation, median charlson comobidity index 4,
• Kidney disease: 20% chronic kidney disease, 6.4% previous renal replacement therapy, at enrollment 51% no kidney disease, 19% stage 1, 11% Stage 2, 12% stage 3
• Labs: lowest median creatinine in 12 months prior to enrollment 0.7 mg/dL, and 1.0 mg/dL at enrollment
• Suspected site of infection: 26% intra-abdominal, 21% lung, 17% skin/soft-tissue, 8% gentitourinary, 8% other, 19% unknown
• median Richmond Agitation-Sedation Scale score 0, 7% coma, 5% delirium

Interventions

• Cefepime was administered as an intravenous push over 5 minutes
• Piperacillin-tazobactam was administered as a bolus over 30 minutes initially followed by an extended infusion over 4 hours for subsequent doses
• Randomization was performed via a best-practice alert embedded into the electronic medical record
• Duration determined by treating clinician

Outcomes

Comparisons are cefepime vs. piperacillin-tazobactam.

Primary Outcomes

Acute kidney injury or death by day 14

25.0% vs. 26.6% (HR 0.95; 95% CI 0.80-1.13; P=0.56)

Secondary Outcomes

Delirium/coma-free days within 14 days

Mean 11.9 vs 12.2 (OR 0.79; 95% CI 0.65-0.95)

Death

7.6% vs. 6.0% (RD 1.6; 95% CI -0.5 to 3.6; P=0.04)

Final creatine level ≥ 2 times the baseline level

1.2% vs 2.1% (RD -0.9; 95% CI -2.0 to 0.2)

Criticisms

• Single center
• Unblinded
• Given that death and AKI trended in opposite directions, the proportional odds ratio assumption is likely violated. This makes the meaning of the primary outcome unclear.
• Clinicians allowed to change, stop or de-escalate antimicrobials
• Patients received median 3 days of therapy so events occurring later in the 14 day follow-up period are more challenging to connect to the antimicrobials
• Most (77%) of total population were exposed to vancomycin in addition to study drug
• Significant cross over between groups

Funding

• NIH grant