JAMA SEPTEMBER 11, 2018

Effect of Piperacillin-Tazobactam vs Meropenem on 30-Day Mortality for Patients With E coli or Klebsiella pneumoniae Bloodstream Infection and Ceftriaxone Resistance

Harris PNA, Tambyah PA, Lye DC, et al. (MERINO Trial Investigators)

Bottom Line

The MERINO trial, a multicenter randomized clinical trial, failed to demonstrate the non-inferiority of piperacillin-tazobactam compared to meropenem for the treatment of ceftriaxone-resistant bloodstream infections due to E. coli or Klebsiella pneumoniae.

Key Findings

1. The primary outcome of 30-day all-cause mortality occurred in 23 of 187 patients (12.3%) in the piperacillin-tazobactam group, compared with 7 of 191 patients (3.7%) in the meropenem group.

2. The absolute risk difference was 8.6% with a 1-sided 97.5% confidence interval of -infinity to 14.5%, which failed to meet the pre-specified non-inferiority margin of 5% (P=0.90 for non-inferiority).

3. The trial was terminated early due to evidence of potential harm and the inability to establish non-inferiority, despite no significant differences in secondary outcomes such as microbiological clearance at day 4 or rates of secondary infections.

Study Design

Design

RCT

Open-Label

Sample

379

Patients

Duration

30 days

Median

Setting

Multicenter, 9 countries

Population Adults with bloodstream infection due to ceftriaxone-nonsusceptible E. coli or K. pneumoniae that remained susceptible to piperacillin-tazobactam

Intervention Piperacillin-tazobactam 4.5 g IV every 6 hours

Comparator Meropenem 1 g IV every 8 hours

Outcome All-cause mortality at 30 days after randomization

Study Limitations

• The trial was open-label, which may introduce observer bias, although objective outcomes like mortality minimize this impact.

• The study was terminated prematurely, limiting the power to detect smaller differences in secondary outcomes or rarer adverse events.

• The results are specific to the population with ceftriaxone-resistant E. coli or K. pneumoniae and may not generalize to other ESBL-producing organisms or different patient cohorts.

• Baseline imbalances between the treatment groups (e.g., immunocompromised status in the piperacillin-tazobactam arm vs. higher APACHE II scores in the meropenem arm) could confound results.

Clinical Significance

The findings suggest that piperacillin-tazobactam is not a suitable carbapenem-sparing alternative for the definitive treatment of bloodstream infections caused by ceftriaxone-resistant E. coli or K. pneumoniae, and support the continued use of carbapenems as the standard of care for these infections.

Historical Context

Prior to the MERINO trial, the clinical community was divided on whether beta-lactam/beta-lactamase inhibitor combinations could safely replace carbapenems for ESBL-related infections to promote antibiotic stewardship. Several retrospective studies yielded conflicting results, often suggesting no difference in mortality, which necessitated this definitive, prospective randomized controlled trial to establish clinical practice guidelines.

Guided Discussion

High-yield insights from every perspective

Med Student

Medical Student

What is the biochemical mechanism of 'ceftriaxone resistance' in the E. coli and Klebsiella pneumoniae isolates studied in the MERINO trial, and why did researchers hypothesize that piperacillin-tazobactam could potentially overcome this resistance?

Key Response

The resistance is primarily mediated by Extended-Spectrum Beta-Lactamases (ESBLs), which hydrolyze third-generation cephalosporins. The hypothesis was that tazobactam, a beta-lactamase inhibitor, would bind to and inhibit the ESBL enzymes, thereby protecting the piperacillin component and allowing it to inhibit cell wall synthesis.

Resident

Based on the MERINO trial results, how should a clinician manage a patient with an E. coli bloodstream infection if the microbiology report shows resistance to ceftriaxone but susceptibility to piperacillin-tazobactam (MIC ≤ 16/4 µg/mL)?

Key Response

The MERINO trial demonstrated that piperacillin-tazobactam failed to meet non-inferiority compared to meropenem, showing a higher 30-day mortality (12.3% vs 3.7%). Therefore, clinicians should avoid piperacillin-tazobactam and use a carbapenem as definitive therapy for ESBL bloodstream infections, even if the isolate appears susceptible to piperacillin-tazobactam in vitro.

Fellow

The MERINO trial results contradicted some earlier observational data that suggested piperacillin-tazobactam was an effective carbapenem-sparing option. Discuss the 'inoculum effect' and how it may explain why piperacillin-tazobactam fails in high-burden bloodstream infections despite favorable MICs.

Key Response

The inoculum effect occurs when the antibiotic's MIC increases significantly as the bacterial density increases. In high-burden infections like bacteremia, the concentration of beta-lactamase enzymes can overwhelm the fixed concentration of the inhibitor (tazobactam), leading to the failure of the piperacillin-tazobactam combination despite standard laboratory testing performed at lower inocula.

Attending

MERINO changed the paradigm for treating ESBL-producing Enterobacterales. How does this trial challenge the traditional clinical teaching that 'susceptibility in the lab translates to efficacy at the bedside' for Gram-negative sepsis?

Key Response

MERINO provides a landmark example where phenotypic susceptibility (meeting CLSI/EUCAST breakpoints) did not prevent clinical failure. It teaches that for certain drug-pathogen-resistance combinations, the pharmacodynamics are too precarious in high-severity disease, necessitating the use of the most robust agent (carbapenems) rather than relying solely on the MIC report.

Scholarly Review

Critical appraisal through the lens of expert reviewers and guideline development

PhD

Analyze the impact of the non-inferiority margin of 5% used in the MERINO trial. If the baseline mortality in the meropenem arm had been higher than the observed 3.7%, how would this have influenced the power and the interpretation of the non-inferiority failure?

Key Response

The 5% margin is relatively tight for mortality. The unexpectedly low mortality in the meropenem arm (3.7%) actually increased the trial's ability to detect a difference. If meropenem mortality had been higher, the 12.3% mortality in the PTZ arm might still have breached the non-inferiority margin, but the absolute risk difference would have been smaller, potentially complicating the 'superiority' signals seen in the post-hoc analysis.

Journal Editor

A major critique of MERINO involves the method of susceptibility testing used at different sites. How might the use of E-tests versus automated systems or broth microdilution introduce bias or limit the internal validity of a trial comparing beta-lactam/beta-lactamase inhibitors?

Key Response

Different susceptibility testing methods can yield varying MICs for piperacillin-tazobactam against ESBL producers. E-tests have been known to underestimate MICs compared to broth microdilution (the gold standard). If the PTZ arm included patients with 'hidden' resistance due to less accurate testing methods, the failure could be attributed to testing errors rather than the drug itself, a critical point for reviewers evaluating the rigor of the enrollment criteria.

Guideline Committee

How did the MERINO trial findings specifically alter the IDSA Guidance on the Treatment of Antimicrobial-Resistant Gram-Negative Infections regarding the use of piperacillin-tazobactam for ESBL-E bloodstream infections compared to urinary tract infections?

Key Response

Following MERINO, guidelines (e.g., IDSA 2020-2023) moved to strongly recommend carbapenems as first-line for ESBL-E bloodstream infections and recommended against PTZ even if susceptible. However, for uncomplicated cystitis, PTZ remains a secondary option because the inoculum is lower and the drug reaches very high concentrations in the urine, highlighting that MERINO's findings are specific to invasive/high-burden disease.

Clinical Landscape

Noteworthy Related Trials

2018

MERINO Trial

n = 379 · JAMA

Tested

Piperacillin-tazobactam

Population

Adults with ceftriaxone-nonsusceptible E. coli or Klebsiella spp. bloodstream infection

Comparator

Meropenem

Endpoint

30-day all-cause mortality

Key result: Piperacillin-tazobactam was found to be inferior to meropenem for the treatment of ceftriaxone-nonsusceptible E. coli or Klebsiella bloodstream infections.

2021

MERINO-2 Trial

n = 392 · Lancet Infect Dis

Tested

Ertapenem

Population

Patients with Enterobacterales bloodstream infections resistant to ceftriaxone

Comparator

Piperacillin-tazobactam

Endpoint

30-day all-cause mortality

Key result: There was no statistically significant difference in 30-day mortality between patients treated with piperacillin-tazobactam and those treated with ertapenem.

2023

DAPPER Trial

n = 252 · Lancet Infect Dis

Tested

Piperacillin-tazobactam

Population

Patients with bloodstream infection due to AmpC beta-lactamase producing Enterobacterales

Comparator

Meropenem

Endpoint

30-day mortality

Key result: Piperacillin-tazobactam was non-inferior to meropenem in the treatment of patients with AmpC beta-lactamase producing Enterobacterales bloodstream infections.

Tailored to your role

Want this tailored to you?

Add your specialty or training stage to get role-specific takeaways and more questions.

Personalize this analysis