The antibiotic cefazolin was associated with fewer kidney injuries than two commonly used antistaphylococcal penicillins in adults with a specific type of Staphylococcus aureus bloodstream infection, according to an international randomised trial. The study also found cefazolin was not unacceptably less effective for 90-day survival under its prespecified comparison.
The findings come from the Staphylococcus aureus Network Adaptive Platform, or SNAP, trial. Published online in the New England Journal of Medicine on June 17, 2026, the research compared cefazolin with flucloxacillin or cloxacillin in adults with penicillin-resistant, methicillin-susceptible S. aureus bloodstream infections.
The study received fresh coverage from European Medical Journal on November 17, but the underlying trial findings were published earlier this year.
Kidney injury occurred less often with cefazolin
Acute kidney injury within 14 days occurred in 92 of 660 patients receiving cefazolin, or 13.9%, compared with 127 of 648 patients receiving flucloxacillin or cloxacillin, or 19.6%.
That represents an absolute difference of 5.7 percentage points—approximately six fewer kidney injuries for every 100 patients treated in the study. The adjusted odds ratio was 0.67, with a 95% credible interval of 0.50–0.89. The researchers’ Bayesian analysis estimated a 99.7% probability that cefazolin was superior for this kidney-safety outcome.
The kidney findings were a secondary safety outcome. The trial’s primary outcome was death from any cause within 90 days of enrolment.
Survival comparison met the trial’s threshold
Among patients evaluable for the primary outcome, 97 of 645 people receiving cefazolin died within 90 days, or 15.0%. In the penicillin group, 109 of 642 died, or 17.0%.
The adjusted odds ratio for mortality was 0.81, with a 95% credible interval of 0.59–1.12. The probability of noninferiority was 99.2%, meeting the trial’s criterion.
Noninferiority means the study met its predefined threshold for concluding that cefazolin was not unacceptably worse than the comparison treatment. It does not mean the two treatments were proved identical.
Although the mortality rate was numerically lower with cefazolin, the study did not establish that it improved survival. Its estimated probability of superiority for mortality was 89.8%, which should not be confused with the stronger kidney-safety finding.
The mortality and kidney-injury analyses used different evaluable patient groups, so their denominators should not be treated as a single study-wide total.
What infections did the study cover?
The original SNAP trial report concerned adults with penicillin-resistant but methicillin-susceptible S. aureus, commonly abbreviated to MSSA, in the bloodstream. The comparison was specifically against flucloxacillin or cloxacillin.
These findings therefore should not be presented as proof that cefazolin is preferable for every bloodstream infection, every bacterial species or every antibiotic comparison. The trial also did not study methicillin-resistant S. aureus, known as MRSA, in this treatment comparison.
The relevant part of the ongoing adaptive trial ran from February 17, 2022, to August 7, 2024, when the criterion for noninferiority had been met.
Open-label design remains an important limitation
The trial was open-label, meaning clinicians and patients knew which treatment was being given. That matters when interpreting safety outcomes as well as treatment decisions.
In correspondence published in October, researchers questioned whether the apparent kidney-safety advantage might have been overstated by surveillance bias—differences in monitoring that could affect how often an outcome is detected. They highlighted the use of serial creatinine measurements to identify kidney injury in an unblinded study. This was a methodological criticism, not a finding that disproved the trial’s results.
The distinction is important: the trial provides randomised evidence supporting cefazolin as an alternative for the infection studied, but the strength and generalisability of its safety advantage still warrant careful interpretation.
What the findings mean for treatment
The study’s conclusion was that cefazolin was noninferior to flucloxacillin or cloxacillin for 90-day mortality and was associated with a lower incidence of acute kidney injury.
For clinicians, the findings add evidence to the choice of antibiotic for adult MSSA bloodstream infection. For patients, they are not a reason to stop or switch a prescribed antibiotic without discussing treatment with their medical team.

