Bloodstream infections caused by Staphylococcus aureus carry a mortality rate that rivals many cancers, yet a fundamental treatment question has remained unresolved for decades: which antibiotic should be the preferred first-line agent? The answer has real consequences for thousands of hospitalized patients each year, particularly those at elevated risk for kidney damage from the current standard drugs.
An international Bayesian adaptive platform trial enrolled adult patients between February 2022 and August 2024, randomizing them to either cefazolin or an antistaphylococcal penicillin — flucloxacillin or cloxacillin — for penicillin-resistant, methicillin-susceptible S. aureus bacteremia. The primary endpoint was all-cause mortality at 90 days. Among evaluable patients, 90-day mortality reached 15.0% (97 deaths out of 645 patients) in one arm, with the trial's prespecified noninferiority criterion — an adjusted odds ratio below 1.2, approximating less than 2.5 percentage points absolute mortality difference — met before the trial's scheduled conclusion. Acute kidney injury within 14 days served as a key secondary safety endpoint, directly addressing a well-documented nephrotoxicity concern associated with antistaphylococcal penicillins.
The clinical significance here is substantial. Antistaphylococcal penicillins like flucloxacillin have long been considered the gold standard for methicillin-susceptible S. aureus bacteremia, but their nephrotoxicity profile creates difficult tradeoffs in patients with pre-existing renal compromise or concomitant nephrotoxic medications. Cefazolin, already widely available and inexpensive, has been used as an alternative, but without robust randomized evidence to anchor that practice. This trial — using a sophisticated adaptive Bayesian design that allows the trial to stop early when evidence thresholds are crossed — provides the strongest randomized evidence to date supporting cefazolin's noninferiority. The adaptive platform methodology is increasingly valued in infectious disease trials because it is both ethically efficient and statistically rigorous. Whether superiority on the kidney injury endpoint was also demonstrated remains to be seen in the full publication, but even a noninferiority finding on mortality alone could shift prescribing practice and guideline recommendations globally.