The results are sufficiently compelling that I immediately made the switch
For decades, a single antibiotic has stood as the default defense against one of medicine's most lethal bacterial threats — but a landmark global trial, spanning more than 150 hospitals, has quietly dismantled that consensus. Researchers have found that cefazolin and benzylpenicillin, drugs already present in the existing pharmaceutical arsenal, offer equal or superior survival outcomes with meaningfully less harm to patients' kidneys. The question of which antibiotic best treats golden staph bloodstream infections — a condition that claims over a million lives each year — has long carried the weight of uncertainty; now, the evidence has arrived, and the harder work of changing entrenched practice begins.
- Golden staph bloodstream infections kill between 15 and 25 percent of those infected, making the choice of antibiotic a matter of life and death at scale.
- Flucloxacillin, the antibiotic that became the global standard after penicillin resistance emerged, has now been shown to carry higher mortality and kidney injury rates than its alternatives.
- Cefazolin reduced 90-day mortality to 15 percent versus 17 percent for flucloxacillin, and cut acute kidney injury rates nearly in half — findings compelling enough that one lead physician immediately changed his own clinical practice.
- Benzylpenicillin, an older drug sidelined by the spread of resistance, proved equally effective for susceptible infections and was associated with 14 percent mortality compared to 22 percent in the flucloxacillin group.
- The evidence is now established, but translating it into routine hospital care will demand increased cefazolin supply chains, reliable laboratory testing infrastructure, and the revision of clinical guidelines worldwide.
More than 150 hospitals across the globe have upended decades of standard practice in treating golden staph bloodstream infections. Published simultaneously in two of medicine's most prestigious journals, the findings suggest that flucloxacillin — the antibiotic that became the default choice after penicillin resistance emerged — may no longer be the best option. Two separate trials identified safer, equally effective alternatives already present in the existing pharmaceutical arsenal.
Golden staph kills more than a million people annually worldwide, and when it enters the bloodstream, mortality rates climb to between 15 and 25 percent. The Staphylococcus aureus Network Adaptive Platform Trial, led by researchers at the Peter Doherty Institute and the University of Newcastle, set out to settle longstanding uncertainty about which antibiotics actually produce the best outcomes across a genuinely global network.
For methicillin-susceptible infections, cefazolin outperformed flucloxacillin on multiple fronts — an 89 percent probability of lower mortality, a 90-day death rate of 15 percent versus 17 percent, and acute kidney injury at half the rate. The evidence was so compelling that infectious diseases physician Steven Tong changed his own clinical practice immediately upon reviewing the results.
A second trial examined penicillin-susceptible cases, testing whether benzylpenicillin — an older drug that fell out of favor as resistance spread — could safely replace flucloxacillin. It could. Mortality in the benzylpenicillin group reached 14 percent compared to 22 percent for flucloxacillin, alongside significantly less kidney damage. Co-lead investigator Todd Lee highlighted the renal protection as a meaningful secondary benefit.
The history matters: flucloxacillin became entrenched in global guidelines precisely because it resisted the enzymes that destroyed regular penicillin. But the trial results suggest that when laboratory testing confirms susceptibility, older or alternative drugs deserve reconsideration. The real work now lies in translating evidence into practice — increasing cefazolin availability, ensuring reliable susceptibility testing, and revising hospital protocols. As Lee put it, trials generate the evidence, but changing practice is the slower, harder task that follows.
More than 150 hospitals across the globe have just upended decades of standard practice in treating golden staph bloodstream infections. The findings, published simultaneously in two of medicine's most prestigious journals, suggest that flucloxacillin—the antibiotic that became the default choice after penicillin resistance emerged—may no longer be the best option. Two separate trials have found safer, equally effective alternatives waiting in the existing pharmaceutical arsenal.
Golden staph kills more than a million people annually worldwide. When the bacterium enters the bloodstream, the stakes climb sharply: mortality rates hover between 15 and 25 percent. For decades, clinicians knew effective antibiotics existed, but uncertainty persisted about which ones actually produced the best outcomes. The Staphylococcus aureus Network Adaptive Platform Trial—a collaboration led by researchers at the Peter Doherty Institute for Infection and Immunity and the University of Newcastle—set out to settle that question by testing different antibiotics and treatment approaches across a genuinely global network.
The first major finding concerns methicillin-susceptible staph infections, or MSSA. Researchers compared cefazolin against flucloxacillin and found cefazolin superior on multiple fronts. There is an 89 percent probability that cefazolin produces lower mortality rates. The numbers tell the story: patients receiving cefazolin died at a rate of 15 percent within 90 days, compared to 17 percent for those on flucloxacillin. Beyond mortality, cefazolin patients experienced acute kidney injury at half the rate—14 percent versus 20 percent. Steven Tong, an infectious diseases physician at the Royal Melbourne Hospital and the Doherty Institute, was so convinced by the evidence that he changed his own clinical practice immediately. "The results are sufficiently compelling," he said, that the shift felt inevitable.
The second trial examined penicillin-susceptible staph, or PSSA—cases where laboratory testing confirms the bacteria remain vulnerable to penicillin itself. Here, researchers tested whether benzylpenicillin, an older antibiotic that fell out of favor as resistance spread, could safely replace flucloxacillin. The answer was yes. Benzylpenicillin proved equally effective while causing less kidney damage. Mortality in the benzylpenicillin group reached 14 percent compared to 22 percent in the flucloxacillin group. Todd Lee, a co-lead investigator at McGill University Health Centre in Canada, emphasized the kidney protection: patients on benzylpenicillin experienced significantly less renal injury alongside the mortality benefit.
The history here matters. Penicillin was once the standard treatment for staph infections until resistance forced a retreat. Flucloxacillin emerged as the replacement—a semi-synthetic penicillin designed to resist the enzymes that destroyed regular penicillin. It became entrenched in clinical guidelines worldwide. But the trial results suggest that when laboratory testing confirms susceptibility to penicillin or when cefazolin is available, those older or alternative drugs deserve reconsideration. Joshua Davis, an infectious diseases physician at the University of Newcastle and global co-lead of the trial, noted that clinicians can now confidently use susceptibility results to guide treatment decisions where testing is available.
The real work begins now. Translating these findings into routine hospital practice will require coordination across multiple fronts. Some countries may need to increase cefazolin production and availability. Laboratories must ensure susceptibility testing reaches clinicians reliably. Hospital protocols and clinical guidelines will need revision. Lee acknowledged the challenge plainly: "Trials generate the evidence, but the next step is making sure that evidence changes practice." This is the largest trial ever conducted on staph bloodstream infections, drawing participation from countries worldwide to answer questions that affect millions of patients annually. The evidence is now in hand. What remains is the slower, harder work of getting hospitals and guideline groups to act on it.
Citas Notables
There is an 89 percent probability that cefazolin is associated with lower mortality in MSSA bloodstream infections— Steven Tong, infectious diseases physician, Royal Melbourne Hospital
Trials generate the evidence, but the next step is making sure that evidence changes practice— Todd Lee, co-lead investigator, McGill University Health Centre
La Conversación del Hearth Otra perspectiva de la historia
Why did flucloxacillin become so entrenched if these alternatives work better?
It wasn't a mistake at the time. Penicillin resistance forced the switch. Flucloxacillin was genuinely better than what came before. But medicine doesn't always revisit old assumptions once a new standard takes hold.
So this trial is essentially saying we should go backwards?
Not backwards—it's saying we should go sideways, or even forwards. We're using the same susceptibility testing that already exists to make smarter choices. If the bacteria is actually vulnerable to penicillin, why use a broader drug? And cefazolin isn't old; it's just been overlooked for this particular use.
What's the kidney injury difference about? Why would one antibiotic damage kidneys more than another?
The mechanisms differ. Flucloxacillin appears to have more direct nephrotoxic effects at the doses needed for bloodstream infections. Cefazolin and benzylpenicillin achieve the same killing power with less collateral damage to the kidneys.
If the evidence is this clear, why hasn't this changed already?
Because a single trial, even a large one, doesn't automatically shift practice. These findings just published. Hospitals have existing protocols, training, supply chains. Guidelines need updating. It takes time for evidence to become routine.
What happens to patients right now, today, in hospitals that haven't heard about this?
They continue receiving flucloxacillin as they have been. Some will do fine. Some will develop kidney injury that might have been preventable. That's why the researchers are emphasizing the urgency of implementation.
Is there any reason a hospital might resist switching?
Cost, supply, inertia, training. Cefazolin might not be as readily stocked everywhere. Clinicians trained on flucloxacillin may need convincing. But the evidence here is strong enough that resistance would be hard to justify.