Comment
Treatment of MSSA bacteraemia: a call for a personalised approach
Carol M Kao, Stephanie A Fritz
Lancet Available online 17 October 2025
https://doi.org/10.1016/S0140-6736(25)01844-6
Staphylococcus aureus infections remain a formidable health threat, responsible for the most infection-related deaths worldwide and a mortality rate of up to 30%, despite contemporary infection prevention practices, antimicrobial stewardship, and novel antibiotics.1, 2 Treatment of S aureus bacteraemia is often complicated by the inability to remove foreign materials (eg, intravascular devices or osteoarticular hardware) and the development of metastatic foci of infection related to persistent bacteraemia.3 For meticillin-susceptible S aureus (MSSA) bacteraemia, treatment with cefazolin versus an anti-staphylococcal penicillin has been a long-standing source of debate among clinicians. Specifically, concern exists regarding the possibility of increased cefazolin resistance in S aureus strains carrying the blaZ gene or infections with a high bacterial burden, also known as the cefazolin inoculum effect.4, 5, 6 Despite this concern, cefazolin is often selected for MSSA bacteraemia treatment given its more convenient dosing and improved safety profile.7 However, there is a paucity of data from randomised controlled trials to inform optimal treatment of S aureus bacteraemia.
In The Lancet, Charles Burdet and colleagues report their findings from CloCeBa, a prospective, open-label, multicentre, non-inferiority, randomised clinical trial in adults with MSSA bacteraemia who were hospitalised, conducted in France.8 Participants were excluded if they had received more than 72 h of antibiotics targeting MSSA, any foreign material suspected to be infected, clinical suspicion for CNS involvement, polymicrobial infection, stage 4 renal failure at baseline, or poor long-term prognosis. Participants were randomly assigned 1:1 to receive either cefazolin or cloxacillin for a minimum of 7 days of assigned therapy, with at least 14 days of total therapy. Participants were followed up for adverse events throughout the 7 days of assigned therapy, at the end of study treatment, and at the end of antibiotic treatment. Of the 315 participants randomly allocated to treatment groups, 292 were enrolled in the intention-to-treat population (146 in each group) and 223 were included in the per-protocol analyses (115 in the cefazolin group, 108 in the cloxacillin group). In the intention-to-treat population, 74% were male and the mean age was 62·7 years (SD 16·4). 83 (28%) of 292 participants were immunosuppressed and 103 (36%) of 284 had a catheter-related infection. Ethnicity and racial data were not reported.
The CloCeBa primary outcome was a composite endpoint for therapeutic success, defined as sterile blood culture at day 3 without relapse of bacteraemia before day 90, clinical success (resolution of infectious signs and symptoms), and survival at day 90. In the intention-to-treat analysis, the primary endpoint was met in 108 (74%) of 146 participants in the cefazolin group and 109 (75%) of 146 participants in the cloxacillin group, thereby achieving non-inferiority. 90-day survival was similar between participants receiving cefazolin and cloxacillin (92% vs 92%), a survival rate higher than previously reported.1, 9 Participants who received cefazolin had better safety outcomes compared with those receiving cloxacillin, with less treatment discontinuation from adverse events (3% vs 9%; p=0·052), fewer severe adverse events by the end of study treatment (15% vs 27%; p=0·01), and decreased acute kidney injury by the end of study treatment (4% vs 17%; p=0·0008).
Despite these encouraging results, an important limitation is the confounding fact that during the 7-day treatment course, a substantially higher number of patients in the cloxacillin group received a second anti-staphylococcal antibiotic compared with those in the cefazolin group (21% vs 8%). These additional agents included aminoglycosides, cephalosporins, and vancomycin, without substantial explanation from the authors regarding this imbalance. Previous studies comparing combination therapy for S aureusbacteraemia showed significantly higher nephrotoxicity when an anti-staphylococcal β-lactam was added for methicillin-resistant Staphylococcus aureus bacteraemia compared with monotherapy with vancomycin or daptomycin alone.10 In the current trial, it would be of interest to conduct secondary analyses to compare individuals who received cloxacillin monotherapy with those who received cefazolin to determine whether the difference between severe adverse event rates remains significant. An additional consideration is that the number of patients in this study with infective endocarditis (14 [5%] of 285) or deep abscess (24 [8%] of 284) was small and limits the generalisability of the findings for these populations.
Similarly to the study by Burdet and colleagues, a recent case series of 216 patients with MSSA infective endocarditis sought to examine whether an inoculum effect and the presence of blaZ were associated with worse clinical outcomes and showed that the presence of an inoculum effect was associated with an increased 30-day mortality.5 In this study by Burdet and colleagues, presence of the blaZ gene was detected in 178 (70%) of 253 MSSA strains and led to abrogation of non-inferiority of cefazolin over cloxacillin in the post-hoc analysis. Altogether, these two studies show a potential association between blaZ detection and the cefazolin inoculum effect and worse clinical outcomes. Thus, personalised risk stratification, accounting for S aureus strain characterisation, could transform our approach to S aureus bacteraemia management. With refined sequencing and computing capabilities, it might soon be feasible to rapidly integrate host and strain-specific biological factors to guide therapy and predict outcomes.11
The study by Burdet and colleagues is one of the few conducted S aureus bacteraemia randomised controlled trials and includes patients who are immunocompromised, an important population at risk for severe S aureus infections that is often excluded from clinical trials. Findings will need to be confirmed in diverse populations, as S aureusstrains in France and Europe differ compared with other parts of the world.12, 13 Overall, the study contributes to the growing body of data suggesting that cefazolin could be used to treat uncomplicated MSSA bacteraemia. These findings will inspire further research that might provide definitive evidence to inform clinical practice.
