Highlights
– In a multicentre randomized non‑inferiority trial (CloCeBa), cefazolin achieved non‑inferior clinical efficacy to cloxacillin for MSSA bacteraemia (75% vs 74% meeting a 90‑day composite endpoint).
– Cefazolin was associated with fewer serious adverse events (15% vs 27%) and markedly lower acute kidney injury (AKI) (1% vs 12%).
– The trial supports cefazolin as a pragmatic, better‑tolerated alternative to anti‑staphylococcal penicillins for many patients with MSSA bloodstream infection, while acknowledging limitations in generalizability.
Background
Staphylococcus aureus bacteraemia (SAB) is a common and potentially life‑threatening infection with substantial morbidity and mortality. For meticillin‑susceptible S. aureus (MSSA), treatment guidelines traditionally recommend anti‑staphylococcal penicillins such as nafcillin, oxacillin, or cloxacillin. Cefazolin, a first‑generation cephalosporin, is widely used as an alternative because of favorable dosing, lower monitoring burden, and an apparent better adverse‑event profile. Observational studies and meta‑analyses have suggested similar efficacy but less toxicity with cefazolin; however, before CloCeBa no large randomized clinical trial had directly compared cefazolin with an anti‑staphylococcal penicillin in MSSA bacteraemia.
Persisting concerns—most notably the potential for a cefazolin ‘inoculum effect’ mediated by type A β‑lactamases and theoretical reduced activity in high‑inoculum deep infections—have limited universal adoption of cefazolin as first‑line therapy in all MSSA bacteraemia scenarios. The CloCeBa trial sought to provide randomized evidence comparing clinical efficacy and safety between cefazolin and cloxacillin in adult patients with MSSA bloodstream infection.
Study design
CloCeBa was a prospective, open‑label, multicentre, randomized non‑inferiority trial conducted at 21 hospitals in France. Adults (≥18 years) with MSSA bacteraemia were eligible, excluding those with intravascular implants or suspected central nervous system infection. Patients were randomized 1:1 to receive intravenous cefazolin 25–50 mg/kg every 8 hours or cloxacillin 25–50 mg/kg every 4–6 hours for the first 7 days. Randomization used computer‑generated blocks with stratification for vascular‑access associated bacteraemia and centre. After day 7, subsequent antibiotic choice and total duration (minimum 14 days) were left to the treating clinician. The primary outcome was a composite measured in the intention‑to‑treat population: sterile blood cultures at day 3 (or day 5 for endocarditis), absence of relapse of bacteraemia, survival, and clinical success at day 90. A non‑inferiority margin of 12% was prespecified. The trial was registered (ClinicalTrials.gov NCT03248063) and funded by the French Ministry of Health.
Key findings
Between Sept 5, 2018, and Nov 16, 2023, 315 participants were randomized: 158 to cefazolin and 157 to cloxacillin. After exclusions (12 and 11 respectively), the analyzed population comprised 146 participants per arm. Baseline characteristics: mean age 62.7 years (SD 16.4), 74% male, median Pitt bacteraemia score 0 (IQR 0–0), indicating mostly low‑severity presentations. Race/ethnicity data were not collected.
Primary outcome
The primary composite endpoint was achieved in 109 of 146 participants (75%) in the cefazolin arm and 108 of 146 participants (74%) in the cloxacillin arm. The absolute treatment difference was −1% (95% CI −11 to 9). The prespecified non‑inferiority analysis met its criterion (p=0.012), indicating that cefazolin was non‑inferior to cloxacillin for the composite 90‑day outcome.
Safety and secondary outcomes
Serious adverse events (SAEs) were less frequent with cefazolin: 22/146 (15%) versus 40/146 (27%) with cloxacillin (p=0.010). Acute kidney injury (AKI) showed a pronounced difference: 1/134 (1%) in the cefazolin group versus 15/128 (12%) in the cloxacillin group (p=0.0002). Other safety signals (hepatotoxicity, hypersensitivity) were not reported as differing substantially in the primary report. Mortality, relapse rates, and culture sterilization (components of the composite endpoint) were integrated into the primary result but individual component frequencies beyond AKI were not detailed in the summary provided here.
Subgroups and contextual features
Important design choices affect interpretation: patients with intravascular implants and suspected CNS infections were excluded. The trial population had low median Pitt scores, so enrolled patients were generally less acutely ill than some contemporary SAB cohorts. Antibiotic therapy after day 7 was at clinician discretion, which reflects pragmatic real‑world practice but introduces heterogeneity in total exposure.
Interpretation and clinical implications
CloCeBa provides the first randomized evidence that cefazolin is non‑inferior to cloxacillin for the management of MSSA bacteraemia using a clinically meaningful composite endpoint measured to 90 days. The similar efficacy combined with substantially lower rates of serious adverse events—most notably AKI—supports cefazolin as a preferred or at least acceptable alternative to anti‑staphylococcal penicillins in many patients.
Practical implications for clinicians include favoring cefazolin when there is elevated risk for nephrotoxicity, when frequent dosing and infusion logistics for cloxacillin are problematic, or when convenience and tolerability are priorities. Cefazolin’s dosing schedule and reliable tolerability profile may facilitate outpatient parenteral antibiotic therapy (OPAT) and reduce hospital resource use.
Expert commentary and limitations
Strengths of the trial include randomized allocation, multicentre conduct, a pragmatic approach to early assigned therapy, and a clinically relevant 90‑day composite endpoint. The randomized design directly addresses the important clinical question left unresolved by observational data.
Key limitations and considerations:
- Open‑label design: clinicians knew the assigned drug, which could influence ancillary care or reporting of subjective outcomes, although hard outcomes (death, sterile cultures, relapse) are less prone to bias.
- Generalisability: exclusion of patients with intravascular implants and suspected CNS infection means results may not apply to device‑associated SAB, cases with prosthetic material, or brain/CNS involvement. The low median Pitt score indicates enrollment skewed toward less severe infections; extrapolation to critically ill patients should be cautious.
- Non‑inferiority margin: the trial used a 12% margin, a choice that should be appraised clinically; some may view it wide for life‑threatening infections, though the observed point estimate difference was negligible (−1%).
- Post‑day‑7 therapy not standardized: differences in subsequent antibiotic use could influence later outcomes; however, initial empirical randomized therapy is often the most clinically consequential period.
- Inoculum effect and microbiologic subgroups: the trial did not report whether failures clustered among isolates with the cefazolin inoculum effect (type A β‑lactamase producers) or other phenotypes. This mechanistic concern remains relevant when treating high‑inoculum deep infections (e.g., endocarditis, large abscesses), where clinicians may consider anti‑staphylococcal penicillins until more data are available.
How this should change practice
For adult patients with MSSA bacteraemia without intravascular implants or CNS infection and who are hemodynamically stable, cefazolin can be considered a first‑line option, particularly where there is baseline renal impairment or high risk of antibiotic‑related nephrotoxicity. Institutions may use these data to revise local protocols and OPAT pathways to incorporate cefazolin more broadly for appropriate patients.
For cases with prosthetic material, CNS involvement, severe sepsis, or other high‑inoculum scenarios, individualized decisions remain appropriate. When there is microbiological evidence of a cefazolin inoculum effect (if testing is available) or high clinical concern for deep‑seated infection, clinicians may still prefer an anti‑staphylococcal penicillin.
Research and guideline implications
CloCeBa fills a major evidence gap and is likely to prompt guideline updates and changes in antibiotic stewardship practice. Further research should address: outcomes in patients with device‑associated SAB and severe illness, the role of cefazolin in endocarditis and deep‑seated infections, the microbiologic correlates of failure (including inoculum effect), and economic analyses comparing hospital and outpatient costs between regimens. Comparative studies in populations with higher illness severity would help broaden applicability.
Conclusion
The CloCeBa randomized trial supports cefazolin as a non‑inferior alternative to cloxacillin for MSSA bacteraemia in adults without intravascular implants or CNS infection, offering similar 90‑day clinical outcomes and substantially lower nephrotoxicity. Clinicians should incorporate these data into individualized treatment decisions and institutional protocols, while recognizing remaining knowledge gaps around high‑inoculum and device‑associated infections.
Funding and registration
The trial was funded by the French Ministry of Health. ClinicalTrials.gov identifier: NCT03248063.
References
1. Burdet C, Saïdani N, Dupieux C, et al.; CloCeBa Study Group. Cloxacillin versus cefazolin for meticillin‑susceptible Staphylococcus aureus bacteraemia (CloCeBa): a prospective, open‑label, multicentre, non‑inferiority, randomised clinical trial. Lancet. 2025 Oct 17:S0140‑6736(25)01624‑1. doi:10.1016/S0140‑6736(25)01624‑1. Epub ahead of print. PMID: 41115439.
2. Tong SYC, Davis JS, Eichenberger E, Holland TL, Fowler VG Jr. Staphylococcus aureus infections: epidemiology, pathophysiology, clinical manifestations, and management. Clin Microbiol Rev. 2015 Jul;28(3):603‑661. doi:10.1128/CMR.00134‑14.
Note: This article summarizes the CloCeBa trial and highlights practical implications derived from the trial data. Clinicians should consult the full manuscript and local guidelines when applying these findings to individual patient care.