Clinical safety and efficacy of anti-staphylococcal penicillins vs cefazolin for methicillin- susceptible Staphylococcus aureus bacteremia in pediatric patients

Abstract Corresponding Author: Zachary Howe: 705 Riley Hospital Drive SFT Room W6111, Indianapolis, IN 46228, 317-948-9562, [email protected] No conflicts of interest. Background Staphylococcus aureus bacteremia (SAB) has the potential to cause severe complications including septic shock, endovascular infection, and disseminated infection. In the setting of a high-inoculum infection (such as endocarditis), heightened production of the BlaZ beta lactamase may increase the effective minimum inhibitory concentration (MIC) at the site of infection, potentially reducing the efficacy of cefazolin. The majority of studies comparing anti-staphylococcal penicillins (ASPCNs) versus cefazolin in methicillin-susceptible S. aureus (MSSA) bacteremia are in adult patients finding that cefazolin is non- inferior to ASPCNs and is associated with fewer adverse effects. Given the significant differences in pathogenesis of infection and rates of complications in children compared to adults with SAB, studies in pediatric cohorts are needed. Thus, we sought to compare safety and efficacy of ASPCNs versus cefazolin for the treatment of MSSA bacteremia in children. Methods A retrospective cohort study was conducted to compare the safety and efficacy of ASPCNs (nafcillin and oxacillin) versus cefazolin in pediatric patients with SAB due to MSSA. The primary endpoint was a composite of treatment failure defined as: mortality due to SAB, recurrence of MSSA bacteremia within 1 year of original SAB, or clinical failure, defined as a blood culture positive for MSSA within a 14-day period following 2 consecutive negative cultures. Secondarily, antibiotic associated adverse effects were compared between groups. Patients with infection of the central nervous system or experiencing mortality within 48 hours of starting antimicrobials were excluded. Results Of 174 patients screened, 127 met inclusion criteria. Of these patients, 64 (50%) received cefazolin and 63 (50%) received an ASPCN (16 oxacillin [12.6%], 37 nafcillin [37.4%]). Patients received a median of 18.5 days of therapy (IQR 19 days). No difference in PRISM score was observed between groups (median 2 vs 3, p=0.267). For the primary endpoint, a statistically significant difference was not observed with cefazolin as compared to ASPCNs (4.7% vs 12.7%, p=0.11). A similar incidence of treatment failure was observed in the ASPCN group across subgroup analyses, including in patients with or without prosthetic material, with or without endocarditis, with full, partial, or no source control, and when comparing cefazolin to nafcillin and oxacillin separately. No differences in incidence of acute kidney injury (17.2% vs 19%, p=0.79) or hepatoxicity (0% vs 3.2%, p=0.24) were observed. Conclusion The current study did not observe an association between use of cefazolin and poorer clinical outcomes as compared to ASPCNs for the treatment of MSSA bacteremia in pediatric patients. ASPCNs have historically been thought to be more likely to lead to clinical success compared to cefazolin for high inoculum infections. In this study, cefazolin was associated with a lower rate of composite mortality, treatment failure, or recurrence in patients with prosthetic material with or without endocarditis, including those without source control. However, as BlaZ genotypic testing is not available for use in clinical practice, it is difficult to determine if such a phenotype was present in a significant proportion of included cases. Future studies would ideally be prospective in nature and have a sample size of patients with infections associated with a high inoculum of bacteria (ex: endocarditis) sufficiently large to determine if a significant improvement in outcomes can be seen with ASPCNs as compared to cefazolin. Should rapid diagnostic testing for BlaZ production become commercially available, studies could also utilize this testing to determine if a difference in clinical outcomes can be observed specifically with infections due to such strains of MSSA.