Persistent Methicillin-resistant Staphylococcus aureus (MRSA) bloodstream infections (BSI) are associated with significant morbidity, mortality, and healthcare expenditures. Vancomycin (VAN) remains the treatment of choice for invasive MRSA BSI. Current guidelines for the treatment of MRSA BSI recommend a VAN AUC 24h/MIC ratio ≥400. The Detroit Medical Center (DMC) implemented an AUC guided dosing strategy. However, data on the association between AUC 24h and clinical outcomes in MRSA BSI are limited. We aimed to evaluate the association between VAN AUC 24h and persistent bacteremia (PB) among patients with BSI.
Multi-center, retrospective cohort study from January 2015 to November 2018. We included adult patients with MRSA bacteremia treated with VAN for which AUC 24h monitoring was performed. AUC was measured using 2-level guided dosing. The primary outcome was PB defined as continued positive cultures >72 hours after VAN initiation. Classification and Regression Tree (CART) analysis was performed to determine the AUC 24h breakpoint (BP) most predictive of PB in the cohort. Mann–Whitney and Fischer exact tests were used for univariate analysis. The independent association between AUC 24h, dichotomized at the CART-derived cut-point, was then examined through multivariable logistic regression analysis.
Overall, 137 patients were included. The median age was 59 (18–85) years, 69.3% male, and 75.2% African American predominance. Most common sources of BSI were skin/soft tissue (39.4%), pneumonia (25.5%), and osteoarticular (16.8%). The median APACHE II score was 13 (8–20). Median time to microbiological clearance was 2.5 days (0.5–12). Patients with AUC 24h ≤ 406.25 were more likely to have PB compared with those with AUC 24h > 406.25 (59.4% and 35.2%, respectively; P = 0.002). After controlling for age, intensive care stay, and concomitant β-lactam therapy; AUC of ≤ 406.25 (aOR 2.767, 95% CI 1.212–6.318) and endocarditis (aOR 2.87, 95% CI 1.079–7.638) were independently associated with PB.