메티실린 내성 황색포도알균 균혈증 소아에서 베이지안 유도 투여 소프트웨어로 추정한 최적의 혈중 반코마이신 곡선하면적

Abstract

Author: Yong Hee Lee Title: Optimal vancomycin area under the concentration-time curve target estimated by Bayesian- guided dosing software in children with methicillin-resistant Staphylococcus aureus bacteremia Background: Despite the widespread use of vancomycin in methicillin-resistant Staphylococcus aures (MRSA) infections, evidence for the target of the area under the concentration-time curve (AUC) for pediatric population is lacking. This study was conducted using Bayesian-based software to derive the target of vancomycin AUC, aiming to provide optimal outcomes for pediatric patients with MRSA bacteremia. Methods: From September 2013 to December 2021, all hospitalized patients with MRSA bacteremia aged ≥ 3 months and < 19 years were analyzed retrospectively. The AUC of vancomycin over 168 hours after the first dose was simulated using Bayesian-guided dosing software, incorporating individual patient data, including age, sex, height, weight, serum creatinine, and all measured vancomycin trough concentrations (Ctrough). Outcomes included persistent bacteremia (≥ 48 hours), recurrence, and 30-day all-cause mortality. Acute kidney injury (AKI) was investigated for 14 days after initiating vancomycin therapy. To identify an appropriate AUC target for anticipating optimal outcomes, I initially examined whether the AUC/minimum inhibitory concentration (MIC) ratio of ≥ 400, as per recent guidelines, demonstrated clinical benefits. In cases where this target was considered inappropriate, alternative AUC targets were explored through receiver operating characteristic (ROC) analysis. In addition, factors potentially influencing outcomes, such as intensive care unit stay, immunocompromised state, clinical severity, polymicrobial bloodstream infection, MIC, source of infection, and conducted interventions, were collectively analyzed. Results: During the study period, 79 cases of MRSA bacteremia were identified. Cases that had received vancomycin 72 hours before the onset of MRSA bacteremia (n=5), had a treatment period of vancomycin <72 hours (n=10), or were receiving renal replacement therapy at baseline (n=8) were excluded, leaving 56 cases finally included in the study. Persistent bacteremia, recurrence, 30-day all-cause mortality, and AKI were observed in 10 (17.9%), 8 (14.8%), 2 (3.7%), and 4 (7.1%) cases, respectively. AUC over the second 24 hours (AUC24-48) ≥ 400 did not demonstrate clinical benefit in the outcomes. On the contrary, persistent bacteremia was frequently observed in patients with higher AUCs. AUC24-48 was a more statistically significant parameter than AUC over the first 24 hours (AUC0-24) in predicting persistent bacteremia. AUC24-48 ≥ 528.2 mg·h/L was associated with persistent bacteremia (P < 0.01). AUC24-48 ≥ 536.8 mg·h/L was associated with AKI (P = 0.01). Variables associated with mortality and recurrence could not be identified. The group with persistent bacteremia exhibited a significantly higher AUC24-48 compared to the group with resolved bacteremia, along with a higher frequency of the presence of noneradicable or not eradicated focus and a higher occurrence of AKI. Conclusion: I suggest that an AUC24-48 < 528.2 mg·h/L is the optimal target for non-dialysis pediatric patients aged ≥ 3 months with MRSA bacteremia. The lower limit of the target range indicating subtherapeutic levels has not been identified. A large-scale multicenter study is necessary to generalize this finding for pediatric patients with MRSA bacteremia. Keywords: MRSA, bacteremia, vancomycin, Bayesian, pediatric