Purpose: Vancomycin is frequently used to treat central nervous system (CNS) infections, yet its cerebrospinal fluid (CSF) penetration remains poorly characterized in patients with intracranial hemorrhage (ICH) requiring external ventricular drains (EVDs). Methods: A prospective observational study was conducted on nine neurosurgical patients with ICH and EVDs receiving intravenous vancomycin. Plasma and CSF samples were collected at predefined time points and analyzed by validated assays. Noncompartmental analysis, regression modeling, and nonlinear mixed-effects pharmacokinetic (PK) modeling were performed. Results: The AUC CSF/plasma ratios ranged from 0.84% to 14.22%. CSF penetration correlated strongly with plasma–CSF concentration ratios at the end of infusion (Spearman r = 0.791; p = 0.004). Linear regression identified urine output (R 2 = 0.51, p = 0.014), WBC/total cell ratio (R 2 = 0.62, p = 0.007), and end-of-infusion concentration (R 2 = 0.45, p = 0.049) as significant predictors of AUC CSF . A two-compartment model provided the best fit for vancomycin population PK, yielding clearance and volume of distribution estimates similar to previous reports, although no significant covariates for CSF penetration were identified. Conclusion: To our knowledge, this is the first study reporting real-world data of vancomycin CSF penetration in Taiwanese neurosurgical patients. These findings provide a critical PK foundation for developing future model-informed precision dosing strategies and validating surrogate TDM markers in this neurocritical care population. Keywords: cerebrospinal fluid, external ventricular drain, intracranial hemorrhage, population pharmacokinetics, therapeutic drug monitoring, vancomycin
Tseng et al. (Sun,) studied this question.