Longitudinal Vancomycin Exposure and PK/PD Target Attainment in Postoperative Central Nervous System Infections: A Single Centre Retrospective Cohort Study

Background/Objectives: Postoperative central nervous system (CNS) infections are severe compli-cations of neurosurgery, and vancomycin is a cornerstone of empirical and targeted therapy; how-ever, longitudinal vancomycin exposure in this population is poorly characterized. Methods: We conducted a single-centre retrospective cohort study of 189 postoperative CNS infection episodes in 174 adults treated between 2020 and 2026, focusing on longitudinal vancomycin exposure and pharmacokinetic.pharmacodynamic (PK/PD) target attainment according to therapeutic drug mon-itoring (TDM) status. A population pharmacokinetic (PopPK) model, developed from 193 concen-trations was used to reconstruct exposure in TDM episodes and simulate exposure for no TDM ep-isodes. Results: The final PopPK model confirmed creatinine clearance (CrCl) as the principal de-terminant of vancomycin clearance; higher baseline CrCl independently predicted a greater propor-tion of treatment time with AUC24 <400 mg·h/L. Underexposure occurred throughout: patients spent 37.2% (TDM) and 47.5% (no TDM) of evaluable treatment time below AUC24 400 mg·h/L, with early underexposure even more common in both groups. The TDM group spent more time within the AUC₂₄ 400–600 mg·h/L range (43.23% vs 29.91%). No statistically significant exposure–outcome associations were identified in the overall population. In an exploratory ICU subgroup analysis, higher mean AUC₂₄ was associated with clinical improvement within 72 h (OR 1.92 per 100 mg·h/L increase, 95% CI 1.12–3.28), although the exposure-by-ICU interaction was not statistically signifi-cant. Conclusions: These findings support early and repeated vancomycin TDM throughout treat-ment to identify and address underexposure, particularly in patients with augmented renal function and those treated in ICU.