Therapeutic Monitoring of Vancomycin and Factors Affecting Survival in ICU Patients with Infections.

Background/

Objectives

Vancomycin remains a cornerstone antibiotic for Gram-positive infections, but its narrow therapeutic window and marked pharmacokinetic variability in critical patients complicate dosing. This study aimed to analyze survival and identify factors associated with mortality in intensive care unit (ICU) patients receiving vancomycin under therapeutic drug monitoring (TDM).

Methods

A single-center, retrospective cohort study was conducted at City Clinical Hospital named after I.V. Davydovsky (Moscow, Russia) between December 2021 and January 2026 (IRB protocol VANCO-2021). A total of 190 adult patients from surgical, therapeutic, and cardiac ICUs receiving vancomycin for at least 3-5 days were analyzed; trough concentrations were determined by HPLC-MS/MS. Cox proportional hazards, spline Cox, and segmented (piecewise) Cox models were applied to identify mortality predictors, with bootstrap subsampling assessing threshold robustness.

Results

Three independent predictors of mortality were identified: ln(minimum vancomycin concentration) (HR = 1.644, p = 0.028), age (HR = 1.019, p = 0.037), and serum creatinine (HR = 1.004, p < 0.001). A data-driven threshold trough concentration (Ctrough) of 25.5 µg/mL was identified, above which mortality increased substantially (63.0% vs. 26.4%; p < 0.05). The adjusted hazard ratio for Ctrough ≥ 25.5 µg/mL was approximately 1.88 (p = 0.054); bootstrap subsampling yielded an uncertainty interval of 23.2-33.6 µg/mL.

Conclusions

Elevated minimum vancomycin concentrations exceeding 25.5 µg/mL are independently and statistically associated with increased mortality in ICU patients, likely reflecting altered drug exposure and illness severity rather than a direct causal effect. Systematic TDM is essential for maintaining vancomycin within the therapeutic range.