Abrar Samman 1,2, Abdullah Alsultan 2, Muneera Al-Jelaify 3, Mohamad Temsah 4
1 Department of Clinical Pharmacy, College of Pharmacy, King Saud University (Riyadh, Saudi Arabia), 2 Department of Clinical Pharmacy, College of Pharmacy, Princess Nora bint Abdulrahman University (Riyadh, Saudi Arabia), 3 Pharmacy Services, King Saud University Medical City (Riyadh, Saudi Arabia), 4 Department of Pediatrics, College of Medicine, King Saud University (Riyadh, Saudi Arabia)
Objective:
This study aimed to determine whether serum creatinine (SCr), cystatin C (CysC), or combined biomarker–based equations provide the most accurate estimation of glomerular filtration rate (eGFR) for predicting vancomycin clearance (CL) in critically ill pediatric patients. Because vancomycin is a renally eliminated antibiotic commonly used to treat serious gram-positive infections in the pediatric intensive care unit (PICU), and renal function in PICU patients is highly variable, determining the most reliable biomarker for assessing kidney function is essential for accurate dosing and optimal therapeutic outcomes.
Methods:
A prospective population pharmacokinetic (PopPK) study was conducted in the PICU for patients aged 1 month to 14 years receiving vancomycin. Peak and trough concentrations were collected, and a nonlinear mixed-effects model was applied using Monolix®. Six eGFR equations [Schwartz (SCr-based), Hoek, Larsson, and Zappitelli (CysC-based), and the Full Age Spectrum (FAS) and Zappitelli combined equations] were evaluated using both covariate modeling and linear regression. CL was allometrically scaled to body weight with an exponent of 0.75, and volume of distribution was scaled with an exponent of 1. Model performance was compared using the objective function value (OFV), Akaike information criterion (AIC), parameter precision, and goodness-of-fit diagnostics. Significant covariates were evaluated on both CL and volume of distribution using stepwise forward inclusion and backward elimination. A bootstrap analysis of 1,000 replicates was conducted to confirm model stability. Monte Carlo simulations (n = 10,000) were performed using Simulx® to evaluate dosing regimens (15–20 mg/kg every 6 or 8 hours) and their ability to achieve AUC₀₋₂₄ 400–600 µg·h/mL and trough concentrations <20 µg/mL.
Results:
Fifty-three patients contributed 102 vancomycin concentrations. A one-compartment model with linear elimination and proportional error best described the data. CysC-based equations produced the most substantial improvements in model fit, with the Hoek equation showing the largest OFV reduction and strongest correlation with CL (R² = 0.26). The weight and the eGFR calculated by Hoek were the only covariates retained as significant predictors of CL. Other covariates, including age, gender, serum albumin, presence of shock, presence of malignancy, steroid use, and vasopressor use, did not remain significant in the final model. Final parameter estimates were CL = 1.66 L/h and Vd = 10.63 L. Parameter precision was acceptable for all fixed and random effects, and bootstrap results demonstrated excellent agreement with model estimates, confirming robustness and reliability. Monte Carlo simulations indicated that 15–20 mg/kg every 6 hours achieved therapeutic targets in patients with normal renal function, while those with impaired renal function (<60 mL/min/1.73 m²) benefited from extended intervals (every 8 hours). Patients with augmented renal clearance (>130 mL/min/1.73 m²) required 20 mg/kg every 6 hours.
Conclusion:
CysC-based eGFR equations, especially the Hoek equation, more accurately predict vancomycin clearance than SCr-based or combined equations in PICU patients. Using CysC-based renal assessment may improve dosing precision, target attainment, and safety, particularly in patients with impaired renal function or augmented renal clearance. These findings support integrating CysC testing into pediatric vancomycin dosing strategies and highlight the need for broader clinical acceptance and additional multicenter research to validate these results further.
References:
• Totapally BR, Machado J, Lee H, Paredes A, Raszynski A. Acute kidney injury during vancomycin therapy in critically ill children. Pharmacotherapy. 2013 Jun;33(6):598–602. doi:10.1002/PHAR.1259 PubMed PMID: 23526674
• Feiten HDS, Okumura LM, Martinbiancho JK, Andreolio C, Da Rocha TS, Antonacci Carvalho PR, et al. Vancomycin-associated Nephrotoxicity and Risk Factors in Critically Ill Children Without Preexisting Renal Injury. Pediatr Infect Dis J. 2019 Sep 1;38(9):934–8. doi:10.1097/INF.0000000000002391 PubMed PMID: 31232892.
• Hartman SJF, Brüggemann RJ, Orriëns L, Dia N, Schreuder MF, de Wildt SN. Pharmacokinetics and Target Attainment of Antibiotics in Critically Ill Children: A Systematic Review of Current Literature. Clinical Pharmacokinetics. 2020. doi:10.1007/s40262-019-00813-w
• Barreto EF, Rule AD, Murad MH, Kashani KB, Lieske JC, Erwin PJ, et al. Prediction of the Renal Elimination of Drugs With Cystatin C vs Creatinine: A Systematic Review. Mayo Clin Proc. 2019 Mar 1;94(3):500–14.
• Dharnidharka VR, Kwon C, Stevens G. Serum cystatin C is superior to serum creatinine as a marker of kidney function: A meta-analysis. American Journal of Kidney Diseases. 2002;40(2). doi:10.1053/ajkd.2002.34487
• Downes KJ, Zane NR, Zuppa AF. Effect of Cystatin C on Vancomycin Clearance Estimation in Critically Ill Children Using a Population Pharmacokinetic Modeling Approach. Ther Drug Monit. 2020;42(6). doi:10.1097/FTD.0000000000000796
• Downes KJ, Zuppa AF, Sharova A, Neely MN. Optimizing Vancomycin Therapy in Critically Ill Children: A Population Pharmacokinetics Study to Inform Vancomycin Area under the Curve Estimation Using Novel Biomarkers. Pharmaceutics. 2023;15(5). doi:10.3390/pharmaceutics15051336
• Samman A, Al-Jelaify M, Alsunaidi A, Temsah MH, Alhasan KA, Alsultan A. Impact of eGFR estimation methods on drug dosing in pediatric ICU: a study of serum creatinine vs. cystatin C. European Journal of Clinical Pharmacology 2026 82:3. 2026 Feb 6;82(3):63-. doi:10.1007/s00228-025-03938-z.
Reference: PAGE 34 (2026) Abstr 12184 [www.page-meeting.org/?abstract=12184]
Poster: Drug/Disease Modelling - Other Topics