Defining Target Vancomycin Trough Concentrations for Treating Staphylococcus aureus Infection in Infants Aged 0 to 90 Days

Target trough concentrations for vancomycin hydrochloride therapy in infants younger than 90 days are poorly defined. In adults, the pharmacodynamic target for Staphylococcus aureus correlates with the ratio of the area under the concentration-time curve over a 24-hour period (AUC₂₄) to the minimum inhibitory concentration (MIC) of the bacteria exceeding 400 (ie, AUC₂₄/MIC>400).^1,2 There are no widely available AUC calculators that we know of for use in infants younger than 90 days to enable clinicians to determine the AUC₂₄ at steady state (approximately 48 hours) based on a trough concentration. We developed a population-based pharmacokinetic model to determine the association between serum trough vancomycin concentrations and AUC₂₄.

Data were collected from 104 young infants enrolled in a randomized clinical trial of vancomycin between September 1, 2014, and December 31, 2017.⁴ Data were analyzed using nonlinear mixed-effect modeling (NONMEM version 7; ICON plc) interfaced with PLT tools from September 1, 2018, to December 31, 2018. The first-order conditional estimation method with interaction was used. The covariates of weight, sex, serum creatinine concentration, and albumin level were added in a stepwise manner to the model. Model accuracy was assessed using goodness-of-fit plots and a visual predictive check. The study was approved by the Royal Children’s Hospital Melbourne Human Research Ethics Committee (HREC 34030) and the South Eastern Sydney Local Health District Human Research Ethics Committee (SSA 16/G/335), and parents provided written informed consent to have blood samples taken.

Simulations were performed using Matlab version 9.1.0.441655 (R2016b) (MathWorks). Covariates from a separate database of 1400 infants admitted to the same hospital over an 18-month period aged 0 to 90 days with normal serum creatinine values were used to generate virtual patients. Institutional review board approval was not needed for this cohort because the data were deidentified. Simulations were stratified to generate 1000 concentration-time profiles for each dosage interval based on the British National Formulary for Children guideline.

The final model was a 2-compartment model with a sigmoidal maturation model (MATMOD) for clearance³ and included weight (WT) and serum creatinine (Cr) as covariates, as follows:

where CL0 is the clearance at postmenstrual age (PMA) 0; θ_WT,CL is the exponent for weight on clearance; θ_Cr,CL is the exponent for creatinine on clearance; and θ_WT,Vc is the exponent for weight on volume of the central compartment (V_central). The θ_WT,CL was fixed at 0.75 because the model estimate was unstable. Goodness-of-fit plots and visual predictive check showed that the model had good predictive performance.

For S aureus with an MIC of 1 mg/L, the probability of attaining an AUC₂₄ greater than 400 mg.h/L for the various dosage intervals are shown in the Figure. The 48-hour trough concentrations that provided a 90% probability of target attainment were greater than 15 to 20 mg/L for the 6-hour dosage interval and greater than 8 to 15 mg/L for the 12-hour dosage interval. Trough concentrations of approximately 30 mg/L were required to achieve an AUC₂₄ greater than 800 mg.h/L (for an MIC of 2) for a 6-hour dosage interval and would not be achievable without increasing the risk of drug-related toxic effects.

To our knowledge, this is the first vancomycin population pharmacokinetic model in young infants developed from prospectively collected data from a randomized clinical trial.⁴ Studies in adults suggest that trough concentrations of 15 to 20 mg/L are required to achieve an AUC₂₄ greater than 400 mg.h/L, and these targets are often applied to young infants.⁵ These recommendations from the adult studies are intended for a 12-hour dosage interval. In contrast, in young infants, dosage intervals range between 6 and 24 hours; therefore, the same trough targets cannot be extrapolated. With less-frequent administration of vancomycin, lower trough concentrations can be targeted and the same AUC₂₄ achieved. Our results differ from a previous pharmacokinetic study of young infants who received the drug at 12- or 24-hour intervals. That study reported that with trough concentrations of 7 to 11 mg/L, an AUC₂₄ greater than 400 mg.h/Lwould be achieved in more than 90% of infants.⁶ This difference may be partially explained by the retrospective data set used in the previous model.

Study limitations include potential sampling or data errors; however, data were prospectively collected after appropriate staff training. Also, because our study population were young infants hospitalized in neonatal units, results can only be generalized to those settings.

The results of this study have important implications for the successful treatment of S aureus infections with vancomycin in young infants. The association between trough concentration and AUC₂₄ for different dosage intervals (Figure) can be used to aid clinicians in the interpretation of trough concentrations in young infants with sepsis.

Accepted for Publication: February 25, 2019.

Published Online: June 10, 2019. doi:10.1001/jamapediatrics.2019.1488

Correction: This article was corrected on August 5, 2019, to correct vancomycin trough concentration values in the Results; and on October 14, 2019, to fix errors in the Results and Figure because of a transcription error for the weight of one of the included infants in the data set.

Corresponding Author: Amanda Gwee, PhD, Department of Paediatrics, The University of Melbourne, The Royal Children’s Hospital Melbourne, 50 Flemington Rd, Parkville, Victoria 3052, Australia (amanda.gwee@rch.org.au).

Author Contributions: Dr Gwee had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Gwee, Cranswick, McMullan, Hunt, Curtis.

Acquisition, analysis, or interpretation of data: Gwee, Cranswick, McMullan, Bolisetty, Duffull.

Drafting of the manuscript: Gwee, Cranswick, Bolisetty.

Critical revision of the manuscript for important intellectual content: Gwee, Cranswick, McMullan, Hunt, Curtis, Duffull.

Statistical analysis: Gwee, Duffull.

Obtained funding: Gwee, Cranswick.

Administrative, technical, or material support: Gwee, Cranswick, McMullan, Bolisetty, Hunt.

Supervision: Cranswick, McMullan, Hunt, Curtis, Duffull.

Conflict of Interest Disclosures: None reported.

Funding/Support: Staff and materials for data collection for this work were funded by the Murdoch Children’s Research Institute. Dr Gwee received stipend support from the Royal Australasian College of Physicians Basser Research Entry Scholarship as well as a National Health and Medical Research Council postgraduate scholarship GNT1150216.

Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.