Article Text

PP-31 Kinetically guided dosing of vancomycin in critical ill neonates and young infants treated for sepsis
  1. Pokorna1,
  2. Síma2,
  3. Cerná1,
  4. Allegaert3,
  5. Tibboel4
  1. 1First Medical Faculty Charles University, PRAGUE 2, Czech Republic
  2. 2Department of Pharmacology, PRAGUE, Czech Republic
  3. 3Department of development and regeneration, KU Leuven Herestraat 49 B, LEUVEN, Belgium
  4. 4Intensive Care and Department of Paediatric Surgery, Erasmus MC, ROTTERDAM, Netherlands


Background Vancomycin (Van) is frequently used in ne-onates and young infants treated for sepsis while a need for prospective dosage validation has been documented in the literature1. Open-label, prospective study includ-ing preterm and term neonates (n=40) and young infants (n=16) treated with Van. A median (IQR) age distribution was 34.1 (24–42) gestational weeks in neonates, 5.5 (1.5 -10) months in infants. The primary goal of the study was to perform a pharmacokinetic (PK) study of Van while the secondary aim was to analyse the influence of covariates on PK (body weight-BW, gestational age-GA, postnatal age-PNA, postmenstrual age-PMA, and glomerular filtration rate estimation (eGFR) according to Schwartz formula)).

Methods Individual PK parameters – volume of distribu-tion (Vd), clearance (CL) were calculated in a one-com-partmental PK model based on individual demographic data and observed Van-plasma levels using MWPharm++ software (MediWare, Prague, Czech Republic). Vancomy-cin population PK one-compartmental model was indi-vidualized to maximise fitting of the simulated PK profile curve with observed concentration points in each patient. AUC24 were computed using individualised PK models in MWPharm++ software. Optimal maintenance doses (MD) were calculated for each patient based on vancomycin clearance values using following formula (MD (mg/day)=24× vancomycin CL (L/hod)×25 mg/L, the value of 25 mg/L was chosen as the midpoint of target therapeutic range for intermittent vancomycin (10–40 mg/L). Descrip-tive parameters median, interquartile range (IQR), mean and standard deviation (SD) were calculated using MS Excel 2010 (Microsoft Corporation, Redmond, USA). Lin-ear regression models were used to evaluate the relation-ships of PK parameters with PK covariates using GraphPad Prism 3.02 (GraphPad Software, Inc., La Jolla, USA).

Results The mean (SD) Vd (L/kg) in neonates was 0.73 (0.31), in young infants 0.74 (0.54). The mean (SD) CL (L/h/kg) was 0.052 (0.02) in neonates, 0.0132 (0.058) in young infants. Linear regression models showed a de-crease in normalised Vd (r2=0.3274, p=0.0001) and in-crease in normalised CL with increasing (GA r2=0.6542, p<0.0001) in neonates, while PMA was a PK covariate for Vd (r2=0.3509,p<0.0001) and CL r2=0.6537, p<0.0001) in neonates and for CL (r2=0.5930, p=0.0005) in young infants. BW was the most predictive for vancomycin CL and consequently MD based on linear regression models. The daily MD calculations using the following formulas have resulted in optimal average vancomycin steady-state concentrations: MD (mg/day)=45.46× BW (kg) –24.64 for neonates, and MD (mg/day)=87.43× BW (kg)– 34.30 for young infants.

Conclusion However, since the practical utility of such an equation is very limited, we propose MD nomograms based on these formulas that can be easily used in clinical settings.

Statistics from

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.