Kim Dao (1), Monia Guidi (1,2), Pascal André (1), Eric Giannoni (3), Thierry Buclin (1), Chantal Csajka (1,2)
(1)Service of Clinical Pharmacology, Lausanne University Hospital, Lausanne, Switzerland, (2) School of pharmaceutical sciences, University of Geneva, University of Lausanne, Geneva, Switzerland.(3)Service of Neonatology, Department Mother-Woman-Child, Lausanne University Hospital, Lausanne, Switzerland.
Objectives:
Several neonatal pharmacokinetic (PK) models of vancomycin and dosing algorithms have been elaborated in the last three decades. (1) Among 20 different dosing approaches, retrieved from international guidelines, protocols used in neonatal intensive care units (NICUs) in Switzerland and published models, most use age, renal function and body weight as covariates to establish the appropriate starting dosage. An absence of consensus regarding which dosing regimen is best suited for neonates is nevertheless being observed, probably in relation with the absence of well-defined therapeutic intervals. Our objectives were first to build a population PK model of vancomycin in a large cohort of neonates and secondly, to compare by simulation vancomycin exposures under existing dosage recommendations with the aim of harmonizing vancomycin dosing in neonates admitted in NICUs in Switzerland.
Methods:
Based on a large neonatal therapeutic drug monitoring population, 1848 vancomycin concentrations from 405 neonates were included in the analysis. A one-compartment model with first-order elimination was developed while testing the influence of age (gestational age (GA), postnatal age (PNA), postmenstrual age (PMA=GA+PNA)), current body weight (WT) and renal function (creatinine (CRT)) related covariates known to influence vancomycin blood levels in neonates (NONMEM®). Vancomycin concentration-time profiles resulting from the 20 identified dosing regimens were predicted applying the final model developed in our population. The exposure target was defined as the ratio of the area under the curve from 0 to 24 h (AUC0-24) over the minimal inhibiting concentration (MIC) after the first dose ≥ 400 based on the recommended target for Methicillin Resistant Staphylococcus Aureus (MRSA) of ≤1 mg/L (2), to the potentially toxic AUC0-24 of ≥ 700 mg·h/L (3). Exposure to vancomycin was evaluated after 1 day of treatment to assess early exposure. AUC0-24 was derived by numerical integration in NONMEM©. Exposures to vancomycin were then expressed as the proportion of patients in the target of AUC0-24/MIC between 400 and AUC0-24 < 700 mg·h/L.
Results:
A one-compartment model adequately described vancomycin concentrations. WT using an allometric scaling was straightaway included on all PK parameters. An interpatient variability was assigned on clearance (CL), but no interindividual variability could be estimated on the volume of distribution. Among all tested covariates, PMA using a sigmoid Emax maturation function and CRT had the most salient impact on vancomycin CL. The significant covariates explained altogether 43% of interpatient variability in CL. An increase of CRT from 52 to 114 µmol/L (for a neonate of 1.2 kg and a PMA of 32 weeks) will reduce CL by 46%. A decrease of PMA from 32 to 28 weeks will reduce CL by 48%, whereas an increase in PMA from 32 to 36 weeks will increase CL by 27%. In the final model, the average CL was 0.268 L/h (CV 22.5%) for a WT of 1.2 kg, a PMA of 32 weeks and a CRT 52 µmol/L and the average volume of distribution was 0.629 L.
Out of the 20 evaluated regimens, the median proportion of neonates predicted within and over the target was 39% (range 4 – 77%) and 1% (range 0 – 68%), respectively. Three of these regimens, Neonatal Formulary 7 (4), Neofax® (5) and the dosing algorithm published by Janssen E et al (6) were associated with the highest proportion of target attainment after 24h of treatment, in 66%, 67% and 77% of patients; the proportion of patients over target was 2%, 2% and 15%, respectively. The first two international guidelines propose simple dose stratifications based on seven different categories of PMA and/or postnatal age (PNA), whereas the very detailed dosing algorithm from Janssen et al includes 19 levels based on PNA and body weight following a loading dose. The simple dosing regimen issued from the summary of product characteristics for vancomycin (i.e. a loading dose of 15 mg/kg, a maintenance dose of 20 mg/kg, every 12 hours for a PNA < 7 days and every 8 hours for a PNA > 7 days) ranks close to them with 65% of patients in target (and none over).
Conclusions:
These results suggest that simpler but also more elaborated vancomycin dosing regimens do only allow the attainment of an optimal AUC0-24h/MIC ≥ 400 and a AUC0-24 < 700 mg·h/L in 65 – 77 % of patients. It remains to determine if a better algorithm can be designed to optimize early vancomycin treatment in neonates without increasing toxicity, while favoring parsimony.
References:
[1] Marsot A, Boulamery A, Bruguerolle B, Simon N. Population pharmacokinetic analysis during the first 2 years of life: an overview. Clin Pharmacokinet. 2012;51(12):787-98.
[2] Rybak MJ, Lomaestro BM, Rotschafer JC, Moellering RC, Jr., Craig WA, Billeter M, et al. Therapeutic monitoring of vancomycin in adults summary of consensus recommendations from the American Society of Health-System Pharmacists, the Infectious Diseases Society of America, and the Society of Infectious Diseases Pharmacists. Pharmacotherapy. 2009;29(11):1275-9.
[3] Neely MN, Youn G, Jones B, Jelliffe RW, Drusano GL, Rodvold KA, Lodise TP. Are vancomycin trough concentrations adequate for optimal dosing? Antimicrob Agents Chemother. 2014;58(1):309-16.
[4] Neonatal Formulary accessed on : http://www.neonatalformulary.com/drugs.asp
[5] Neofax 2011®, Montvale, NJ, Thomson Reuters, 2011, pp 102-3.
[6] Janssen EJ, Valitalo PA, Allegaert K, de Cock RF, Simons SH, Sherwin CM, et al. Towards Rational Dosing Algorithms for Vancomycin in Neonates and Infants Based on Population Pharmacokinetic Modeling. Antimicrobial agents and chemotherapy. 2015;60(2):1013-21.
Reference: PAGE 27 (2018) Abstr 8599 [www.page-meeting.org/?abstract=8599]
Poster: Drug/Disease Modelling - Paediatrics