Helder Duarte(1), Ana Mirco(2), Paulo Paixão(1), José A.G. Morais(1), Fátima Falcão(2), Nuno Silva(1)
(1) Faculdade de Farmácia da Universidade de Lisboa, Lisboa, Portugal (2) Centro Hospitalar Lisboa Ocidental, Lisboa, Portugal
Objectives: To develop a population pharmacokinetic model of amikacin in premature newborns babies. From the estimated population model, establish new therapeutic protocol for initial therapy.
Methods: The population pharmacokinetic model was developed from retrospectively collected data of 90 premature newborn babies, with a total of 302 individual concentration-time points, using a nonlinear mixed effects model with Monolix® software(1). Demographic, clinical and analytic data were used to explain between subject variability on each PK parameter. Simulations from the developed popPK model were conducted in order to evaluate the current therapeutic protocol in the clinical setup, as well as to improve the achievement of the therapeutic goals.
Results: Pharmacokinetic profile of amikacin after short term IV infusion was adequately described with a one compartment model with first order elimination and an additive and proportional residual error model. Body weight was identified as predictive of both volume of distribution and serum clearance. Creatinine clearance, post menstrual age and the need of parenteral nutrition were also identified as predictive of individual clearance. In the final model, volume was estimated as 32.7 (L/70 kg). Clearance was estimated as 2.07 (L/h/70kg) for patients without parenteral nutrition and 1.52 (L/h/70 kg) for those with parenteral nutrition needs. These figures are consistent with current literature(2). Unexplained between subject variability was 37.6% for volume and 36.9% for clearance. Residual variability was 0.335 mg/L with a proportional component of 35.4%. Evaluation through simulations of the currently implemented therapeutic protocol in the clinical setup showed that more than 50% of the patients had the potential to present either sub therapeutic or toxic peak concentrations. New recommendations improved the percentage of peak concentrations within therapeutic goal to more than 65%, without worsening of the expected trough concentrations.
Conclusions: Our final popPK model showed a 40% and 50% reduction of unexplained inter-individual variability on volume and clearance respectively. New initial therapeutic recommendations were able to improve, from ca. 40% to ca. 60% the number of peak plasma concentrations within therapeutic target. An easily readable table based on the new recommendations is proposed for clinical use.
References:
[1] Lavielle M, Mentré F. Estimation of Population Pharmacokinetic Parameters of Saquinavir in HIV Patients with the MONOLIX Software. J Pharmacokinet Pharmacodyn [Internet]. 2007 Mar 30;34(2):229–49. Available from: http://link.springer.com/10.1007/s10928-006-9043-z
[2] Murphy J. Aminoglycosides. Clinical Pharmacokinetics: Pocket Reference. 1993. p.1-20
Reference: PAGE 25 (2016) Abstr 5976 [www.page-meeting.org/?abstract=5976]
Poster: Drug/Disease modeling - Paediatrics