Dosing in Children - The First and Next Dose

Abstract

Background: Pharmacokinetic (PK) in children is different from those in adult, particular in neonates and infants. Children primarily differ from adults in terms of size while infants and neonates differ also in their maturation of structure and function. These differences should be considered in optimizing individualised dosage regimens, particularly, in drugs that have narrow therapeutic index. A single target has a better balance between benefit and adverse effects. Population PK models can restore old experience of drug behaviour to the care of new patient. Design the initial and revised dosage regimens for the next patient who appears to belong to that particular population. FirstDose (initial) and NextDose (revised) dosing methods are taking into account the variability in children, infants and neonates for dosing individualisation with using population PK models Aim: 1) To evaluate the antibiotic dosing calculator clinical trial with using FirstDose web based and how closed the target concentration is achieved when comparing a department protocol for dosing with the web based. 2) To find the best methods for initial and revised doses for busulphan that hit the target concentration of busulphan. Methods: FirstDose and NextDose are web base tools that used for dosing individualisation. FirstDose was used existing PK models to calculate the doses regimen and it showed expected time course of concentration and times of target concentration measurements. The NextDose was used Bayesian mixed effect model, which incorporating between subject and occasion variability, and busulphan audit data was used to develop and test this method. Results: The prospective study from the antibiotic dosing clinical trial showed no significant different between department protocol and dose calculator arms p<0.05. However, more than half of clinicians surveyed found the antibiotic dose calculator was better than existing methods. The retrospective study for busulphan initial dose was shown a significant different between allometric scaling and current initial dosing methods, absolute mean error is 21.29 ± 12.53 mg/d (p<0.05). The daily dose from the actual method used by clinicians for dosing adjustment of busulphan and the AUC predicted dose was observed a significant different in the mean percentage difference (-27.16 ± 50.05) p<0.05. However, there is no significant difference between the actual method and the Bayesian method (-17% ± 20%) p>0.05. Conclusion: Web based tools would be linked with scientific theory and the clinical application and might be a future for dosing individualisation. First and Next Dosing method had reduced dosing error by removing calculations usually performed manually.