The daily warfarin dose for each patient was calculated using the published regression equation: ��dose?= ?0��009?+ 0��011 (height)?+?0��357 (VKORC1)???0��478 (CYP2C9*3)?? 0��277 (CYP2C9*2)?+?0��186 (Indication), with the following keys: input height in centimetres; VKORC1 genotype: input 0 for AA, 1 for AG, and 2 for GG; CYP2C9*3 genotype: input 0, 1, or 2 for the number of Bleomycin
*3 alleles; CYP2C9*2 genotype: input 0, 1 or 2 for the number of *2 alleles; Indication: input 0 for Fontan procedure, 1 for other indications (Biss et?al, 2012). The square root of the predicted daily warfarin maintenance dose was compared to the square root of the actual daily dose using Pearson's correlation analysis. A P value of <0��05 was taken as statistically significant. Statistical analysis was performed using MiniTab v15.0 (Coventry, UK). Forty-nine children were studied. A summary of the demographic, <a href="http://www.selleckchem.com/products/Rapamycin.html
">Rapamycin datasheet clinical and genetic characteristics of the derivation and the validation cohorts is shown in Table?1. Median age was 11��4?years (range: 1�C18?years) for the derivation cohort and 7��2?years (range: 0�C17?years) for the validation cohort. All genotypes were in Hardy�CWeinberg equilibrium. The predicted mean (��SD) daily warfarin dose was 3��3?��?1��8?mg (range: 0��2�C8��1?mg) and the actual mean daily warfarin dose was 3��3?��?2��0?mg (range: 0��6�C9��1?mg). Pearson's correlation analysis showed a close and highly significant relationship between the square root of the actual warfarin dose and the square root of the predicted dose (r?=?0��833, P?<?0��001) with regression equation: ��actual warfarin dose (mg)?=?0��329?+?0��825 ��predicted warfarin dose (mg). The square roots of the doses provide a better statistical correlation (as this transformation stabilises the variance and gives an approximately Normal <a href="http://www.selleckchem.com/products/MG132.html
distribution). However, to aid data interpretation Fig?1 shows the relationship between the absolute values of predicted warfarin dose and actual dose. There were some differences between the derivation and the validation cohorts (Table?1), particularly the younger age of the validation cohort. A greater proportion of the validation cohort was anticoagulated for a prosthetic heart valve and this is reflected in the higher number with a higher target INR range. Despite this, the pharmacogenetics�Cbased warfarin dosing equation performed well in the prediction of warfarin dose. This study provides further support for pharmacogenetics-guided warfarin dosing in children. The relation of maintenance warfarin dose to loading dose has not been explored in children and the equation may need to be modified if it is to be used to predict warfarin dose during initiation of anticoagulant therapy. The clinical utility of a pharmacogenetics-based warfarin dosing, in terms of a reduction in adverse events, would require prospective randomised controlled trials.