PARAMETRIZATION BY NONLINEAR-REGRESSION ANALYSIS OF THE ACTIVE ACID METABOLITE OF EBASTINE USING DIFFERENT WEIGHTING METHODS

The study of the analytical error function has been applied to evaluat e the pharmacokinetics of ebastine active acid metabolite after a sing le oral dose of 10 mg of metabolite in 6 healthy volunteers. Plasma co ncentrations were measured using an automated solid phase extraction m ethod. The analytical method was fully validated and the function of t he analytical error subsequently determined. The parametric approach w as performed by nonlinear regression analysis comparing the results ob tained after the application of different weighted least square method s including the homoscedastic method (W = 1) and heteroscedastic metho ds using weights of 1/C, 1/C-2 and the inverse of the concentration va riance calculated through the function of the analytical error. The re sults obtained in the study showed no influence of the different weigh ting methods used on the estimation of the pharmacokinetic parameters of ebastine acid metabolite.