A LIMITED SAMPLING MODEL WITH BAYESIAN-ESTIMATION TO DETERMINE INULINPHARMACOKINETICS USING THE POPULATION-DATA MODELING PROGRAM P-PHARM

This study describes the methodology used to calculate the individual clearance (CL) and volume of distribution (Vd) of inulin using 1 or 2 blood samples taken during the disposition and elimination phase after a single intravenous perfusion, and the population parameters. The me an population parameters and their interindividual variability were ob tained from an initial group of 90 patients including 38.5% who had di abetes, 49% who were obese, and 12.5% who were diabetic and obese. Amo ng these patients, 44.5% had normal renal function (creatinine clearan ce ranging from 70 to 150 ml/min/1.73m(2)) and 20% showed renal insuff iciency with a creatinine clearance ranging from 15 to 60 ml/min/1.73m (2). A 2-compartment model was fitted to the population data using P-P HARM. The population parameter estimates of CL and Vd were 6.85 +/- 1. 04 L/h and 4.95 +/- 0.84L, respectively. The interindividual variabili ty of CL was explained by a linear dependency between serum creatinine and body area. The interindividual variability of Vd was explained by a linear dependency with body area. A test group of 25 additional pat ients was used to evaluate the predictive performance of the populatio n parameters. Seven blood samples were collected from each individual in order to calculate individual parameter estimates using standard fi tting procedures. These values were compared with those estimated by m eans of a Bayesian approach using population parameters and 1 or 2 sam ples selected from the individual observations. The results show that the bias of CL and Vd, estimated using either 1 or 2 samples, was not statistically different from zero, and that the precision of these par ameters was excellent.