A Generalized Numerical Deconvolution Procedure for Computing Absolute Bioavailability — Time Profiles

Abstract

The use of either body fluid measurements of drug concentration or pharmacological response intensity data for the computation of drug absorbed as a function of time requires a mathematical model for the system and the ability to perform a deconvolution of the response. Deconvolution procedures can also be used in verifying a model’s accuracy to represent any given system by comparing an actual input into the system to that calculated by using these methods. For simple one, two and three-compartment models, analytical expressions have been presented in the literature to determine drug bioavailability. Such expressions are specific and restricted to a particular configuration and order of the system. A generalized numerical deconvolution procedure has been developed to compute the amount of drug absorbed for any order system. Its use has been illustrated with pharmacological data, however, it applies equally well to drug concentration measures from body fluids. The procedure described here is simple and is easily implemented on a digital computer. While some deconvolution schemes tend to be unstable, the algorithm presented here is inherently stable. Furthermore, this procedure requires a minimum of core memory and central processor time to execute.