The Wagner-Nelson method applied to a multicompartment model with zero order input.

It is shown that the Wagner-Nelson absorption method provides zero-order input rate constants exactly, or with small error, in a large number of cases where the two-compartment open disposition model applies. Factors affecting the accuracy of the method were studied with error-free simulated data. The method was applied to real data for three drugs. With ethanol infused over 2 h in eight human trials the estimated rate constant averaged 99.6 per cent of the known rate constant with a coefficient of variation of 6.86 per cent. With pindolol infused over 3 h five human subjects the estimated rate constant averaged 98.7 per cent of the known rate constant with a coefficient of variation of 22.5 per cent. With theophylline administered orally in a sustained-release form to seven human subjects the Wagner-Nelson method provided estimated zero-order rate constants which averaged 95.8 per cent of those estimated by an exact two-compartment absorption equation with a coefficient of variation of 38.1 per cent (in this case bolus intravenous data were available for the same subjects).