Comparison of fitting methods for the analysis of plasma concentration‐time data resulting from constant rate intravenous infusion

Plasma concentration‐time data resulting from constant rate intravenous infusion may be analysed in two ways: Samples may be collected both during and after infusion and fit to an infusion model. Samples may be collected after infusion is complete and the data may be fit as an i.v. bolus. The purpose of this study was to contrast these two fitting procedures in terms of the accuracy of the parameter values obtained. Concentration‐time data were computer‐generated with the introduction of random error to simulate the disposition profiles of two model drugs. The results of these simulations indicate that satisfactory values for area‐dependent parameters may be obtained without fitting data during the infusion phase. The exception to this is the apparent steady‐state volume whose values become less accurate with longer infusion times. The parameters most affected by ignoring data points in the infusion phase are the central volume of distribution, and the coefficient and disposition rate constant associated with the initial, rapid phase of disposition. The equation which describes the entire concentration‐time profile provides the most accurate parameter estimates of the model equation. In addition, we also describe the influence of the fitting method on the intercompartmental transfer rate constants.