As lignocaine clearance is influenced by factors such as cardiac failure and liver impairment, clinical pharmacokinetic principles should be used to account for kinetic variability so that target concentrations are achieved consistently throughout the course of intravenous therapy. Two groups of patients with ischaemic heart disease, who received lignocaine, were studied: a control group with no feedback or intervention from therapeutic drug monitoring, and an intervention group in which strict guidelines for lignocaine administration were introduced. Lignocaine plasma concentrations were measured by EMIT (Syva), and rapid feedback of concentration data in the intervention group allowed adjustment of infusion rates using the Chiou equation. The mean concentration in the intervention group remained within the therapeutic range (2-5 micrograms/ml) at all times, whereas it exceeded 5 micrograms/ml after the first 7 h in the control group. The distribution of concentrations in the intervention group was always narrower than that in the control group. The study also included a comparison of the ability of the Chiou equation and a Bayesian optimisation procedure to estimate pharmacokinetic parameters and to forecast lignocaine concentrations over various periods of time. There was no significant difference between prediction errors determined by the two methods at various points throughout a 32-h period; both methods were associated with a negative prediction bias beyond the first 12 h of infusion. It is likely that this reflects assumptions made about lignocaine clearance and indicates the need for more sophisticated kinetic models.