To investigate the bioavailability of rivastigmine, an approved therapy for patients with mild to moderate dementia of the Alzheimer's type, at the highest approved single dose of 6 mg.Randomised, two-period crossover, single-centre, non-blinded, inpatient study.Eleven patients (five females and six males) with mean age 69.5 years.The 6 mg oral dose was compared with a 2 mg intravenous dose of rivastigmine infused over a 1-hour period. Plasma concentrations of rivastigmine and its metabolite NAP 226-90 were measured with a gas chromatographic/mass spectrometric method.Following oral administration of a single 6 mg capsule, rivastigmine is rapidly absorbed with an average time to peak plasma concentration of about 1 hour and an average peak concentration of about 25.6 g/L. By a noncompartmental approach, the absolute bioavailability of the 6 mg oral dose of rivastigmine was 71.7% when compared with a 2mg intravenous infusion normalised for dose. By using a population pharmacokinetic model with Michaelis-Menten elimination, absolute bioavailability was estimated at 60.2%. The average terminal elimination half-life of rivastigmine ranged from 1.4 to 1.7 hours for both treatments. Plasma concentrations of the major metabolite, NAP 226-90, formed by the hydrolysis of rivastigmine by cholinesterase are lower than those of the parent compound following oral and intravenous administration.A noncompartmental approach and a compartmental approach based on a population pharmacokinetic model with Michaelis-Menten elimination yielded comparable values, 71.7% and 60.2% respectively, for the absolute bioavailability of a single 6 mg oral dose of rivastigmine. Comparison with previous studies confirmed that the oral form of the drug exhibits increased bioavailability with increasing dose, consistent with its nonlinear pharmacokinetics..