Reboxetine is a novel selective norepinephrine inhibitor that has been evaluated in the treatment of patients with depression. Reboxetine is a racemic mixture, and the (S,S)-(+)-enantiomer appears to be the more potent inhibitor. However, the ratio of the areas under the concentration-time curves of the (S,S)-(+)- and (R,R)-(-)-enantiomers in vivo is approximately 0.5. There is no evidence for chiral inversion. Differences in the clearances of the 2 enantiomers may be explained by differences in protein binding. The pharmacokinetics of reboxetine are linear following both single and multiple oral doses up to a dosage of 12 mg/day. The plasma concentration-time profile following oral administration is best described by a 1-compartment model, and the mean half-life (approximately 12 hours) is consistent with the recommendation to administer the drug twice daily. Reboxetine is well absorbed after oral administration. The absolute bioavailability is 94.5%, and maximal concentrations are generally achieved within 2 to 4 hours. Food affects the rate, but not the extent, of absorption. The distribution of reboxetine appears to be limited to a fraction of the total body water due to its extensive (>97%) binding to plasma proteins. The primary route of reboxetine elimination appears to be through hepatic metabolism. Less than 10% of the dose is cleared renally. A number of metabolites formed through hepatic oxidation have been identified, but reboxetine is the major circulating species in plasma. In vitro studies show that reboxetine is predominantly metabolised by cytochrome P450 (CYP) 3A4; CYP2D6 is not involved. Reboxetine plasma concentrations are increased in elderly individuals and in those with hepatic or renal dysfunction, probably because of reduced metabolic clearance. In these populations, reboxetine should be used with caution, and a dosage reduction is indicated. Ketoconazole decreases the clearance of reboxetine, so that the dosage of reboxetine may need to be reduced when potent inhibitors of CYP3A4 are coadministered. Quinidine does not affect the in vivo clearance of reboxetine, confirming the lack of involvement of CYP2D6. There is no pharmacokinetic interaction between reboxetine and lorazepam or fluoxetine. Reboxetine at therapeutic concentrations has no effect on the in vitro activity of CYP1A2, 2C9, 2D6, 2E1 or 3A4. The lack of effect of reboxetine on CYP2D6 and CYP3A4 was confirmed by the lack of effect on the metabolism of dextromethorphan and alprazolam in healthy volunteers. Thus, reboxetine is not likely to affect the clearance of other drugs metabolised by CYP isozymes.