OBJECTIVE: The purpose of this pharmacokinetic study was to investigate the dose-dependent inhibition of model substrates for CYP2D6, CYP2C19 and CYP1A2 by four marketed selective serotonin reuptake inhibitors (SSRIs): citalopram, fluoxetine, fluvoxamine and paroxetine.

METHODS: The study was carried out as an in vivo single-dose study including 24 young, healthy men. All volunteers had been identified as sparteine- and mephenytoin-extensive metabolisers. The volunteers received in randomised order, at weekly intervals, increasing single oral doses of one of the four SSRIs, followed 3 h later by sparteine (CYP2D6), mephenytoin (CYP2C19) and caffeine (CYP1A2) tests. Fluoxetine was given at 3-week intervals because of the long half-life of fluoxetine and its metabolite norfluoxetine. Citalopram, fluoxetine and paroxetine were given in doses of 10, 20, 40 and 80 mg and fluvoxamine was given in doses of 25, 50, 100 and 200 mg.

RESULTS: With increasing doses, there was a statistically significant increase in the sparteine metabolic ratio (MR) (P < 0.01, Page's test for trend) for all four SSRIs. The increase was modest after intake of citalopram and fluvoxamine, while the increase was more pronounced after fluoxetine intake, although no volunteers changed phenotype from extensive metabolisers to poor metabolisers. Three of the six volunteers changed phenotype from extensive metabolisers to poor metabolisers after intake of 40 or 80 mg paroxetine. There was a statistically significant increase in the mephenytoin S/R ratio (P < 0.01, Page's test for trend) with increasing doses of fluoxetine and fluvoxamine, but not after citalopram and paroxetine. However, no volunteers changed phenotype from extensive to poor metabolisers of S-mephenytoin. After intake of fluvoxamine, the urinary excretion of the metabolites related to N3 demethylation of caffeine were below the limit of quantification, whereas there were no significant changes in the urinary caffeine metabolic ratios after intake of the other three SSRIs.

CONCLUSION: This investigation confirms that paroxetine and fluoxetine are potent inhibitors of CYP2D6, that fluvoxamine and fluoxetine are moderate inhibitors of CYP2C19 and that fluvoxamine is a potent inhibitor of CYP1A2 in humans in vivo. The clinical prediction of interaction from single-dose experiments may have to take the degree of accumulation during steady-state after multiple doses into account.