The recently introduced antidepressants, the selective serotonin reuptake inhibitors (SSRIs) [citalopram, fluoxetine, fluvoxamine, paroxetine and sertraline], are known for their clinical efficacy, good tolerability and relative safety. They differ from each other in chemical structure, metabolism and pharmacokinetic properties. Therapeutic drug monitoring of these compounds is not widely used, as the plasma concentration ranges within which clinical response with minimal adverse effects appears to be optimal are not clearly defined. Almost all recent assays developed for the quantitative determination of SSRIs and their metabolites in blood are based either on the separation of SSRIs by high performance liquid chromatography (HPLC) or gas chromatography (GC). Citalopram and fluoxetine have been introduced as racemic compounds. There are some differences in the pharmacological profile, metabolism and pharmacokinetics between the enantiomers of the parent compounds and their demethylated metabolites. Stereoselective chromatographic methods for their analysis in blood are now available. With regard to the SSRIs presently available, no clearcut plasma concentration-clinical effectiveness relationship in patients with depression has been shown, nor any threshold which defines toxic concentrations. This may be explained by their low toxicity and use at dosages where serious adverse effects do not appear. SSRIs vary widely in their qualitative and quantitative interaction with cytochrome P450 (CYP) isozymes in the liver. CYP2D6 is inhibited by SSRIs, in order of decreasing potency paroxetine, norfluoxetine, fluoxetine, sertraline, citalopram and fluvoxamine. This may have clinical consequences with some but not all SSRIs, when they are taken with tricyclic antidepressants. Except for citalopram and paroxetine, little is known about the enzymes which control the biotransformation of the SSRIs. There have been many reports on marked pharmacokinetic interactions between fluoxetine and tricyclic antidepressants. Fluoxetine has a stronger effect on their hydroxylation than on their demethylation. Interactions observed between fluoxetine and alprazolam, midazolam and carbamazepine seem to occur on the level of CYP3A. Fluvoxamine strongly inhibits the N-demethylation of some tricyclic antidepressants of the tertiary amine type and of clozapine. This may lead to adverse effects but augmentation with fluvoxamine can also improve response in very rapid metabolisers, as it increases the bioavailability of the comedication. Fluvoxamine inhibits with decreasing potency, CYP1A2, CYP2C19, CYP2D6 and CYP1A1, but it is also an inhibitor of CYP3A. Fluoxetine and fluvoxamine have shown to increase methadone plasma concentrations in dependent patients. Some authors warn about a combination of monoamine oxidase (MAO) inhibitors with SSRIs, as this could lead to a serotonergic syndrome. Studies with healthy volunteers suggest, however, that a combination of moclobemide and SSRIs, such as fluvoxamine, should not present serious risks in promoting a serotonin syndrome. A combination of moclobemide and fluvoxamine has successfully been used in refractory depression, but more studies are needed, including plasma-concentration monitoring, before this combined treatment can be recommended. Paroxetine is a substrate of CYP2D6, but other enzyme(s) could also be involved. Its pharmacokinetics are linear in poor metabolisers of sparteine, and non-linear in extensive metabolisers. Due to its potent CYP2D6 inhibiting properties, comedication with this SSRI can lead to an increase of tricyclic antidepressants in plasma, as shown with amitriptyline and trimipramine. CYP3A has been claimed to be involved in the biotransformation of sertraline to norsertraline. Clinical investigations (with desipramine) confirmed in vitro findings that CYP2D6 inhibition by sertraline is only moderate. (ABSTRACT TRUNCATED)