In recent years, potential new antidepressants have been developed that inhibit serotonin (5-hydroxytryptamine; 5-HT) and noradrenaline (norepinephrine) reuptake in a selective manner. Examples of these serotonin/noradrenaline reuptake inhibitors (SNRIs) are duloxetine, milnacipran and venlafaxine. All 3 compounds effectively inhibit the serotonin and noradrenaline transporter, as judged from in vitro and in vivo tests. However, potency and selectivity differ between them. The concentration of drug that inhibits uptake by 50% (IC50) for serotonin and noradrenaline, respectively, in rat brain preparations are 2.6 and 7 nmol/L for duloxetine, 203 and 100 nmol/L for milnacipran and 210 and 640 nmol/L for venlafaxine. Duloxetine and venlafaxine are extensively metabolised to demethylated compounds that also inhibit serotonin and noradrenaline uptake, whereas milnacipran lacks active metabolites. Venlafaxine, but not milnacipran, downregulates beta-adrenoceptor-mediated responses after single dose and repeated administration. This has been suggested to contribute to a rapid antidepressant action of the former agent. However, since other established antidepressants, such as several selective serotonin reuptake inhibitors and milnacipran, do not cause this effect on beta-adrenoceptors after single dose and repeated administration, the relationship between changes in these receptors and antidepressant action is unclear. Controlled studies in depressed patients have shown an efficacy of milnacipran and venlafaxine superior to placebo and comparable to that of reference antidepressants. Furthermore, SNRIs are devoid of some of the undesirable adverse effects common to first generation antidepressants (i.e. tricyclics and monoamine oxidase inhibitors). This may be due to a lack of interaction with aminergic receptors, and results in better compliance (although the incidence of adverse events of SNRIs, mainly gastrointestinal, increases with dose). The rapid onset of antidepressant effects has been reported in patients with major depression who were treated with venlafaxine and, to a lesser extent, with milnacipran. Although promising, these results need to be interpreted with caution, given the methodological difficulties in measuring appropriately the onset of antidepressant action and the few studies specifically addressing this issue.