A major depressive episode can be categorised as severe based on depressive symptoms, scores on depression rating scales, the need for hospitalisation, depressive subtypes, functional capacity, level of suicidality and the impact that the depression has on the patient. Several biological, psychological and social factors, and the presence of comorbid psychiatric or medical illnesses, impact on depression severity. A number of factors are reported to influence outcome in severe depression, including duration of illness before treatment, severity of the index episode, treatment modality used, and dosage and duration of and compliance with treatment. Potential complications of untreated severe depression include suicide, self-mutilation and refusal to eat, and treatment resistance. Several antidepressants have been studied in the treatment of severe depression. These include tricyclic antidepressants (TCAs), selective serotonin reuptake inhibitors (SSRIs), serotonin-noradrenaline (norepinephrine) reuptake inhibitors, noradrenergic and specific serotonergic antidepressants, serotonin 5-HT(2) receptor antagonists, monoamine oxidase inhibitors, and amfebutamone (bupropion). More recently, atypical antipsychotics have shown some utility in the management of severe and resistant depression. Data on the differential efficacy of TCAs versus SSRIs and the newer antidepressants in severe depression are mixed. Some studies have reported that TCAs are more efficacious than SSRIs; however, more recent studies have shown that TCAs and SSRIs have equivalent efficacy. There are reports that some of the newer antidepressants may be more effective than SSRIs in the treatment of severe depression, although the sample sizes in some of these studies were small. Combination therapy has been reported to be effective. The use of an SSRI-TCA combination, while somewhat controversial, may rapidly reduce depressive symptoms in some patients with severe depression. The combination of an antidepressant and an antipsychotic drug is promising and may be considered for severe depression with psychotic features. Although the role of cognitive behaviour therapy (CBT) in severe depression has not been adequately studied, a trial of CBT may be considered in severely depressed patients whose symptoms respond poorly to an adequate antidepressant trial, who are intolerant of antidepressants, have contraindications to pharmacotherapy, and who refuse medication or other somatic therapy. A combination of CBT and antidepressants may also be beneficial in some patients. Electroconvulsive therapy (ECT) may be indicated in severe psychotic depression, severe melancholic depression, resistant depression, and in patients intolerant of antidepressant medications and those with medical illnesses which contraindicate the use of antidepressants (e.g. renal, cardiac or hepatic disease).