The primary hypothesis of this confirmatory study is that lithium therapy will acutely decrease suicidal ideation and/or suicidal behaviour in inpatients with a major depressive episode (MDE, unipolar and bipolar disorder according to DSM IV criteria). The specific aim is to test the hypothesis that lithium plus treatment as usual (TAU), compared to placebo plus TAU, results in a significantly greater decrease in suicidal ideation and/or behaviour over 5 weeks in inpatients with MDE.

The study will consist of 254 adult patients, hospitalized for a major depressive episode with suicidal ideation and/or behaviour who will be randomized to two groups of each 127 participants. All study participants will receive treatment as usual in the hospital, including (psycho)pharmacological treatment and, if needed, psychotherapy plus a 5-week course of lithium or placebo. Rating scales will be used daily resp. weekly to measure suicidal ideation and/or behaviour, depression, anxiety, mixed/manic features and impulsiveness. Experienced clinicians will rate suicidal ideation and/or behaviour as well as depression twice a week and anxiety, mixed/manic features and impulsiveness once a week.

Inclusion Criteria:

• Major depressive episode; inpatient at screening visit; suicidal ideation/behaviour present defined by a clinical rating of 8 on the Sheehan Suicidality Tracking Scale (S-STS) and a rating of ≥20 on the Montgomery Asberg Depression Scale (MADRS) at both screening and baseline visits; both gender, age 18 years.

Exclusion Criteria:

• Contraindication for and history of lithium treatment within the past 6 months; patient unable to tolerate lithium treatment in the past; comorbid borderline/antisocial personality disorder, currently active substance dependency; patients with acute or unstable severe medical conditions, patients unable to understand the informed consent or involuntary inpatients, positive toxicology screen (illegal drugs), pregnancy and lactation