Résumé
The increased risk of suicidal behaviours during the first weeks of an antidepressant treatment remains a controversial issue while treating depression. This risk led regulatory bodies to prompt warnings, although untreated major depressive disorder still represents the main risk factor of suicidal behaviour. These warnings may not represent an adequate solution. Thus, we need to identify the subgroup of depressed patients that present an increased risk of suicidal behaviour after the initiation of an antidepressant. Predictors of risk are mainly related to the depressive disorder on the one hand and to the suicidal behaviour, on the other: (1) severe depression, with early age at onset, and long duration, as well as a lack of response or a worsening of depressive or anxious symptoms after the initiation of the treatment; (2) the past history of suicide attempt, the past history of emergence or worsening of suicidal thoughts or behaviours with a previous antidepressant treatment, and comorbid misuse of substances, personality disorders or psychosocial adversity. Recent data suggest also that when instituting a selective serotonin reuptake inhibitors, using higher doses than recommended may be harmful. Moreover, depressed patients presenting at baseline an active suicidal ideation or a history of suicide attempt seem to display a poorer short-term response to antidepressants. Thus, depressed patients with the highest risk of suicidal ideation or attempts may be poor responders to available antidepressants. This highlights the urgent need to develop efficacious treatments for patients at risk. New perspectives are offered with treatment targeting biological systems involved in the pathophysiology of suicidal behaviour (inflammation, glutamatergic, opioids).