Certains psychotropes (alcool, drogues illicites, médicaments) ont été incriminés, chez l'homme, dans l'apparition de troubles majeurs du comportement par levée d'inhibition psychomotrice. La désinhibition comportementale induite par la prise thérapeutique de psychotrope ou lors d'un mésusage peut faciliter le risque de passage à l'acte auto ou hétéro-agressif. Nous avons donc conduit un travail bibliographique visant à relever les liens de causalité ou d'imputabilité entre la prise d'une substance psychoactive chez l'homme sain et l'apparition conjointe d'un trouble majeur du comportement, d'une part les comportements violents et, d'autre part, les conduites suicidaires. L'imputabilité de certaines substances seulement (alcool, antidépresseurs, benzodiazépines, cocaïne) a pu être significativement établie dans notre étude. Toutefois, cette imputabilité peut être critiquée car des facteurs individuels, sociaux et psychiatriques semblent jouer un rôle tout aussi important que l'effet pharmacologique du psychotrope dans la genèse de ces comportements violents, et en particulier pour les conduites suicidaires. Une prise en charge thérapeutique spécifique vis-à-vis du toxique devra être envisagée à l'avenir par les professionnels de santé, pour réduire l'apparition de ces troubles du comportement iatrogènes.

In humans, some psychotropic agents (alcohol, drugs, illicit substances) have been suggested to play a role in the occurrence of major behavioural disorders, mainly due to the suppression of psychomotor inhibition. Behavioural disinhibition is a physiological mechanism which allows humans to behave appropriately according to a given environmental situation. The behavioural disinhibition induced by either thera-peutic dosage or misuse involves the loss of restraint over certain types of social behaviour and may increase the risk of auto or hetero-aggression and acting out. The increased use of psychotropic agents in recent years and the occurrence of unwanted effects are worrying and must be detected and evaluated. The objective of the present study was to establish a causal relationship between psychoactive substance use and occurrence of major behavioural disorders, such as paradoxical rage reactions and suicidal behaviour, based on a literature analysis. It consisted of reviewing reports of drug-induced violent reactions in healthy volunteers and demonstrating, where possible, a cause-effect relationship. Patients with schizophrenia and psychopathic personalities were not included in our study since psychiatric comorbidity could influence behavioural responses. Psychotropic agents included drugs, licit and illicit substances already associated with violence in the past. Many reports used the « Go/No Go test » to evaluate the disinhibiting effect of psychotropic substances ; this allows the « cognitive mapping » of drugs. The results suggest that only alcohol, antidepressants, benzodiazepines and cocaïne are related to aggressive behaviour. The best known precipitant of behavioural disinhibition is alcohol, which induces aggressive behaviour. However, there are large differences between individuals, and attentional mechanisms are now recognised as being important in mediating the effects of alcohol. Suicidal tendency as an adverse antidepressant reaction is rare, especially with atypical antidepressants. However, the risk of acting out exists and the responsibility of antidepressant agents in the genesis of suicidal tendencies is now esta-blished. The disinhibiting effects of benzodiazepines are well-known and proven by clinical trials. It's a « model » of acting out, and the causal relationship is undeniable. That cocaïne is related to violent behaviour is demonstrated by its pharmacological actions on CNS. The chronic use of cocaïne induces « a limbic dyscontrol syndrome » based on the altered activity of limbic structures. On the contrary, we could not demonstrate a causal relationship between aggression and either cannabis, ecstasy or phencyclidine. Cannabis abusers look particularly for euphoria and relaxing effects. Aggression as an adverse cannabis reaction is very rare and occurs in most cases in association with other drugs and in predisposed individuals. Ecstasy use may lead to long-term alterations of neuronal function in the human CNS and cause psychiatric disorders. However, there is insufficient information about long-term use of ecstasy to estimate its role in the occurrence of behavioural disorders. Clinical and forensic assumptions about phencyclidine and violence were not warranted. However, the substance-effect relationships can be criticized in the case of alcohol, antidepressants, benzodia-zepines and cocaïne. In fact, individual, social and psychiatric factors exert an influence on behaviour that is superior to the pharmacological effect of psychotropic agents. The most important parameter in drug-induced behavioural disinhibition is dosage, but mode of administration is also important. In addition, polysubstance abuse is very common. Substances may be taken simultaneously and alcohol is frequently combined with drugs. The combinations of substances result in multiple interactions, and very little is known about the effects of these interactions on violence in humans. Co-occurrence of substance abuse and other mental disorders is also very frequent. Multiple substance abuse should be avoided, because potential interactions between two or more drugs are more likely to cause violent behaviour. In the future, a specific treatment of these deleterious phenomena will have to be considered in order to reduce drug-induced iatrogenic behavioural disorders.