Le switch est communément admis comme une des options à notre disposition en cas d’échec d’un traitement antidépresseur, cependant des incertitudes existent quant à sa mise en œuvre : quel délai respecter avant de switcher, faut-il effectuer un switch direct ou progressif, faut-il changer de classe, comment minimiser le risque d’interactions ? La revue de la littérature montre qu’il est recommandé d’attendre quatre à huit semaines avant de changer de traitement si la réponse reste insuffisante. Un switch précoce est cependant possible en cas d’absence totale de réponse à deux et quatre semaines. Le changement de classe thérapeutique est le plus logique en cas d’absence de réponse et pourrait être légèrement plus efficace que le switch au sein d’une même classe. Le switch direct est possible et bien toléré dans de nombreux cas, à l’exclusion des situations faisant intervenir un IMAO ou un antidépresseur tricyclique. Le switch augmente les chances de réussite du traitement et permet de minimiser les effets indésirables par rapport à la polymédication. Les recherches actuelles sur la physiopathologie de la dépression et les mécanismes d’action des médicaments font espérer de nouvelles perspectives de prise en charge des patients.

The switch is generally admitted as one of the available options in the event of non-response to an antidepressant treatment, despite uncertainties about its implementation in current practice: what time window before switching? Is it necessary to proceed with a direct or with a gradual switch? Is it necessary to change for a different pharmacotherapeutic class? How to minimize interaction risks? If a treatment fails because of poor compliance due to intolerance, it is possible to remain within the same therapeutic class and select another treatment with a more favourable safety profile for the patient. In the remaining non-response cases, changing therapeutic class is the more logical course and may be slightly more efficacious than the switch within the same class.

A review of the literature shows that it is recommended to wait 4 to 8 weeks before changing treatment if the response is insufficient. However, an early switch is possible in case of non-response at 2–4 weeks. Direct switch is possible and well tolerated in most instances, except for situations implicating a monoamine oxidase inhibitor (MAOI) or a tricyclic antidepressant. Direct switch is easy and, therefore, compliance issues associated with the complexity of treatment tapering can be avoided.

From the pharmacologic standpoint, the lack of effect on the cytochrome P450 isoenzymes, the absence of active metabolites, and the poor binding to plasmatic proteins are all important elements to be identified in order to minimize the risk of interaction. Current research on physiopathology of depression and mechanisms of action of drugs both support expectations for new perspectives for patients’ care. The switch increases the chances for a treatment to be successful with response rates of 20 to 70 % in the open-labelled clinical studies. It also has the advantage of minimizing adverse effects compared to polytherapy.

A great number of depressed patients require more than one treatment protocol to obtain or maintain a response. Switching is part of the therapeutic pattern of depression and is recommended by the French authorities. The available data allow the specification of switch modalities as function of the evolution of the initial treatment.