Même si le taux de réponse aux traitements médicamenteux, et notamment aux antidépresseurs inhibiteurs de recapture de la sérotonine (IRS) et de la noradrénaline (IRSN), est relativement élevé dans la plupart des troubles anxieux, il reste un part non négligeable de patients résistants à une première ligne de traitements. De ce fait, nous avons réalisé une analyse de la littérature portant sur les alternatives pharmacothérapeutiques en cas d’échec des traitements IRS, en nous concentrant sur les études contrôlées en double insu. Le plus grand nombre d’études porte sur le trouble obsessionnel-compulsif. Plusieurs essais contrôlés ont conduit à des résultats positifs, en associant aux IRS des antipsychotiques (rispéridone, aripiprazole, halopéridol), des anticonvulsivants (lamotrigine, topiramate) ou d’autres molécules (pindolol, N-acétylcystéine, mémantine). Dans le trouble anxieux généralisé, aucune stratégie pharmacologique n’a fait la preuve de son efficacité chez les patients résistant aux IRS ou IRSN dans des études contrôlées. C’est le cas également des phobies sociales. En revanche, des résultats significatifs, même s’ils sont souvent partiels, ont été obtenus dans le traitement des états de stress post-traumatique résistants avec l’adjonction d’antipsychotiques (rispéridone ou olanzapine) ou de prazosine. Au total, quelques stratégies pharmacothérapeutiques sont en partie validées comme alternatives aux traitements de première intention, dans les TOC en particulier, mais les résultats sont le plus souvent insuffisants et des recherches spécifiques sont encore nécessaires pour améliorer la prise en charge des troubles anxieux sévères et réfractaires.

Anxiety disorders are widespread psychiatric conditions with significant social and professional disability, poor quality of life, an increased risk of suicide, and frequent attendance of medical services. Serotonin reuptake inhibitors (SRI) and serotonin and norepinephrine reuptake inhibitors (SNRI) have demonstrated a rather robust efficacy for the treatment of most of anxiety disorders. Nevertheless a substantial number of patients are resistant or still suffer from residual symptoms despite this first line treatment. The objective of our paper is to review relevant studies for the pharmacologic management of anxiety disorders resistant to the first line treatment. For this purpose, we conducted a pubmed/medline search for double-blind placebo-controlled trials of treatment-resistant anxiety disorders. An adequate trial for a SRI in the treatment of obsessive-compulsive disorder (OCD) should continue for at least 12 weeks. Special considerations of the comorbidities and symptom profile could help in the choice of an appropriate pharmacotherapy. Several trials have highlighted the efficacy of antipsychotics as an add-on to SRI in treatment-resistant OCD such as haloperidol more so when comorbid with a tic disorder, or risperidone that can reduce OCD as well as depressive symptoms. Aripiprazole has been shown efficacious in two placebo-controlled double-blind trials, while the efficacy of quetiapine and olanzapine remains controversial. Other trials showed some efficacy of anticonvulsants (lamotrigine, topiramate), pindolol, memantin and N-acetylcystein as an adjunctive treatment to SRI for resistant OCD. Few trials have investigated selective serotonin reuptake inhibitors (SSRI) or SNRI resistant generalized anxiety disorder showing a failure of adjunctive therapy with olanzapine, quetiapine, ziprasidone and risperidone. These studies were underpowered and very limited in number. Adjunctive risperidone for resistant post-traumatic stress disorder (PTSD) showed benefit in some but not all trials. Olanzapine was beneficial for the reduction of the CAPS score in addition to the improvement of sleep disturbances. Furthermore, prazosin was efficacious by reducing PTSD symptoms, sleep disturbances, nightmares, and psychological distress. One double-blind placebo-controlled study was conducted to investigate treatment-resistant social phobia showing no benefit of pindolol add-on paroxetine. Our results demonstrate that the pharmacological management of treatment-resistant anxiety disorders is not sufficiently investigated in double-blind placebo-controlled trials, despite a growing evidence in favor of antipsychotics and some other pharmacological agents in resistant OCD and, to a lesser extent, PTSD. Hence, there is a crucial need for larger double-blind placebo-controlled trials for resistant anxiety disorders. Finally, being out of the scope of our review, we omitted studies of non-pharmacologic therapies.