À efficacité comparable, la tolérance constitue un critère de choix majeur du traitement antidépresseur. Or, l’évaluation des effets indésirables (EI) des antidépresseurs se heurte à de nombreuses difficultés méthodologiques. Ainsi, nous avons mené une étude dont l’objectif était de comparer 2 outils de mesure des EI des antidépresseurs : la notification spontanée (NSp) et l’échelle UKU. Cette étude a été menée auprès de 50 patients traités depuis plus de 4 semaines par un antidépresseur appartenant à l’une des dernières classes commercialisées : les inhibiteurs sélectifs de la recapture de la sérotonine et de la noradrénaline, les inhibiteurs de la sérotonine et de la noradrénaline, les antidépresseurs noradrénergiques et sérotoninergiques spécifiques et la tianeptine. Le pourcentage de patients déclarant au moins un EI est significativement plus important avec l’échelle UKU qu’avec la NSp (84 % vs 58 %, p ≪ 0,01). Il en est de même pour le nombre total d’EI recueillis (n = 177 vs n = 47, p ≪ 0,001). Concernant le nombre d’EI d’ordre sexuel, les chiffres obtenus sont 4 fois plus importants avec l’échelle UKU qu’avec la notification spontanée (n = 36 vs n = 10, p ≪ 0,001). Le score moyen d’évaluation globale de l’interférence des EI avec la vie quotidienne du patient montre que, globalement, ce dernier n’est pas influencé par les EI dans sa vie quotidienne. Ces résultats soulèvent la question de la standardisation de l’évaluation des EI.

Background and aim of the stydy. Overall, the efficacy of the newer antidepressants: serotonin selective reuptake inhibitors (SSRI), selective serotonin/norepinephrine reuptake inhibitor (SNRI), noradrenergic and specific serotonergic antidepressant (NaSSA) and tianeptine is similar to that of the tricyclics, and so their acceptability/safety becomes a selection criterion for the clinician. However, side-effect assessment comes up against several difficulties : distinguishing between somatic symptoms caused by the depression and those caused by the treatment – which assessment tool to use (spontaneous notification, standardised scales that are not specific for the side effects caused by psychotropic drugs, standardised scales specific for the side effects caused by psychotropic drugs, meta-analysis, etc.) – which data sources to consult (anecdotal reports, reviews, prospective studies), and which data set to use, etc. As a result, the question of the exhaustiveness and reliability of the data consulted by the clinician can arise. We therefore conducted a comparative study in patients treated with these newer antidepressants, of 2antidepressants side-effect assessment tools : spontaneous notification (SN) versus the UKU scale, a standardised scale specific for the side effects of psychotropic drugs. Methodology – The depressed outpatients were selected from a psychiatric unit in a French psychiatric hospital and from a non-hospital consulting room. The main inclusion criteria were : male or female subjects, suffering from major depression without melancholia or psychotic features or suffering from mood disorders (according to DSM IV criteria), who had been treated for at least 4 weeks with one of the newer antidepressants. The main exclusion criteria were : any other psychiatric disorder, a serious physical disorder, treatment with neuroleptics, mood-changing drugs or other antidepressants, and patients who were not able to understand the questionnaire. The investigation was carried out by a clinical pharmacist. Results – Fifty patients were included in the study. There were 18 men and 32 women. The mean age was 53.5 15.9 years [22 – 77], the mean period of treatment was 24 30.5 months [1 – 127] and 52 % of the patients received concomitant medication with anxiolitic or hypnotic drug(s). The percentage of patients who reported at least one side effect was significantly higher for the UKU scale than for SN (84 % vs 58 %, p ≪ 0.01). The ratio between SN and UKU scale scores was 2/3. A similar pattern was found for the total number of side effects (n = 177 vs n = 47, p ≪ 0.001). The ratio between the total number of side effects for the SN and UKU scale was 1/4. The side effects were divided into five subgroups : psychiatric, neurovegetative, sexual, neurological and others. In all these subgroups, the number of side effects reported was significantly higher when the UKU scale was used than when SN was used. The values were as follows : psychiatric (n = 44 vs n = 15, p ≪ 0.001), neurovegetative (n = 59 vs n = 15, p ≪ 0.001), sexual (n = 36 vs n = 10, p ≪ 0.001), neurological (n = 11 vs n = 2, p ≪ 0.001) and other side effects (n = 27 vs n = 5, p ≪ 0.001). Nineteen side effects were only reported when SN was used (for example : dry eyes, incompatibility with alcohol, euphoria…). Twenty-four side effects were only reported when the UKU scale was used (for example : increased libido, loss of bodyweight…). The side effects had no impact on daily life in most of 80 % of the patients ; there was no significant difference between the patient’s assessment of the discomfort caused by side effects and the clinician’s assessment. In 90 % of cases, the side effects did not lead to any change in the treatment. Discussion – The findings of this study show that the collection of data regarding side effects depends on the assessment tool used : the number of side effects reported was significantly higher when the UKU scale was used than when SN was used. However, this finding must viewed with caution, because it has been showed that checklists can induce symptoms in suggestible patients. Neurovegetative troubles are the most commonly reported side effects, and neurological troubles the least often reported. This matches the tolerability profile of these antidepressants. The disorders that were least frequently spontaneously reported were the neurological, sexual and « other » side effects. These emerged only when the clinician asked the patient about them. The 19 side effects that were only reported when SN was used were side effects that were not included in the UKU scale or that had not been present during the three days before we started the investigation. The 34 side effects that were only reported when the UKU scale was used were either side effects with no apparent link with the treatment (for example : micturition troubles) or embarrassing effects (such as increased libido). Conclusion – Our findings show that the collection of data on side effects depends on the assessment tool used. These findings need to be confirmed by large-scale comparative studies, and the standardisation of the assessment of side effects is a question that needs to be raised.