¹University of Gothenburg, Pharmacology, Gothenburg, Sweden

Background: The possible efficacy of selective serotonin reuptake inhibitors (SSRIs) and serotonin and noradrenaline reuptake inhibitors (SNRIs) in depressed children and adolescents remains highly controversial, with some arguing that these drugs are virtually devoid of efficacy in non-adults while others advocate an opposite stance. We have now conducted what we believe is the first comprehensive meta-analysis with access to patient-level data addressing the impact of SSRIs and SNRIs on depression in the young.

Aims & Objectives: Based on 22 double-blind placebo-controlled clinical trials (n=5,496) of SSRIs (citalopram, escitalopram, fluoxetine, sertraline, paroxetine) and SNRIs (desvenlafaxine, duloxetine, venlafaxine), most of which used the Children’s Depression Rating Scale (CDRS-R) for assessment of outcome, this study was undertaken to shed light on antidepressant responsiveness in those aged 18 or below with special focus on the possible impact of age (6-11 vs 12-14 vs 15-18) and sex.

Method: Trial- and patient-level item-based meta-analyses.

Results: An initial trial-level meta-analysis including all trials suggested SSRIs but not SNRIs to be superior to placebo while revealing no difference between different SSRIs in terms of efficacy. A subsequent patient-level, item-specific analysis of pooled data from all SSRI trials using the CDRS-R rating scale showed effect sizes versus placebo in those aged 15-18 on key measures between 0.26 and 0.44. In girls aged 12-14, SSRIs again outperformed placebo, especially with respect to the symptom irritability, whereas only miniscule effects were observed in boys of the same age. In children aged 6-11, improvement on placebo was high, and differences between treatments virtually none (boys) or modest (girls). In those below the age of 15, a response to SSRIs, when at hand, was usually observed already after one week of treatment, irritability usually being the symptom for which the largest effect was observed. A similar pooled analyses of all SNRI trials using the CDRS-R rating scale confirmed that these drugs are less effective than SSRIs and barely superior to placebo (except for in girls below the age of 12).

Discussion & Conclusions: It is concluded i) that age and sex – and tentatively puberty – are strong predictors for the antidepressant effect of SSRIs in subjects aged ≤18 years, the effect being weakest in boys below the age of 11 and largest in subjects of both sexes above the age of 15, ii) that significant differences between active drug and placebo – when at hand – are often observed already at the first week of treatment, iii) that the previous suggestion that fluoxetine be superior to other SSRIs in non-adult depression appears unfounded, iv) that irritability is not only one of the most common symptoms in depressed children but also a symptom particularly likely to respond to SSRIs, and iv) that SNRIs are generally less effective than SSRIs in the young. While this abstract is based on a preliminary analysis, the definite outcome will be presented at the congress.