The aim of the systematic review and meta-analysis by Liang et al (2022) «Liang SC, Sun CK, Fan HY ym. Therapeutic effects o...»1 was to investigate the efficacy of antidepressants for restricted and repetitive behaviours, their effects on global impairment and other symptom subdomains in patients with ASD. The influence of age and dosage on potential adverse effects, tolerability and acceptability was evaluated in patients with ASD.
Search strategy included PubMed, Embase, ClinicalKey, Cochrane CENTRAL, ScienceDirect, Web of Science and ClinicalTrials.gov from database inception to June 2021. Only randomized controlled trials in patients with ASD were included. Interventions included antidepressants and comparators. Outcome assessment used behavioural rating scales for primary or associated symptoms in patients with ASD. From the 1103 articles initially retrieved from the databases, total of 16 studies were included in the meta-analysis. 2 RCTs were unpublished studies from ClinicalTrials. The number of participants ranged in the studies between 6 - 78 in the intervention group and 7 - 80 in the placebo group.
The meta-analysis included a total of 899 participants, mean age of 12.57 years (range 2.5–60 years), 4 (25%) of the studies also included an adult population. 18.62% were female (range 0%−33.3%). Seven data sets from 6 studies provided IQ data (range 49,6-103). The duration of treatment varied from 2 to 48 weeks.
Antidepressants were prescribed in 457 patients (50.8%), including selective serotonin reuptake inhibitors (SSRIs; e.g., fluoxetine, citalopram or sertraline), selective serotonin and norepinephrine reuptake inhibitors (e.g., venlafaxine), buspirone, mirtazapine and tricyclic antidepressants (e.g., clomipramine, desipramine).
15 of the studies used placebos as controls. In one study haloperidol was used as control.
In this meta-analysis, there were two types of behavioural rating scales as outcomes :1) questionnaires mainly for restricted and repetitive behaviours, such as the Children's Yale–Brown Obsessive–Compulsive Scale or the National Institute of Mental Health Global Obsessive–Compulsive Scale Others 2) more general behavioural rating scales that contain different symptom subdomains, including stereotyped behaviours, irritability, social withdrawal, hyperactivity and inappropriate speech, such as the Aberrant Behavior Checklist, the Behavioural Assessment Scale, the Children's Psychiatric Rating Scale and visual analogue scales.
A meta-analysis of 13 data sets from 12 studies found significant improvement for restricted and repetitive behaviours in patients who received antidepressants compared to controls: Effect size using Hedges g was 0.27 (0.10 to 0.44), p= 0.002. For SSRI medications (fluoxetine, sertraline citalopram and fluvoxamine, 7 studies), Hedges g was 0.24, (95% CI 0.04 to 0.44), p= 0.02. For SNRI (e.g., venlafaxin, 1 study) Hedges g was 0.19 (–0.86 to 1.24), p= 0.72, for tricyclic antidepressants (clomipramine, desipramine, 4 studies) Hedges g was 0.37 (–0.03 to 0.77), p= 0.07, buspirone (1 study) Hedges g 0.33 (–0.20 to 0.86), p=0.22.
For global improvement from 11 data sets (11 studies), Hedges g was 1.00 (95% CI 0.30 to 1.69) p=0.01. However, the results were not statistically significant for the groups of SSRI, SNRI, tricyclic antidepressants or buspirone.
For patients with ASD who received antidepressants the improvement in irritability (7 studies) was significantly better (Hedges g = 0.33, p < 0.03) as well as in hyperactivity (7 studies)(Hedges g = 0.22, p < 0.03), but not in inappropriate speech (7 studies) (Hedges g = 0.24, p = 0.1), social withdrawal (7 studies) (Hedges g = 0.03, p = 0.88) or anxiety (4 studies) (Hedges g = 0.44, p = 0.31).
Subgroup analyses showed significant effects of type of medication (i.e., SSRI v. clomipramine) on restricted and repetitive behaviours p= 0.005 as well as on global improvement p= 0.007 (higher efficacy of clomipramine compared to SSRIs), and age group (i.e., children and adolescents v. adults) favouring adults (p= 0.001 on restricted and repetitive behaviors, p= 0.040 on global improvement). A meta-regression using a mixed-effects model demonstrated significant associations between the therapeutic effects of antidepressants for the symptoms of ASD and the severity of the Clinical Global Impression scale (regression coefficient −0.95 for restricted and repetitive behaviours, p = 0.02) suggesting that more severe forms of ASD may be less responsive to treatment
Most studies were assessed to be of fair quality. 75.8% (81/112) had an overall low risk of bias, 18.8% (21/112) unclear and 5.4% (6/112) had high risk of bias. Two studies received financial support from pharmaceutical companies. Publication bias was evaluated to be low.
A former Systematic review and meta-analysis by Deb et al (2021) «Deb S, Roy M, Lee R ym. Randomised controlled tria...»2 aimed to assess the efficacy of antidepressant and anti-anxiety medications in people with ASD. In this study 15 RCT-studies for a total of 958 people with ASD were included. In this meta-analysis there were contradictory findings among the studies, with larger studies mostly showing a non-significant difference in outcomes between the treatment and the placebo groups. Meta-analysis of pooled Yale-Brown Obsessive Compulsive Scale and Clinical Global Impression Scale data from nine studies (60%) did not show any statistically significant intergroup difference on either of the outcome measures (effect size −0.23, 95% CI (−0.53, 0.07)). The adverse effects reported were mild and, in most studies, their rates did not show any significant inter-group difference. The authors concluded that it is difficult to draw any definitive conclusion about the effectiveness of either antidepressant or antianxiety medications to treat either ASD core symptoms or associated behaviours due to the methodological flaws in the most included studies and contradictory findings.
12 of the 15 studies of this meta-analysis are included also in the systematic review and meta-analysis of Liang et al (2022) «Liang SC, Sun CK, Fan HY ym. Therapeutic effects o...»1. Additional three studies in this meta-analysis evaluated the effect of venlafaxine, buspirone and agomelatine.
General comments: Comment 1: In Finland, the use of different antidepressants is restricted to depression and anxiety (depending on the drug). Use of antidepressants for other indication is off-label.
Comment 2: The results suggest that antidepressants may have a positive effect in treating the global symptoms of ASD. The results of the subgroup analyses in the most recent meta-analysis «Liang SC, Sun CK, Fan HY ym. Therapeutic effects o...»1 suggest that adults and patients with less severe overall ASD may have a better response to antidepressants than adolescents and children and patients with more severe ASD. This finding suggests that medication treatments need to be tailored for different subgroups of patients with ASD.
Comment 3: The effect sizes of the meta-analyses were quite similar for the restricted and repetitive behaviours in the two meta-analyses. In the most recent meta-analysis «Liang SC, Sun CK, Fan HY ym. Therapeutic effects o...»1 the efficacy was somewhat higher and it reached significance. This may be due to inclusion of more studies in meta-analysis and also to a possible contribution of more studies evaluating clomipramine which had higher efficacy than SSRI's. However, the effect sizes were still very small and the results for tricyclic antidepressants were not statistically significant.
Comment 4: None of the studies investigated the therapeutic effects on depressive symptoms, and only 4 included symptoms of anxiety. Thus, the role of possible comorbid depression or anxiety cannot be ruled out. Due to these, our confidence in the results is moderate.
Comment 5: In addition, there is a former systematic review and meta-analysis by Williams et al (2013) «Williams K, Brignell A, Randall M ym. Selective se...»6 which included nine RCTs with a total of 320 participants. However, most of the studies in this systematic review and meta-analysis were included in the later studies by Liang et al, (2022) «Liang SC, Sun CK, Fan HY ym. Therapeutic effects o...»1 and Deb et al (2021) «Deb S, Roy M, Lee R ym. Randomised controlled tria...»2. There were two RCTs on fenfluramine, which is not available in Finland.
Comment 6: In addition, there is a systematic review in Cochrane database by Iffland et al (2023) «Iffland M, Livingstone N, Jorgensen M, ym. Pharmac...»7 about pharmacological intervention for irritability, aggression, and self-injury in autism spectrum disorder (ASD). As a part of this systematic review the authors compared also short-term effect of antidepressants and placebo on irritability. Three studies (267 participants) which were included in the analysis are also included in the systematic review and meta-analysis by Liang et al (2022) «Liang SC, Sun CK, Fan HY ym. Therapeutic effects o...»1. No data for aggression or self-injury were reported or could be included in the analysis.