LAUSR

INTRODUCTION Emotional dysregulation and irritability are common in individuals with autism spectrum disorder (ASD). We conducted the first meta-analysis assessing the efficacy of a broad range of pharmacological interventions for emotional dysregulation and irritability in ASD and predictors of response. METHOD Following a pre-registered protocol (PROSPERO: CRD42021235779), we systematically searched multiple databases until 01/01/2021. We included placebo-controlled randomized controlled trials (RCTs) and evaluated the efficacy of pharmacological interventions and predictors of response for emotional dysregulation and irritability. We assessed heterogeneity using Q statistics and publication bias. We conducted sub-analyses and meta-regressions to identify predictors of response. The primary effect size was the Standardized Mean Difference. Quality of studies was assessed using the "Cochrane Risk of Bias Tool" (RoB2). RESULTS 2,856 individuals with ASD in 45 studies were included, of which 26.7% of RCTs were at high risk of bias. Compared to placebo, antipsychotics (1.028, 0.824 to 1.232) and medications used to treat ADHD (0.471, 0.061 to 0.881) were significantly better than placebo in improving emotional dysregulation and irritability, while evidence of efficacy was not found for other drug classes (p>0.05). Within individual medications, evidence of efficacy was found for aripiprazole (1.179, 0.838 to 1.520) and risperidone (1.074, 0.818 to 1.331). Increased rates of comorbid epilepsy (β=-0.049, p=0.026) were associated with a lower efficacy. CONCLUSION Some pharmacological interventions (particularly risperidone and aripiprazole) have proved efficacy for short-term treatment of emotional dysregulation and irritability in ASD and should be considered within a multimodal treatment plan, taking into account also tolerability profile and families' preferences.