In a recent meta-analysis «Schwartz S, Correll CU. Efficacy and safety of ato...»1, efficacy and safety of atomoxetine (ATX) was compared with placebo in the treatment of ADHD in children and adolescents. Study inclusion criteria were as follows: ATX monotherapy versus placebo; double blinded randomized controlled design; primary ADHD of any type, with oppositional defiant disorder (ODD), anxiety, and depression comorbidities allowed; patient age < 18 years; patient IQ of > 75. The primary outcome measure was baseline to end-point change in parent-rated, 18-item ADHD efficacy scale (ADHD-RS IV P/I), Swanson, Nolan, and Pelham rating scale-revised-IV ADHD (SNAP-RS IV ADHD), ADHD-RS Japanese version, investigator-administered (ADHD-RS J/I Total) or Parent/Investigator ADHD checklist score. Data from the safety outcomes were also extracted.
Altogether 25 RCTs, with 3 928 participants were included in the analysis. The mean follow-up time was 8,6 weeks (range: 4 – 18 weeks). The average ATX dose was 1,17 mg/kg/day. Mean age was 10,3 years, with 68,2 % being <13 years old. Altogether, 51,3 % were anti-ADHD medication naïve, and 44,6% had comorbid ODD/CD.
Based on all 25 studies with 3 928 participants, ATX outperformed placebo regarding overall ADHD symptoms (Effect Size = -0,67; CI 95% -0,56 to -0,71; p<0,0001), hyperactivity / impulsivity (ES = -0,67; CI 95% -0,53 to -0,81; p<0,0001), and inattention (ES = -0,59; CI 95% -0,51 to -0,67; p<0,0001). Based on 14 studies with 2 384 participants, 44,4% versus 21,4% of patients improved by ≥40% (NNT=4), the relative risk for not improving being 0,69 (CI 95% 0,63 – 0,75). Respectively, based on 16 studies with 2 630 participants, 39,9% versus 65,9% improved by <25% (NNT=4), relative risk of not improving being 0,60 (CI 95% 0,55 – 0,67).
All-cause discontinuation with ATX was similar to that of placebo. However, ATX had higher discontinuation rate due to adverse effects (AEs) (RR = 1,89; CI 95% = 1,08 – 3,31; p=0,03). At least 1 AE (70,4% versus 56,1%; p<0,01) and ≥1 psychiatric AE (21,5% versus 7,4; p<0,01) were more frequent with ATX, whereas serious AEs, aggression, and suicidal ideation were not different from placebo.
Another meta-analysis «Stuhec M, Munda B, Svab V ym. Comparative efficacy...»2, having more strict inclusion criteria than «Schwartz S, Correll CU. Efficacy and safety of ato...»1, included 17 high quality studies with 2 459 participants in analysis of efficacy of atomoxetine (ATX) in core symptoms of ADHD in children and adolescents. Double-blind, randomized, parallel studies with placebo, lasting from 2 weeks at least to 3 months at most, and including children and adolescents under the age of 18 were included. Studies with comorbid conditions other than ODD were excluded. Studies where research methodology was not clear and affected the final clinical score, and studies where demographic data about children and adolescents and clinical outcomes was not clearly presented, were also excluded.
The following scales were used to measure outcomes in the included studies: ADHD Rating Scale-IV(ADHD-IV), Conners' Parent Rating Scale (CPRS), Conners' Teacher Rating Scale (CTRS), IOWA inattentive-impulsive-overactive scale (IOWA-IO), Swanson, Nolan, and Pelham, Version IV(SNAP-IV), Strengths and Weaknesses of ADHD Symptoms and Normal Behavior (SWAN). All-cause discontinuation was used as a measure of acceptability.
Compared to placebo, treatment with ATX resulted in larger decrease in total ADHD symptom scores (Standard Mean Difference = -0,68; CI 95% -0,59 to -0,76). Compared to placebo, discontinuation with ATX treatment was not statistically different (OR=0,91; CI 95% 0,66 – 1,24).