Alpha-2 agonists for attention-deficit/hyperactivity disorder in youth: a systematic review and meta-analysis of monotherapy and add-on trials to stimulant therapy

Objective: To meta-analyze the efficacy and safety of α-2 agonists in pediatric attention-deficit/hyperactivity disorder (ADHD).

Method: We searched MEDLINE, EMBASE, Cochrane Library, CINAHL, and PsycINFO until May 2013 for randomized trials comparing α-2 agonists with placebo in ADHD youth. Primary outcome was reduction in overall ADHD symptoms. Secondary outcomes included hyperactivity/impulsivity, inattentiveness, oppositional defiant disorder symptoms (ODD symptoms), all-cause discontinuation, specific-cause discontinuation, and adverse effects. Standardized mean differences (SMD), relative risk (RR), and number-needed-to-treat/number-needed-to-harm (NNT/NNH) were calculated. Data were analyzed separately in monotherapy and as add-on to psychostimulants.

Results: Altogether, 12 studies (N = 2,276) were included. Across 9 studies (n = 1,550), α-2 agonist monotherapy significantly reduced overall ADHD symptoms (SMD = -0.59, p < .00001), hyperactivity/impulsivity (SMD = -0.56, p < .00001), inattention (SMD = -0.57, p < .00001), and ODD symptoms (SMD = -0.44, p = .0004). Similarly, α-2 agonist add-on treatment (3 studies, n = 726) significantly reduced overall ADHD symptoms (SMD = -0.36, p < .0001), hyperactivity/impulsivity (SMD = -0.33, p < .0001), and inattention (SMD = -0.34, p < .0001), but effect sizes were lower than in monotherapy trials (p = .03-0.04). As monotherapy, α-2 agonists had lower all-cause (RR = 0.70, p = .01, NNT = 10) and inefficacy-related (RR = 0.39, p < .0001) discontinuations than did placebo; however, intolerability-related discontinuation was similar, despite significantly more common fatigue (NNH = 10), sedation (NNH = 17), and somnolence (NNH = 4) and significantly greater hypotensive (clonidine-IR), bradycardic (clonidine-IR), and QTc prolonging (guanfacine-XR) effects. Added to stimulants, α-2 agonists had all-cause and specific-cause discontinuations that were comparable to those of placebo, but somnolence (NNH = 10) was more common, and hypotensive and bradycardic effects (clonidine-XR and guanfacine-XR) were greater than with placebo.

Conclusions: α-2 Agonist monotherapy and, possibly to a lesser extent, co-treatment, are significantly superior to placebo for overall, hyperactivity, and inattentive ADHD symptoms. Efficacy advantages need to be balanced against fatigue, somnolence/sedation, hypotension, bradycardia, and possibly QTc prolongation.