Differences in maintenance of response upon discontinuation across medication treatments in attention-deficit/hyperactivity disorder

The attention-deficit/hyperactivity disorder (ADHD) treatment literature has been focused on onset-of-effect and short-term effect size, with little exploration of ADHD symptoms upon medication discontinuation. The objective of this narrative review and analysis was to better understand the relapse of ADHD symptoms upon discontinuation of medication treatment in children, adolescents, and adults with ADHD who have responded to medication treatment and to explore differences among different medications in maintaining treatment response. Randomized withdrawal studies of dexmethylphenidate hydrochloride (d-MPH), methylphenidate modified-release (MPH-LA), lisdexamphetamine dimesylate (LDX), guanfacine extended-release (GXR), and atomoxetine (ATX) in both children/adolescents and adults with ADHD were reviewed. The percentage of relapse was significantly higher and the time-to-relapse significantly shorter with placebo compared to active treatment in patients who were previously stable on 5 weeks to 1 year of active treatment, suggesting clinically significant benefit with continued long-term pharmacotherapy. However, percentage of relapse at each time point studied after discontinuing stimulants and GXR appears substantially higher than observed when discontinuing ATX, suggesting longer maintenance of response after discontinuing ATX than after stimulants and GXR. Additionally, slope of relapse percentages over time appears to be more rapid with stimulants or GXR than with ATX. These differences in maintenance of response among ATX, GXR, and stimulants may reflect differences in mechanisms of action and persistence of the medication effect. Alternatively, they may be due to methodological differences, including study design and response/relapse definitions. Continued investigation is needed regarding factors that affect risk of symptom relapse upon discontinuation of pharmacotherapy.