HONOLULU— Few psychiatric therapies can challenge the 80% response rate typically seen when attention-deficit/hyperactivity disorder (ADHD) is treated with methylphenidate or amphetamine. But although the pharmacotherapy of ADHD constitutes a remarkable success story, the fact remains that nearly a third of patients treated with a stimulant either fail to respond to the medication or are unable to tolerate it.

Although many such treatment failures can be attributed to biological differences between individuals, another major contributor to poor clinical outcome is comorbidity. Three quarters of patients with ADHD meet criteria for one or more additional DSM-IV diagnoses—even among preschoolers, comorbidity is the rule—and these concurrent conditions must be taken into account when formulating a treatment plan. In many cases, the order in which the disorders are treated can substantially affect outcome. “By treating the comorbidity, not only do you improve the person’s well-being, but you also improve the likelihood of a response for ADHD symptoms,” noted Timothy E. Wilens, MD, at the annual meeting of the American Academy of Child and Adolescent Psychiatry. Dr. Wilens, an Associate Professor of Psychiatry at Harvard Medical School, reviewed the pharmacological arsenal for treating ADHD in adolescents and discussed strategies for treating refractory cases.

THE CAST OF CHARACTERS

In general, medications that have dopaminergic and/or noradrenergic properties are useful in the treatment of ADHD. The stimulants—short- and long-acting forms of both methylphenidate and amphetamine—remain the first-line treatment. These medications appear to be similarly effective and, as with other ADHD pharmacotherapies, seem to work equally well whether the patient is six years old or 60, Dr. Wilens said. The tricyclic antidepressants are slightly less effective than the stimulants for treating ADHD; other options include (in approximate order of decreasing efficacy) bupropion, pemoline, clonidine, and guanfacine.

Because methylphenidate and amphetamines are by far the most widely used agents for ADHD, most treatment-refractory patients will have already tried at least one of them. However, Dr. Wilens noted that methylphenidate and amphetamines exert their effects via somewhat different mechanisms: Methylphenidate binds fairly selectively to the dopamine transporter (thereby blocking reuptake) but does have some affinity for the norepinephrine transporter; amphetamines block reuptake of dopamine and norepinephrine as well but, more importantly, also facilitate release of these neurotransmitters from presynaptic storage granules. These mechanistic differences may explain why a patient who doesn’t respond to methylphenidate may benefit from amphetamine, and vice versa.

Both types of medication are now available in extended release forms (see Table) that may increase compliance and eliminate the need for in-school dosing; some patients may require higher doses (up to 2 mg/kg for methylphenidate; 1.5 mg/kg for amphetamine) when using these versions. In general, higher doses are more effective but increase the risk of adverse effects; an alternative to “pushing the dose” is to augment with a tricyclic or bupropion.

Most patients tolerate stimulants quite well. Side effects that do occur can often be prevented or minimized by taking the drugs with meals (to avoid nausea) or by using short-acting forms (to avoid insomnia). Adjunctive use of an antihistamine, a tricyclic, or clonidine can also reduce insomnia. If dysphoria is a problem, a low-dose mood stabilizer can be very helpful, Dr. Wilens said. Asking the patient about the timing of the adverse effect can be instructive; if problems such as dizziness or agitation typically occur around the time that drug levels are peaking or rapidly escalating, switching to a long-acting form may eliminate the problem.

The view that stimulants cause tics and that they should be avoided in patients with tic disorders is eroding; the emerging consensus is that the medications exacerbate tics in perhaps a third of patients but do not cause de novo tics. (See, for example, the editorial by Roger Kurlan, MD, in the January 2002 issue of Movement Disorders.) Adding clonidine can be helpful if tics become a problem. Because the age of onset of tic disorders is generally a few years after that of ADHD, the appearance of tics in newly medicated children is sometimes coincidental.

TRY A TRICYCLIC?

If stimulants are ineffective or poorly tolerated, the tricyclics are a good alternative, having shown efficacy in at least 14 controlled trials. Typical response rates are about 68% with desipramine and slightly less with nortriptyline. Patients who have failed a course of methylphenidate or amphetamine do just as well on tricyclics as do drug-naïve patients. “That’s a nice finding to tell parents who are feeling somewhat demoralized” about their child’s poor response to stimulants, Dr. Wilens noted. He emphasized that the tricyclics (as well as bupropion and atomoxetine) produce improvement in all ADHD symptom clusters (attention, distractibility, hyperactivity, impulsivity), just as the stimulants do.

One drawback to using desipramine is the need for cardiac monitoring. Dr. Wilens suggested, however, that obtaining an electrocardiogram is probably a good idea when using any tricyclic for extended periods, especially those that are structurally similar to desipramine, such as nortriptyline, amitriptyline, and imipramine. Nonetheless, he noted that a recent American Medical Association review found no association between desipramine and the occurrence of sudden death.

Although nortriptyline is easy to use, weight gain is sometimes a problem. In a recent nine-week study, Dr. Wilens and colleagues found a mean weight increase of 6 lb. However, for patients who are underweight or who have lost weight while using stimulants, this apparent drawback could be an advantage, Dr. Wilens noted: “We find that nortriptyline helps stabilize and normalize their weight.”

OTHER OPTIONS

The data for other medications are more sketchy. The response rate for the atypical antidepressant bupropion is lower than that for the tricyclics; clinicians should keep in mind that bupropion can have stimulant-like side effects and may exacerbate tics.

Pemoline, a stimulant that is pharmacologically unrelated to methylphenidate and the amphetamines, produced a moderate response rate (55%) in a recent controlled trial at Harvard, as well as in several previous studies. Current labeling for pemoline recommends a maximum dose of 112.5 mg/d, but Dr. Wilens noted that participants in the Harvard study received an average of 181 mg/d without undue adverse effects and that dosages of up to 3 mg/kg/d may provide optimal response. Clinicians should be alert for the possibility of hepatotoxicity.

Because clonidine is an alpha-2 noradrenergic agonist, it is often used for children who are hyperactive/impulsive or aggressive, or as adjunctive treatment for patients who have sleep problems. Three controlled trials have found benefits for patients with ADHD. Depression occurs in about 8% of users but usually resolves if the medication is discontinued; sedation can also be a problem. The question of whether there is an interaction between clonidine and methylphenidate remains somewhat controversial; however, recent investigations of this issue by the American Medical Association, the Food and Drug Administration, and others have concluded that the two drugs do not interact.

Guanfacine is pharmacologically similar to clonidine. In a controlled trial of children with ADHD plus tics, guanfacine improved symptoms in both domains, though the improvements in attention fell short of what is generally seen with standard medications. Side effects are similar to those of clonidine, though guanfacine is less sedating.

Modafinil showed benefits in one small study but was no better than placebo in an unpublished multicenter trial.

COMPLICATED BY COMORBIDITY

Whatever the chosen agent, pharmacotherapy is usually more likely to succeed if comorbid conditions are diagnosed and treated. In the case of comorbid substance abuse, for example, “most people in the field now agree that you have to stabilize the substance abuse first” if ADHD treatment is to succeed, Dr. Wilens stated. One peril of trying to “treat through the substance abuse” is illustrated by an open trial that Dr. Wilens recently conducted in young adults with ADHD, bipolar illness, and substance abuse problems. Subjects received medication and referrals for substance abuse counseling; although there was some reduction of ADHD symptoms, 43% of the subjects dropped out of the trial.

Data suggest that antidepressants should be the first-line treatment for patients who have ADHD plus a comorbid substance abuse disorder, according to Dr. Wilens. Stimulants are the next best choice.

Several other factors should be taken into account when treating ADHD patients who have a substance-abuse disorder. Adverse drug interactions (delirium, tachycardia) have been reported when marijuana users were treated with tricyclics, so other agents may be more appropriate in this population. Extended release formulations may be a wise choice for patients at risk for stimulant abuse; a recent case report described three patients who were unable to get “high” on Concerta despite their extensive efforts to grind, snort, and otherwise abuse the drug. Among stimulants, amphetamine has the highest abuse potential and pemoline the lowest.

Depression and anxiety can also interfere with treatment of ADHD: Most of the relevant stimulant studies have found poorer response rates in children with comorbid depression compared to those with ADHD alone. Tricyclics, on the other hand, seem to reduce ADHD symptoms just as well in children with depression as they do in those without a mood disorder. (The depression and anxiety symptoms do not necessarily improve, though.)

For most depressed children with ADHD, Dr. Wilens said he first focuses on treating the depression (which, unlike ADHD, can be lethal). He typically begins with a selective serotonin reuptake inhibitor; only later does he tackle the ADHD.

Bupropion is a reasonable option for patients with comorbid mood disorders. In a recently completed study, Dr. Wilens and colleagues found that bupropion reduced symptoms of ADHD and bipolar disorder (especially bipolar II) in children with both disorders. In another (uncontrolled) study, 24 children with depression and ADHD were treated with bupropion (up to 3 mg/kg bid); response rates were good for depression (87%) as well as for ADHD (62%). Both disorders responded in 58% of subjects.

Finally, two open studies have found that adding fluoxetine to a regimen of methylphenidate improved symptoms of both ADHD and depression. No drug interactions were observed.

MORE OF THE SAME

There is also a “very poor likelihood of having any ADHD response” if the patient has untreated mania. In systematic chart reviews, Dr. Wilens and colleagues have found that patients who receive stimulants without mood stabilization see little or no reduction in ADHD symptoms. (Fortunately, the stimulants don’t seem to trigger psychosis or worsen mood.) Conversely, treating the patient’s mania facilitates improvement in ADHD.

For ADHD patients who have “manageable” bipolar illness, Dr. Wilens said he typically starts with lithium or an anticonvulsant (divalproex or carbamazepine); if the patient is already manic, he begins with an atypical antipsychotic and later adds a mood stabilizer. Once bipolar symptoms are under control, he treats the ADHD symptoms.

Bipolar ADHD patients treated with bupropion show reductions not only in ADHD symptoms but also in mania and depression, so the medication may have some mood stabilizing properties, Dr. Wilens said. Thus, it may be worth considering for mild, evolving cases of bipolar II.

WHAT LIES AHEAD

Treatment options for ADHD will further expand as new medications enter the armamentarium. Atomoxetine is already under review by the Food and Drug Administration; it is unclear whether it will emerge as a first-line treatment, though Dr. Wilens suspects that it will at least be near the top of the hierarchy. There is promising anecdotal data for some nicotinic and cholinergic agents, and reboxetine may also prove useful in this setting. Finally, new forms of existing medications (eg, MethyPatch, the patch form of methylphenidate currently in phase III trials) will likely appear.

In the meantime, clinicians should be systematic in their approach but also creative, Dr. Wilens emphasized. Finally, he noted that “diagnosis drives the treatment; have a good sense of what you are treating before you start adding on medications.”

—Peter Doskoch

Suggested Reading
1. Spencer T, Biederman J, Wilens T. Pharmacotherapy of attention deficit hyperactivity disorder. Child Adolesc Psychiatr Clin N Am. 2000;9:77-97.

Return to table of contents