Dr. Armando Filomeno —who had talked with Dr. Gerald Erenberg about atomoxetine in November 2002, at the biennial Tourette Syndrome Association meeting in the US—  recently asked the distinguished pediatric neurologist for an article about his personal experience with atomoxetine almost a year and a half after its introduction in his country. This is Dr. Erenberg’s article written as a letter.

June 20, 2004

Dear Armando,

The following is my experience with atomoxetine for ADHD:

Atomoxetine represents a departure from the traditional treatment of ADHD. Such treatment has traditionally been with the psychostimulant class of drugs, and the major effect is thought due to an increase in dopamine activity. Use of psychostimulants dates back to the 1930’s when benzedrine was first used. Amphetamines were already available in the 1940’s, and methylphenidate has been in use since 1954. Psychostimulants have been extensively studied, and their usefulness has been well established. Concerns about psychostimulants include their addictive potential (hence they are controlled substances in the United States), appetite suppression, sleep difficulties, possible effect on growth, and possible causing or worsening of tics.

Atomoxetine is a norepinephrine reuptake inhibitor, and its effect on dopamine is unclear. It has been suggested that dopamine may be increased in the frontal lobes because of the effect on tracts connecting the deeper structures and the frontal lobes. How norepinephrine helps the symptoms of ADHD is not understood, but the drug clearly showed efficacy when tested in phase 3 studies. Separate studies showed that there was no danger of the medication increasing tics. Safety was good including a lack of cardiovascular side-effects. This was important to show since desipramine, a tricyclic antidepressant with effect on norepinephrine, was linked to sudden death due to its effect on the cardiac conduction system.

Atomoxetine came into wide use in the United States early in 2003. The public as well as many physicians were intrigued with the idea of a non-stimulant medication with no addictive potential and no risk of tics. Within 6 months of introduction, atomoxetine represented over 16% of new prescriptions for the treatment of ADHD. The phase 3 testing had shown side-effects such as sedation, gastrointestinal distress, dizziness, dry mouth, and decreased appetite, but in relatively small percentages of patients administered the drug.

Instructions for beginning treatment suggested that the initial dose be 0.5 mg/kg/day taken as a once a day dose in the morning. The full dose of 1.2 mg/kg/day could be started as soon as 3 days after introduction of the medication. The manufacturer’s studies had shown no advantage to a dose above 1.2 mg/kg/day. Some effect would be seen quite early, but full effect might take 4-6 weeks, similar to the pattern seen when starting the SSRI medications.

In my practice, our initial experience with atomoxetine was when we began switching patients from their stimulant medication to this new drug. Thus, our first group of patients had actually been treated with psychostimulants or atomoxetine at different times. Subsequently, we began starting our newly diagnosed ADHD patients with atomoxetine, representing a group who had never received psychostimulants. This was especially attractive when the patient being treated either already had tics or had a first degree relative with tics.

By now, we have treated over 200 patients with atomoxetine. Unfortunately, the results have been disappointing. The incidence of side-effects has been higher than in the patients treated with psychostimulants. This has led to our reducing the highest dose to not more than 1.0 mg/kg/day. Sedation and GI distress were so common that we now instruct patients to take their medication at night rather than in the morning, and to never take the medication on an empty stomach. When atomoxetine is found helpful, the time of administration should not be important since the effect lasts for over 24 hours, another advantage over the psychostimulants. When families compared the improvement seen with psychostimulants to that seen with the switch to atomoxetine, almost every one of them requested that their children be placed back on a psychostimulant. And in those patients started initially on atomoxetine, reports from parents and teachers generally showed only minor improvement in the ADHD symptoms. The results were rarely as robust as are the results when treating with psychostimulants.

Based on these results, we are again starting most newly diagnosed patients on a 12 hour psychostimulant preparation. The exceptions are those who have previously failed treatment with a psychostimulants, those who had severe side-effects on psychostimulants, or those patients who have ADHD and active tics. I would hope that further refinements will lead to better medications for the treatment of ADHD, but for now, psychostimulants remain the drugs of choice in treating ADHD.

Gerald Erenberg, MD
Senior Pediatric Neurologist
Cleveland Clinic Foundation
Cleveland, OH
USA

Immediate past Chairman, Tourette Syndrome Medical Advisory Board