Avoiding Stimulants May Not Prevent Manic Switch: A Case Report With Atomoxetine

To the Editor: Attention deficit hyperactivity disorder (ADHD) affects 5.3% of children and adolescents worldwide, and it is associated with significant impairment in academic and social functioning. Stimulants are first-choice treatment for the disorder.¹ However, this psychoactive drug may promote psychotic and manic-like symptoms in approximately 0.25% of children with ADHD.² In the presence of bipolar disorder (BD), current guidelines suggest mood stabilization with antimanic agents before treatment of ADHD.

Nonetheless, when subthreshold or family history of manic/hypomanic symptoms are present in patients with ADHD, guidelines fail to provide clear and evidence-based data to orient treatment. Current indications are a close follow-up and adequate treatment of ADHD. Other options are the use of antimanic agents, which has not yielded significant positive results, or the use of atomoxetine or other nonstimulant agents.³

Atomoxetine, a selective inhibitor of norepinephrine reuptake, has been shown to improve ADHD symptoms. It promoted a significant response in both children and adolescents according to two recent meta-analyses. Although a small percentage of subjects have experienced symptoms of hostility and aggression, the rate of manic switch has not been determined.⁴

We describe the case of a 14-year-old patient who presented manic symptoms after the use of atomoxetine. We also take the opportunity to discuss the recent studies of manic symptoms after ADHD treatment.

Case Report

“Kathy” is a 14-year-old female teenager. When she was in preschool, she was diagnosed with ADHD by a child neurologist. Parent training was indicated at preschool instead of medication. During the school years, her notebooks were incomplete, and she daydreamed and avoided repetitive activities or activities where concentration was required. She was very disorganized. Also, she was hyperactive, constantly “on the go,” and dancing and jumping all the time. She was stubborn and did not obey orders (only at home). She was easily frustrated and annoyed and sometimes slammed her bedroom door, but in school, she could control those behaviors. Conduct issues were not present. She was retained two grades (first and second grade).

Sometimes she became “silly.” She laughed easily, her thought was accelerated, and she was more generous than usual with her friends. Most moments of excessive happiness preceded parties or times when receiving gifts. However, these periods would never last more than a couple of hours, and neither the parents nor the patient see any impairment in these changes. No hypersexuality, increased energy, or decreased need for sleep was reported. She said she also felt sad every day and would not see fun in everyday life, but did not report this to impair her life. She feared horror movies, nighttime, darkness, and would not be home alone. No developmental or medical abnormalities were reported. Her mother had chronic treatment-resistant depression, and her brother had a multiple substance use disorder.

Previous treatment trials included carbamazepine (which was reported to increase her irritability), oxcarbazepine (no response), risperidone (due to oppositional behavior toward parents, without clinical response), and methylphenidate (short and long-acting formulations, without clinical response, and small increase in irritability).

Inattention, impulsivity, and mood instability persisted, with significant impairment in academic and interpersonal functioning. Although at baseline assessment some indicators of BD were observed, parental assessment of mania according to the Child Mania Rating Scale for Parents score was 10, and the Child Mania Rating Scale for Teachers score was 6.⁵ ADHD symptoms, according to scores on the Swanson, Nolan, and Pelham IV (SNAP-IV) questionnaire, were as follows: inattention, 2.44; hyperactivity, 1.77; opposition, 0.25; and total score, 1.53 by her parents’ assessment. In school, scores were as follows: inattention, 2.22; hyperactivity, 0.11; opposition, 0.25; total score, 0.88.

Due to the absence of a response to methylphenidate, atomoxetine 25 mg (0.4 mg/kg per day) was given, and her inattention and hyperactivity presented partial improvement. No mood complaints were reported. When the atomoxetine dosage was increased to 25 mg twice a day (0.8 mg/kg per day), the patient had a significant improvement in inattention, hyperactivity, oppositional symptoms, and school performance, but she became very irritated, thought that her classmates were “childish,” and bragged on them. She spent around $500.00 (U.S.) on cell phone calls to her friends and boyfriend. She became very suspicious of her boyfriend. She was increasingly more scared of sleeping alone and needed the lights on. No visual or auditory hallucinations and no suicidal thoughts or ideation were present.

Risperidone was started and raised to 1 mg, and paranoid thoughts and irritation decreased, but she cried easily, had increased symptoms of separation anxiety, and would not even sleep alone. Lamotrigine was started and titrated to 50 mg/day. Because of significant weight gain, risperidone was switched to aripiprazole. The patient is currently stable, using aripiprazole 15 mg, lamotrigine 50 mg, and atomoxetine 25 mg. The current Child Mania Rating Scale for Parents score is 15, and the Child Mania Rating Scale for Teachers score is 1. Current SNAP-IV scores by her parents’ assessment are as follows: inattention, 1.22; hyperactivity, 0.77; opposition, 1.00; total score, 1.00. In school, the scores are as follows: inattention, 0.89; hyperactivity, 0; opposition, 0; total score, 0.31.

Discussion

The main purpose of this report is to illustrate a manic episode onset with the use of atomoxetine, an effective agent for the treatment of ADHD. In addition, we would like to remind clinicians about the need for close observation and the consideration of risks and benefits when treating patients with subthreshold mood symptoms.

The symptoms discussed here are common in clinical practice; ADHD symptoms are clearly observed, and mood instability can be detected, but a diagnosis of BD-not otherwise specified may not be ascertained because duration criterion is not present. Manic/hypomanic symptoms may be easily confounded with ADHD (in this case represented by impulsivity and event-related mood changes).

Although ADHD and BD are highly comorbid (11%−22% of comorbidity with BD has been mentioned in subjects with ADHD and 40%−89% of ADHD in samples of BD),⁶ treatment options for ADHD in this context are not well defined. Studies regarding stimulants, considered the first-line medication intervention for ADHD, exhibit controversial results. Current data suggest the use of stimulant agents as safe and well tolerated, and mood destabilization is an exception (but not rare), with low rates of the development of manic symptoms. However, it remains a concern that the use of stimulants may hasten the onset of BD.

Atomoxetine, considered a second-line agent for the treatment of ADHD, has shown moderate efficacy.¹ In the presence of BD or ADHD with subsyndromal BD, it could be considered as a pharmacological option due to the smaller risk of manic switch. However, this case and a few studies have shown otherwise. Most published trials have reported the occurrence of aggression and irritability.⁴ In double-blind, placebo-controlled trials previously conducted, irritability and mood swings were reported in 8% and 2% of subjects, respectively.⁴ However, a history of mood symptoms was an exclusion criterion for these studies. Thus, the occurrence of these adverse events should most likely be higher in general practice. Mania and hypomania are also reported as possible events.

Due to the high rates of comorbidity with BD and all the risks inherent to a manic episode onset, monitoring mood symptoms is crucial when starting children and adolescents with ADHD on atomoxetine. This care should be emphasized when personal or family history of mood disorders or symptoms are present, although mood worsening may not happen exclusively in the context of these risk factors.