Article Detail

International Journal of Risk and Safety in Medicine 12 (1993) 3-35 IOS Press Psychostimulants in the treatment of children diagnosed with ADHD: Risks and Mechanisms of Action

Peter R. Breggin

Director, International Center for the Study of Psychiatry and Psychology (ICSPP) Abstract: Millions of children in North America are diagnosed with attention deficit/heractivity disorder and treated with psychostimulants such as methylphenidate, dextroamphetamine, and methamphetamine. These drugs produce a continuum of central nervous system toxicity that begins with increased energy, hyperalertness, and overfocusing on rote activities. It progresses toward obsessive/compulsive or perseverative activities, insomnia, agitation, hypomania, mania, and sometimes seizures. They also commonly result in apathy, social withdrawal, emotional depression, and docility. Psychostimulants also cause physical withdrawal, including rebound and dependence. They inhibit growth, and produce various cerebral dysfunctions, some of which can become irreversible. The 'therapeutic' effects of stimulants are a direct expression of their toxicity. Animal and human research indicatinos that these drugs often suppress spontaneous and social behaviours while promoting obsessive/compulsive behaviors. These adverse drug effects make the psychostimulants seemingly useful for controlling the behavior of children, especially in highly structured environments that do not attend to their genuine needs. 1.) Introduction The diagnosis of attentiondeficit/hyperactivity disorder (ADHD) in children, and the use of stimulant medication for behavioral control, has become very common in North America, and is spreading to Europe and Australia. In 1995, the International Narcotics Control Board (INCB) showed concern that "10 to 12 percent of all boys between the ages of 6 and 14 in the United States have been diagnosed as having ADD and are being treated with methylphenidate" (p. 2). Recently, the US Drug Enforcement Administration (DEA) announced an eight-fold increase in production quotas for methylphenidate (MPH) from 1,768 kg in 1990 to 14,442 kg in 1998 (Feussner, 1998). In addition, the use of stimulant medication has further escalated with the vigorous marketing of amphetamines. No official data afe available, but probabbly 4-5 million children receive pscyhostimulants in the United States each year (Breggin, 1998a). Drawing largely on double-blind placebo-controlled trials, this report examines adverse drug reactions (ADRs) associated with dextroamphetamine (AMPH)(Dexadrine, Adderall), methamphetamine (M-AMPH) (Desoxyn, Gradumet), and MPH (Ritalin). Special attention will be given to ADRs affection the central nervous system (CNS). The report also examines the mechanism of stimulant drug action. The behavioral or clinical effects of stimulants may be understood as a continuum of CNS toxicity. The drugs suppress spontaneous and social behaviors while promoting obsessive/compulsive or perseverative behaviors. These adverse drug effects make children more manageable in structured or controlled situations, especially those that lack sufficient adult supervision and attention. The efects are independent of any diagnosable disorder and occur in entirely normal animals and children. 2.) Overview of stimulant-induced adverse drug reactions (ADRs) 2.1 The continuum of psychostimulant toxicity Psychostimulants produce a continuum of toxicity based on generalized CNS exitation with direct effects on various neurotransmitter systems, including dopamine, norepinephrine, and serotonin. The continuum begins with feelings of increased energy, hyper-alertness, and an intensified focus on rote activities. It progresses toward insomnia, obsesive/compulsive or perseverative activities, agitation, hypomania, mania, and sometimes seizures.

Other psychostimulant ADRs -- such as somnolence, fatigue, lethargy, social withdrawal, and mental depression -- probably result from a combination of direct drug actions and the brain's compensatory reactions to these effects. Compensatory reactions became especially apparent during reductions in the blood concentration of the drug during wihdrawal or between doses. Rebound is a worsening of symptoms above baseline as direct drug effects wear off and compensatory CNS reactions become mroe dominant.

Table 1 (available in the PDF version of this article) summarizes the ADRs caused by MPH and AMPH as compiled from several well-recognized sources. In addition to familiar psychiatric ADRs such as nervousness, irritability, anxiety, depression, and increased emotional sensitivity or easy crying, there are infrequently emphasized ADRs such as impaired cognitive performance, compulsions, decreased social interest, and, in the extreme, a "zombie-like" constriction of affect and spontaneity mentionedby name and described by Arnold and Jensen (1995), Swanson, Cantwell, Lerner, McBurnett, Pfiffner et al. (1992). and Fialkov and Hasley (1984).

The eight studies listed in Table 2 were double-blind and (with one partial exception) placebo-controlled, and were selected because they are relatively recent and make an attempt to evaluate ADRs (Table 2).

Firestone, Musten, Pisterman, Mercer and Bennett (1998) found statistically significant MPH-induced ADRs in younger children across treatment conditions on the broad categories of "Somatic Complaints" and "Sociability", including inhibition or suppression of behavior such as Sad/Unhappy, Drowsiness, Talks less with others, and Uninterested in others, as well as Nightmares, and Decreased appetite. Obsessive/compulsive ADRs were not included in the list of potential ADRs.

In comparing placebo to the higher dose there were striking findings in regard to ADRs that suppress behavior: "Talks less with others" increased from 21.9 to 50% with a rise in severe cases from 3.1 to 9.4%. "Uninterested in others" increased from 31.2 to 75% with a rise in severe cases from 0 to 12.5%; "Sad/Unhappy" rose from 47 to 84% with a rise in severe cases from 3.1 to 15.6%. "Nightmares" increased from 28 to 62% with an increase in severity from 0 to 6%. "Tics or nervous movements" increased from 3.1  to 12.5% with a rise in severe cases from 0 to 3.3%,

The authors also made a separate calculations of the percentage of children who "deteriorated" in regard to various symptoms when comparing the 0.5 mg/kg dose to placebo: Sad/Unhappy - 69% (p=0.01); Drowsiness - 62% (p=0.001); Uninterested in others - 62% (0.0002). In addition, there was a deterioration in appetite in 75% (p=0.001) of the children on 0.5 mg/kg compared to placebo.

Four of 41 children (10%) withdrew from treatment (reasons unspecified in report). As a conservative estimate, at least 4 children had severe ADRs.

The authors raise the possibility that observers might unintentionally consider the social dampening ADRs as improvements in the children's behaviors. However, they also noted: "This social dampening effect reported by parents is of some concern, especially considering the claims that methylphenidate is used as a 'chemical billy club' or 'straightjacket' (p. 20). These findings, indicating severe ADRs among very young children, are consistent with an earlier study by Schleifer, Weiss, Coen, Elman, Cregic et al. (1975) who reported "less social behavior and interaction", as well as "sadness, irritability, excessive hugging and clinging, and increased solitary play, as well as the more usual side effects of poor appetite and difficulty getting to sleep..."(p. 49). The treating physician and the parents discontinued treatment in 25 of 28 children because of ADRs.

Mayes, Crites, Bixler, Humphrey, and Mattison (1994) conducted double-blind placebo-controlled MPH trials involving preschoolers but trials involving older children were single blind. There was a substantial rate of behavior-suppressing ADRs: 18.8% of the children suffered from lethargy. "Children with lethargy were variously described by raters as tired, withdrawn, listless, depressed, dopey, dazed, subdued, and inactive" (p. 1104). In 58% there was in increase or emergence of "stereotypical behaviors, including hand-wringing, arm-waving, teeth-grinding, and foot-tapping" (p. 1104). Obsessive/compulsive activities (stereotypy) were also observed.

Mayes et al. reported that 26.1% of the children suffered from "irritability" during treatment. Five children (7%) displayed disturbing ADRs, including one manic-like reaction with "incessant talking", one "wild" and "out of control", and one "aggressive behavior" (p. 1105). Two of these five also developed abnormal movements. Mayes et al. also described more typical MPS adverse effect, including insomnia (13%); stomachache, nausea or vomiting (11.6%); loss of appetite (20.3%); and headache (4.3%).

Allowing for overlapping reports of more than one ADR per child in study, probably more than 50% of the children suffered from lethargy or other adverse CNS reactions. Six were discontinued due to ADRs and that number will be used to make a conservative estimate of severe ADRs.

Schachar, Tannock, Cunningham and Corkum (1997) found that 5 of 46 children (>10%) dropped out due to ADRs in a 24-week long MPH study. These 5 children will be used to calculate the number of severe ADRs. Their drug-induced symptoms included behavioral aberrations such as "sadness and  behavioral deterioration, irritability, withdrawal, lethargy, and violent behavior", "withdrawal and mild mania", and "withdrawal and dysphoria" (p. 760). Parental ratings by phone indicated a statistically significant overall increase in physiological symptoms (commonly, anorexia and stomachaches) and affective symptoms (commonly, withdrawal, sadness and crying).

The authors concluded, "Affective symptoms were significantly associated with MPH, but they tended to develop later in the course of treatment" (p. 761). These delayed ADRs will be missed in typical drug trials which last only a few weeks.

Barkley et al. also found that "the percentage of children experiencing proneness to crying also increased by at least 10% at the high dose of medication" (p<0.05) (p. 187). Finally, Barkley et al. reported that three children (3.6%) "were unable to complete the protocol because of serious adverse reactions to medication ... One child had a nervous facial tic, dizziness, and headache; a second had dizziness, headache, and increased hyperactivity; and the third had excessive speech and disjointed thinking" (p. 186).  Even in this brief, relatively low dose study, one child developed manic-like symptoms with "excessive speech and disjointed thinking". Again choosing a relatively conservative estimate, Barkely et al. study had three children with severe ADRs.

Borcherding, Keysor, Rapoport, Elia, and Amass (1990) focused on perseverative, obsesssive/compulsive or overfocused ADRs (for details, see Table 3 in the PDF version of this article). The treatment included both MPH and AMPH. Observations were made on the day hospital ward, in school, and by the families. This close scrutiny probably accounts for the "extraordinarily high rates of obsessive/compulsive behaviors, movement abnormalities, or both" (p. 92). Most of these ADRs "were seen only by staff sensitive to these possible effects" (p. 92).

Bocherding et al. found a strong connection between abnormal movements and obsessive/compulsive behaviors in association with MPH (p=0.009). Tics, overfocusing, and other compulsive behaviors were observed in 34 (76%) of the 45 participants who completed the study, plus one subject with severe tics who was dropped. Abnormal movements were observed in 26 of 45 children (58%). Obsessive/compulsive or perseverative ADRs (summarized in Table 3) were observed in 23 of 45 children  (51%). The authors reported, "When compared to placebo, both drugs increased the likelihood (p<0.o1) of repetitious, perfectionistic, overfocused behaviors" (p. 90) Of these 23 children, 14 (60.8%) suffered one or more of the following abnormal movements: orofacial, stereotypy, or other tics. Twelve of the 23 had orofacial tics and 6 had stereotypy, including 4 who had both. At least three children developed severe drug-induced obsessive/compulsive symptoms (one on MPH, two on AMPH), including one child who played Legos for a 36-hour period without breaking to eat or sleep and another who "became compulsive about raking leaves and did so for 7 consecutive hours, after which he felt compelled to rake individual leaves as they fell" (p. 87).

One child had to stop the trial "due both to the severity of the tic he developed during his initial treatment phase (AMPH) and exacerbated symptoms of separation anxiety. This child also lost 2 pounds during treatment" (p. 85). At one point the tics "increased to occur over 10 times per hour" (p.87). The tics did not fully clear. Conservatively, at least 4 children in this trial had severe ADRs.

Solanto and Wender (1989) studied cognitive function using one daily dose of MPH for 3 days. They found that 42% of the children became "overaroused" with "cognitive perseveration". Compulsive, perseverative behaviors thus begin with the first doses of stimulant medication, accounting for its immediate "therapeutic" effect.

Castellanos, Giedd, Elia, Marsh, Ritchie et al. (1997) studies the effects of AMPH and MPH on children comorbid for ADHD and Tourette's syndrome. While the investigators focused on tics rather than on perseverative/obsessive ADRs, they reported: "Largely transient obsessive-compulsive symptoms were also noted (n=5 on MPH, 1 on AMPH) including retracing letters, excessive erasing, rearranging and collecting compulsions, and obsessional sexual thoughts" (p. 593). The rate of obsessive ADRs for MPH was 25% during a three-week exposure.

Castellanos et al (1997) reported that one child on AMPH dropped out due to vomiting and another due to worsened behavior. Three more had "greater tic severity scores on all doses of both stimulants than at baseline" and were discontinued from stimulants at the conclusion of the study. This leads to a conservative estimate of 5 severe ADRs.

Stimulant-induced obsessions and compulsions have been reported as long as 4 years after the beginning of drug treatment (Kouris, 1998). Therefore, even the high rates found in these studies are likely to underestimate these ADRs for long-term treatment.

Firestone et al. found an increase in "Tics or nervous movements" from 3.1% on placebo to 12.5% on 0.5 mg/kg MPH, with an increase in severe cases from 0% on placebo to 3.1% on 0.5 mg/kg. Borcherding et al. (1990), as noted, reported the appearance of abnormal movements in approximately 58% of their children, including one seemingly irreversibel case. Barkley et al. (1990) found a  10% increase in tics in children treated with the higher dose of MPH. With both MPH and AMPH, Castellanos et al. (1997) found a dose-dependant worsening of tics in a "substantial minority" of patients comorbid for ADHD and Tourette's syndrome. As already noted, three discontinued medication at the conclusion of the trials due to increased tic severity on both MPH and AMPH. They observed, "a substantial proportion of our small sample (one third) continued to have stimulant-associated exacerbations of their tic disorder which outweighed the clinical benefits of stimulants" (p.594).

 Lipkin, Goldstein and Adesman (1994) (not 1 of the 8 controlled trials) found a 9% rate of abnormal movements in a retrospective evaluation of 122 children diagnosed with ADHD currently or recently treated with stimulants. One child developed a very severe and irreversible Tourette's syndrome involving "facial twitching, head turning, lip smacking, forehead wiping, and vocalizations". Other tics and dyskinesias found in the study included mouth movements; eye blinking, rolling or deviation; eye "bugging"; neck turning; and face rubbing. Five of the children had more than one type of dyskinesia. There were no differences in rates on MPH and AMPH. Children developed the tics or dyskinesias with drug exposures varying from less than 1 week to 23 months.

Schmidt, Kruesi, Elia, Borcherding, Elin et al. (1994) recorded changes in calcium and magnesium concentrations in the blood during treatment with MPH and AMPH that they believe may contribute to the abnormal movements.

Tics can be stigmatizing, embarrassing,  and even disfiguring. Many children would probably prefer to suffer from "ADHD-like" symptoms rather than endure tics.

3.5 Summary of findings in clinical trials

Even though most of these clinical trials were short-term and low dose (Table 2, in the PDF version of this article ), many serious ADRs were reported. The total number of severe ADRs is 30 out of 359 children (8%). Using broader criteria, the rate rises to probably between 10 and 20%.

If clinically observable potentially significant ADRs are included, the rate is much higher, in the 20 to 50% (or more) range. For example in the three studies that examined obsessive/compulsive ADRs (including overfocusing and perseveration), these ADRs were extraordinarily common: 25, 42, and 51%, respectively, for Castellanos et al. (1997), Borcherding et al. (1990) and Solanto and Wender (1989).

Despite such high rates for serious, severe ADRs, the rates and severity of ADRs should be expected to be much higher under routine clinical conditions. These conditions include much longer exposures to stimulants (months or even years instead of 1 to 3 weeks in most of the controlled trials), often higher doses (more than the 0.3 - 0.6mg/kg MPH in most of the controlled trials), polypharmacy, less adequate medical evaluations and supervision, and parents and teachers who are not educated to identify ADRs and to terminate treatment before they worsen.

 3.6 Lessons from stimulant-induced psychosis

Many studies have compared stimulant-induced psychoses to the symptoms of schitzophrenia (Ellinwood and Tong 1996; Murry, 1998;  Rebec and Bashore, 1984; Segal, Weinberger, Cahill, and McCunney, 1980) MPH is used experimentally to produce or worsen psychotic symptoms in adults diagnosed schitzophrenic (Koreen, Lieberman, Alvir, and Chakos, 1997; Lieberman, Kane, and Alvir, 1987). Stimulant abuse is also known to cause a disorder that may remain chronic and become indistinguishable from schitzophrenia (Flaum and Schultz, 1996).

3.7 Effects of selective serotonin reuptake inhibitors (SSRIs) in children

Psychoactive drugs will probabaly tend to produce mental disorders, including psychosis, at a higher rate in children than adults. For example, the rate for mania/hypomania induced by the SSRI-type antidepressant fluoxetine (Prozac) in all US clinical trials with adults was 0.7% (Physician's Desk Reference, 1998, p. 860). In many of the short placebo-controlled clinical trials, it was even less (range of 0 to 0.8%). However, in a recent placebo-controlled clinical trial of fluoxetine in children and adolescents (Emslie, Rush, Weinberg, Kowatch, Hughes, et al. 1997), three out of 48 children dropped out due to "manic symptoms" (6.2%).

King, Riddle, Chappell, Hardin, Anderson et. al (1991) described the "emergence of self-destructive phenomena in children and adolescents, ages 10 to 17, during fluoxetine treatment". They found "self-injurious ideation or behavior appeared de novo or intensified" in 6 of 47 patients being treated with fluoxetine for obsessive/compulsive disorder. Four of the cases required hospitalization and three required "restraints, seclusion, or one-to-one nursing care". Riddle, King, Hardin, Scahill, Ort et al. (1990/1991) found that 12 of 24 children and adolescents, ages 8 to 16, developed two or more behavioral side effects in reactions to fluoxetine. Most of the youngsters were being treated for obsessive compulsive symptoms. The drug-induced effects included motor restlessness sufficient to cause concern to parents or teachers, incomina, social disinhibition manifested by garrulousness or subtle impulsivity, and a subjective sense of discomfort due to restlessness, agitation, or excessive energy. The group included three children with ADHD, all of whom became worse. The behavioral abnormalities remained stable for weeks until the fluoxetine was reduced or stopped, and were easily confused with the children's original emotional problems. The seven children on placebo developed no such effects.

 4. ADR reports from the FDA Spontaneous Reporting System

A review of the 2,821 reports of adverse drug events to the Spontaneous Reporting System for MPH (1985-March 3, 1997) revealed some potential often-ignored ADRs (Food and Drug Administration, 1997). Here are some highlights (analyzed by Breggin, 1998b; methodology of analysis discussed in Breggin, 1998c; Kessler, 1993; Leber, 1992):

 (1) More than 150 reports of liver abnormalities, mostly abnormal liver function tests.