Abstract
Objective:
This review aims to detail specific psychotropic medication treatment differences in adverse drug events (ADEs) between children, adolescents, and adults.
Method:
A thorough data-based search of ADEs was made, augmented by findings from approved and updated U.S. Food and Drug Administration (FDA) drug labels, controlled clinical trial reports, and from FDA information on ADEs at scheduled public hearings.
Results:
Children were found to generally exhibit more ADEs to psychotropic medication than adolescents and adults. These ADEs primarily include altered growth velocity, rash, vomiting, dystonia, tics, affect lability, activation, metabolic blood test abnormalities, sedation, sialorrhea, and electrocardiogram irregularities.
Conclusion:
Children have more biological vulnerabilities than older individuals, which could account for their greater rate of ADEs to psychotropic medication treatment.
Introduction
Only three review articles were found on ADEs from psychotropic compounds that described age-grouped differences. The review by Gupta and Waldhauser (1997) focused on pediatric medication ADEs in infants and toddlers. Anderson (2002) limited her review to anticonvulsant ADEs in youth, and Brent et al. (2004) briefly chronicled numerous drugs and toxicants that have adverse effects in youth and/or adults.
In this review, the major age-related differences in psychotropic ADEs reported in the literature will be identified and briefly described by specific drug class, subclass or drug entity. Then, a more detailed discussion follows on factors associated with these age-grouped ADE differences. Last, a brief summary will be presented.
Method
Publication databases used for this review included Medline, PsycInfo, Embase, and International Pharmaceutical Abstracts. Psychiatric textbooks were also reviewed for age-grouped ADE findings. The literature search focused on age, psychotropic drug entities, ADEs/reactions, and ADEs in association with controlled trials of psychotropic medication treatment. Search terms included individual and combinations of terms such as child, youth, age, ADEs, adverse drug reactions, side effects, psychotropic, antidepressant, antipsychotic, stimulant, anticonvulsant, and the names of specific drugs. ADE findings were organized categorically and described by drug class, subclass, and drug entity. Excluded from the age-grouped search were geriatric adults and neonates. Case series reports, open drug trials, and clinical trials involving relatively few subjects were included very infrequently.
The great majority of the published studies used in this review were large, industry-sponsored, short-term, controlled clinical drug trials that employed systematic reporting of ADEs. In instances where useful ADE findings came from case series reports, these are noted as such. A few findings came from open trials, reviews, and meta-analyses. These also are noted. In cases where the age-grouped ADE data were incomplete or equivocal, the ADE was not included in this review. For example, disinhibition in youth was reported to be more frequent with benzodiazepines than with placebo in one of two small controlled studies (Simeon et al. 1992; Graae et al. 1994). However, it has also been reported in adults (Saias and Gallarda 2008). That possible ADE–age relationship was therefore deemed inconclusive and was not included in the review. Likewise, increased blood levels of prolactin secondary to risperidone treatment were reported variably in youth as well as in adults and were therefore not included as age-disparate. However, prolactin levels after olanzapine treatment were reported in a company study and in a U.S. Food and Drug Administration (FDA) Web site to be higher in adolescents than adults. Consequently, that finding was assumed to be well founded and was included.
Many of the reported age-grouped findings listed in this review were obtained from FDA-authorized drug labels that were available on their website. As far as is evident, the age-grouped findings on the drug labels are from industry-sponsored, controlled clinical trials or open-ended extensions of these. Likewise, age-grouped differences in psychotropic drug ADEs presented by the manufacturer at the FDA hearings to request an approved indication were included in this review. Such data were based on double-blind, placebo-controlled (DB-PC) clinical drug trials. ADE drug–placebo differences by age group when prominent were also a source of reported findings. The emphasis in this review was on evidence-based increases in psychotropic ADEs that reflected an increased level of vulnerability in youth. ADEs that occur more frequently in adults than in children do exist (Brent et al. 2004), but these were touched on in this review only when pertinent.
The available literature varied in the ages of subjects experiencing ADEs. In this review, youth represents both children and adolescents, preschool < age 6, children ages 5 or 6 to 12 or 13, and adolescents ages 12 or 13 to 18 or 19. In the olanzapine registration trials, for example, the subjects were listed as adolescents, not as youth. In the Woods et al. (2002) report on olanzapine, cases were listed as youth because both children and adolescents were included.
Major Age-Related (Child, Adolescent, and Adult) Differences in ADEs After Treatment with Psychotropic Medications
Hypnotics and sedatives
(a) Zolpidem in a multisite, double-blind, placebo-controlled clinical trial (n = 201) for youth (aged 6–17 years) resulted in hallucinations in 7.4% of the drug-treated subjects as compared with none in the placebo group. These ADEs were higher in children than in adolescents. In premarketing studies of zolpidem in adults, less than 1% experienced hallucinations (Ambien FDA Approved Label 2008; Ambien (zolpidem) BPCA Clinical Review Summary 2008; Blumer et al. 2009).
(b) Diphenhydramine can diminish mental alertness in both children and adults. However, “[I]n the young child, particularly, [it] may produce excitation” (Diphenhydramine Official FDA Information 2009).
Medications for attention-deficit/hyperactivity disorder
(a) Methylphenidate and amphetamine compounds
(1) Decreased growth velocity
Stimulants tend to lessen growth velocity in weight and height in youth with attention-deficit/hyperactivity disorder (ADHD) (Ahmann et al. 2001; Spencer et al. 2004; Swanson et al. 2007, 2009; Faraone et al. 2010). This effect is relatively more prominent in preschoolers (Ghuman et al. 2001; Swanson et al. 2006; Wigal et al. 2006). As the duration of this medication treatment increases in youth, the decrement in height and weight percentile growth velocity tends to increase (Swanson et al. 2007, 2009). Adults on stimulants also often lose weight—but understandably not height (Paterson et al. 1999; Kooij et al. 2004).
(2) Reduced appetite and complaints of abdominal pain
In a very large comparison of children, adolescents, and adults diagnosed with ADHD who participated in controlled DB-PC clinical trials of lisdexamfetamine dimesylate (n = 1020), the investigators found that children experienced a significantly greater incidence of decreased appetite and upper abdominal pain, whereas adults reported significantly more headache and dry mouth (Goodman et al. 2010).
(3) Euphoria
Adults in controlled trials who received customary doses of stimulants can experience euphoria (Rapoport et al. 1980; deWit et al. 2002). School-aged children with ADHD, on the other hand, experience a reduction in euphoria when treated with stimulants (Barkley et al. 1990; Efron et al. 1997; Ahmann et al. 2001; Stein 2003; Sonuga-Barke et al. 2009).
(4) Other differences in affect
Children given stimulants for ADHD are prone to appear emotionally subdued if the dose is high (Daughton et al. 2009), but this effect has not been reported for adults. Tearfulness during stimulant treatment has been noted frequently in very young children (Ghuman et al. 2001; Stein 2003), whereas dysphoria—but not tearfulness—has been reported in adults as a stimulant ADE (Wilens et al. 1995).
Preschool children compared with older children experience more ADEs from stimulants—including irritability and anxiety (Firestone et al. 1998; Ghuman et al. 2001; Greenhill et al. 2008). Like preschoolers, adults may experience increased levels of anxiety with stimulant medication (Spencer et al. 2001; Biederman et al. 2005b; Wender et al. 2011). However, school-aged children with ADHD were reported to experience a reduction in anxiety when treated with methylphenidate (Barkley et al. 1990; Ahmann et al. 1993, 2001; Efron et al. 1997; Sonuga-Barke et al. 2009).
(5) Hallucinations
Hallucinations, mostly visual, occur at an estimated rate of 1/1,200 in youth with ADHD during clinical trials with stimulants (Mosholder et al. 2009), but this has not been reported as a problem for adults given customary therapeutic doses during clinical trials.
(6) Tics
Varley et al. (2001) in a chart review of 555 youth with ADHD treated with stimulants reported that younger children had significantly more tic ADEs than older children. An earlier large chart review by Lipkin et al. (1994) revealed a similar pattern. No reports of stimulant-induced tics in adults were located.
(b) Modafinil. The incidence of rash from modafinil which resulted in discontinuation during clinical trials was 0.8% (13/1,583) for pediatric patients versus 0% (0/4,264) for adults (Provigil FDA Approved Label 8/17/2007).
(c) Atomoxetine
(1) Decreased growth velocity
Atomoxetine can cause a reduction in growth velocity in youth. This is particularly the case for preschoolers and youth aged 6 and 7 years (Kratochvil et al. 2006), but to a somewhat lesser extent in older youth (Spencer et al. 2005; Donnelly et al. 2008). Children younger than age 9 given atomoxetine in clinical trials gained on average less in weight (2.1 kg), and in height (1.2 cm) than predicted after 3 years of treatment (Strattera FDA Approved Label 12/22/2009). Obese adults can lose weight on atomoxetine (Gadde et al. 2006), but this ADE has not been reported for adults with ADHD (Simpson and Plosker 2004).
(2) Somatic and sexual symptoms
Atomoxetine is associated with more ADEs in early childhood than in adolescence (Kratochvil et al. 2006), including a higher frequency of vomiting (Donnelly et al. 2008; Hazell et al. 2008) and somnolence (Wilens et al. 2006). In large clinical trials in adult males, 6%–9% on atomoxetine reported sexual dysfunctions versus 1%–2% on placebo (Simpson and Plosker 2004). Sexual dysfunction ADEs have not been reported for youth in clinical trials with atomoxetine.
Antidepressants
(a) Selective serotonin reuptake inhibitors
(1) Decreased growth velocity
Pediatric subjects (n = 49) treated with fluoxetine for 19 weeks gained on average 1.1 cm less in height (p < 0.004) and 1.1 kg less in weight (p < 0.008) than those on placebo (n = 47) (FDA Pediatric Labeling Changes Prozac 2/18/2005; Nilsson et al. 2004). Selective serotonin reuptake inhibitors (SSRIs) do not affect height growth in adults, but some can increase body weight over time, a response particularly associated with paroxetine (Dannon et al. 2007; Demyttenaere and Jaspers 2008).
(2) Activation and increased sedation
Activation (behavioral disinhibition) as an ADE in response to SSRI compounds occurs on average in 11% of children in DB-PC clinical trials, but only in 2%–4% in adolescents and adults (Carlson and Mick 2003; Safer and Zito 2006; Harada et al. 2008). Activation with SSRIs is particularly common in preschoolers (Safer and Zito 2006; Coskun and Zoroglu 2009). Sedation is reported more often in clinical trials as an SSRI side effect in youth compared with adolescents and adults (Safer and Zito 2006).
(3) Suicidality
SSRIs and other contemporary antidepressants cause a significantly higher average rate of suicidal ideation and suicidal behavior than placebos in youth (Hammad et al. 2006; Barbui et al. 2009). The level of suicidality resulting from these antidepressants remains a statistically significant risk until age 25, after which it prominently declines as a risk (Friedman and Leon 2007; Barbui et al. 2009). Among antidepressants, this ADE in youth is more evident with venlafaxine and paroxetine (Hammad et al. 2006).
(4) Sexual dysfunction
In clinical trials of SSRIs, adult males experience sexual dysfunction ADEs that range from 24% to 73% (median 58%) (Werneke et al. 2006). In a consecutive chart review (n = 34), adolescent males aged 15 and16 years treated with SSRIs reported a 23% rate of sexual dysfunction when queried about this (Scharko and Reiner 2004). This ADE has not been reported in children given these medications.
(b) Mirtazapine. In an 8-week clinical trial for youth given mirtazapine in doses of 15 to 45 mg/day, 49% had a weight gain of at least 7% compared with 5.7% of placebo-treated patients. The mean weight gain for youth on mirtazapine was 4 kg, but1 kg on placebo. In a similar clinical trial of adults with mirtazapine, 7.5% experienced a weight gain that was > 7% above baseline versus 0% for subjects on placebo (Remeron FDA Approved Label 2/12/2008).
(c) Venlafaxine. At the end of a 6-month open label clinical trial of venlafaxine, weight and height increases were significantly less than expected for preadolescents compared with adolescents (FDA Pediatric Labeling Changes, Effexor 2/18/2005). Statistically significant decrements in expected weight as well as in expected height in youth relative to placebo were also recorded by Rynn et al. (2007) following two large 8-week trials of venlafaxine. Again, height decrements have not been reported for venlafaxine treatment in adults.
Antipsychotic (neuroleptic) medications
(a) Antipsychotics: Sedation and withdrawal patterns
Antipsychotics induce more sedation in youth than in adults in clinical trial and postmarketing reports (Realmuto et al. 1984; Woods et al. 2002; Cheng-Shannon et al. 2004). Withdrawal emergent dyskinesia from these medications has been reported mostly in youth (Campbell et al. 1997; Owens 1999); it usually resolves shortly after drug discontinuation (Rancurello et al. 1992; Correll et al. 2007). Tardive dyskinesia—usually with associated oral-buccal movements—is a neuroleptic-induced ADE whose presence and persistence is most common in adults, particularly after age 45 (Owens 1999; Kane 2002).
(b) Haloperidol and other first generation antipsychotics were found in a large chart review (n = 135) to induce dystonic reactions far more frequently in youth than in adults. Also, youth aged 10–19 were found to clearly benefit more than their elders from anticholinergic prophylaxis to prevent this ADE (Keepers and Casey 1987).
(c) Olanzapine
(1) Increased weight gain
Weight gain induced by olanzapine is particularly prominent in preschoolers. In one open trial, the 15 children on this drug (mean age 5.0 years) gained an average of 3.17 kg in the first 8 weeks of treatment (Biederman et al. 2005a). Compared with adult subjects administered olanzapine in clinical trials, adolescents experienced a greater proportion (65.1% vs. 35.6%) of olanzapine-induced increases of ≥ 7% in baseline body weight (Kryzhanovskaya et al. 2009).
(2) Metabolic and laboratory changes
Olanzapine-treated adolescents compared with olazapine-treated adults had significantly greater increases in abnormal liver function tests, and prolactin blood levels (Kryzhanovskaya et al. 2009). On the other hand, adults to a greater degree than adolescents responded to olanzapine with larger mean increases in fasting glucose, and fasting triglycerides, and a greater incidence of abnormal changes from normal to high levels in total cholesterol during treatment (Alfaro 2007; Kryzhanovskaya et al. 2009). In a recent FDA summation—which in some respects is at variance with published reports—the following was written: “Compared to patients from adult clinical trials, adolescents treated with oral Zyprexa were likely to gain more weight, experience increased sedation, and have greater increases in total cholesterol, triglycerides, LDL cholesterol, prolactin and hepatic aminotransferase levels” (FDA Pediatric Labeling Changes, Zyprexa 12/4/2009).
(3) Abnormal postmarketing somatic and behavioral reports
In postmarketing reports to the FDA from the year 2000 that were organized in terms of risk per 10,000 persons, it was found that significantly more olanzapine ADE reports were received for youth than for adults with respect to rash, convulsions, sedation, weight gain, vomiting, agitation, and hostility (Woods et al. 2002).
(d) Clozapine can cause seizures in both adults and children, but the rate during clinical trials has been reported to be higher in children (Kumra et al. 1996; Sporn et al. 2007). Likewise, neutropenia and tachycardia with clozapine treatment have been reported to be more common ADEs in children than in adults (Sporn et al. 2003).
(e) Risperidone
(1) Increased weight gain
Risperidone causes more weight gain relative to baseline body weight in youth than in adults in analyses of pooled DB-PC clinical trials (Safer 2004). Risperidone-induced weight gain is most prominent in young children; this ADE progressively decreases in degree with advancing age (Safer 2004; Maayan et al. 2010). It is of interest that children (aged 9 and 10 years) given risperidone for 1 year also experienced greater average gains than expected in height, from the 48th to the 53rd percentile, averaging a 1.2 cm increase (Dunbar et al. 2004; Risperidal FDA Approved Label 12/8/2009).
(2) Somatic symptoms
Sialorrhea (drooling) was reported at a rate of 29%–35% in risperidone-treated youth found to have autism compared with a 4% rate at baseline (Aman et al. 2005). In adults and in youth without autism treated with risperidone, the reported rate of sialorrhea is 2% (Fleischhacker et al. 2001; Cheng-Shannon et al. 2004; PDR 2008).
(3) Metabolic and laboratory changes
Children with autism whose median age was 6 years and who were treated with risperidone for 16 weeks (n = 124) experienced statistically significant blood level increases in fasting insulin, insulin resistance, fasting glucose, leptin, triglycerides, and liver enzymes (Boorin et al. 2010). Statistically significant laboratory changes in an 11-week study and two 8-week trials with risperidone in adolescents yielded notable but clearly less striking results. In these studies, significant increases occurred in triglycerides (Correll et al. 2009), serum transaminase (Sikich et al. 2004), and insulin (Findling et al. 2010), as well as significantly decreased HDL cholesterol levels (Findling et al. 2010). In adolescents given risperidone for over 1 year (n = 19), triglycerides became significantly elevated (Laita et al. 2007). Such blood level deviations have been far less commonly found in adults treated with risperidone (de Leon et al. 2007, PDR 2008).
(f) Quetiapine
(1) Increased weight gain
In two major 8-week trials of quetiapine (300–600 mg) in adults, the mean weight gain ranged from 1.1 to 1.7 kg (McElroy et al. 2010; Young et al. 2010). In the 10.8-week study by Correll et al. (2009), the mean weight gain for antipsychotic drug-naïve youth (n = 36) was 6.1 kg.
(2) Metabolic and laboratory changes and dystonia
Youth in clinical trials given quetiapine experienced increases in thyroid-stimulating hormone levels, acute dystonia, and prolactin blood levels to a greater degree than adults (Seroquel FDA Approved Label 12/2/2009).
(3) Cardiovascular changes
Official label statements on quetiapine-induced cardiovascular changes are as follows: “Increases in blood pressure and potentially clinically significant increases in heart rate (>110 bpm) occurred in children and adolescents and did not occur in adults” (FDA Pediatric Labeling Changes, Seroquel 12/2/2009). “Orthostatic hypotension occurred more frequently in adults (4%-7%) compared with children and adolescents (<1%)” (Seroquel FDA Approved Label 12/2//2009). Of course, heart rate is characteristically higher and more variable in children than in adolescents and adults (Yeragani et al. 1994; Silvetti et al. 2001).
(4) Electrocardiogram recorded differences
In electrocardiogram (ECG) recordings, youth given quetiapine in short-term clinical trials had a twofold higher rate of increased levels (>20 mm Hg) of supine systolic blood pressure than adults (Seroquel, Pediatric Advisory Meeting Materials Dec. 8, 2009). Average heart rate increases in adults given quetiapine were 6 bpm (7–1 bpm) (quetiapine minus placebo) in 3–6-week clinical trials following doses of 300–600 mg (Seroquel FDA Approved Label 12/2/2009). By comparison, youth given quetiapine in 3–6-week PC clinical trials had average ECG recorded heart rate increases of 7.1 (3.8 quetiapine (-) − 3.3 placebo) bpm and 14.5 (12.8 (-)-1.7) bpm following daily doses of 400 mg of quetiapine, and average increases of 15.1 (13.4 (-) − 1.7) bpm on doses of 600 mg/day. In a 6-week clinical trial of 800 mg/day of quetiapine compared with placebo, youth had a mean heart rate increase of 14.5 (11.2 (-)-3.3) bpm (Seroquel FDA Pediatric/Advisory Committee, briefing information 12/8/2009). When the clinical trial youth population was divided by age group, the 10–12-year-olds had greater supine heart rate increases than adolescents aged 13–17 (Seroquel FDA Pediatric/Advisory Committee, briefing information 12/8/2009).
(g) Aripiprazole induced weight gain in drug trials of 3 to 11 weeks duration in youth ranged from 0 to 4.4 kg (n = 451) (Correll et al. 2009; Fraguas et al. 2010). In adults, weight gain with aripiprazole in a 6-week trial was less than 1 kg (Abilify FDA Approved Label 2009).
(h) Ziprasidone led to a mean prolongation of the QTc interval (mean = 23–28 msec) in two open label prospective trials in youth (n = 20 and 29) that included regular ECG recordings (Blair et al. 2005; Correll et al. 2008); this is higher than the mean increase in adults, which ranges from 11 to 21 msec (Blair et al. 2005; Miceli et al. 2010).
Lithium
In clinical trials, vomiting has been reported frequently (e.g., > 10%) for youth, but not for adults (Campbell et al. 1991; Silva et al. 1992; Malone et al. 2000; Amsterdam and Shults 2008 2010).
Anticonvulsant drugs
(a) Valproic acid has been associated with increases in liver toxicity—particularly in infancy, but also from ages 2 to 10. The rate of this ADE is lower for adolescents and adults (Dreifuss et al. 1987; Koenig et al. 2006). Valproic acid is also more likely to induce pancreatitis in youth than in adults (Yazdani et al. 2002; Gestner et al. 2007). A number of studies indicate that teenage girls are more likely than their elders to experience a valproate-induced polycystic ovarian syndrome (Isojarvi et al. 1993; Vainionpaa et al. 1999; Joffe et al. 2006).
(b) Gabapentin has been reported in open label and DB trials to frequently induce behavioral dyscontrol in youth (Khurana et al. 1996; Lee et al. 1996; Glauser 2004; Wolf et al. 1995). Adult studies have not reported this ADE (Weintraub et al. 2007).
(c) Phenytoin causes more gingival hyperplasia in children than in adults in findings from community- and outpatient-based research (Livingston et al. 1979; Casetta et al. 1997; Majola et al. 2000).
(d) Zonisamide in postmarketing reports has been reported to cause a greater frequency and more severe episodes of oligohydrosis, hyperthermia, and metabolic acidosis in youth than in adults (Zonisamide FDA Safety Alert 2/23/2009).
(e) Topiramate
(1) Growth retardation
In open extended trials of topiramate for seizure disorders, growth retardation has occurred with greater frequency in toddlers compared to older pediatric patients (FDA Pediatric Labeling Changes, Topamax 12/22/2009). Significant decreases in body mass index in 53 children (mean age = 7.3 years) who were treated with topiramate for at least 1 year have also been reported (Reiter et al. 2004).
(2) Somatic and cognitive changes
More FDA postmarketing reports on topiramate have been received for youth than for adults on decreased sweating and elevated body temperature, a finding noted because “the majority of case reports [for these ADEs] have been in children” (Topamax FDA Safety Alert 7/1/2003). The manufacturer reported that in clinical trials with topiramate for epilepsy “… the incidences of cognitive/neuropsychiatric adverse events in pediatric patients were generally lower than observed in adults” (Topamax FDA Approved Label 12/22/2009).
(f) Lamotrigine and carbamazepine are approximately twice as likely to induce a rash in youth as in adults on evidence from multicenter trials (Pellock 1987; Mackay et al. 1997; Guberman et al. 1999; Calabrese et al. 2002). Rash is also a frequent occurrence (often at a rate of 10% or more) with other anticonvulsants, such as divalproex, oxcarbazapine, and topirimate (Cueva et al. 1996; DelBello et al. 2005; Hellings et al. 2005; Wagner et al. 2006), supporting the possibility that this ADE may represent an age-related, drug class vulnerability (Aronson 2006).
(g) Phenobarbital has been repeatedly found to induce agitation in young children more than in older individuals based on large open-label and DB-controlled clinical trials (Hardin 2000; Glauser 2004). Long-term use also can impair intellectual development in infants and toddlers based on a large PC trial and a 2-year follow-up (Farwell et al. 1990).
(h) Levetiracetam treatment resulted in a 38% rate of untoward behavior patterns in clinical trials in children compared with a 13% rate for adults (Benjamin et al. 2009).
Factors That Influence Drug-Induced, Age-Related Vulnerabilities
(1) Bodily growth. Growing children who take stimulant drugs are particularly sensitive to its anorexic effects, the mechanism that is most associated with the suppression of growth velocity (Safer et al. 1972; Poulton 2005; Swanson et al. 2009). This appetite suppression ADE is most pronounced in preschoolers (Greenhill et al. 2008). At the other extreme, drugs that increase appetite will likely lead in children—more than in adults—to an abnormal increased rate of weight gain (Safer 2004).
(2) Activity level. Children are naturally more restless than adults. Thus, drugs such as SSRIs that can create activation are more likely to induce behavioral disinhibition in children than in adolescents and adults (Safer and Zito 2006; Harada et al. 2008). At the other extreme, stimulants that can decrease activity level will have a more prominent inhibiting effect on those most active, who are usually younger (Stein 2003).
(3) Vomiting, fevers, crying, and rash in association with infections occur more frequently in children than in adults (Hay et al. 2005; Liu et al. 2010), and thus represent potential vulnerabilities that are often accentuated in youth after exposure to psychotropic medication. Similarly, since tic disorders are most common in childhood (Snider et al. 2002), it is likely that children would experience more drug-induced tic movements than adults (Varley et al. 2001).
(4) Developmental disabilities particularly in the preschool years increase vulnerability to psychotropic ADEs. In the Dreifuss et al. (1987) study, for example, preschool children with seizure disorders were more at risk for liver toxicity from valproic acid than were older children with seizure disorders. Further, those with both seizure disorders and congenital abnormalities were most at risk to valproic acid liver toxicity, particularly if they were younger and were receiving other anticonvulsants concomitantly.
(5) Brain maturation is clearly pronounced in childhood. During early childhood, there is massive synaptic pruning and selective strengthening of most favored synapses (Wiznitzer and Findling 2003), an increased level of myelination (Benes et al. 1994) and increasing white matter density (Paus et al. 1999). Also, during this developmental period, levels of neurotransmitters dramatically change. For instance, serum dopamine beta hydroxylase and plasma norepinephrine increase with advancing age (Young et al. 1984), whereas CSF homovanillic acid decreases with advancing age (Leckman et al. 1980). An example of the influence of brain maturation can be found in age-grouped responses to nicotine cigarette use which leads more readily to dependence when smoking is initiated in adolescence rather than in adulthood (Jamner et al. 2003; DiFranza and Richmond 2008).
(6) Pharmacokinetic drug responses generally differ between childhood and adulthood, whereas adolescents and adults usually have similar pharmacokinetic profiles (Kearns et al. 2003; Green 2007). For the most part, in childhood as compared with adulthood, the half-lives of psychotropic drugs are shorter, skin absorption is greater and faster, plasma drug levels have sharper peaks, and the hepatic biotransformation of drugs into active metabolites is more rapid (Heim 1987; Rancurello et al. 1992; Cohen and Jermain 1998; Kearns et al. 2003). How these above-mentioned differences influence ADE patterns in young children has been the product of little research. An example of ADE age differences in enzymatic metabolism is the less efficient metabolism of ethanol with advancing age (Inaba and Cohen 2000). This partly explains why adolescents generally have less dysphoric effects from alcoholic beverages than their elders (Locke and Newcomb 2001; Schuckit and Smith 2004).
(7) Laboratory blood level and cardiovascular norms are often different for youth than for adults. For example, the upper level of the normal range of blood triglyceride is lower in youth than adults (110 mg/dL vs. 150 mg/dL), as is the upper level of the total cholesterol blood range (170 mg/dL vs. 200 mg/dL) (Laita et al. 2007; Correll et al. 2009). Likewise, cardiovascular norms differ substantially in youth, who characteristically have a higher heart rate and a lower blood pressure compared with adults (Silvetti et al. 2001). Thus, laboratory and cardiovascular norms for youth will need to be considered when cardiac and metabolic ADEs are reported.
(8) Body and hormonal developmental differences. During their early developmental years, children experience more rapid biological changes than their elders and are particularly vulnerable to neurotoxic hazards such as lead and mercury (Weiss 2000). Children also differ biologically from adults in that they have comparatively more total body water, less adipose tissue, relatively less gastric acidity, a greater liver mass to total body mass ratio, and more efficient renal function (Heim 1987; Rancurello et al. 1992; Murry et al. 1995; Cohen and Jermain 1998). Later—as children enter puberty, estrogen and testosterone blood levels markedly increase (Belgorosky and Rivarola 1987; Cutler 1997), which could explain the new onset of vulnerability by adolescents to SSRI sexual function ADEs.
(9) Preschool years. Preschool youth in general have more ADEs with medication treatment than school-aged youth. In a very large national outpatient and emergency department dataset, the incidence of ADE visits per 1,000 persons was nearly twice as great in youth under age 5 than in youth aged 5–17 years (Bourgeois et al. 2010). Preschool youth were found to be particularly vulnerable to allergic reactions after first-time exposures to medications (Bourgeois et al. 2009).
Conclusion
Children during their preschool years are generally more vulnerable to ADEs after the administration of most psychotropic medications than are adolescents and adults. Preschool children are particularly vulnerable to ADEs from valproic acid, diphenhydramine, stimulants, antipsychotics, phenobarbital, and SSRIs. During development, children are also more vulnerable to compounds that can impair normal development, such as steroids, tetracycline, and lead intake (Witkop and Wolf 1963; Reinisch et al. 1995; Yeh et al. 2004). Further, young children with developmental impairments are more prone to liver toxicity from valproate than are children without such disorders (Dreifuss et al. 1987).
Children's disorders, vulnerabilities, and body development differ from adults in numerous respects. Compared with adults, children have higher rates of disorders such as enuresis, encopresis, separation anxiety, simple phobia, stuttering, dyslexia, overactivity, and tic movements, and they are less at risk for disorders such as dementia, persistent tardive dyskinesia, and melancholia. As a group, they are more prone to experience febrile seizures, vomiting, rash, crying, temper tantrums, and undue restlessness. They have a number of lower blood level laboratory ranges of normal values, and children commonly have more rapid pharmacokinetic responses to medications. In clinical trials of antidepressant drugs, they also have a greater response to placebo than do older persons (Bridge et al. 2009). Age differences thus often alter children's responses to psychotropic medications.
Clinical Significance
Long-term medication safety studies for youth are few and ADEs in children are under-reported (Hazell and Shakir 2006; Benjamin et al. 2009). Consequently because of this and because of children's extra vulnerability to many psychotropic medications, it behooves pediatric physicians to be particularly careful in their treatment decisions that pertain to the prescribing of psychotropic medication.
Footnotes
Disclosure
Daniel J. Safer has no conflicts of interest or financial ties to disclose.
Acknowledgment
I would like to thank Julie M. Zito, Ph.D., for her many helpful suggestions during the development of this manuscript.
