Psychopharmacotherapy in Early Life

The problems associated with drug therapy during the childbearing and perinatal periods, as well as the rapidly developing field of child and adolescent psychopharmacotherapy, require increased attention. We consider the role of psychotropics in the early life period (i.e., from conception through adolescence), dividing the chapter into two major areas:

Psychopharmacotherapy during the childbearing and perinatal periods
Psychopharmacotherapy in children and adolescents

Psychopharmacotherapy during the Childbearing and Perinatal Periods

Perhaps, no other area of psychiatry raises the anxiety of a treating clinician more than the use of drugs in this population. Thus, the riskbenefit ratio must be considered from the perspectives of

The embryo, fetus, and neonate:
- Spontaneous abortion or premature labor
- Toxicity or withdrawal symptoms
- Morphological teratogenicity
- Breastfeeding from a mother on a psychotropic
- Future behavioral teratogenicity, because psychotropics target the brain and the effects on fetal brain morphogenesis may not be apparent for several years
The mother:
- Premenstrual dysphoric disorder (PMDD)
- Polycystic ovarian syndrome (PCOS)
- Decreased or increased likelihood of conception
- Emotional stress associated with pregnancy (1)
Both:
- Alterations in stress-related systems (e.g., neuroendocrine, immune/inflammatory, cardiovascular) perhaps mediated by placental corticotropin-releasing hormone (CRH) (2)
- The consequences of an untreated or inadequately managed mental disorder
- Postpartum episodes (e.g., depression, psychosis)

These issues are growing in importance in part because of the increasing number of pregnancies in women with more severe and chronic mental disorders (3,4). Further, studies indicate that 50% to 65% of pregnancies in the United States are not planned (5), and up to one third of women receive psychotropics during pregnancy (6). For example, women are at risk for substantial mood dysregulation during the perinatal period, including

Clustering of mood and anxiety disorders during the childbearing years
High prevalence of mood, behavioral, and cognitive symptoms during pregnancy, although depression prevalence during pregnancy is comparable to that of the nongravid state (7)
Bipolar disorder (BPD), which may destabilize during pregnancy
Postpartum mood dysregulation, particularly in patients with a prior history of mood disorder, potential genetic predisposition, or elevated CRH levels during late pregnancy (8,9)

Thus, the need for various psychotropics may be even greater during this period (10), but these drugs may also impact the potential to
become pregnant or produce adverse effects in both mother and child during and after pregnancy.

TABLE 14-1 FOOD AND DRUG ADMINISTRATION RISK CATEGORIES

Category	Interpretation
A	Controlled studies show no risk
	Adequate, well-controlled studies in pregnant women have failed to demonstrate risk to the fetus.
B	No evidence of risk in humans
	Either animal findings show risk, but human findings do not, or if no adequate human studies have been done, animal findings are negative.
C	Risk cannot be ruled out
	Human studies are lacking, and animal studies are either positive for fetal risk or lacking as well. However, potential benefits may justify potential risk.
D	Positive evidence of risk
	Investigational or postmarketing data show risk to the fetus. Nevertheless, potential benefits may outweigh risks.
X	Contraindicated in pregnancy
	Studies in animals or humans or investigational or postmarketing reports have shown fetal risks that clearly outweigh any possible benefit to the patient.

It is also important to weigh both known and potential risks of psychotropics against the risks of untreated symptoms (11). Thus, the disease itself or associated unhealthy behaviors (e.g., substance use or abuse) may contribute to negative outcomes. If used during pregnancy, medication with the lowest possible risk should be employed (e.g., Food and Drug Administration [FDA] risk categories in Table 14-1). Further, a single medication at a higher dose is preferable to polypharmacy. In thi s context, a series of excellent reviews are published (10,11,12 and 13) to help guide clinicians in the most appropriate management of these patients.

ANTIPSYCHOTICS

The issue of antipsychotics during pregnancy warrants careful scrutiny, given the apparent increasing birth rate in patients with psychotic disorders and the need for chronic drug therapy to maintain optimal functioning (14). In this context, many patients will require antipsychotic therapy during pregnancy. Although the existing literature does not identify a potential for abnormal fetal developmental or later behavioral teratogenicity with antipsychotic exposure, large well designed, prospective studies are not available to guide our choices (15). Clozapine is listed in category B, and all other second-generation antipsychotics (SGAs) are in category C. Further, placental passage may differ among agents. For example, one study found that the placental passage ratio was highest for olanzapine (72%) followed by haloperidol (66%), risperidone (49%), and quetiapine (24%) (16). In this context, we have modified the clinical guidelines of Miller (3,4) as follows:

Avoid low-potency phenothiazine antipsychotics, if possible, during the period of highest risk for teratogenicity (i.e., 4 to 10 weeks after conception) and near term because of possible maternal hypotension and possible neonatal adverse effects
Taper antipsychotics, if possible, 2 weeks before the estimated date of confinement to minimize withdrawal effects in the neonate
Resume the antipsychotic immediately postpartum in chronically psychotic women to minimize the possibility of a postpartum episode
Use higher potency agents (e.g., haloperidol) to minimize sedation, orthostasis, gastrointestinal slowing, tachycardia, and morphological teratogenicity (17)
Avoid clozapine, if possible

Immediately discontinue the antipsychotic and use bromocriptine to manage a neuroleptic malignant syndrome, if it develops
During breastfeeding, SGAs such as risperidone and quetiapine are preferable while clozapine and olanzapine should be avoided (18,19)
Avoid routine prophylaxis with antiparkinsonian agents, including benztropine, diphenhydramine, and amantadine, all of which are associated with congenital anomalies. Calcium supplementation may be a useful alternative, and propranolol or atenolol may be used for akathisia if cardiovascular status is stable (12,20).

ANTIDEPRESSANTS

Serious depression develops in approximately 10% of pregnant women and may occur at a higher rate in women with severe obstetric risk
(e.g., diabetes) versus those with a low-risk pregnancy (21,22 and 23). Women on antidepressants who stop when they become pregnant may be at substantially greater risk of relapse than those who remain on their medication (24). Further, a recent national survey found a significantly higher prevalence of major depression in postpartum versus nonpregnant women (25). In this regard, it is important to remember that depression itself may also have a deleterious impact on later development.

Most data on teratogenicity involve the tricyclic antidepressants (TCAs) and the selective serotonin reuptake inhibitors (SSRIs). Thus far, no major deficits are identified; however, neonatal toxicity; withdrawal symptoms; and a possible increased risk for gestational hypertension, preeclampsia, and persistent pulmonary hypertension of the newborn are reported (26,27). Fluoxetine may pose a greater risk in infants for perinatal adverse and withdrawal effects but not for major fetal anomalies. Monoamine oxidase inhibitors (MAOIs) are known animal teratogens, but the lack of human data limits any conclusions. Insufficient experience also precludes any firm suggestions about bupropion, trazodone, venlafaxine, desvenlafaxine, duloxetine, nefazodone, or mirtazapine. Clinical guidelines for the management of depression during pregnancy include

Nondrug approaches (e.g., cognitive-behavioral therapy (CBT), interpersonal psychotherapy, bright light therapy) are always preferable (28,29 and 30)
If necessary, agents that have been better studied and have fewer adverse effects are preferred (e.g., fluoxetine, citalopram, escitalopram) (31,32 and 33)
If possible, paroxetine should be avoided due to a greater risk for congenital cardiac malformations with first trimester exposure (34,35). If exposed to this agent during early pregnancy, fetal echocardiography should be considered.
Nortriptyline and desipramine may be the TCAs of choice, given their extensive assessment during pregnancy and well-known therapeutic concentrations, which should be monitored (36)
Dose increase may be necessary due to increased turnover in the second half of pregnancy (37)
If an antidepressant is stopped during pregnancy, it should be gradually tapered to avoid maternal or fetal withdrawal syndromes
If clinically possible, drug tapering or dose reduction should begin 3 weeks before the estimated date of conception (EDC) (12,38)
Electroconvulsive therapy (ECT) modified for pregnancy may be a reasonable choice for acutely suicidal or psychotically depressed patients (39) (see Chapter 8)
Those with a prior history may be more susceptible to postpartum depression; psychotherapy alone or combined with maintenance antidepressant therapy may benefit such patients and should be done with a team approach (e.g., psychiatrist, obstetrician, pediatrician) to obtain the optimal outcome (40,41,42,43 and 44)

MOOD STABILIZERS

BPD is a recurring illness whose course is even more complicated during pregnancy and the postpartum period (45,46). Further, many effective drug therapies (e.g., lithium, valproate [VPA], carbamazepine [CBZ]) carry significant pharmacokinetic, physiological, and teratogenic risks and are classified as pregnancy category D. The effects of lithium during the first trimester on the developing fetal heart (i.e., Ebstein’s anomaly) are well publicized, but it appears to be a weak cardiovascular teratogen (47). Dosing requirements usually increase during pregnancy but should be decreased (e.g., 50%) during labor because of the large fluid loss, which may precipitate lithium toxicity. Detailed discussions of these issues are found in Chapter 10. Although teratogenicity is well documented for VPA or CBZ alone, their combined use may be particularly detrimental (48).

The following are clinical strategies to minimize the risks to mother and fetus, either from drug-induced anomalies or the potential ravages of the disease process:

There should be frank discussion about family planning
The risk-to-benefit ratio must be carefully weighed when contemplating whether to initiate, continue, or withdraw drug therapy (49)
Lithium, VPA, and CBZ (pregnancy category D) should be avoided, especially during the first trimester, if clinically feasible
Consider alternative therapies, such as lamotrigine (pregnancy category C), SGAs, or ECT modified for pregnancy
If lithium is necessary during the first trimester, sonography can clarify the presence and severity of anomalies such as Ebstein’s tricuspid valve defect
Lithium should be given in smaller, divided doses to avoid higher peak plasma levels
If the patient is exposed to VPA or CBZ, the presence of neural tube defects should be evaluated (e.g., serum alpha protein, amniocentesis, ultrasound), especially if they are used together
If anticonvulsants are used, daily folate (up to 4 mg/day) may decrease the risk of neural tube defects, and vitamin K (20 mg/day) may prevent drug-induced bleeding
With BPD, the increased risk of a postpartum episode warrants the resumption of medication compatible with lactation soon after delivery (50).

ANTIANXIETY AGENTS

The experience of pregnancy itself is often anxiety provoking, and symptoms sufficient to warrant drug therapy are common in this group. Although this discussion primarily focuses on the benzodiazepines (BZDs), antidepressants and buspirone are also used for women of childbearing age with certain anxiety-related disorders. The best-documented adverse effect of the BZDs is a neonatal withdrawal syndrome that can occur with several of these agents (e.g., diazepam, alprazolam, triazolam). In addition, there is a weak positive relationship between diazepam exposure and oral clefts (51).

Clinical considerations and approaches should include

Alternative, nondrug management of anxiety whenever possible (e.g., behavioral therapies, interpersonal psychotherapy, relaxation techniques, cessation of stimulants such as caffeine)
Consider an SSRI (except paroxetine) during the first and second trimesters
Avoid BZDs in first trimester, if possible
If a BZD is necessary, consideration of lorazepam because of its possible lower accumulation in fetal tissue; but avoid the use of BZDs until after the 10th week of gestation (particularly diazepam) to preclude oral defects
Gradual tapering of a BZD before delivery to minimize any neonatal withdrawal phenomena
Avoidance of diphenhydramine because of both fetal teratogenic and withdrawal complications (52)
Reversal of fetal or neonatal toxicity with flumazenil (53)

Interspersing treatments (i.e., psychotherapy, SSRIs, BZDs) during the three trimesters may help achieve the best outcome while posing lower risks (54).

NURSING

Since relapse rates for various psychiatric disorders are often increased in the postpartum period, it is often necessary to continue psychotropic drug therapy. The safest strategy is to not breastfeed if the mother is taking a psychotropic. Breastfeeding, however, is desirable for a variety of reasons (e.g., decreased infections, lower neonate mortality rates); and in this context, certain issues should be considered when psychotropic(s) are used, including

Risks associated with continued breastfeeding
Risks associated with stopping medication
Polypharmacy, which should be avoided and the lowest effective dose used
Weaning if the neonate experiences adverse effects and/or sustained drug plasma levels (typically a <10% neonate-calculated dose exposure is considered safe)

Based on very limited data:

Drugs to avoid include
- Lithium (alternatives may be VPA, CBZ, or SGAs)
Drugs of concern include
- Lamotrigine
- Clozapine (olanzapine may be an alternative) (55)
Drugs that appear to be acceptable include
- SSRIs (e.g., sertraline) (56,57 and 58)
- Venlafaxine/desvenlafaxine (59)

PREMENSTRUAL DYSPHORIC DISORDER

Premenstrual dysphoric disorder (PMDD) affects 3% to 8% of women in their reproductive years,
often requiring the use of various psychotropics. Unlike the much milder and more prevalent premenstrual syndrome (PMS), PMDD is characterized by

Persistent irritability or anger
Tension, marked anxiety
Dysphoria, hopelessness, self-deprecation
Mood lability, fatigue, decreased energy
Dyssomnia
Appetite fluctuations
Feelings of being out of control
Other physical symptoms

Further, these symptoms have a significant impact on a woman’s interpersonal relations and general lifestyle, often requiring the help of a mental health professional. Typically, at least five symptoms are present in most menstrual cycles within the previous year, usually during the last week of the luteal phase. These symptoms begin to remit within a few days after the onset of the follicular phase and are typically absent during the week after menses (60). Pharmacological management focuses on several different agents, including (61)

Lithium
Alprazolam
Various antidepressants

Although trials with lithium were disappointing, evidence indicates that alprazolam (0.75 to 4 mg/day) during the luteal phase in women with a well-defined symptomatic period can help (62,63,64 and 65). The majority of positive studies found that later-generation antidepressants, especially those with strong serotonergic effects, are helpful and may produce a more rapid onset of action when used for PMDD than for depression (66). Thus, studies with SSRIs (e.g., fluoxetine, sertraline, paroxetine), nefazodone, venlafaxine, and clomipramine all show promise. In several trials, lacteal phase treatment was at least as effective as continuous treatment in women with PMDD (67).

Effective nonpharmacological strategies include exercise, relaxation training, calcium supplementation, increased complex carbohydrate consumption, and CBT (68).

POLYCYSTIC OVARIAN SYNDROME

This is a heterogeneous syndrome of hyperandrogenic, chronic anovulation of unknown cause that differs from polycystic ovaries (PCO) (69). It affects young women of childbearing age, decreasing the likelihood of conception and predisposing the patient to significant medical complications. PCO syndrome (PCOS) has a prevalence of 5% to 10% (70,71), but only a small percentage of these individuals clinically exhibit menstrual irregularities or hyperandrogenism. This syndrome is highly correlated with excessive weight (i.e., up to 80% of women with the syndrome are overweight or obese) and is manifested by

Reproductive dysfunction:
- Ovulatory dysfunction
- Menstrual disorders
- Infertility
- Miscarriage
Metabolic derangements:
- Insulin resistance, hyperinsulinemia
- Impaired glucose tolerance
- Dyslipidemia
- Noninsulin-dependent diabetes
Emotional sequelae due to factors such as
- Hirsutism
- Female pattern, androgenic alopecia
- Acne
Long-term health effects, such as
- Cardiovascular disease
- Endometrial hyperplasia or malignancy
- Obstructive sleep apnea

Although the pathophysiology of PCOS is uncertain, several important factors include

Obesity
Hereditary disorder of insulin action (72)
Hypothalamic-pituitary axis dysfunction
Ovarian defects
Epilepsy
Antiepileptic drugs (e.g., VPA, others)

Treatment goals in women of childbearing age involve regulation of the menstrual cycle, management of peripheral hyperandrogenism, and prevention of diabetes and subsequent complications (e.g., cardiovascular disease) (69). Approaches include

Diet and exercise to reduce weight
Supportive or group psychotherapy to manage the psychological impact
Agents such as oral contraceptives to reverse hyperandrogenism
Agents such as clomiphene to increase fertility
Insulin-lowering medications (e.g., metformin) to regulate glucose levels
If clinically feasible, consideration of agents other than VPA to manage epilepsy or BPD (see Chapter 10)

Psychopharmacotherapy in Children and Adolescents

Arguably, no area in clinical psychopharmacotherapy has greater potential benefits than for children and adolescents. Study methodology and the scope of pharmacotherapy research in youth, however, is limited in comparison to adults. Of prescription drugs marketed in the United States, most are not approved by the FDA for use in children (73). Stimulants such as methylphenidate were approved for children ≥6 years of age. Some antidepressants, such as sertraline and fluvoxamine, are approved for obsessive-compulsive disorder (OCD) but not for depression in school-aged children (74). Although there is an increased trend in prescribing psychotropic medications for children between 2 to 5 years of age (75), most have not received regulatory approval for children <6 years of age or been adequately tested for efficacy or safety in this younger age group (76).

Thus, child and adolescent psychiatrists are faced with the conundrum of either depriving younger patients of potentially effective medications or prescribing such treatment “off-label,” without optimal information on dosing, efficacy, and safety in this population. Although placebocontrolled trials were once considered difficult to conduct in youth, more recently, there have been several large, multisite, controlled studies and others are in progress that will better inform treatment decisions.

A critical aspect of this situation is that early and effective treatment intervention in a disease process may lessen long-term sequelae. It is equally important to consider the unknown, long-term effects of medications that may alter intracellular neurochemistry and genetic activity (77). Thus, the long-term effects of intracellular changes on growth, development, and disease trajectory are virtually unknown (78). Weighing the costs and benefits of various medications and making judicious choices in the absence of clear evidence is a continuous challenge in child psychiatry. For example, if left untreated, symptoms can interfere with the development of self-esteem, family and peer relationships, and school and workplace performance. Thus, consequences are likely to persist even after the resolution of the acute episode, possibly profoundly affecting further psychosocial adjustment (79).

There is a growing awareness on the part of the government, the medical profession, the pharmaceutical industry, and the public about the importance of having empirical data on which to base medication treatment decisions for children and adolescents. Both the National Institutes of Mental Health (NIMH) and the FDA are taking steps to increase the amount of information available on the optimal medication treatment in this population. At the request of the director of NIMH, the Institute of Medicine formed a committee to assess the status of research in mental disorders affecting youth. The resulting report, “Research on Children and Adolescents with Mental, Behavioral, and Developmental Disorders,” led to a 5-year plan to stimulate a wide range of clinical research (including psychopharmacological) and to develop young investigators in this area (80,81). During the same time frame, the American Academy of Pediatrics issued its “Guidelines for the Ethical Conduct of Studies to Evaluate Drugs in Pediatric Populations” (82). To encourage research, the FDA in the late 1990s offered 6 months of patent extension to manufacturers of selected approved drugs if they conducted appropriate studies in children and adolescents. This approach meant that a company could more than recoup the cost of the study by the additional revenue generated during this extended period.

As a result of these efforts, more research is under way in child and adolescent clinical
psychopharmacology than ever before. The full impact of this increased activity, however, will take several years to realize due to the lag time between initiating and completing such trials. As data from these pediatric clinical trials are published, they are subjected to more scrutiny, leading to augmented scientific rigor (83,84 and 85). Some of the minimum standards include providing a consort chart (83), details of randomization methods (84), explanations of varying methodology in calculating effect sizes (85), and improved methods of eliciting adverse events that are central to patient safety (84). Several newly completed and published studies are summarized in this chapter.

TREATMENT ISSUES INFORMING PHARMACOTHERAPY PRACTICE IN PEDIATRIC PSYCHIATRY

There are several psychosocial and pharmacokinetic differences between adults and youth that impact decision making in pediatric psychiatry

Psychosocial Aspects

First, there are important differences between adults and children or unemancipated adolescents in the physician-patient relationship. Thus, the child psychiatric patient is often brought to the clinician because someone else (e.g., parent, teacher) is concerned or annoyed by certain behaviors. Accordingly, the patient may be a passive, if not reluctant, participant in treatment. Second, there is a greater possibility with children than with adults that a beneficial medication may have deleterious effects on growth and development. Therefore, clinicians need to consider the following questions carefully before initiating treatment:

Do the needs and desires of the child or adolescent and the parent conflict or are they in synchrony?
What is the patient’s social and family situation and how will it influence treatment outcome?
Will the parent or guardian be able to assist with the administration and monitoring of the medication?
What other forms of treatment may be needed (e.g., education about the condition and about better behavioral management techniques, family therapy, or individual psychotherapy)?
What does the patient think about his or her condition, the need for treatment, and the specific treatment being recommended?
How will the treatment affect the patient’s selfconcept and relations with others?
How is the patient doing in school? (e.g., if there are significant changes in level of functioning, the time course and magnitude of the changes should be carefully assessed).

Pharmacokinetic Issues

Ideally, drug dose in children and adolescents is based on systematic studies in this age group. Although such pharmacokinetic data are always an important factor, optimal dosing should also be based on efficacy and safety studies in these populations, since children may be more sensitive to the beneficial or adverse effects of specific medications. Until sufficient data are available, optimal dosing in children and adolescents will be difficult and by necessity often based on extrapolations from adult studies and clinical experience. Hence, as with adults, the guiding principle is to aim for the lowest effective dose. Other important related issues to consider are

Does the child or adolescent have a condition of sufficient severity to warrant medication?
Have all the options been reasonably discussed and their relative merits and liabilities weighed?
What outcome parameters will be used to document the potential beneficial and adverse effects of the medication?
Is the addition of a second or third medication necessary (e.g., if the first treatment fails, should it be discontinued rather than resorting to polypharmacy)?

Although the concentration-response curves for efficacy and safety may differ in children and adolescents compared with adults, the prescriber can use therapeutic drug monitoring (TDM) for at least some newer medications to determine whether the patient is achieving a concentration proven effective and safe in adults (see Chapter 2).

There are a number of pharmacokinetic differences between children and adolescents in comparison with adults, frequently leading to the need for higher doses on a milligram-perkilogram basis to achieve the same effective adult drug concentrations. These differences are summarized in the following paragraphs.

Most psychotropic medications are highly lipophilic. The percentage of total body fat, which is a reservoir for these lipid-soluble compounds, increases during the first year of life then decreases until prepuberty (79). Thus, children at different ages have different volumes of fat storage that can affect the residual time a drug remains in the body.

The acquisition of adult levels of both cytochrome P450 (CYP 450) and phase II drugmetabolizing enzymes is enzyme and isoform specific (86). Research demonstrates that the traditional view of a locked-step progression of drug-metabolizing capacity is overly simplistic. Still, some generalizations are possible. Most drug-metabolizing enzyme activity is absent in the fetus but rapidly increases over the first years of life. Toddlers and older children have levels of several, but not all, drug-metabolizing enzymes exceeding those of adults. These levels decline from that point until “usual” adult levels are achieved by the end of puberty. Developmental changes in the activity of specific CYP isoenzymes over the first 2 decades of life are reflected in the increase and then decrease in theophylline clearance (1A2); a decrease in phenytoin clearance (2C9/10, 2C19); and a decrease in the ratio of carbamazepine-10,11-epoxide to CBZ (3A3/4) (87,88). Increasingly, the role of the P-glycoprotein transporter is also being considered, and variants may play an important role in drug response (e.g., guanfacine) (89). Further details are provided in Table 14-2.

In part, the rate of drug metabolism also depends on liver mass. Relative to body weight, the liver of a toddler is 40% to 50% greater and that of a 6-year-old child is 30% greater than that of an adult (79). This difference in size is another reason that children tend to clear drugs more rapidly than adults and frequently need higher doses on a milligram-per-kilogram basis to achieve the same plasma levels and clinical effect.

TABLE 14-2 DEVELOPMENTAL PATTERNS FOR SPECIFIC DRUG-METABOLIZING ENZYMES

Cytochrome P450	Development Pattern
	Phase I Isoenzymes
1A2	Adult level reached by 4 months but exceeded by age 1-2 years. Decline to adult levels by the end of puberty. Gender differences are possible during puberty.
2C9, 2C19	Adult activity reached by 6 months but exceeded by 1.5 to 1.8 times by age 3-4 years. Decline to adult levels by end of puberty.
2D6	Adult levels obtained by 3-5 years of age.
3A4	Adult levels reached by 6-12 months but then exceeded by 1-4 years of age. Decline to adult levels by the end of puberty.
	Phase II Enzymes
NAT2	Adult activity present by 1-3 years of age.
TPMT	Adult activity achieved by 7-9 years of age.
UGT	Adult activity by 6-18 months of age.
ST	May exceed adult levels during early childhood.
NAT2, N-acetyltransferase-2; ST, sulfotransferase; TPMT, thiopurine methyltransferase; UGT, glucoronosyltransferase.

By 1 year of age, glomerular filtration rate and renal tubular mechanisms for secretion reach adult levels; however, fluid intake may be greater in children. Thus, lithium has a shorter half-life and more rapid renal clearance in children as compared with adults (90).

Nevertheless, it should not be surprising that the interindividual differences in the clearance of psychiatric medications are as great in children and adolescents as in adults. For example, genetically determined differences in CYP 2D6 isoenzyme function are expressed at birth. Hence, 5% to 10% of children and adolescents of northern European origin are deficient in CYP 2D6 activity and will develop four to six times higher levels of drugs that are predominantly metabolized by this isoenzyme than individuals with a functional copy.

Of all psychiatric medications, the psychostimulants are the best studied in terms of their pharmacokinetics in children and adolescents (Table 14-3). The next most studied are the antidepressants, and only a few studies exist with antipsychotics and anxiolytics in this age group.

As discussed in Chapter 2, TDM is used to assess the ability of younger patients to clear a drug. The prescriber then uses the results to adjust the dose to achieve adult concentrations. Still, this approach is only an approximation of
what might be optimal for a child or adolescent patient. This is because younger patients are either more or less sensitive to the beneficial or adverse effects of the drug (and thus might need a higher or lower concentration than adults to achieve an optimal response). Nevertheless, TDM serves as a reasonable reference point in the absence of more definitive efficacy and safety data in children and adolescents. Using TDM in this manner, the prescriber then carefully adjusts the dose based on clinical assessment of safety and efficacy. It is also prudent to divide the daily dose more frequently than in adults to avoid excessively high peak plasma drug concentrations that may be associated with increased tolerability and safety problems.

TABLE 14-3 IMPORTANT FEATURES OF COMMON STIMULANTS USED TO TREAT ATTENTIONDEFICIT DISORDER IN CHILDREN AND ADOLESCENTS

Feature		Methylphenidate	Dextroamphetamine	Atomoxetine
Elimination half-life		2-3 hours	6-7 hours	5
Time to peak plasma concentration (T_max)		1-3 hours	3-4 hours	1-2
Onset of behavioral effect		1 hour	1 hour	4-6 weeks
Duration of behavioral effect		3-4 hours	4 hours	Sustained after the onset
Daily dose range
	mg/kg/day	0.6-1.7	0.3-1.25	0.5-1.4
	mg/day	10-60	5-40	40-120

Psychosis in Children and Adolescents

DESCRIPTION OF THE DISORDER

Psychosis is characterized by impairment in reality testing, hallucinations, delusions, and severe disturbances in cognition and/or bizarre behaviors. It occurs in approximately 1% of children and adolescents (91,92 and 93). In clinical settings, the prevalence of delusions and hallucinations in youth is approximately 4% and 8%, respectively (93,94). These figures are consistent with reports in adult studies (e.g., 3% to 5% experience hallucinations before 21 years of age).

The studies that examined the phenomenology of psychosis in children and adolescents found that hallucinations (particularly auditory) are the most common symptom (93,94,95,96,97,98,99 and 100). In general, delusions and formal thought disorder are rare and less elaborate in children, but they increase in frequency during late adolescence. The low frequency of delusions and formal thought disorder are accounted for by the lack of cognitive maturity in children, the low prevalence of schizophrenic spectrum disorders in youth, and because most children referred to treatment are usually experiencing their first episode (92,93 and 94,97,98 and 99). As the disease progresses, the prevalence of thought disorder appears to increase (93,94,96). This information needs to be interpreted with caution, however, because it came from retrospective studies with small samples and used nonstandardized methods of clinical evaluation.

The most common diagnoses in children and adolescents who are experiencing hallucinations and/or delusions are major depressive disorder (MDD) and BPD. By contrast, the diagnosis of schizoaffective and schizophrenic disorders is less frequent (91,92,94,97,99,100 and 101). In fact, auditory hallucinations are relatively common in pediatric depression and, when associated with a positive family history, increase the likelihood of developing BPD (100). One study of youth presenting to mood and anxiety disorder clinics used standardized interviews and showed that approximately 4.5% of children and adolescents (n = 2,031) had significant psychotic symptoms (94). These symptoms were primarily auditory hallucinations (80%), delusions (22%), and thought disorder (3.3%). Subjects were diagnosed
with a major depression (41%), BPD (24%), or, less frequently, schizophrenia (14%). The presence of formal thought disorder (e.g., loose associations) in the absence of a developmental disorder, particularly in children older than 7 years, may indicate a schizophrenic spectrum illness (96). In this context, a retrospective study of adult schizophrenia (95) and two prospective studies of younger subjects (ages 16 to 30 years and 12 to 45 years) suggested that the presence of brief psychotic episodes, perceptual disturbances, thought disorganization, unusual thoughts, poor functioning, and blunted or inappropriate affect combined with a family history of psychosis increased the risk for schizophrenia and related psychotic disorders (102,103). An important caveat is that patients with mood disorders may present with schizophrenic-like thought disruption (91,92,99,100), making appropriate diagnosis and treatment more challenging. Evidence indicates that childhood-onset schizophrenia demonstrates distinct cortical gray matter loss during adolescence, which appears to be an exaggeration of the normal cortical gray matter developmental pattern and eventually mimics the pattern seen in adult-onset cases as the children become young adults (104). These cortical gray matter changes in childhood-onset schizophrenia may be diagnostically specific, unrelated to the effects of medications, and also are shared by healthy full siblings. This suggests a genetic influence on the abnormal brain development.

PRODROMAL SYMPTOMS

Early Identification

Recognizing the prodromal features of psychosis has important clinical implications, including

Early antipsychotic medication intervention may be associated with a better long-term outcome (105,106,107,108 and 109)
The use of SGAs with their lower rate of particular adverse effects makes early intervention more acceptable
The deficit process (or negative symptoms) is usually present at the first psychotic episode (110). This may mean that the symptoms begin prior to the onset of illness as currently defined. In this context, accurate early identification and aggressive intervention may arrest the transition into a full disorder
Early presentation of psychosis often causes severe distress on the family system, which also requires early treatment interventions

In an attempt to identify prodromal symptoms, Alaghband-Rad et al. (95) retrospectively examined the premorbid clinical characteristics of younger patients. They reported that the most common symptoms were social withdrawal, impaired school performance, markedly peculiar behavior, impaired hygiene, blunted inappropriate affect, vague/poverty of speech, odd magical thinking or perceptions, and marked lack of initiative.

Subsequently, research efforts focused on the prospective identification of prodromal symptoms (102,109,111). Theoretically, the term “psychotic disorder” includes schizophrenia and psychotic mood disorders. But there is a bias toward applying this term to schizophrenia as reflected by the content of the screening instruments designed to identify early psychotic and prepsychotic features. McGlashan (102) and McGorry (103) used one of three criteria to identify such prodromes:

Brief intermittent psychosis. One or more positive symptoms such as hallucinations, suspiciousness, or unusual thought content (scores of >3 to 4 on these Brief Psychiatric Rating Scale [BPRS] items) in the last 3 months, for at least several minutes a day, at least once a month.
Attenuated positive symptoms. At least one symptom of schizotypal disorder and one symptom of psychosis such as hallucinations, suspiciousness, or unusual thought content in subdued form (lower scores of 1 to 2 on these BPRS items) in the past year and present at least once per week in the last month.
Genetic risk and recent deterioration. Having a first-degree relative with a history of any psychotic or schizotypal personality disorder and a drop of >30 points on Global Assessment o f Functioning (GAF) for at least 1 month.

Using these criteria, 43% of these patients in McGorry’s sample and 50% in McGlashan’s sample eventually converted to the full disorder over 24 months. There are two major differences, however, between these two studies. McGorry (103) focused on ages 16 to 30 years and included affective psychosis, while McGlashan (102) included a wider age range (12 to 45 years) and
excluded affective psychosis. The term “ultra-highrisk” subjects is favored over prodrome by McGorry (103), as it does not imply inevitable progress to a full disorder, whereas the term “prodrome” is typically used by McGlashan (102). In addition, neither of these studies included measures to identify affective symptoms despite compelling evidence that psychosis is common in pediatric BPD (98,112,113,114,115 and 116).

Treatment Interventions

Woods and colleagues (117) conducted a multisite, double-blind, placebo-controlled, randomized trial for prodromal symptoms (n = 60) using olanzapine flexibly dosed between 5 and 15 mg for 8 weeks. The olanzapine-placebo difference in attenuating prodromal symptoms reached significance (p < 0.05) at week 8. Olanzapine patients, however, gained 9.9 lb versus 0.7 lb for placebo patients (p < 0.001). Extrapyramidal symptoms (EPSs) were minimal in both groups. McGorry and colleagues (118) compared crisis or needs-based versus specific preventive intervention (i.e., risperidone, mean dose of 1.3 mg/day) in youth (14 to 28 years). At the end of 6 months, those who received the specific intervention did better than those who received needs-based intervention (p < 0.03). Once the specific intervention was discontinued, however, protection against progression did not persist at 12 months (p < 0.24). This study underlined the importance of more specific pharmacotherapy and psychotherapy in reducing the risk of transition to psychosis among those at high risk. Cornblatt and colleagues (104) studied subjects between 14 and 22 years with negative symptoms-attenuated positive symptoms and schizophrenia-like psychosis. Using a naturalistic design, preliminary results over 6 months indicated that antidepressants in combination with mood stabilizers and/or anxiolytics were as effective as antipsychotics. Given the weight gain with many antipsychotics, they proposed antidepressants as a possible alternative.

In a naturalistic study (n = 116), a substantial number of patients who met the criteria for a prodromal syndrome for first psychosis were treated with antipsychotic and/or antidepressant medications (119). This study highlighted the relationship of these medications with prodromal symptom severity at baseline and 6-month follow-up. Participants who met the criteria for the prodrome were evaluated at eight centers as part of the North American Prodrome Longitudinal Study (NAPLS). Symptom ratings (positive, negative, disorganized, and general) and data on antipsychotics, SSRIs, and other antidepressant medications were obtained at baseline and 6-month follow-up. Analyses revealed that all symptom dimensions declined in severity over time, but there were differences in the magnitude of the decline as a function of antipsychotic medication (i.e., those never on antipsychotics showed less reduction in positive and disorganized symptoms over time). SSRIs and other antidepressants were not linked with declines in symptom severity. Consistent with findings from earlier studies with small samples, these latest results suggest that SGAs may be effective in reducing the severity of symptoms associated with the prodrome to psychotic disorders.

In summary, while more definitive results are awaited, antipsychotics may be a viable option for youth at high risk for psychosis.

TREATMENT OF PSYCHOSIS

This section reviews medications specifically used for the management of psychosis as well as the treatment of disorders most commonly associated with psychotic symptoms in youth (e.g., mood disorders).

The treatment of children and adolescents with psychosis is divided into acute, maintenance (to prevent relapse), and prophylactic (to prevent recurrence) phases. Acute and maintenance treatment should be offered to all patients, and prophylaxis is reserved for those with recurrent or more severe illnesses. Although the treatment of children and adolescents with psychosis depends on the underlying cause, it should always include psychosocial as well as pharmacological management for both the primary disorder, as well as other comorbid psychiatric, medical, or neurological conditions.

First-Generation Antipsychotics

The improvement in psychotic symptoms obtained with first-generation antipsychotics (FGAs) is associated with blockade of central dopamine-2 (D₂) receptors. The FGAs are divided into low- and high-potency agents based on their affinity for these receptors. Low-potency compounds include chlorpromazine and thioridazine.

In comparison to the high-potency FGAs, these agents have a lower affinity for the dopaminergic receptor but also impact cholinergic, histaminergic, and adrenergic receptors. This accounts for adverse effects such as dry mouth, constipation, tachycardia, somnolence and increased weight, lower blood pressure, changes in the cardiac conduction (e.g., increased QTc interval), and arrhythmias. Conversely, higher potency FGAs (e.g., haloperidol, fluphenazine) have a greater affinity for dopaminergic receptors accounting for their increased incidence of EPS such as tremor, bradykinesia, stiffness, cogwheel rigidity, dystonia, akasthisia, tardive dyskinesia (TD), and withdrawal dyskinesia (see also Chapter 5).

In adult schizophrenic patients, all the FGAs administered at equivalent doses (e.g., 300 to 800 mg/day of chlorpromazine) for at least 6 weeks are effective for the acute treatment of positive symptoms (e.g., hallucinations, delusions, and agitation) and the prevention of relapse and recurrences (120,121). Higher doses usually do not convey further benefit but do increase the likelihood of adverse effects. In vivo neuroimaging studies with positron emission tomography (PET) and single photon emission computed tomography (SPECT) in adults; data from clinical studies in adult patients with schizophrenia; and data from animal models all indicate that low doses of haloperidol (e.g., 2.5 to 10 mg/day) block at least 65% of striatal D₂ receptors, producing a therapeutic response. Higher doses of haloperidol, however, will produce greater D₂ receptor occupancy (e.g., >70%), which is associated with increased risk for EPS (121).

Small, randomized, controlled trials (RCTs) compared clozapine with haloperidol; loxitane with haloperidol; and placebo with haloperidol to assess the efficacy of FGAs for early-onset psychosis (92,120,122,123). Similar to the adult literature, these trials suggest that FGAs are useful for the short-term treatment of positive symptoms but are not as efficacious for negative symptoms (e.g., lack of motivation, blunted affect). They also can cause significant adverse effects. In addition to EPS, they increase prolactin secretion that can produce menstrual disturbances, galactorrhea, gynecomastia, impotence, and sexual dysfunction due to blockade of tuberoinfundibular D₂ receptors. These antipsychotics (in particular thioridazine and pimozide) can also prolong PR and QT intervals. The prolongation of the QT interval is associated with serious arrhythmias such as torsade de pointes. The warning signs that an arrhythmia may be present include dizziness, palpitations, nausea, vomiting, and possibly syncope. These agents are also associated with

Gastrointestinal side effects (nausea, constipation, stomatitis)
Cognitive disturbances (in particular the low-potency antipsychotics)
Blood dyscrasias (e.g., agranulocytosis, thrombocytopenia)
Hepatic abnormalities (e.g., increase in the liver transaminases, cholestatic jaundice)
Dermatological problems (e.g., atopy, photosensitivity)
Eye complications (e.g., chlorpromazine and thioridazine at high doses may induce lenticular and retinal pigmentation and retinitis pigmentosa)
Dyslipidemia
Neuroleptic malignant syndrome (NMS)

Although rare, NMS can be life threatening, producing autonomic instability, hyperthermia, and rigidity. The last problem may cause significant muscle breakdown leading to myoglobinuria and acute renal tubular necrosis (see Chapter 5).

To prevent or to control EPS, it is usually necessary to use an anticholinergic agent. These drugs, however, produce their own significant adverse effects (e.g., dry mouth, constipation, urinary hesitation, tachycardia), as well as cognitive disturbances that may affect a child’s psychosocial functioning. Many adverse effects are dose dependent and thus can be avoided by low initial doses, slow upward titration, and taking the medications at certain times (e.g., at bedtime to avoid somnolence during the day). Although patients may develop tolerance to some of these adverse effects, their management is of prime importance because they often lead to nonadherence and parental reluctance to administer them to their children.

Paralleling the experience with adults, the use of FGAs has substantially diminished in favor of SGAs. This is in part due to a decreased risk for acute EPS and TD; a decreased use of anticholinergic medications to counteract these effects; and possible decreased long-term adverse effects on the developing brain (123). The FGAs, however, are still useful for the temporary management of agitation, hallucinations and delusions, and for the treatment of conditions such as Tourette disorder.

If a patient develops dyskinesia while taking an FGA, immediate discontinuation is warranted when clinically feasible. The manifestations of TD should be documented and the patient examined to exclude other causes. Although symptoms usually subside within several weeks, they may recur if the offending antipsychotic is reintroduced. Thus, an SGA is usually the most appropriate next choice.

Second-Generation Antipsychotics

In the United States, these presently include clozapine, risperidone, olanzapine, quetiapine, ziprasidone, aripiprazole, paliperidone, iloperidone, and asenapine. In adults, these antipsychotics are efficacious for positive symptoms, and possibly negative, cognitive, and mood symptoms. These agents are also useful for the treatment of mood disorders in adults such as acute mania, major depression, and associated agitation. The increased efficacy of these compounds and decreased incidence of EPS may be related to their greater affinity for serotonergic receptors (e.g., 5-HT2), producing a higher ratio of serotonergic/dopaminergic antagonism. In addition to open studies and case reports (92,122,124), there are several controlled studies that support the use of SGAs for youth with schizophrenia.

In the first RCT of childhood-onset schizophrenia, clozapine (176 [±149] mg) was more effective than haloperidol (16 [±8] mg) for both positive and negative symptoms, while inducing significantly less EPS (125). In another 12-week RCT, 39 subjects (15.6 [±2.1] years) were treated with clozapine (403.1 [±201.8] mg) or olanzapine (26.2 [±6.5] mg) (126). Using a ≥30% reduction on BPRS-C scores, 66% improved on clozapine compared with 33% on olanzapine. Although clozapine also demonstrated superiority over haloperidol and olanzapine in early-onset schizophrenia (EOSS) disorders (i.e., before 18 years), it is usually reserved for treatment-resistant cases (127). This is because its advantages are limited by such issues as increased appetite and weight gain, sialorrhea, sedation, seizures, and neutropenia.

Risperidone (4 [±1.2] mg), olanzapine (12.3 [±3.5] mg), and haloperidol (5.0 [±2] mg) were compared in an 8-week RCT involving 50 subjects (14.7 [±2.7] years) with a broad range of psychotic disorders (128). All treatments significantly reduced BPRS-C scores, with a ≥20% reduction in 74% on risperidone, 88% on olanzapine, and 54% on haloperidol. Of note, several subjects were on concomitant medications, (e.g., antidepressants and mood stabilizers). An important observation from this study is that EPS and weight gain were more substantial in youth compared to published data from adult studies.

There are three subsequent, double-blind, placebo-controlled trials in youth with schizophrenia (129,130 and 131). Findling, et al. (129) compared aripiprazole (10 or 30 mg) with placebo in 302 subjects (13 to 17 years) and reported a reduction of 26.7, 28.6, and 21.2 points on the total baseline PANSS score, respectively. Adverse events included EPS, akathisia, and somnolence. Of note, there was no significant weight gain with aripiprazole. Haas et al. studied 160 subjects (15.6 [±1.3] years) who received risperidone high dose (4 to 6 mg), low dose (1 mg), or placebo. They reported a 20.7, 19.9, and 7.8 point reduction on the total baseline PANSS score, respectively (130). Higher doses of risperidone produced a greater incidence of EPS, dizziness, and hypertonia compared with the lower doses. In a 6-week trial, Kryzhanovskaya et al. compared olanzapine (11.1 [±4] mg) with placebo in 107 subjects (16.2 [±1.3] years) (131). There was a significant reduction in BPRSC scores from baseline to end point (p < 0.003) with olanzapine, and more patients on olanzapine completed the trial versus placebo (i.e., 68% vs. 43%; p < 0.02). Weight gain was higher in the olanzapine group at 4.3 (±3.3) kg and was greater than in adult studies. Prolactin and triglyceride levels also increased to a significantly greater degree in the olanzapine group versus the placebo group.

Second-Generation Antipsychotics versus First-Generation Antipsychotics

The Treatment of Early-onset Schizophrenia Spectrum Disorders Study (TEOSS) considered the relative merits of two SGAs, olanzapine (2.5 to 20 mg/day) and risperidone (0.5 to 6 mg/day), as compared with the FGA, molindone (10 to 140 mg/day) (132). The major findings indicated comparable benefit withall three agents in the 8-week, multicenterRCT involving 116 youth. Further, adverse effects differed with the SGAs producing more metabolic-related
and molindone more neuromotor-related complications.

Second-Generation Antipsychotics: Adverse Effects

In addition to blocking serotonergic and dopaminergic receptors, some SGAs also impact other receptors including muscarinic, histaminergic, and adrenergic, accounting for their anticholinergic effects, weight liability, and hypotension, respectively. Depending on the dose, particularly at treatment initiation, all SGAs can produce sedation, fatigue, gastrointestinal symptoms (e.g., loose stools and nausea), and headaches. Except for ziprasidone and aripiprazole (although no data are available from drug-naive subjects to assess if aripiprazole is truly “weight neutral”), they typically increase appetite and body weight. Patients with lower pretreatment body mass index, younger age, and female gender are at higher risk to experience an increase in weight, which in turn increases the risk for various medical illnesses, impairs quality of life, induces low self-esteem, and increases the likelihood of nonadherence. In addition to the usual societal ostracism for people with psychosis, increased weight adds to the stigma associated with obesity. SGAs, such as clozapine and olanzapine, are also associated with hyperlipidemia, new-onset diabetes mellitus, and diabetic ketoacidosis (DKA) (120,122,127,133). It is not clear, however, whether these complications are secondary to the weight gain, or the result of glucose dysregulation with these agents. Although the SGAs may improve cognition in adult patients, some cause sedation and anticholinergic effects that may interfere with academic performance in younger patients. Risperidone also induces sustained hyperprolactinemia potentially leading to gynecomastia, galactorrhea, menstrual and sexual disturbances, rashes, sweating, and photosensitivity (120,122).

Since several of these adverse effects are dose dependent, it is imperative to start treatment at low doses and increase slowly only as necessary. This not only minimizes the risk for adverse effects but increases patient and/or their parents’ adherence to treatment. Similar to the use of FGAs in adults, low doses of SGAs (e.g., risperidone 2 to 4 mg/day; olanzapine 7.5 to 15 mg/day) are sufficient to produce 65% to 70% D2 receptor occupancy and achieve a therapeutic response (121), while minimizing the risk of adverse effects.

Since clozapine is associated with lifethreatening events including myocarditis, neutropenia (absolute counts under 1,500/mm ³), and agranulocytosis (neutrophil counts less than 500/mm ³), it requires close laboratory monitoring. Clozapine also increases the risk for seizures, abnormal EEGs, myoclonus, eosinophilia, fever, persistent tachycardia, and may be associated with pulmonary embolism. Ziprasidone, paliperidone, and iloperidone are associated with a greater increase in the QT interval requiring the need for baseline ECG and subsequent monitoring in individuals with cardiac problems. Clinicians should avoid using these agents with other medications that also increase the QT interval, as well as in patients with congenital long QT syndrome, prior cardiac arrhythmias, or a family history of torsade de pointes or sudden death. The family history is particularly important if the death occurred when the person was young.

Continuation and Maintenance Therapy

To avoid relapse and recurrence, adults with schizophrenia require ongoing maintenance antipsychotic treatment. Although not systematically studied, children with schizophrenia almost certainly need this approach as well. It is useful, however, to consider a trial off medication to reevaluate the possible cause of such episodes. For example, if the child had a major depression with psychosis, the need for an antipsychotic is usually temporary.

Although the length of treatment in youth is not clarified, to consolidate response and prevent relapse, all patients should continue treatment for at least 6 months after complete symptom remission. Patients, who do not achieve remission, have early-onset psychosis, and/or a family history of schizophrenia or other psychotic disorders should be offered a longer treatment period (e.g., 1 year). Children and adolescents with psychotic mood disorders may be slowly tapered off their antipsychotic when symptoms remit. Children with an unambiguous diagnosis of schizophrenia must be offered indefinite maintenance therapy to avoid future psychotic episodes and additional deterioration in their cognitive, emotional, and social skills. Also, those with difficult-to-treat psychosis, severe agitation,
and/or who pose a suicidal or homicidal threat should be offered indefinite maintenance therapy (120,122,123).

Since we do not know the potential for longterm adverse effects associated with SGAs in this age group, clinicians must monitor for the development of significant but as yet unappreciated problems.

Treatment of Comorbid ADHD

In a case series report, five subjects were treated with stimulants plus an SGA for childhood-onset schizophrenia (134). Symptoms of attentiondeficit hyperactivity disorder (ADHD) improved on the Conners Rating Scale (p < 0.05). Further, there was no worsening of psychotic symptoms over the 2-year follow-up despite continuation on stimulants.

Treatment of Affective Psychosis

Psychotic symptoms (especially auditory hallucinations) are frequently associated with mood disorders in children and adolescents. Taking into account the child’s developmental stage, clinicians need to carefully evaluate such symptoms at intake and maintain a high index of suspicion because children may be experiencing psychosis but not communicating these symptoms to their parents. The presence of psychotic symptoms in a child with depression increases the risk of BPD and the risk for suicide. By contrast, the presence of thought disorder in the absence of a developmental disorder and in children older than seven years with a family history of schizophrenia is more indicative of a schizophrenia spectrum disorder.

Although the use of antipsychotics is not well studied in this context, they are utilized for the treatment of psychosis in this population. Given their adverse effect profile and mood stabilizing properties (at least in adults), SGAs are usually the first choice. Studies evaluating their longterm adverse effects, however, are absent and are needed. For youth with BPD or MDD, specific treatment(s) (e.g., lithium) are indicated. Since youth with psychotic depression and a family history of BPD are at greater risk to develop BPD (100), there must be careful monitoring of their mental status. These disorders are usually comorbid with other psychiatric disorders (e.g., substance abuse, ADHD, anxiety) that also require treatment. As with other psychotic disorders, psychosocial treatment is indispensable to improve adherence, provide support, and educate families.

Bipolar Disorder. There are no large, controlled trials of mood stabilizers for pediatric BPD. The extant literature from case reports, open studies, and small, controlled trials, however, suggests that these agents are effective (135). Although younger bipolar patients with psychotic symptoms may respond to a standard mood stabilizer alone, preliminary reports suggest that they also need adjunctive antipsychotics (136). This is particularly true if they are too agitated to cooperate with treatment or as a temporary measure until the primary mood stabilizer exerts its full effect (see “Treatment of Bipolar Disorder” for more detailed description of trials that included subjects with affective psychosis). There are also a few reports in children, as well as controlled trials in adults, suggesting that SGAs alone can successfully treat mania and psychosis (137).

Major Depression. The SSRIs are efficacious and well tolerated for children and adolescents with MDD (138,139 and 140). Although clinical experience suggests that youth with MDD and psychotic symptoms may respond to an SSRI alone, there are no studies to support this strategy. In adults, only 20% to 40% with psychotic MDD respond to antidepressant monotherapy, and placebo response is quite low (141). Even if monotherapy with antidepressants is effective, recovery appears to be greater and more rapid when antidepressants are combined with an antipsychotic. Although these agents can be slowly tapered after remission of psychosis, it is not clear how long patients should be maintained on an antipsychotic. For adults who do not respond to an antidepressant-antipsychotic combination, ECT is particularly effective (141) and should also be considered for younger patients.

RECOMMENDED TREATMENT STRATEGIES

We first evaluate the severity of psychosis and baseline motor status with the Brief Psychiatric Rating Scale (BPRS) for Children (142) and the Abnormal Involuntary Movement Scale (AIMS) (143), respectively. The evaluation of abnormal
movements before starting an antipsychotic is important because dyskinesias or other movement disorders may exist before treatment, indicating an increased propensity to develop EPS and/or TD (120,121 and 122). Before prescribing an SGA, baseline and subsequent neurological exams are conducted to rule out dyskinesia. In addition, we recommend a baseline CBC plus differential, fasting glucose level, liver function tests, lipid profile, ECG, and ascertainment of body mass index. For clozapine, we suggest a baseline EEG and ECG with routine monitoring (e.g., every 6 months). Since the SGAs are metabolized by CYP 450 isoenzymes, it is important to monitor potentially relevant pharmacological interactions with other medications metabolized by this system.

We recommend starting with a low dose of an SGA, gradually increasing it to a tolerable level while monitoring for adverse effects. In this context, it is important to complete an adequate trial before switching to an alternate agent.

CONCLUSION

Since psychotic symptoms in youth are more common in affective illness than schizophrenia, a careful differential diagnosis is difficult but critical. The term “early-onset psychosis” is interchangeably used for schizophrenia spectrum disorders. Auditory hallucinations are the most common symptom. SGAs are the first-line treatment, with mood stabilizers and antidepressants often used as adjuncts when psychotic symptoms present in the context of bipolar or unipolar disorders. The risk-benefit ratio must be reconsidered on a periodic basis when using SGAs plus mood stabilizers and/or antidepressants.

Depression in Children and Adolescents

DESCRIPTION OF THE DISORDER

The existence of MDD, much like the more recent discussions on BPD in children, was vigorously debated in 1980s (144). Subsequently, a consensus was achieved with estimated prevalence rates of 0.9% in preschoolers, 2% in children, and 4% to 8% in adolescent population (145,146 and 147). Unlike adults, an equal number of young males and females are affected with MDD. Also, twice as many suffer from MDD in adolescence compared with younger age groups. Approximately, 30% of the youth in outpatient settings and twice that number in inpatient settings present with depressive disorder (148,149).

The Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition, Text Revision (DSM-IV-TR) accommodates developmental specificity in describing pediatric depression such that (60)

Irritable mood can be present instead of depressed mood
Failure to gain expected weight equals weight loss criteria
Dysthymic disorder (DD) can be 1-year in duration in children as opposed to 2 years in adults

Symptom expression varies widely based on age. Children younger than 7 years do not have well-developed verbal skills to express their internal feelings or thoughts (90,150,151). Thus, they frequently present with a depressed facial expression, monotonous voice, and decreased psychomotor activity. They may also degrade themselves, feel unworthy of being loved, complain of boredom, disengage from peers, and withdraw from their favorite sports or other activities. Somatic problems may become more apparent and academic performances worsen. Children who present with psychotic depression usually report auditory hallucinations and, more rarely, delusions (145,152). With increasing age, there is a decrease in somatic complaints and phobias and an increase in anhedonia, diurnal variation, hopelessness, psychomotor retardation, and delusions (153). Suicide attempts, completed suicide, and drug abuse are common in adolescent depression (154).

It is important to consider other medical or psychiatric conditions that may produce depressive symptoms, before making a definitive diagnosis of MDD. Further, in determining if a child’s behavior is abnormal, one needs to know what is normal for that age. Thus, the clinician should consider situations or diagnoses where depressed mood may be a symptom of other problems such as abuse, neglect, or separation anxiety disorder (SAD). Substance abuse may present with depressed mood, especially in adolescents. Negative symptoms of early-onset psychosis may masquerade as MDD. In this context, it is useful to assess premorbid functioning and behavior to ascertain the onset of insidious schizophrenia.
Since a mixed presentation of depressive plus manic symptoms is common in pediatric BPD, clinicians can misinterpret it as MDD (152). Thus, it is particularly important to inquire about manic or hypomanic symptoms that would dictate treatment with a mood stabilizer as opposed to an antidepressant.

TREATMENT OF DEPRESSION

In 2004, the FDA required a box warning for all antidepressants used in children and adolescents, based on the Psychopharmacologic Drugs and Pediatric Advisory Committee recommendations (155,156). This was later expanded to include young adults younger than 25 years. The original decision was based on a review of the data from 23 trials involving 4,300 subjects. The advisory committee concluded that there was an increased risk of suicidality in pediatric patients for all antidepressants compared with placebo in these trials. Of note, there were no completed suicides. The American Academy of Child and Adolescent Psychiatry (AACAP), however, considered this decision unjustified as only 78 out of 4,300 children reported increased suicidal ideation and/or self-injurious behavior (157). This represented 4% of reported adverse events among those that received active treatment, compared with 2% in those who received placebo. There are also expert reviews including a preliminary report by the Task Force on SSRIs and Suicidal Behavior in Youth (January 21, 2004) posted by the American College of Neuropsychopharmacology (ACNP). This task force (including experts in adult and pediatric depressive disorders) reviewed the available published and unpublished clinical, epidemiological, and autopsy data to evaluate the benefits and risks of SSRIs in youth (<18 years of age). The ACNP report described toxicological analysis in adults and youth suggesting that, in those completing suicide, the majority had not been taking an antidepressant prior to death. Further, they cite international epidemiological data that support declining rates of suicide in youth over the past decade, coincident with increasing antidepressant prescription rates. The task force concluded that SSRIs do not increase the risk of suicidal thinking or attempts and that their benefits for depression in youth outweigh the risks (158). In support of these findings, an American Journal of Psychiatry review (159) examined a large sample (48,000) of youth receiving SSRIs across numerous clinical trials. Seventy-seven suicides were documented, but the rates were comparable to those on placebo or antidepressants. Several experts in the pediatric psychiatric field also rejected the notion that there is a significant increased risk of suicidality associated with these agents (157).

Although there is a definite role for antidepressants in children, this controversy serves the purpose of raising awareness about the careful monitoring for any treatment-emergent suicidal ideation or activation. This is especially important in depressed children with a family history of BPD (160). The data summarized here reviews the efficacy and safety of antidepressants, emphasizing their use in conjunction with careful clinical monitoring.

Selective Serotonin Reuptake Inhibitors

Results from randomized, placebo-controlled trials in children and adolescents indicate that the SSRIs are first-line agents due to their positive efficacy signal, favorable adverse effect profile, and wide therapeutic index.

For example, Keller and colleagues (140) conducted an 8-week, double-blind trial comparing paroxetine (20 to 40 mg), imipramine (200 to 300 mg), and placebo in 275 adolescents with MDD. Paroxetine demonstrated significantly greater improvement compared with placebo based on change in the Hamilton Depression Rating Scale (HDRS) scores (p < 0.05); the depression items on the Kiddie-Schedule for Affective Disorders and Schizophrenia (K-SADS-L) (p < 0.05); and a Clinical Global Impression (CGI) score of 1 or 2 (p < 0.05). By contrast, response to imipramine was not significantly different from placebo on any of these measures. Further, discontinuation rates for adverse events were 9.7% for paroxetine, 6.9% for placebo, and 31.5% for imipramine. Of note, a third of those discontinuing from imipramine had cardiovascular adverse effects. The role of paroxetine, however, remains inconclusive, with two subsequent negative studies conducted in South Africa (161) and the United States (162). The first was a multisite, 12-week, double-blind, placebo-controlled trial (DBPCT) that examined the efficacy, safety, and tolerability of paroxetine (20 to 40 mg/day) in 286 adolescents with MDD (161). The proportion
of Montgomery-Asberg Depression Rating Scale (MADRS) responders (i.e., at least 50% reduction from baseline) on paroxetine and placebo were similar and not statistically different at end point (p < 0.70). A similar result was obtained for change from baseline score on the Kiddie-Schedule for Affective Disorders and Schizophrenia for SchoolAge Children (K-SADS-L) depression subscale. In general, results differed by age, with patients older than 16 years demonstrating a greater response to active treatment. This age group also reported more adverse experiences relative to placebo than younger adolescents. Emslie et al. (162) also studied youths (7 to 17 years) with MDD who were randomized to receive paroxetine (10 to 50 mg/day) or placebo for 8 weeks. The primary efficacy measure was change from baseline in the Children’s Depression Rating Scale-Revised (CDRS-R) total score. Adjusted mean change scores from baseline for patients receiving paroxetine and placebo were —22.58 and —23.38 points, respectively (p < 0.684). Increased cough (5.9% vs. 2.9%), dyspepsia (5.9% vs. 2.9%), vomiting (5.9% vs. 2.0%), and dizziness (5.0% vs. 1.0%) occurred in ≥5% of the paroxetine group and at least twice that of the placebo group. In this trial, paroxetine was no more efficacious than placebo for treating pediatric MDD.

Earlier, Emslie and colleagues (163) examined youth (8 to 18 years; n = 168) with MDD in an 8-week, DBPCT of fluoxetine (20 mg/day). This SSRI produced a greater mean improvement in CDRS-R scores compared with placebo after 1 week and throughout the study period (p < 0.05). Although more fluoxetine-treated (65%) than placebo-treated (53%) patients met the a priori response criterion of ≥30% decrease in the CDRS-R score, this difference was not significant (p < 0.093). In a depression relapse prevention study, Emslie et al. (164) randomly assigned 102 subjects to continue treatment with fluoxetine (n = 50) or switch to placebo (n = 52). Of these, 21 (42.0%) participants in the fluoxetine group relapsed, compared with 36 (69.2%) in the placebo group. Time to relapse was also significantly shorter in the placebo group (p < 0.007). The authors concluded that continuing treatment with fluoxetine was superior to placebo in preventing relapse and in increasing time to relapse.

Wagner and colleagues (165) reported data on sertraline from two multicenter, RCTs conducted at 53 centers in five countries. Youth (6 to 17 years) with MDD received sertraline (n = 189) at a flexible dose of 50 to 20 mg/day or placebo (n = 187) for 10 weeks. Sertraline was superior to placebo in decreasing CDRS-R scores (p < 0.001). Further, using a 40% decrease in symptoms on the CDRS-R as criterion for response, higher rates were noted with sertraline (69%) compared with placebo (59%; p < 0.05). Adverse events included diarrhea, vomiting, anorexia, and agitation in >5% of those who took sertraline, with an incidence twice that for those who received placebo. Time to first response and time to first persistent response with sertraline were evaluated in children (6 to 11 years; n = 177) and adolescents (12 to 17 years; n = 199) with MDD (166). This 10-week, DBPCT was followed by 24 weeks of open-label sertraline treatment. The study was not adequately powered to detect efficacy differences between age groups. A post hoc analysis found no statistically significant differences in time to first response or first persistent response between sertraline and placebo in children tested on the CDRS-R. Sertraline had a significantly faster time to first persistent response, however, in adolescents compared with placebo. Within treatment groups, children had a significantly faster time to first response than adolescents, whether treated with placebo or sertraline but not on time to first persistent response. Both age groups had similar improvement over 34 weeks of treatment.

In another report, Wagner and colleagues (167) examined citalopram (20 mg/day) in an 8week, DBPCT in younger subjects (7 to 17 years) at 21 sites across the United States. Based on a criterion of ≤28 on the CDRS-R, 36% responded with active drug compared with 24% on placebo (p < 0.05). Rhinitis, nausea, and abdominal pain occurred at ≥10% frequency in both groups, with no significant difference between them.

Emslie et al. (168) conducted a DBPCT of escitalopram (10 to 20 mg) in youths (12 to 17 years) with MDD (n = 312). A total of 83% patients (260/313) completed 8 weeks of doubleblind treatment. Significantly greater improvement occurred in the escitalopram group relative to the placebo group at end point based on the CDRS-R change score (i.e., —22.1 vs. —18.8; p> 0.022; last observation carried forward). Adverse events occurring in at least 10% of escitalopram patients were headache, menstrual cramps, insomnia, and nausea. Influenza-like symptoms occurred in at least 5% of escitalopram patients and at least twice the incidence of placebo
(i.e., 7.1% vs. 3.2%). The discontinuation rate due to adverse events was 2.6% with escitalopram and 0.6% with placebo. Serious adverse events were reported in 2.6% and 1.3% of escitalopram and placebo patients, respectively. The incidence of suicidality was similar for both groups.

The Treatment of Adolescent Depression Study (TADS) was an NIH-Sponsored, 12-week, 13site trial, which randomized subjects to fluoxetine, cognitive-behavioral therapy (CBT), fluoxetine plus CBT, or placebo (n = 327) (169). Response (based on the CDRS-R) was 73% with combination therapy, 62% with fluoxetine, and 48% with CBT at week 12 and 86% with combination therapy, 81% with fluoxetine, and 81% with CBT at week 36. Of note, there were no completed suicides. Suicidal ideation was present in 29% of patients at baseline and in 10% at the end of 12 weeks, with the lowest incidence among those on fluoxetine. Suicidal behavior, however, was greater among those who received fluoxetine. Adding CBT to medication was superior to monotherapy, with benefits outweighing the risks at the end of active treatment (i.e., week 36) and persisting during 1-year follow-up on all measures of depression and suicidality. In contrast to earlier reports of worsening with short-term treatments, longer treatment was associated with persistent benefits over 1 year of naturalistic follow-up.

Other Antidepressants

In a DBPCT, over 6 weeks, 33 subjects (8 to 17 years) who met DSM-IV criteria for major depression were treated with either venlafaxine and psychotherapy or placebo and psychotherapy. This study showed no additional advantage for venlafaxine (170). Results were similarly disappointing in two, subsequent, multicenter, DBPCTs of venlafaxine extended release (XR) that examined safety, efficacy, and tolerability in subjects (7 to 17 years) with MDD (171). Participants received venlafaxine XR (n = 169; flexible dose, based on body weight) or placebo (n = 165) for up to 8 weeks. The primary efficacy variable was change from baseline on the CDRS-R score. There was no significant difference between venlafaxine XR and placebo on the CDRS-R in either study. A post hoc, age subgroup analysis of the pooled data, however, showed greater improvement on the CDRS-R with venlafaxine XR compared with placebo (24.4 vs. 19.9; p < 0.022) among adolescents (12 to 17 years) but not among children (7 to 11 years). The most common adverse events were anorexia and abdominal pain. Although hostility and suicide-related events were more common in venlafaxine ER-treated participants than in placebo-treated participants, there were no completed suicides. Thus, venlafaxine XR may be effective in depressed adolescents, but its safety and efficacy in pediatric patients is not established.

Findling and colleagues (172) examined the pharmacokinetics of nefazodone and two of its metabolites based on age. They were more variable with shorter half-lives in children and adolescents compared with adults. Nefazodone was well tolerated, however, and associated with significant reductions (p < 0.001) in depressive symptoms.

Another 8-week, open trial with bupropion SR in adolescents with ADHD and comorbid MDD or DD reported a significant improvement in depressive symptoms from baseline (p < 0.05) (173).

There have been more than a dozen controlled trials of TCAs in pediatric MDD, which generally yielded negative results (101,144). Earlier studies suggested that amitriptyline, desipramine, and imipramine were superior to placebo, but also produced substantial adverse effects. The relative immaturity of noradrenergic pathways in children and adolescents was thought to underlie this poor tolerability (174). Further, reports of sudden death with desipramine highlighted the potential cardiovascular risk associated with TCAs in youth (175).

Although an open trial with MAOIs in 23 adolescents who had failed treatment with TCAs yielded positive results, these agents are rarely used in this age group because of bothersome and potentially life-threatening adverse effects (176). In addition, the rigid tyramine-free diet may be particularly difficult for children to follow.

RECOMMENDED TREATMENT STRATEGIES

Important medication-related issues for commonly used antidepressants are summarized in Table 14-4. This information and the results from controlled trials are incorporated into the following treatment recommendations:

Patients should be observed carefully for any emerging suicidal ideation, especially in the first few weeks of treatment.
No baseline laboratory work is currently indicated (140).
Except for lower initial doses, the administration of SSRIs in children and adolescents is similar to adults.
Start at lowest possible dose, increasing weekly by 1 mg/kg/day for fluoxetine and citalopram and 3 mg/kg/day for sertraline and fluvoxamine, or less based on tolerability. Negative results with paroxetine suggest that this drug should not be used.
Optimal time period for desired response is at least 6 weeks.

TABLE 14-4 IMPORTANT ISSUES WITH COMMONLY USED ANTIDEPRESSANTS IN CHILDREN AND ADOLESCENTS

Drug	FDA Pediatric Labeling	Dose (mg/day)	Dosing Schedule	Common Adverse Effects	Available Formulations
Fluoxetine	Above 8-18 years, major depression, 7-17 years in OCD	5-60	Once a day	Irritability; anxiety; insomnia; akathisia; Gl disturbance; headache; flu-like symptoms on rapid discontinuation; CYP inhibition	10, 20, 40 mg, Elixir-20 mg/5 mL mint flavored
Paroxetine	—	10-30	Once a day		10, 20, 40 mg, Oral solution 10 mg/mL
Sertraline	6 years and above, for OCD	25-200	Once a day		25,50, 100 mg, Elixir 20 mg/mL, not flavored
Fluvoxamine	8 years and above for OCD	12.5-200	Twice a day		25,50, 100 mg
Citalopram	—	10-40	Once a day		10,20,40 mg, Elixir-10 mg/5 mL
Escitalopram	—	5-20	Once a day		5, 10,20 mg
Venlafaxine	—	37.5-150	Twice a day; extended release (XR) can be given once a day	Treatment-emergent anxiety; increased diastolic blood pressure	Immediate release (IR): 25, 37.5, 50, 75, 100 mg;XR: 37.5, 75, 150mg
Bupropion	—	50-300	Twice a day; sustained release (SR) can be given once a day	Insomnia; irritability; seizures (avoid with seizure disorder)	IR: 75-100 mg; SR: 100, 150, 200 mg

TREATMENT-RESISTANT DEPRESSION

There are several factors that contribute to treatment resistance, including misdiagnosis, comorbid diagnoses, bipolar depression, an inadequate drug dose or treatment duration, inappropriate psychotherapy, medication nonadherence, and complicating factors such as abuse (see Chapters 6 and 7). Strategies to address these problems in the pediatric population are not different from adults with depression. Optimization of current medication (which includes adjusting the dose and/or extending the trial) and switching to another medication in the same or different class are options. Although anecdotal information reports value with TCA augmentation of lithium and the use of phenelzine, this is inconsistent with current recommendations to avoid TCAs or MAOIs in children and adolescents (141).

The Treatment of SSRI-resistant Depression in Adolescents (TORDIA) trial involved 334 youth (12 to 18 years) with MDD inadequately responsive to 2 months of monotherapy with an SSRI (177). In this study, patients were randomly switched to one of four groups for 12 weeks of additional treatment:

A different SSRI
A different SSRI plus CBT
Venlafaxine (150 to 225 mg)
Venlafaxine plus CBT

The combination of medication plus CBT generated significantly higher response rates (p < 0.009) compared with medication alone. Increased diastolic BP and pulse, as well as adverse skin reactions, were more frequent with venlafaxine versus the SSRI.

If other treatments fail, ECT is an option. Two psychiatrists experienced in treating children and not otherwise involved with the case should concur with the recommendation for ECT prior to its initiation (178). Preliminary reports also indicate that transcranial magnetic stimulation (TMS) may be safe and effective in adolescents with MDD (179).

Psychotic Depression

A combination of an SSRI and an SGA is the firstline intervention. As long-term adverse effects of SGAs such as TD and weight gain are unknown, we suggest tapering them after remission of psychotic symptoms (150).

Bipolar Depression

It is difficult to differentiate bipolar from unipolar depression for several reasons. BPD may present with one or more depressive episodes at the onset. Further, mixed mania is common in the pediatric population (152,180). Also, the irritability seen in unipolar depression can mimic the irritability seen in a manic episode or vice versa (160). As in adults, a bipolar diathesis is considered when there are more severe or psychotic symptoms. When bipolar depression is confirmed, the choice of medication is a mood stabilizer.

The efficacy of lamotrigine was examined in an 8-week, open-label trial as adjunctive or monotherapy in 20 adolescents (12 to 17 years; mean age 15.8; 7 boys, 13 girls) with BPD I, II, or not otherwise specified (NOS), depressive episode (181). Lamotrigine was initiated at 12.5 to 25 mg/day The mean final dose was 131.6 mg/day. Primary response criteria were a 1 or a 2 on the Clinical Global Impression-Improvement Scale at week 8. A secondary criterion was at least a 50% decrease in the baseline CDRS-R score. Seven subjects were on other psychotropic medications. Sixteen subjects (84%) responded based on the primary criterion, and 12 (63%) responded based on the secondary criterion. There was no significant weight change, rash, or other adverse effects during the trial. Adolescents with bipolar depression appeared to respond to lamotrigine treatment, whether as adjunctive therapy or monotherapy

Bipolar adolescent girls should be educated about birth control measures, in part because of their impaired judgment and in part because of the potential teratogenic effect of the mood
stabilizers (see Chapter 10). Further, it is critical to adjust the dose of oral contraceptives when combined with agents such as CBZ and lamotrigine to avoid an inadvertent pregnancy.

Dysthymic Disorder

Studies in adults find that SSRIs, TCAs, and MAOIs at the same doses used for MDD are efficacious for the treatment of dysthymic disorder (DD) (182). Using traditional psychotherapies is also appropriate. Data on treatment of DD in children and adolescents is sparse, but the same practice used for adults is recommended (150).

MAINTENANCE TREATMENT TRIALS

In view of the chronic nature of MDD, treatment should be continued for at least 6 to 12 months after remission of acute symptoms. Emslie and colleagues (183) compared the efficacy of fluoxetine (20 to 60 mg/day) with placebo in the prevention of relapse in MDD over 32 weeks. Subjects included those that achieved a score of ≤28 on the CDRS-R on fluoxetine therapy in an acute trial that preceded the maintenance phase. Relapse occurred in 34% of those on fluoxetine compared to 60% on placebo (p ≤ 0.05). Further, the investigators demonstrated that adverse events and tolerability did not differ between fluoxetine and placebo. Monthly cognitive behavior therapy sessions may also prevent relapse in youth (184). If a patient initially responds to an antidepressant, the physician should continue the same dose for at least 8 months, unless there are significant adverse effects. If planning to stop the antidepressant, the physician should taper the drug over a period of 6 weeks to avoid withdrawal symptoms.

The practice guidelines for treatment of depressive disorders in children and adolescents suggest that youth with two or three episodes of MDD receive maintenance treatment for at least 1 to 3 years (150). Thus, maintenance antidepressant treatment may extend from 8 to 12 months to an indefinite period based on the severity of MDD, presence of psychosis, suicidal risk, level of impaired functioning, and number of past episodes.

SAFETY AND TOLERABILITY

The initial workup prior to antidepressant therapy includes a CBC with differential, primarily to rule out infection and anemia; serum electrolytes; and BUN and creatinine to rule out kidney disorders. Liver and thyroid function (T3, T4, TSH) tests can rule out disorders that may present as MDD or accentuate depressive symptoms. With a history of a seizure disorder, an EEG is appropriate. Similar to adults, the dexamethasone suppression test (DST) does not have sufficient specificity to be used routinely.

CONCLUSION

There is increasing evidence for the efficacy of SSRIs in treating pediatric major depression. There is no clear evidence of increased suicidal ideation in the reported, controlled trials. Combined medication plus psychotherapy appears to be superior to medication alone, particularly in resistant depression. Caution must be exercised if there is an identified risk for BPD in view of the activating effects of SSRIs, including risk of suicidal ideation. In our opinion, the efficacy and safety data support a role for pharmacotherapy in youth with depression.

Bipolar Disorder in Children and Adolescents

DESCRIPTION OF THE DISORDER

The clinical presentation of BPD in the preadolescent and early adolescent age groups is often debated. Although symptoms with onset in midto late adolescence are similar to that of adult disorder, the NIMH Research Roundtable on Prepubertal BPD (185) agreed that pediatric BPD can present with “narrow” as well as “broad” phenotypes. Children and adolescents with the “narrow” phenotype have recurrent periods of major depression and mania or hypomania, fitting the classic definitions of BPD type I or II described in the DSM-IV-TR (60). These episodes are often characterized by rapid cycling (186,187). Despite having classic symptoms of mania or hypomania, many fall short in either the number of items or in duration criteria and therefore qualify for BPD, NOS (i.e., the broad phenotype) (188). These children usually manifest symptoms with or without clear episodicity, including severe irritability and temper outbursts, depression, anxiety, hyperactivity, poor
concentration, and impulsivity (189). Recent data also indicate that preschool offspring of parents with BPD have an elevated risk of ADHD and greater levels of subthreshold mania and depressive symptoms (190).

Geller and colleagues (191,192) compared the clinical characteristics of a sample of bipolar children (7 to 16 years) with an ADHD group and healthy controls referred from community practitioners. They found that grandiosity, elated mood, hypersexuality, flight of ideas, and decreased need for sleep differentiated children with BPD from the other two groups. In contrast, Biederman and colleagues reported the phenomenology of BPD in a sample of children 12 years and younger as presenting with elevated levels of irritability rather than elevated mood (180,193,194). Longitudinal studies are needed to clarify whether core symptoms of elation, grandiosity, and symptom periodicity are required or whether severe irritability without elation and/or grandiosity is sufficient to diagnose pediatric BPD.

Geller and colleagues (195) use the term complex cycling to describe the presence of short cycles embedded within a more prolonged cycle or episode. The terms “ultra rapid cycling” (5 to 364 cycles per year) and “ultradian cycling” (more than 365 cycles per year; mania occurs for 4 or more hours per day) can help document rapid mood fluctuations (196,197). According to DSM-IV-TR criteria, an ultradian cycle cannot be considered an episode or a cycle of mania, hypomania, or depression. Although such rapid mood fluctuations are common in bipolar youth, the extent to which this is specific to the disorder is unknown. Psychosis is reported in 16% to 60% with pediatric BPD (187), with auditory hallucinations the most common symptom.

There are no existing data on the prevalence of BPD in the pediatric population that considers these current views in the prepubertal population. One community study using the Schedule for Affective Disorders and Schizophrenia (SADS) evaluated the rates of bipolar spectrum disorders in adolescents and reported a lifetime prevalence of approximately 1% in youth (14 to 18 years) (198). The majority had BP-II disorder or cyclothymia. An additional 5.7% of the cases had subsyndromal symptoms of BPD defined as a “distinct period of abnormally and persistently elevated, expansive, or irritable mood.” Retrospective studies in adults with BPD report that up to 60% had their onset before age 20. Among clinic samples, BPD is noted in 0.6% to 15% depending on the instrument utilized to ascertain diagnosis, the referral source, and the program’s specialization (192,197,198 and 199).

TREATMENT OF BIPOLAR DISORDER

Mood Stabilizer Monotherapy

An open-label, randomized trial of divalproex sodium, lithium, and CBZ for bipolar I and II was conducted in 42 children (8 and 18 years) (135). Response was defined as at least a 50% reduction from baseline Young Mania Rating Scale (YMRS) scores and CGI-I scores ≤2. Using these criteria, response rates were 53% with divalproex sodium, 38% with lithium, and 38% with CBZ with no significant difference between the groups. In this trial, nausea was the most commonly reported adverse event primarily occurring with CBZ. Sedation was the next most common adverse event with CBZ and divalproex sodium.

Another multisite, open-label, 8-week trial of divalproex sodium examined outcome in 40 subjects (7 to 19 years) with mania or hypomania in the context of bipolar I or II disorder (200). Based on a 50% reduction in baseline YMRS scores, the response rate was 61%. Adverse events occurring in >10% of subjects included headache, nausea, vomiting, diarrhea, and sedation.

A 6-month, prospective study of divalproex sodium specifically examined its effects on mixed mania in pediatric BPD (n = 34) (201). A total of 73.5% of subjects met response criteria (i.e., 50% reduction in baseline YMRS scores and ≤40 on CDRS-R). It is important to note, however, that rescue paradigms were used to sustain subjects through the 6-month trial period, and those with comorbid ADHD continued on psychostimulants (n = 13). Further, breakthrough psychotic symptoms were treated with risperidone (n = 17). This study illustrates the complexity of conducting a clinical trial in youth with BPD frequently complicated by chronicity, comorbid conditions, and breakthrough symptoms. A double-blind, randomized trial of divalproex XR consisted of a 28-day acute phase followed by a 6-month open-label extension phase to examine its safety and efficacy in 150 patients (manic or mixed episode; 10 to 17 years) (202). A treatment effect was not
observed with divalproex XR based on change in mean YMRS baseline score (i.e., divalproex ER, -8.8 [n = 74]; placebo, -7.9 [n = 70]). Divalproex was also similar to placebo based on the incidence of adverse events. In the long-term study, YMRS scores decreased modestly (2.2 points from baseline) in 66 participants. The most common adverse events were headache and vomiting. These results do not provide support for the use of divalproex XR in the treatment of youths with bipolar I disorder, mixed or manic state.

A DBPCT of oxcarbazepine involved 116 outpatients (7 to 18 years) with bipolar I disorder, manic or mixed episodes (203). Flexibly dosed treatment with oxcarbazepine (mean dose = 1515 mg/day) or placebo did not significantly improve YMRS scores at end point (adjusted mean change: oxcarbazepine, —10.90; placebo,

— 9.79). Dizziness, nausea, somnolence, diplopia, fatigue, and rash were reported in at least 5% of the patients in the oxcarbazepine group with an incidence at least twice that of placebo.

A DBPCT of topiramate (n = 29, 52%) compared with placebo (n = 27, 48%) involved 56 children and adolescents (6 to 17 years) with a diagnosis of bipolar I (204). The authors reported a significant difference between slopes of the linear mean profiles of the YMRS (p < 0.003) using a post hoc repeated measures regression as well a change in Brief Psychiatric Rating Scale for Children at day 28 (—14.9 vs.

— 5.9; p < 0.04) using observed data. Although topiramate was generally well tolerated, adverse events included decreased appetite, nausea, diarrhea, and paresthesias. The results were inconclusive, however, because premature terminations resulted in a limited sample size.

Second-Generation Antipsychotic Monotherapy

Olanzapine. This agent was FDA approved for the treatment of mania in adults in 2004 but is not approved for the treatment of juvenile mania. Olanzapine has a rapid onset of action for mixed and manic episodes. An 8-week, open trial of olanzapine in 5- to 14-year-olds (n = 23) with the narrow phenotype of BPD was conducted (205). Using a lower threshold for response (i.e., 30% reduction in baseline YMRS scores and ≤3 on CGI-S), 61% met criteria. In addition, 35% of the subjects received stimulants during the entire trial. Olanzapine led to an increase in appetite in 61% of the subjects, somnolence in 43.5%, and abdominal pain and weight gain in 30.4%.

A large, multisite, 3-week, DBPCT of olanzapine for 161 adolescents (mean age = 15 years) with manic or mixed episodes reported a significantly greater reduction in symptoms and higher rates of response and remission in patients on drug versus placebo (206). In this study, participants were randomly assigned to receive either placebo or olanzapine (2.5 to 5.0 mg/day) titrated in a response-dependent fashion up to 20 mg/day. The mean dose of olanzapine was 10.7 ± 4.5 mg/day. After a week of treatment, patients on olanzapine had an 18-point reduction in their baseline YMRS score, whereas patients assigned to placebo reduced their score by 10 points (p < 0.001). At end point, the group treated with olanzapine showed higher response (48% vs. 22%; p < 0.002) and remission (35% vs. 11%; p < 0.001) rates compared with placebo. Based on these findings, the number needed to treat (NNT) for response with olanzapine was 3.8, and 4.2 for remission. Olanzapine was associated with a greater mean weight gain (i.e., 3.7 kg) compared with placebo (i.e., 0.3 kg) and an incidence of clinically significant (i.e., at least 7%) weight gain in 42% of patients on olanzapine and 2% on placebo (p < 0.001, number needed to harm (NNH) = 2.5; CI = 2.0 to 3.3). The olanzapine group also showed increases in prolactin levels relative to the placebo group, particularly among boys (females: 25.7% vs. 0%, p < 0.007, NNH = 3.9, 95% CI = 2.5 to 8.9; males: 62.5% vs. 5%, p < 0.001, NNH = 1.7, 95% CI = 1.4 to 2.5).

Thus, with olanzapine, the likelihood of harm (weight gain; NNH of 2.5) offsets the likelihood of benefit (response, remission; NNTs of 3.8 and 4.2, respectively). Based on data from the above trial and our clinical experience in pediatric populations, if olanzapine is used, it should be initiated at 2.5 mg daily and titrated to a target dose of 5 to 20 mg once daily or divided twice daily depending on the child’s weight and treatment response. Longer-term data in pediatric BPD is less available, but a few open-label prospective studies show promising results. For example, after completing the 3-week trial, 146 adolescents continued taking olanzapine in an openlabel fashion for up to 26 weeks (207). At the end of this extended treatment period, study participants showed a 63% response rate with olanzapine (i.e., <50% reduction in YMRS score and
a Clinical Global Impressions-Bipolar Disorder [CGI-BP] severity score of 3). However, 69% of the adolescents had a 7% or greater increase in body weight and 40.5% of participants had abnormally high prolactin levels. Thus, metabolic adverse effects limit the utility of olanzapine as an acute and maintenance agent for youth withBPD (225).

Risperidone. In 2009, this agent was the first SGA to receive FDA approval as a monotherapy for the short-term treatment of mania or mixed episodes in BPD youths (10 and 17 years) (208).

An open trial considered risperidone in 6- to 17-year-olds with BPD (n = 30). Based on a YMRS baseline score decrease of 30% or CGI-I score ≤2, 70% responded (209). In this study, a fourfold increase in prolactin levels, as well as weight gain, was reported. In a 3-week, multicenter, DBPCT, 50 youths (10 to 17 years) with manic or mixed episodes received risperidone (0.5 to 2.5 mg/day; mean modal dose, 1.9 mg/day); 61 received risperidone (3 to 6 mg/day; mean modal dose, 4.7 mg/day); and 58 received placebo (210). At end point, risperidone demonstrated a greater decrease in YMRS scores (p < 0.001) and a greater increase in response rates compared with placebo (i.e., 59% in the lowdose group; 63% in the high-dose group; and 28% in the placebo group). Mean weight gain was 2 kg with risperidone (0.5 to 2.5 mg/day), 1.5 kg with risperidone (3 to 6 mg/day), and 0.65 kg with placebo. Rate of discontinuation with risperidone was 12% (13/111) versus 7% (4/58) with placebo. The most common adverse events leading to discontinuation included somnolence (5%), nausea (3%), abdominal pain (2%), and vomiting (2%). The authors concluded that the 0.5 to 2.5 mg/day dose had a better riskbenefit profile than the 3 to 6 mg/day dose.

In pediatric BPD and schizophrenia trials, prolactin increases are reported in 82% to 87% with risperidone and in 3% to 7% with placebo, yielding an NNH of 1.3 (211).

The recommended starting dose of risperidone in children and adolescents with acute manic or mixed episodes is 0.25 mg/day, and increases of 0.5 to 1 mg/day with a target dose of 2.5 mg/day (QD or divided twice or three times daily) (208). The FDA recommends a 2.5 mg/day maximum dose, as there is no evidence that higher doses are more effective. Since risperidone is commonly associated with weight gain and hyperprolactinemia, physicians should assess for these issues prior to treatment and monitor every 6 months thereafter.

Quetiapine. In 2004, this agent was approved by the FDA for the treatment of acute mania in adults but has not been approved for use in youths with mania.

In a 3-week, multicenter, DBPCT of acute mania in youths (10 to 17 years), the YMRS response rate was higher for those receiving quetiapine (300 or 600 mg) (n = 188) versus those receiving placebo (n = 89) (i.e., 61% vs. 37%, p < 0.002), yielding an NNT of 4.2. (212). Common adverse effects with quetiapine were somnolence (quetiapine 30%, placebo 10%, NNH = 5.0), sedation (quetiapine 25%, placebo 4.4%, NNH = 4.9), and dizziness (quetiapine 18.1%, placebo 2.2%, NNH = 6.3). Mean weight gain with placebo was 0.4 kg, and for both quetiapine groups was 1.7 kg. The rates of clinically significant (>7%) weight gain were 12.2% with quetiapine and 0% with placebo, resulting in an NNH of 8.2.

The recommended dosing of quetiapine in youths with acute manic or mixed episodes is to start with 50 mg/day taken at bedtime, increased to 100 mg/day after 2 days, and then increased to a target dose of 300 to 600 mg after 2 additional days QD or BID. Quetiapine is associated with sedation and orthostatic hypotension, presumably secondary to its affinity for histaminergic and a-adrenergic receptors, respectively. Weight gain in children on quetiapine appears to be slightly less than that for olanzapine or risperidone (213).

Aripiprazole. This agent has FDA approval as a monotherapy for the treatment of acute manic and mixed episodes in children and adolescents (10 to 17 years), as well as for maintenance treatment of BPD and adjunctive therapy for major depression in children (208). Aripiprazole is also approved for the treatment of schizophrenia in adolescents (13 to 17 years).

A 4-week, DBPCT studied adolescents (10 to 17 years) with acute manic or mixed episodes (214). The YMRS pooled response rate in 197 subjects receiving either 10 or 30 mg/day of aripiprazole was higher (i.e., 54.2%) than in 99 subjects receiving placebo (i.e., 26.1%). The NNT for YMRS response to aripiprazole at end point was 3.6, and a significant decrease in YMRS was seen with aripiprazole at week 1 and thereafter. Common adverse effects with aripiprazole were
somnolence, EPS, fatigue, nausea, akathisia, blurred vision, salivary hypersecretion, and dizziness. Rates of somnolence (aripiprazole, 22.9%; placebo, 3.1%; NNH = 5.1); extrapyramidal disorder (aripiprazole, 19.8%; placebo, 3.1%; NNH = 6.0); akathisia (aripiprazole, 9.7%; placebo, 2.1%; NNH = 13.2); salivary hypersecretion (aripiprazole, 5.6%; placebo, 0%; NNH = 17.9); and clinically significant (≥7%) weight gain appeared to increase with the aripiprazole dose (i.e., 10 mg, 4.0%; 30 mg, 12.3%; placebo, 4.6%; NNH = 28). Mean weight gain was modest and statistically similar across groups (10 mg, —0.55 kg; 30 mg, —0.90 kg; placebo, —0.54 kg). Discontinuation rates due to adverse effects were 7% with aripiprazole and 2% with placebo.

A recent, small (n = 43), 6-week, DBPCT assessed the role of aripiprazole in adolescents with acute mania and comorbid ADHD (215). Aripiprazole was superior to placebo in reducing baseline YMRS scores (p < 0.02) and did not worsen symptoms of ADHD.

The recommended dosing of aripiprazole in children and adolescents (10 to 17 years) with acute manic or mixed episodes is to start with 2 mg/day, increase to 5 mg/day after 2 days, and increase to the target dose of 10 mg/day after 2 additional days (QD or BID). Subsequent dose increases should be administered in 5 mg/day increments up to 30 mg/day (QD or BID).

Ziprasidone. In 2006, this agent was approved by the FDA for the treatment of acute manic episodes in adults but has not been approved for pediatric mania.

A 4-week, multicenter, DBPCT in children and adolescents (10 to 17 years) with acute manic or mixed episodes assigned 149 subjects to ziprasidone (80 to 160 mg/day) or 88 subjects to placebo (216). Ziprasidone produced a greater mean YMRS score decrease than placebo (—13.8 vs. — 8.6, respectively). Response rates using a less stringent observed cases analysis were 62% for ziprasidone and 35% for placebo, yielding an NNT for ziprasidone compared with placebo of 3.7. The most common adverse effects with ziprasidone were sedation (33%), somnolence (25%), headache (21%), nausea (13%), fatigue (13%), and dizziness (11%). Mean weight change was —0.6 kg and —0.2 kg for the ziprasidone and placebo groups, respectively

Clinically significant (≥7%) weight gain on ziprasidone was 7% versus 4% on placebo, yielding an NNH of 33.3. There were no significant changes in mean body mass index, lipids, liver enzymes, or glucose levels. One ziprasidone-treated subject had a QT prolongation of 460 ms. Ziprasidone is among several SGAs that block potassium currents (217) and prolong ventricular repolarization (218,219). This may lead to prolonged QT duration, torsades de pointes, or fatal arrhythmias (220,221). Prior to initiating treatment with ziprasidone, a careful personal history of congenital long QT syndrome should be obtained, along with a family history of sudden cardiac death (222). Ziprasidone should be avoided in individuals who are on other medications that also prolong the QT interval. Although there are currently no official recommendations for monitoring the QT interval by ECG, an ECG should be obtained at baseline and following attainment of the ziprasidone target dosage (223,224).

Pediatric populations may be given an initial daily dose of 20 mg/day, which may than be titrated to 80 to 160 mg QD or BID. Administration of ziprasidone is complicated by the phenomenon of increased risk of akathisia at lowers doses, and the doubling of absorption when administered with food.

Mood Stabilizer Plus Antipsychotic

Kafantaris and colleagues (226,227) examined a small sample (n = 35) of manic adolescents with or without psychosis. They reported that antipsychotics plus lithium were superior to lithium monotherapy, with subjects worsening after discontinuation of their antipsychotic.

DelBello and colleagues (228) conducted a double-blind, placebo-controlled augmentation trial of divalproex sodium plus quetiapine or placebo in 30 manic adolescents. Based on a ≥50% reduction in baseline YMRS scores, the response rate was 87% with combination therapy versus 53% with divalproex sodium plus placebo (p < 0.001). There appeared to be no group differences in adverse events except an increased incidence of sedation in the combined medication group.

Pavuluri and colleagues (229) conducted a 6-month, prospective trial of lithium, plus risperidone versus divalproex sodium plus risperidone in youth (7 to 17 years). With response defined as ≥50% YMRS change score from baseline, 82.4% met criteria with lithium plus risperidone and
80% with divalproex sodium plus risperidone. Defining remission as ≥50% YMRS change score from baseline; an end point CGI-I of ≤2; and an end point C-GAS of ≥51; 45% and 60% remitted, respectively. Weight gain, nausea, sedation, and gastrointestinal upset were the most frequent adverse effects, with no significant group differences.

In summary, the limited data indicate that acute combination therapy with a mood stabilizer plus antipsychotic may be effective and tolerated in children and adolescents with BPD.

Combined Mood Stabilizers

Findling and colleagues (230) studied bipolar I and II patients using lithium plus divalproex sodium for mood stabilization. All subjects were switched to this combination (Phase I of the study) before being randomized to either lithium or divalproex sodium monotherapy (Phase II of the study). At the point of entry into Phase I, 90 subjects (5 to 17 years) were in a mixed, manic, depressed, hypomanic, or euthymic phase. Stimulants were used in 53%, a₂-agonists in 22%, SGAs in 19%, and SSRIs in 10%. Response rate was 70.6% and remission rate 46.7% for those on combination therapy. A total of 16.7% withdrew due to intolerance of study medication. Common adverse effects included nausea, emesis, stomach pain, enuresis, tremor, thirst, headache, sedation, appetite increase, and diarrhea.

Kowatch and colleagues followed an original sample from their acute trial mentioned earlier into a seminaturalistic study for up to 16 weeks (231). Rescue algorithms included augmentation of any of the three mood stabilizers used in the acute trial with another mood stabilizer (if bipolar); antipsychotic (if psychotic); stimulant (if ADHD was predominant); or antidepressant (if depressed; SSRI or bupropion). Twenty of 35 subjects required combination therapy in the follow-up phase. Response was 80% based on a YMRS change score of ≥50% improvement from baseline.

In summary, accruing data indicate that response rates are higher with combination regimens.

COMORBIDITY TRIALS

Geller and colleagues (232) conducted a doubleblind, lithium and placebo-controlled, 6-week trial in youth with bipolar I (n = 12), bipolar II (n = 5), and MDD with bipolar risk factors (n = 8). Subjects also met criteria for substance abuse (i.e., alcohol and/or marijuana). Weekly random urine samples for drug assays were obtained. Six of 13 on active treatment with flexibly dosed lithium were responders with only 10% positive on their urine drug screen. By contrast, 40% were drug positive on placebo. Adverse events included thirst, polyuria, nausea, vomiting, and dizziness.

Scheffer and colleagues (233) conducted a 4-week trial in subjects (n = 40) randomized to a combination of divalproex sodium plus placebo or divalproex sodium plus a mixed amphetamine salt. Subjects (6 to 17 years) were randomized to one regimen for 2 weeks, then crossed over to the other for 2 weeks. Diagnoses included bipolar I (77.5%) and bipolar II (22.5%) disorder plus ADHD. The mean age was 11.4 years. Response defined as a YMRS change score ≥50% from baseline and CGI score of 1 or 2 occurred in 32 patients, 30 patients then crossed over to receive amphetamine salt and experienced improvement in their ADHD without worsening of mania. Adverse events included abdominal pain, increased appetite, and drowsiness. This study is important because it demonstrated that there was no worsening of mood symptoms with stimulant treatment of comorbid ADHD.

DISCONTINUATION TRIAL

One study examined the effects of lithium monotherapy discontinuation in adolescent BPD. Forty subjects (12 to 18 years) who responded to lithium were then randomized to either lithium or placebo (234). At the end of 2 weeks, 52.6% (n = 10/19) of those on lithium and 61.9% (n = 13/21) of those on placebo had an exacerbation (i.e., did not maintain response achieved at study entry). The interpretation of these results is limited, however, by the study’s short duration.

MAINTENANCE TRIAL

Given the chronic nature of BPD, continuation treatment trials to examine the ability of pharmacotherapy to prevent or postpone relapse of manic and depressive episodes are important. One such study was conducted by Findling and colleagues (235) using a double-blind design
comparing lithium to divalproex sodium over 18 months in 60 bipolar I and II, stable, euthymic subjects. Thirty children and adolescents (5 to 17 years) were enrolled, in each arm. Stimulants were used in 58.3%. At the end of 18 months, only three subjects remained in each arm. Time to relapse and premature discontinuation for any reason did not differ between the lithium or divalproex sodium groups. Median survival in the study was 114 days on lithium and 112 days on divalproex sodium. Subjects with higher baseline YMRS scores discontinued earlier. Adverse effects with lithium included emesis and enuresis (30% each). Stomach pain and headache (23.3% each) were the most common adverse effects in the divalproex sodium group.

RECOMMENDED TREATMENT STRATEGIES

Although the pharmacotherapy trials outlined above systematically address the issue of drug efficacy for pediatric BPD, they do not directly guide clinicians how to incorporate the data in a cogent fashion (i.e., translate science to service). To assist in this regard, Kowatch and colleagues (236) published treatment guidelines for BPD in children and adolescents. The model of pharmacotherapy developed in these guidelines was subsequently tested by our group in an algorithm study to establish its feasibility (237).

Algorithm Study

This was a prospective, 18.6-month, outpatient study comparing the algorithm group with a “treatment as usual” (TAU) group. Subjects (n = 34) included those with an initial bipolar manic or mixed episode. Their mean age was 11.74 (±3.36) years. The algorithm allowed for multiple tactics to treat mood, psychosis, ADHD, sleep difficulties, and attend to partial or nonresponse on follow-up. In the entire sample, a final CGI-I BP of ≤2 was reported in 68.3% with a final mean Children’s Global Assessment Scale (C-GAS) score of 56.19 (±7.62). Mean weight gain was 6.3 lbs in the algorithm group and 5.7 lbs in the TAU group. This study highlights the need to intervene as complicated problems arise, with the development of tolerance, with a partial outcome, or when comorbid conditions are present.

CONCLUSION

Pediatric BPD is a complex clinical entity. Treatment involves the management of mood instability across various phases of the illness. Manic and mixed symptoms require mood stabilizer monotherapy or combined therapy with an SGA or second mood stabilizer. Depression may require lithium, but there are no controlled trials to address depressive symptoms in this population. Comorbid ADHD can be safely and effectively treated with a mood stabilizer plus stimulant. Lithium can potentially alleviate substance abuse, but benefit is inconclusive. More short- and long-term controlled trials are needed. Algorithms are a practical approach to treating additional symptoms such as sleep difficulties, psychosis, and breakthrough mania or depression.

Anxiety Disorders in Children and Adolescents

DESCRIPTION OF THE DISORDER

The DSM-IV-TR includes the following major anxiety disorders for youth:

Separation anxiety disorder
Generalized anxiety disorder (GAD)
Social phobia (i.e., social anxiety disorder)
Specific phobia
Panic disorder (with and without agoraphobia)
Agoraphobia without panic disorder
Posttraumatic stress disorder (PTSD)
Obsessive-compulsive disorder

In addition, most children with selective mutism also meet criteria for social phobia, though selective mutism may have a multifactorial etiology (238). Each of these disorders has its own distinct features, yet shares a common foundation of excessive, irrational fear, and dread. This section will address the anxiety disorders with the exception of PTSD and OCD. These two differ relative to etiology and treatment considerations and will be discussed later.

The DSM-IV-TR only categorizes SAD and selective mutism as disorders “usually first diagnosed in infancy, childhood, or adolescence.” Regardless of referral concerns, it is important to inquire about symptoms of each type of disorder due to the multiplicity of ways in which anxiety
manifests in children and adolescents. For example, anxious symptoms may present as crying, tantrums, irritability, oppositionality, argumentativeness, and other behavioral problems in addition to the core features of fear, worry, and avoidance. The DSM-IV-TR adjusts to account for these differences between child and adult manifestations of anxiety disorders. For example, children can be diagnosed with specific phobia if anxious symptoms include crying and tantrums. Also, it is not necessary for children to recognize that their fear is excessive or unreasonable to meet criteria for social phobia or specific phobia. Furthermore, the criteria for GAD require three associated symptoms for adults but only one for children. School-aged children and adolescents display the discrete subtypes of anxiety disorders proposed in the DSM-IV-TR, but anxiety symptoms in preschoolers may be more diffuse and less distinct from one another (239).

It is a rule in pediatric psychiatry that anxiety disorders commonly co-occur with other anxiety and other psychiatric disorders. Thus, more than two thirds of children with an anxiety disorder meet criteria for at least one other anxiety disorder, 28% to 69% also suffer from major depression, and 15% to 30% have ADHD. Other common comorbid conditions include oppositional defiant disorder, learning disabilities, and language disorders (240,241,242,243,244 and 245). Children with anxiety disorders are also at increased risk for alcohol abuse in adolescence (246). It is important to carefully assess for the presence of comorbid conditions so that they can be properly identified and treated early in conjunction with the primary anxiety disorder.

Prevalence rates for anxiety disorders in children and adolescents vary from 10% to 20% depending on the stringency of diagnostic criteria and impairment. Although rates of anxiety disorders are similar in prepubertal boys and girls, there is a slightly higher occurrence in pubescent girls than boys (247,248).

The typical age of onset varies across the anxiety disorders and approximates the developmental progression of normal fears in childhood, such that

SAD often presents at ages 6 to 9 years
GAD presents at any age, but most often at ages 10 to 12 years
Social phobia presents at age 12 years and older

Further, the same child may experience different anxiety disorders over time (249).

TREATMENT OF ANXIETY DISORDERS

Cognitive-Behavioral Therapy

Although the focus of this section is on pharmacotherapy, cognitive-behavioral therapy (CBT) is often the first choice in managing youth with anxiety disorders (250). There are several controlled studies underway to examine the comparative efficacy of medication versus CBT as well as other psychotherapeutic interventions alone or in combination. Until evidence from these comparative studies can inform clinical practice, treatment of childhood anxiety disorders usually begins with CBT. The exception occurs when severe anxiety, impairment, or comorbidity warrant medication to sufficiently stabilize the child for participation in CBT. Thus, medication is initiated or combined with CBT for reduction of intense, acute symptoms in a severely impaired child or when there is partial response to CBT and potential for improved outcome with augmentation (251). Despite symptom reduction with treatment, residual anxiety can significantly impact functioning and increase the risk for symptom persistence over time (242,244,252,253). Thus, a combination of several intensive treatments may be needed to achieve sufficient remission of symptoms, to improve functioning in older youth, and to manage comorbid depression or social withdrawal (254,255).

Psychopharmacologic Treatments

Selective Serotonin Reuptake Inhibitors.

SSRIs are the first choice of medication to treat childhood anxiety disorders. Several RCTs support their short-term efficacy and safety. Studies included

Fluvoxamine for GAD, SAD, and social phobia (245)
Fluoxetine for GAD, social phobia, and selective mutism with social phobia (242,256)
Sertraline for GAD (257)

Adverse effects with these medications were generally transient and mild and included stomachaches, increased activity level, insomnia, and agitation/disinhibition at higher doses.

A multicenter, placebo-controlled trial of fluvoxamine for youth (n = 128) with social phobia, SAD, or GAD with substantial symptom severity and global impairment reported significant improvement in the fluvoxamine (76%)
versus placebo (29%) group (p < 0.05) (245). Fluvoxamine was dosed to a maximum of 250 mg/day in children younger than 12 years and 300 mg/day in adolescents. Social phobia and severity of illness predicted a less favorable outcome (258). In a subsequent controlled trial, youth with GAD, SAD, and/or social phobia and significant functional impairment were randomized to fluoxetine (20 mg/day) or placebo (242). Youths (n = 74) with social phobia and GAD responded significantly better to fluoxetine than placebo (p < 0.05). Youths with severe SAD, however, showed only a trend toward improvement on fluoxetine. Overall, there was only partial resolution of symptoms in about half of the treatment group in this fixed dose study. Fluoxetine was well tolerated.

There are currently no controlled studies for medication treatment of panic disorder in youth, but a small, open trial of various SSRIs in children and adolescents (n = 12) reported significant reduction in symptoms from baseline (259). Adjuvant BZDs were also used as required.

Although randomized controlled trials have established the safety and efficacy of SSRIs in acute treatment of youth with anxiety disorders, their benefits and risks with long-term exposure have not been studied. Pine (253) recommends that clinicians consider a medication-free trial for children who have a marked reduction in anxiety or depressive symptoms on an SSRI and maintain stability for 1 year. Medication tapering should occur during a low-stress period, and if the child or adolescent begins to relapse, the SSRI should be reinitiated.

There is currently no empirical support for a specific SSRI for a particular childhood anxiety disorder. Clinically, the choice is often based on potential adverse effects, medication half-life, or positive response to a specific SSRI in a close relative with anxiety (260). There are no other dosing guidelines at this time, but clinicians are encouraged to start with low doses in children and adolescents, to monitor adverse effects closely, and to increase the dose slowly based on treatment response and tolerability until there is symptom resolution and improvement in functioning (242,278,261). Clinicians should consider increasing the dose if significant improvement does not take place by week 4 of treatment (242,245). Of note, there was no increase in suicidal ideation/behavior in these trials for anxiety disorders.

Other Antidepressants. Although SSRIs are the drugs of choice for a variety of childhood anxiety disorders, dual mechanism antidepressants (e.g., venlafaxine, TCAs) are potential alternatives. There are no controlled trials in childhood anxiety disorders to guide treatment with medication combinations when a single medication does not adequately manage anxiety symptoms. Comorbid diagnoses are also a factor to consider in selection. A pooled analysis from two multicenter placebo-controlled trials with venlafaxine XR suggests it is effective and well tolerated in youth with GAD (262). In one trial, a total of 320 subjects (6 to 17 years) received venlafaxine XR or placebo. Response rates were 69% with active drug versus 48% with placebo adverse effects included asthenia, anorexia, pain, and somnolence. A second DBPCT with venlafaxine XR also demonstrated benefit for pediatric social anxiety disorder (263).

Clomipramine, a TCA with serotonergic properties, is used alone or to augment the effect of an SSRI, especially when OCD symptoms are present. Clomipramine should be introduced at low doses and carefully monitored, as it may increase the blood level of the concurrent SSRI and/or produce adverse effects typical of TCAs. In general, TCAs such as clomipramine are used less often because they require ECG monitoring and can be fatal in overdose.

Buspirone. Although this agent is used for GAD in youth, efficacy data from controlled trials does not exist. An open, dose-escalation study reported this agent was safe and well tolerated up to 30 mg BID in adolescents (264). Higher peak plasma levels were achieved in children (vs. adolescents and adults), and some had to withdraw because of adverse effects such as lightheadedness, headache, and dyspepsia.

Benzodiazepines. BZDs are often used as adjunctive treatment for severe anxiety or in specific, time-limited anxiety-producing situations (e.g., a child phobic of flying). Although the few controlled studies yielded conflicting results, these agents can rapidly reduce anxiety symptoms while waiting for the full effect from an SSRI (259). Due to the possibility of physiological dependence (220), BZDs are contraindicated in a youth with a substance abuse or dependence diagnosis and should also be used cautiously with a family history of substance abuse (261).
α-Agonists. In clinical practice, guanfacine (1 mg up to three times a day) or clonidine (0.1 mg up to three times a day) is combined with an SSRI when there is significant associated autonomic arousal and/or restlessness, though no controlled studies exist. An ECG and careful monitoring of pulse and blood pressure are recommended with these medications. It is important to discuss the possibility of rebound hypertension on their discontinuation with patients and families, so that medication is not stopped abruptly. Youth with comorbid Tourette disorder, trichotillomania, other impulse control disorders, ADHD, BPD, or PTSD may benefit from these medications as well.

RECOMMENDED TREATMENT STRATEGIES

The following clinical approach is recommended for the treatment of anxiety in young patients:

Regardless of presenting problems, screen routinely for anxiety disorders
Anxiety disorders in children and adolescents have frequent comorbidities that must also be addressed
Major risk factors (e.g., parental anxiety, temperament, parenting styles, attachment, negative life events, and stressors) and protective factors (e.g., coping style) are identified for childhood anxiety disorders
Treatment should incorporate empirically supported interventions such as CBT and SSRIs as detailed earlier
No specific SSRI is recommended, and there is no apparent difference in efficacy for anxiety disorders among the SSRIs
SNRIs may be an alternative approach.

Obsessive-Compulsive Disorder in Children and Adolescents

DESCRIPTION OF THE DISORDER

OCD is a common psychiatric illness with lifetime prevalence in pediatric populations ranging from 1% to 3%. Persistence of OCD meeting the full diagnostic criteria after initial diagnosis, however, was observed in only 41% at follow-up (265). Compulsions are more common than obsessions, and predictive factors for persistence include an earlier age of onset and inpatient status. Poor initial treatment response seems to signal an increased risk for OCD persistence and greater severity. Given the primary role of SSRIs in the management of pediatric OCD, as well as recent concerns about their safety, these data will be carefully summarized.

TREATMENT OF OCD

Selective Serotonin Reuptake Inhibitors

The evidence for the treatment of pediatric OCD includes more than 19 studies representing more than 1,000 children and adolescents. These trials support the short- and medium-term efficacy of serotonergic medications (266,267), including

Fluoxetine (268,269)
Fluvoxamine (270)
Paroxetine (271,272,267)
Sertraline (273,274,275,276 and 277)

Five studies also support the long-term efficacy (>12 months) of SSRIs for OCD in youth (i.e., continuing benefit without loss of efficacy). The most recent of these long-term studies reported on a 12-month extension phase with sertraline. In this trial, Wagner et al. (277) found that the Children’s Yale-Brown Obsessive Compulsive Scale (CY-BOCS) scores continued to improve throughout the duration of treatment. At end point, 53% of children and 41% of adolescents met criteria for full remission (i.e., CY-BOCS ≤8), while 22% of children and 28% of adolescents met criteria for partial remission (i.e., CY-BOCS score of 9 to 15, considered minimally impaired). These data indicate that about 70% of youth realized substantial benefit from long-term treatment.

Table 14-5 presents the overall response rates, defined by ≥25% reduction in the CY-BOCS, which are about 50% for each SSRI (267,268,269 and 270,272,273 and 274,279,280,281,282,283 and 284).

TABLE 14-5 RANDOMIZED PHARMACOTHERAPY TRIALS IN PEDIATRIC OCD

Length of Tx (weeks)

Dose Mean mg/day (Range)

% Completers

Mean Age in Years (Range)

% Responders

Author (year)

Drug

N (Drug/PBO)

Drug

PBO

Drug

PBO

Measure

Drug

PBO

Liebowitz (2002) (268)

Fluoxetine

43(21/22)

65.5(20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79 and 80)

13.0

12.3

CY-BOCS NIMH-GOCS CGI-S^o

NR^a

(6,7,8,9,10,11,12,13,14,15,16,17 and 18)

Geller (2002) (266)

Paroxetine

203(98/105)

23.0(10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49 and 50)

11.3

CY-BOCS

65^b

(7,8,9,10,11,12,13,14,15,16 and 17)

Geller (2003) (267)

Paroxetine

193(95/98)

32.2(10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59 and 60)

11.8

11.6

CY-BOCS

30^b

(8,9,10,11,12,13,14,15,16 and 17)

Geller (2001) (259)

Fluoxetine

103(71/32)

24.6(20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59 and 60)

11.4

CY-BOCS NIMH-GOCS^b CGI-S^b

49^a

(7,8,9,10,11,12,13,14,15,16 and 17)

Riddle (2001) (270)

Fluvoxamine

120(57/63)

165(50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199 and 200)

13.4

12.7

CY-BOCS NIMH-GOCS^a CGI-S^a

42^a

(8,9,10,11,12,13,14,15,16 and 17)

March (1998) (274)

Sertraline

187(92/95)

167(25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199 and 200)

12.6¹

CY-BOCS NIMH-GOCS^a CGI-S^a

53^b

(6,7,8,9,10,11,12,13,14,15,16 and 17)

Riddle (1992) (279)

Fluoxetine

13(7/6)

20 (20)^c

11.8¹

CY-BOCS CGI-S^bLOI-CV

(8,9,10,11,12,13,14 and 15)

DeVeaugh-Geiss (1992) (280)

Clomipramine

60(31/29)

NR (75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199 and 200)

14.5

14.0

CY-BOCS NIMH-GOCS^a

50^a

(10,11,12,13,14,15,16 and 17)

Leonard^d (1991) (281)

Clomipramine Desipramine

11 (NA) 9(NA)

123(50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199,200,201,202,203,204,205,206,207,208,209,210,211,212,213,214,215,216,217,218,219,220,221,222,223,224,225,226,227,228,229,230,231,232,233,234,235,236,237,238,239,240,241,242,243,244,245,246,247,248,249 and 250)

100 90

14.7^e

NIMH-GOCS^a LOI-CV

(8,9,10,11,12,13,14,15,16,17,18 and 19)

March (1990) (282)

Clomipramine

16(8/8)

190(50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199 and 200)

100

15.0^e

CY-BOCS NIMH-GOCS^a

(10,11,12,13,14,15,16,17 and 18)

Leonard^d(1989) (283)

Clomipramine

23 (NA)

50(25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199,200,201,202,203,204,205,206,207,208,209,210,211,212,213,214,215,216,217,218,219,220,221,222,223,224,225,226,227,228,229,230,231,232,233,234,235,236,237,238,239,240,241,242,243,244,245,246,247,248,249 and 250)

92^e

13.9^e

NIMH-GOCS^b LOI-CV

(7,8,9,10,11,12,13,14,15,16,17,18 and 19)

Desipramine

25 (NA)

153(25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199,200,201,202,203,204,205,206,207,208,209,210,211,212,213,214,215,216,217,218,219,220,221,222,223,224,225,226,227,228,229,230,231,232,233,234,235,236,237,238,239,240,241,242,243,244,245,246,247,248,249 and 250)

92^e

Flament (1985) (284)

Clomipramine

38(19/19)

141(50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79,80,81,82,83,84,85,86,87,88,89,90,91,92,93,94,95,96,97,98,99,100,101,102,103,104,105,106,107,108,109,110,111,112,113,114,115,116,117,118,119,120,121,122,123,124,125,126,127,128,129,130,131,132,133,134,135,136,137,138,139,140,141,142,143,144,145,146,147,148,149,150,151,152,153,154,155,156,157,158,159,160,161,162,163,164,165,166,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199 and 200)

91^e

14.5^e

NIMH-GOCS³ LOI-CV^a

(10,11,12,13,14,15,16,17 and 18)

^ap <0.05.

^bp <0.01.

^cFixed dose.

^dDesipramine used as comparator.

^ePooled data provided for drug and placebo.

Act, Active; CGI-S, Clinical Global Impressions scale-Severity; CY-BOCS, Children’s Yale-Brown Obsessive Compulsive Scale; LOI-CV, Leyton Obsessional Inventory-Child Version; NIMH-GOCS, National Institute of Mental Health Global Obsessive-Compulsive Scale; NR, not reported; NA, not applicable; PBO, placebo.

CMI data supplied by the neuroscience department at Novartis Pharmaceuticals Corporation; Responders defined as >25% reduction on CY-BOCS.

Only gold members can continue reading. Log In or Register to continue

Stay updated, free articles. Join our Telegram channel

Tags: Principles and Practice of Psychopharmacotherapy

Aug 27, 2016 | Posted by drzezo in PHARMACY | Comments Off

Basicmedical Key

Fastest Basicmedical Insight Engine

Psychopharmacotherapy in Early Life

Like this:

Related

Stay updated, free articles. Join our Telegram channel

Full access? Get Clinical Tree

Basicmedical Key

Fastest Basicmedical Insight Engine

Psychopharmacotherapy in Early Life

Share this:

Like this:

Related

Related posts:

Stay updated, free articles. Join our Telegram channel

Full access? Get Clinical Tree