Chapter 93. Affective Disorders and Suicide

Some experiences of sadness, grief, or depressed mood arise during the course of most children’s lives. Divorces, the death of a grandparent, the departure of a close friend, or a failed hope—all are common and upsetting. However, there has been increasing recognition that some children and adolescents suffer from serious and pervasive disorders related to their mood and that these disorders are associated with significant morbidity (ie, impairment in psychosocial function, low self-esteem) and mortality (ie, intended suicide or “accidental” death secondary to an impulsive behavior such as driving while intoxicated). Consequently, the critical clinical task is to differentiate children and adolescents with serious disorders from the larger group of children and adolescents who have some symptoms of sadness or grief but who are not clinically depressed. Since the diagnosis and risk assessment depends on emotional information gleaned from an interview, it can be extremely difficult for a clinician to differentiate between those children and adolescents with serious mood disorder and those who are profoundly and acutely upset in response to a combination of adverse circumstances.

Because our understanding of the pathophysiology of emotional dysregulation is just emerging, diagnosis of major depressive disorders relies primarily on clinical history and observed mental status. There are few reliable tests to assist or to confirm a diagnosis. Current knowledge is based on the best available data from clinical studies, epidemiological research, and careful observations of longitudinal course.

Currently, depressive illnesses are classified into three broad categories: depressive disorders, dysthymic disorder, and bipolar affective disorder or cyclothymic disorder (see Table 93-1). Estimates of the point prevalence of major depression are approximately 1.5% to 2.5% in prepubertal children, increasing to 3% to 8% during adolescence. Bipolar affective illness is estimated at a point prevalence of 0.2% to 0.4% among prepubertal children, increasing to approximately 1% among adolescents. Depressive illnesses do not show gender differences in prepubertal children; however, at puberty, there is a significant increase in major depression among females, resulting in a female-to-male ratio of 3:1 during adolescence. Some research suggests that early onset of puberty increases the risk of depression in girls.1 Bipolar disorders are equally common in males and females throughout the life cycle.

Anxiety disorders, attention deficit hyperactivity disorder (ADHD), substance abuse, and conduct disorder can be concurrent with mood disorders. Some may show more genetic transmission (eg, ADHD). However, other comorbid states may represent bidirectional causality. Evidence from one longitudinal study suggests that anxiety disorders often antedate depressive symptomatology, whereas conduct symptoms generally follow an affective illness.2

Most etiologic models of early onset depression focus on the interactions between individual vulnerabilities and adverse life events. The central question of why some children grow up amidst chronic adverse conditions with no signs of significant mood disturbances, while others develop severe affective illnesses following relatively mild difficulties remains unanswered. Emotional trauma (eg, from physical and sexual abuse, loss of a parent, loss of a sibling or close friend) has been associated with the early onset of depressive disorders. Increased risk for early onset depression is associated with greater familial loading for depression, having a parent who developed depression at an early age (highly relevant to the recognition of postpartum depression), family history of either bipolar affective illness or recurrent unipolar depression, and major affective illness present in three generations. However, the single most important risk factor associated with developing an early depressive illness is having at least one depressed parent, with quadruple the risk for those with two depressed parents. Parental depression contributes to illness in the children through both genetic and shared environmental sources of transmission, including disruption of the parental role, decreased family support, and increased parent–child discord. Parent–child discord was shown to be significantly worse in families of depressed children when compared with those of nondepressed children in a control group. Additionally, emerging data suggest that toddlers and infants are at increased psychiatric risk secondary to relatively higher exposures to noxious and stressful stimuli such as abuse or neglect.3

Twin studies and familial patterns of depression strongly support the arguments for a genetic basis of depression. Concordance rates for depression are greater than 2:1 in monozygotic versus dizygotic pairs. A longitudinal study in New Zealand demonstrated that the interplay of life stress and a single allele difference in the serotonin transporter gene results in the development of depression in susceptible individuals but these conclusions are now being questioned.4

Child and adolescent depression has been associated with the use of medications such as glucocorticoids, immunosuppressives, antiacne treatments, antimalarials, and antivirals. Chronic illnesses such as cancer, cystic fibrosis, epilepsy, juvenile diabetes, sickle cell anemia, and organ transplantation may also predispose the child and adolescent to developing an adjustment or mood disorder.

The relationship between mood disorders and neuroendocrine changes in children and adolescents is still not clear. It may be that abnormalities in the hypothalamic-pituitary axis (HPT; demonstrated by dexamethasone-suppression test) and blunted response to growth hormone (GH) increase the risk for depressive disorders. Some of these abnormalities have been found in nondepressed children who have very high rates of affective illnesses on both sides of the family, suggesting that these changes may represent a vulnerability trait for major depressive disorder.5

Functional neuroimaging in children and adolescents is still in its infancy, and the risk of exposing children to radiation limits further study. However, early studies corroborate results found in similar studies of adults: depressed subjects have a smaller prefrontal cortex and basal ganglia. This preliminary work suggests structural differences may be genetically transmitted and, like neuroendocrine changes, serve as trait marker for depression.6

Depressed mood in children and adolescents can present as sadness, irritability, or boredom. Depressed children also are likely to have somatic complaints (eg, abdominal pain, headaches). A small number of very ill children present with psychotic features related to their mood disorder (ie, delusions of worthlessness, hopelessness, sin, or guilt; self-deprecatory or auditory hallucinations; and paranoid ideation).

An episode of mania is usually characterized by euphoria or grandiosity (ie, an unrealistic sense of being grand, powerful, or famous) and indicates a bipolar affective disorder (BPAD). Such episodes can be associated with anger and irritability as well. Mania can be differentiated from attention deficit hyperactivity disorder, because mania is much more likely to present with increased energy, increased sexuality, euphoria, and grandiosity. Clinically, it can be quite difficult to diagnose mania or hypomania in very young children; however, in its most severe forms, children and adolescents with mania display bizarre behaviors and irrational ideas (such as the belief that they can fly) that are not present in other syndromes. Like adults, adolescents in manic episodes show periods of prolonged excitation, euphoria, rapid speech, grandiosity, sensation seeking, and promiscuity. When children and adolescents with bipolar disorder are depressed, they often present as anergic (ie, low energy and slow), hypersomnic, and psychotic.

The diagnosis of juvenile BPAD has garnered significant attention and has stirred up much controversy over the last decade. Some researchers believe that intense bouts of very aggressive behavior in response to relatively minor provocations (eg, absence of a favorite snack) are suggestive of early mania. Indeed, studies have shown that when children with these behavioral patterns undergo standardized assessments, they often meet DSM-IV criteria for mania. Additionally, there is a much higher incidence of bipolar disorder in first-degree relatives of these children, and current prospective studies show that a significant percentage of these children eventually look more classically manic as they age.7,8

Much of the controversy stems from the seeming absence of euphoric and grandiose behavior in juvenile mania. Additionally, because not all of these children show manic symptoms as they age, the question remains as to whether these explosive episodes represent a true bipolar diathesis. To this end, terms such as broad spectrum bipolar disorder have emerged that allow clinicians to differentiate children with these presentations from those who fall within the more narrowly defined, classic descriptions of mania. Finally, pharmacological studies have shown that many of these “broad spectrum” bipolar children respond more quickly to atypical antipsychotic agents than they do to more classic mood stabilizers. This increased use of antipsychotics in juvenile populations, with some attendant success accompanied by significant side effects, has also contributed to clinicians’ discomfort with this prescribing practice and thus with this diagnosis.7-9

For the pediatrician, it is important to remember that depression or very irritable or explosive behavior in a child may actually represent an early presentation of BPAD that might require child psychiatric consultation. The use of antidepressants in this population is associated with potential worsening of the presumed mania, a risk that alone justifies existing recommendations from both the American Academy of Pediatrics (AAP) and the American Academy for Child and Adolescent Psychiatry (AACAP) for close monitoring of all children who take antidepressants.

Any persistent mood disturbance that is associated with functional impairment (ie, deterioration of school or social function) should raise the suspicion of a major depressive disorder. Children are usually accurate reporters of symptoms that reflect their internal state, including depressed mood, guilt, worthlessness, and suicidal thoughts. Parents most often note externally validated symptoms such as irritability, decline in school performance, and withdrawal from social and other pleasurable activities. Affective illness in children often is associated with sad mood and other pervasive, negative moods such as being grouchy, mad, or bored.

The primary goals of clinical assessment are to determine (1) the presence of symptoms of mood disorder, (2) the extent that these symptoms are interfering with school or social functions, and (3) the potential self-destructive behavior (suicide, risk-taking, substance abuse, etc). Definitive diagnosis of child and adolescent depression usually is best achieved by taking a careful history from the child and using parent and collateral sources; this can be supplemented by structured or semistructured diagnostic interviews. Self-report screens for depressive disorder include the Children’s Depression Inventory (CDI) for 8- to 13-year-olds, and the Beck Depression Inventory, Reynolds Adolescent Depression Screen, and the Mood and Feeling Questionnaire for older adolescents. These well-validated questionnaires list depressive symptoms and ask the child or adolescent to rate the severity of such symptoms.

The most common mood problems among children and adolescents are depressive symptoms that are associated with an adjustment disorder secondary to specific stressors (eg, family discord, school failure, or difficulty with peers). Often, such difficulties will respond to a few sessions of supportive counseling, with the pediatrician and adolescent working together to focus on identifying the problem and developing strategies that ease the stress. If family discord is prominent, problem-solving interventions should involve the entire family. Family psycho-education is designed to improve compliance, to reduce tensions of living with an affectively ill individual, and to sensitize the family to early signs of recurrence of the disorder.

Children and adolescents with a severe mood disorder are probably best managed by a child psychiatrist. Three approaches have been empirically validated: psychotropic medication, specifically selective serotonin reuptake inhibitors (SSRIs); cognitive behavioral therapy (CBT); and interpersonal therapy (IPT).

Pharmacotherapy

Pharmacotherapy with SSRIs has been and continues to be the first-line treatment for early onset depression. At present, fluoxetine is the only SSRI that is approved by the Federal Drug Administration (FDA) for treating children and adolescents with mood disorders, although some studies have shown other SSRIs such as sertraline, paroxetine, and citalopram to be efficacious. In 2006, an FDA meta-analysis of children and adolescents taking SSRIs for depressive symptoms determined that patients taking the drug had a higher risk for suicidal tendencies than those taking the placebo. However, a subsequent meta-analysis, which included additional studies not used by the FDA, found a 14:1 ratio between those children who benefited from SSRI treatment and those who became suicidal during treatment. The temporal relationship between the recent increase in adolescent suicide attempts and clinicians’ decrease in prescribing SSRIs suggests that SSRI treatment of depression might be lifesaving.10

Suicide attempts and suicidal ideation are increased in many patients who suffer from depression. Patients who are at greatest risk for morbidity and mortality are those who are usually treated pharmacologically, which may explain some of the association between antidepressant medications and suicidal behavior in young patients. However, due to the black box warning, in the United States it is prudent to use fluoxetine as a first choice for medication management of depression and to follow a protocol of weekly tracking until there is clear positive improvement in mood.11

Antidepressants carry other risks, as do all medications. The risk of prescribing antidepressants needs to be compared with the risk of allowing significant affective disorders to go untreated. The AAP, the AACAP, and the FDA have advised careful monitoring of all patients who are given antidepressants.12 At the very least, clinicians should inform patients and parents that any increase in agitation associated with antidepressants, especially that which occurs during the first week of treatment, should prompt an immediate call to the prescribing physician. In these instances, the antidepressant should be stopped because of the risk for secondary mania, and a child psychiatrist should be consulted.

Naturalistic studies indicate that untreated major depressive disorders in children and adolescents can last as long as a year. Between 25% and 40% of children will fail to respond to psychotherapeutic interventions and may require pharmacotherapy instead or in addition to psychotherapy. Treatment with psychotropic medication should be continued until the patient’s symptoms have remitted; however, if there has been little improvement after 3 to 4 months, the treatment approach should be augmented or changed. After remission is achieved, therapy should be continued for another 6 to 9 months in order to prevent recurrence. Despite treatment, the risk of relapse and recurrence within 5 years is extremely high.13 Bipolar disorder requires prophylaxis with a mood-stabilizing agent—either lithium or an anticonvulsant, such as valproic acid or carbamazepine, or an atypical antipsychotic—prior to introducing a selective serotonin reuptake inhibitor.

Cognitive-Behavior and Interpersonal Therapy

Clinical trials have demonstrated the efficacy of cognitive-behavior therapy (CBT) and interpersonal therapy (IPT) for child and adolescent depression.14 In CBT, treatment focuses on correcting maladaptive and negative thinking patterns that predispose to and reinforce depression, whereas IPT focuses on improving interpersonal interactions that may be related to depressive experiences. Psychotherapy should be aimed at ameliorating interpersonal and social deficits that are associated with depressive symptoms.

There is also evidence that suggests that these approaches to the patient hold significant therapeutic value in treating pediatric depression and that the extent to which patients feel trusted and understood is the best predictor of a particular intervention’s efficacy. Thus, referral to a skilled therapist might be just as helpful as a referral to a therapist who has specific training in CBT or IPT.15 However, recommendations for CBT or IPT have been frustrating for referring physicians, as clinicians skilled in performing these kinds of interventions are not as numerous or available as are clinicians who practice more “open-ended” therapies by asking questions that encourage further exploration.

Most patients who are affectively ill can be managed outside the hospital. Inpatient psychiatric referrals should be considered for those who are psychotic, acutely suicidal, manic, abusing substances, or resistant to outpatient treatment.

In the United States, suicide is the third leading cause of death among adolescents ages 10 to 24 years. According to data from the Centers for Disease Control (CDC),16 the suicide rate in 2003 to 2004 was 7.32 per 100,000. This represents an increasing number in three groups: females age 10 to 14 years and 15 to 19 years, and males age 15 to 19 years. The most common cause of death in females was hanging or suffocation and in males was a gunshot wound. Prepubertal children, those ages 5 to 14 years, have a much lower rate at 0.5 per 100,000. However, those prepubertal children who suffer from major depression are at a high risk for suicide during adolescence or young adulthood. Approximately 50% of completed suicides occur following previous suicidal threats or attempts.

In addition to successful suicides, suicidal behavior has also become increasingly common. Only a fraction of adolescent suicide attempts ever come to medical attention, but according to the Youth Risk Behavior Surveillance System (YRBSS) of 2005—a survey given to high school students across the United States—16.9% of adolescents had considered suicide and 8.4% had attempted it within the year. Given these alarming statistics, identifying children and adolescents at risk for suicidal ideation or attempts is crucial. The Teen-Screen is a two-part survey, a self-report questionnaire and an open-ended interview, with a mental health professional, which was administered to 55,000 high school students in 2005. Participation was voluntary but required the consent of the teenager’s parent or guardian. If the student demonstrated psychiatric symptoms on both the questionnaire and the clinical interview, he or she was then referred for further evaluation. The Teen-Screen has been criticized for presumed violation of parental rights and for its high false-positive rates. Supporters contend that although the survey has a high sensitivity and a low specificity, the risk of missing a truly suicidal teenager is greater than the risk of misdiagnosing an unaffected one.

Common risk factors for suicidal behavior are listed in Table 93-2. There is also growing evidence that a homosexual or bisexual orientation is a risk factor for suicidal behavior in males; this is possibly secondary to familial or social stigma. Furthermore, any adolescent who engages in high-risk behaviors such as drinking, smoking, using drugs, or having unprotected sex is also at increased risk of depressive disorder, suicide ideation, and attempts. Recent evidence suggests that substance abuse (particularly binge drinking and marijuana use in boys) may predispose teenagers to depression, rather than representing a form of self-medication for a mood disorder.17

A child or adolescent who is suspected of being at risk should be directly questioned concerning suicidal ideation, moving from nonspecific to specific questions if the answers are positive. Examples of this line of questioning are included in Table 93-3.

Pediatricians should avoid broad promises of confidentiality that they would likely break in order to protect and properly treat the child or adolescent. The patient’s parents must be given some feedback about the assessment, because parental involvement is usually a determining factor in the child’s or adolescent’s compliance with treatment. It is critical to ensure that potential means of suicide, particularly firearms, are removed from the home and that potentially lethal medications are well secured.

“No suicide” contracts, in which children or adolescents agree not to harm themselves and that identify ways in which they can help themselves if they experience suicidal ideation, are not helpful in preventing attempts. Children or adolescents who are actively suicidal are in no position to understand the contract’s ramifications or to abide by them as they face unforeseen stressors. In addition, parents or guardians who are caring for the child or adolescent might be tempted to relax their level of vigilance once such a contract is signed.

With children and adolescents who express active suicidal ideation, who have attempted suicide, or who are acting recklessly, it is usually necessary to admit them to a psychiatric inpatient unit for containment; to seek psychotherapeutic care (medication or therapy or both); and, ultimately, to effect transition to more intensive outpatient care.

Adolescent suicide attempts have the potential to become “contagious,” usually defined as three or more suicides in the same community that occur within a limited period, although there is controversy as to whether one suicide actually precipitates others. The suicide of any young person is a tragedy of momentous proportions that stuns and saddens both fellow students and involved adults. Many school systems have developed protocols for managing such events; these usually involve school assemblies, consultation with local mental health providers, and easy access for students and faculty to identified personnel within the high school. Media coverage (television, Internet, radio, or newspaper) can be detrimental, as it sometimes serves as a catalyst to action in vulnerable adolescents with previously latent ideation. In such circumstances, the need for information must be balanced with the risk for imitation.

Given the prevalence of both normal sadness and serious mood disorders in the life of a child, virtually every pediatrician will undoubtedly be called upon to recognize, assess, and contribute to the treatment of depression, risk-taking behavior, and substance abuse.