Chapter 18. Mood Disorders

Essentials of Diagnosis

DSM-IV-TR Diagnostic Criteria

Major Depressive Episode

1. Five (or more) of the following symptoms have been present during the same 2-week period and represent a change from previous functioning; at least one of the symptoms is either (1) depressed mood or (2) loss of interest or pleasure.

   Note: Do not include symptoms that are clearly due to a general medical condition, or mood-incongruent delusions and hallucinations.

   1. depressed mood

   2. markedly diminished interest or pleasure

   3. significant weight loss when not dieting or weight gain, or decrease or increase in appetite

   4. insomnia or hypersomnia

   5. psychomotor agitation or retardation

   6. fatigue or loss of energy

   7. feelings of worthlessness or excessive or inappropriate guilt

   8. diminished ability to think or concentrate, or indecisiveness

   9. recurrent thoughts of death or suicide

2. The symptoms do not meet criteria for a Mixed Episode.

3. The symptoms cause significant distress or impairment in social, occupational, or other important areas of functioning.

4. The symptoms are not due to the direct physiological effects of a substance (e.g., a drug of abuse, a medication) or a general medical condition (e.g., hypothyroidism).

5. The symptoms are not better accounted for by Bereavement, i.e., after the loss of a loved one, the symptoms persist for longer than 2 months and are characterized by marked functional impairment, morbid preoccupation with worthlessness, suicidal ideation, psychotic symptoms, or psychomotor retardation.

(Adapted, with permission, from the Diagnostic and Statistical Manual of Mental Disorders, 4th edn., Text Revision. Washington DC: American Psychiatric Association, 2000.)

General Considerations

Epidemiology

The separation into bipolar and non-bipolar disorder has proved clinically and diagnostically useful. It is supported by family studies, twin studies, and biological studies. It is supported further by differential clinical responses to treatment and differential disease onsets and outcomes. To these factors, we can add the epidemiologic risk factors detailed in Table 18–1.

Symptoms and disorders of the depression spectrum are rather common. Lifetime prevalence rates for significant depressive symptoms are 13–20% and for major depressive disorder 3.7–6.7%. Major depressive disorder is about two to three times as common in adolescent and adult females as in adolescent and adult males. In prepubertal children, boys and girls are affected equally. Rates in women and men are highest in the 25–44-year-old age group.

Etiology

Despite intensive attempts to establish its etiologic or pathophysiologic basis, the precise cause of major depressive disorder is not known. There is consensus that multiple etiologic factors—genetic, biochemical, psychodynamic, and socioenvironmental—may interact in complex ways and that the modern-day understanding of depressive disorder requires an understanding of the interrelationships among these factors.

Life Events

Recent evidence confirms that crucial life events, particularly the death or loss of a loved one, can precede the onset of depression. However, such losses precede only a small (though substantial) number of cases of depression. Fewer than 20% of individuals experiencing losses become clinically depressed. Although other major life events may occur prior to the onset of depression, many patients become depressed with little or no apparent provocation. These observations argue strongly for a predisposing factor, probably genetic, developmental, or temperamental in nature.

Biological Theories

Neurotransmitters

Associations between mood and monoamines (i.e., norepinephrine, serotonin, and dopamine) were first indicated serendipitously by the mood-altering effects of isoniazid (used initially for the treatment of tuberculosis) and later by reports that isoniazid (a monoamine oxidase inhibitor [MAOI]) affects monoamine concentrations in the brains of laboratory animals. We now know that all clinically-effective antidepressants affect postsynaptic signaling of serotonin, norepinephrine, or both at the postsynaptic membrane. This action has led to the hypothesis that depression is caused by a neurotransmitter deficiency and that antidepressants exert their clinical effect by treating this imbalance.

In the late 1970s, emphasis shifted from acute presynaptic to delayed postsynaptic receptor—mediated events after it was shown that most chronic antidepressant treatments (including pharmacotherapy and electroconvulsive therapy [ECT]) cause subsensitivity of the norepinephrine receptor—coupled adenylate cyclase system in brain. This desensitization of norepinephrine receptor systems was linked to a decrease in the density of β–adrenoceptors. More importantly, it paralleled the delayed onset of action common to all antidepressants. More recent concepts have evolved from the original deficiency hypotheses, emphasizing the integration of multiple intracellular signals that regulate neuronal response (i.e., changes in G protein, cyclic adenosine monophosphate, or protein kinase and the induction of gene transcription). These cellular mechanisms are thought to ultimately affect the expression of specific genes. Therefore, abnormalities of intracellular signal transduction and/or gene expression are now thought to underlie the physiology of depression. Other neurotransmitters (e.g., acetylcholine, gamma amino butyric acid, melatonin, glycine, histamine), hormones (e.g., thyroid and adrenal hormones), and neuropeptides (e.g., corticotropin-releasing hormone, endorphins, enkephalins, vasopressin, cholecystokinin, substance P) may play significant roles in the modulation of mood.

Neuroendocrine Factors

Emotional trauma sometimes immediately precedes the onset of depression. Emotional trauma can also precede the onset of endocrine disorders such as hyperthyroidism and Cushing's disease, both of which are commonly associated with psychological disturbance, most commonly in mood and cognition. When endocrine changes are associated with psychological disturbance it is often unclear whether such changes are precipitants, perpetuating influences, or secondary effects.

The two endocrine systems most extensively studied in psychiatry are the hypothalamic-pituitary-adrenal (HPA) axis and the hypothalamic-pituitary-thyroid (HPT) axis. About half of patients with major depression exhibit cortisol hypersecretion that returns to normal once the depression is cured. The evidence of cortisol hypersecretion includes increased adrenocorticotropin (ACTH) secretion, increased plasma concentration and urinary excretion of cortisol and its metabolites, alterations in the normal circadian rhythm of cortisol secretion, relative resistance to glucocorticoid negative feedback inhibition of ACTH secretion, and blunted ACTH responses to corticotropin-releasing hormone. However, not all depressed patients exhibit evidence of significant hypercortisolism. Several studies have attempted to document the psychological effects of manipulating plasma glucocorticoid concentrations in depressed patients. They reported antidepressant effects in some depressed patients after administration of such antiglucorticoid medications as aminoglutethimide, metyrapone, and ketokonazole.

Human studies have also demonstrated a profound effect of thyroid hormones on brain development, maturation, and connectivity. The effects of thyroid hormones on mature brain function, as they pertain to mood, are less marked. The most common effects are as follows: (1) Depression and cognitive decline are the most frequently observed psychiatric symptoms in patients who have adult hypothyroidism, (2) A small dose of thyroid hormone, preferably triiodothyronine (T3), will accelerate the therapeutic effect of various antidepressants, particularly in women, and can convert antidepressant nonresponders into responders in both sexes, (3) Most longitudinal studies have revealed dynamic reductions in serum thyroxine (T4) concentrations in depressed patients during a wide range of somatic treatments, including various antidepressants, lithium, sleep deprivation, and ECT, (4) Administration of thyrotropin-releasing hormone (TRH) may induce an increased sense of well-being and relaxation in normal subjects and in patients with neurologic and psychiatric disease, especially depression, and (5) Although overt thyroid disease is rare in major depression, subtle forms of thyroid dysfunction are common—for example, absence or flattening of the diurnal thyroid-stimulating hormone (TSH) curve, often caused by a reduction in the nocturnal TSH surge; a blunted TSH response after administration of TRH (discussed in greater detail later in this section); and subclinical hypothyroidism or positive antithyroid antibodies.

Early Life Stress

In recent years, an increasing number of studies have evaluated both the immediate and long-term neurobiological effects of early childhood trauma (e.g., physical or emotional abuse, neglect, or parental loss). Studies of traumatized persons or animal models of early life stress suggest long-lasting effects on neuroendocrine, psychophysiological, and neurochemical systems. Together, these effects may present the biological basis of an enhanced risk for psychopathology, including depression. In addition to having added childhood trauma to the existing list of risk factors for depression, these studies have created a burgeoning awareness and consensus that the high number of children who are exposed in our society to early trauma is unacceptable. Hopefully, further studies may elucidate the many factors that determine individual vulnerability or resilience to the neurobiological effects of early trauma and help to prevent their deleterious neurobiological and psychopathological consequences.

Psychosocial Theories

Psychoanalytic and Psychodynamic Models

In the early 20th century, psychoanalytic interpretations of mental illness were prominent. Karl Abraham (1911) wrote the first important psychoanalytic paper on depression, stating that depression is unconsciously motivated and the result of repressed sexual and aggressive drives. Concurrent writings of Sigmund Freud suggested that depression and, to a lesser extent, mania were understood as precipitated by loss and manifested by regressions to anal and oral phases of libidinal development. These regressions in adult life resulted from an unfortunate combination of predisposition (constitutional overaccentuation of oral eroticism) and critical childhood disappointments. Armed with Freud's concepts on the role of introjection in normal mourning and in melancholia, Abraham wrote of the “severe conflict of ambivalent feelings from which [the patient] can only escape by turning against himself the hostility he originally felt towards his object.”

The precepts that depression is the result of a loss and that it symptomatically represents “anger turned against the self” are part of the enduring legacy of classic psychoanalytic thinking. Freud believed that the real, threatened, or imagined loss of a narcissistic object choice (meaning that the individual's love of the object was equivalent to love of the self) would trigger a withdrawal of libido away from the object and back into the self through introjection of the ambivalently cathected object. The patient then would attack himself (depressive symptoms) as though for misdeeds that were the doing of the lost object. In other words, the depressed patient experiences a loss as a narcissistic wound; suicide becomes an unconscious attempt to destroy the now hated object that dwells in the patient's ego by means of introjection. Like Abraham, Freud also emphasized the importance of somatic factors in predisposing the individual to depression and in the clinical picture itself. Diurnal variation of mood, for example, was beyond psychological explanation. Later psychoanalytic writers (e.g., M. Klein, D. W. Winicott, E. Bibring) sought to expand on Freud's thinking, incorporating further developments in ego-psychology, object relations theory, and self-psychology. Although widely held, traditional psychodynamic theories have not held up to careful study. Moreover, psychoanalysis or most forms of psychodynamic therapy either have not been formally tested or have not been shown to be more effective than simple interpersonal contact.

Behavioral Models

In the mid-1900s, three models based on behavioral theory emerged (see Chapter 10). Peter Lewinsohn showed that depression can be caused by inadequate or insufficient positive reinforcement. In everyday life, this can occur in two main ways: (1) if an environment lacks positive reinforcement (e.g., through individual or mass unemployment) or (2) if the person is not able to take advantage of reinforcement (e.g., through isolation induced by poor social skills). Inadequate positive reinforcement may lead to a self-perpetuating cycle consisting of dysphoria, a reduction in behaviors that would normally obtain the reinforcement, lowered self-esteem and increased hopelessness, and increased isolation.

Martin Seligman developed the theory of learned helplessness as he was searching for an animal model of depression. Laboratory animals given random shocks from which they cannot escape develop apathy to any stimulus. Generalizing this observation to humans, Seligman's theory suggests that depression can result from situations in which a person has lost (actual or imagined) control over negative life events.

The cognitive-behavioral model of depression developed by Aaron Beck suggests that depression develops when the patient cognitively misinterprets life events. The conceptual core of this model consists of the cognitive triad of depression: (1) a negative self-view (i.e., “things are bad because I am bad”), (2) a negative interpretation of experience (i.e., “everything has always been bad”), and (3) a negative view of the future (i.e., “everything will always be bad”). It is a basic tenet of this theory that a depressed person interprets the world through depressive schemata that distort experiences in a negative direction. Typical cognitive distortions include arbitrary inference (in which the person assumes a negative event was caused by himself or herself), selective abstraction (in which the person focuses on the negative element in an otherwise positive set of information), magnification and minimization (in which the person overemphasizes negatives and underemphasizes positives), and inexact labeling (in which the person gives a distorted label to an event and then reacts to the label rather than to the event).

Genetics

Mood disorders are familial, but the exact mode of transmission is not understood (see Chapter 3).

Clinical Findings

Signs & Symptoms

Major Depressive Episode

The cardinal feature of a major depressive episode is a depressed mood or the loss of interest or pleasure (see Depressed Mood and Anhedonia below) that predominates for at least 2 weeks and causes significant distress or impairment in the individual's social, occupational, or other important areas of functioning. During this time, the individual must also exhibit at least four additional symptoms (i.e., other than depressed mood or anhedonia), drawn from the following common features of depression:

Depressed Mood

Depressed mood is the most characteristic symptom, occurring in over 90% of patients. The patient usually describes himself or herself as feeling sad, low, empty, hopeless, gloomy, or down in the dumps. The quality of mood is likely to be portrayed as different from a normal sense of sadness or grief. The physician often observes changes in the patient's posture, speech, facies (e.g., a melancholic expression known as facies melancholica), dress, and grooming consistent with the patient's self-report. Many depressed patients state that they are unable to cry, whereas others report frequent weeping spells that occur without significant precipitants.

A small percentage of patients do not report a depressed mood, sometimes referred to as masked depression. These patients are usually brought to their physician by the family members/coworkers who have noticed the patients’ social withdrawal or decreased activity. Patients may associate depression with feelings of sadness. However, depression as often involves emotional numbness or lack of positive reactivity. Similarly, some children and adolescents do not exhibit a sad demeanor, presenting instead as irritable.

Anhedonia

An inability to enjoy usual activities is almost universal among depressed patients. The patient or his or her family may report markedly diminished interest in all, or almost all, activities previously enjoyed such as sex, hobbies, and daily routines.

Change in Appetite

About 70% of patients observe a reduction in appetite with accompanying weight loss; only a minority of patients experience an increase in appetite, often associated with cravings for particular foods such as sweets.

Change in Sleep

About 80% of depressed patients complain of some type of sleep disturbance, the most common being insomnia. Insomnia is usually classified as initial (i.e., problems in falling asleep), middle (i.e., problems of staying asleep with frequent awakenings throughout the night), or late (i.e., early morning awakening). The most common and unpleasant form of sleep disturbance in major depressive disorder is late insomnia, with awakenings in the early morning (usually around 4–5 am) and significant worsening of depressive symptoms in the first part of the day. In contrast, initial insomnia is especially common in those with significant comorbid anxiety. Some patients complain of hypersomnia rather than insomnia; hypersomnia is common in atypical depression and seasonal affective disorder (SAD) and is often associated with hyperphagia.

Change in Body Activity

About one-half of depressed patients develop a slowing, or retardation, of their normal level of activity. They may exhibit a slowness in thinking, speaking, or body movement or a decrease in volume or content of speech, with long pauses before answering. In about 75% of depressed women and 50% of depressed men, anxiety is expressed in the form of psychomotor agitation, with pacing, an inability to sit still, and hand-wringing.

Loss of Energy

Almost all depressed patients report a significant loss of energy (anergia), unusual fatigue or tiredness, and a general lack of efficiency even in small or elementary tasks.

Feelings of Worthlessness and Excessive or Inappropriate Guilt

A depressed individual may experience a marked (and often unrealistic) decrease in self-esteem. In European cultures, well over half of depressed patients exhibit some guilt, ranging from a vague feeling that their current condition is the result of something they have done, to frank delusions and hallucinations of poverty or of having committed an unpardonable sin. In other cultures, shame or humiliation is experienced.

Indecisiveness or Decreased Concentration

About one-half of depressed patients complain of or exhibit a slowing of thought. They may feel that they are not able to think as well as before, that they cannot concentrate, or that they are easily distracted. Frequently they will doubt their ability to make good judgments and find themselves unable to make even small decisions. On formal psychological testing, the patient's accuracy is usually retained, but speed and performance are slow. In severe forms, called pseudodementia, particularly among the elderly, memory deficits may be mistaken for early signs of dementia. In contrast to dementia, pseudodementia usually reverses after treatment of the underlying depression. However, when cognitive symptoms are comorbid with depression, they may represent an emerging dementia that can still be present after the depression has resolved.

Suicidal Ideation

Many depressed individuals experience recurrent thoughts of death, ranging from transient feelings that others would be better off without them, to the actual planning and implementing of suicide.

Suicide is the eighth leading cause of death in the United States, accounting for more than 30,000 deaths each year. Major depression accounts for roughly 50% of suicides, and 15% of patients with depression eventually die by suicide. The risk of suicide is present throughout a depressive episode but is probably highest immediately after initiation of treatment and during the 6–9-month period following symptomatic recovery. Table 18–2 lists common predictors of suicide risk.

Alcohol and drug dependence account for roughly 25% of suicides, and psychosis is present in 10% of suicides. For each person who completes suicide, 8–10 people attempt suicide, and for every completed suicide, 18–20 attempts are made. Patients with a history of suicide attempts account for 50% of completed suicides. Medical illnesses are associated with as many as 35–40% of suicides, and with as many as 70% of suicides occurring in those over the age of 60 years. As much as 10% of general medical admissions result from failed suicide attempts.

Melancholic Subtype

The diagnosis of major depression with melancholic features is made when the patient has either loss of pleasure in all, or almost all, activities, or lack of reactivity to usually pleasurable stimuli, along with three or more of the following: (1) distinct quality of depressed mood (i.e., the mood is different than that experienced after the loss of a loved one); (2) depression regularly worse in the morning, (3) early morning awakening (at least 2 hours prior to the usual time of awakening), (4) marked psychomotor retardation or agitation, (5) significant anorexia or weight loss, and (6) excessive or inappropriate guilt. Other characteristics sometimes associated with melacholia include the absence of personality disturbance before first episode, the occurrence of one or more previous episodes followed by complete remission, and prior good response to specific and adequate somatic antidepressant therapy (e.g., ECT, tricyclic antidepressants [TCAs], MAOIs, lithium).

Atypical Depression

Although the term depression often calls to mind the pattern of symptoms associated with melancholia (especially insomnia and loss of appetite), some patients experience a reversed pattern – that is, symptoms that are opposite to those seen in melancholia. The Diagnostic and Statistical Manual of Mental Disorders, fourth edition (DSM-IV) diagnostic criteria for major depression with atypical features include mood reactivity (that is, mood that brightens in response to a positive event) along with two or more of the following: (1) significant weight gain or increase in appetite, (2) hypersomnia, (3) leaden paralysis (i.e., heavy, leaden feelings in arms or legs), (4) a long-standing pattern of sensitivity to real or perceived interpersonal rejection (that are not limited to episodes of mood disturbance) that results in significant social or occupational impairment. In the case of rejection sensitivity, emotional reactions may occur after imagined, anticipated, or real rejection by others and may be extreme at times. This diagnostic subcategory was derived in the 1960's by British psychiatrists based on differential responses to TCAs and MAOIs. They posited that melancholic depression responded preferentially to the former, while atypical depression to the latter. Subsequent research has confirmed that persons with atypical features respond better to MAOIs than to tricyclics.

Seasonal Affective Disorder

Occasionally patients experience depressive episodes at characteristic times of the year. Most commonly the episodes of SAD begin in fall or winter and remit in spring, but occasionally they can also be observed in summer. SAD with prevalence during the winter months (winter depression) appears to vary with latitude, age, and sex. SAD is more common in higher latitudes (i.e., closer to the north or south poles) and among younger people, particularly in females. SAD is characterized clinically by hypersomnia, anergia, and a craving for sweets. SAD responds particularly well to light therapy and typically to serotonergic agents (i.e., selective serotonin reuptake inhibitors [SSRIs]).

The seasonal pattern specifier is given if all of the following are true: (1) there has been a regular temporal relationship between the onset of major depressive episodes in bipolar I or bipolar II disorder or major depressive disorder, recurrent, and a particular time of the year (e.g., fall or spring), (2) full remissions (or a change from depression to mania or hypomania) also occur at a characteristic time of the year.

Psychological Testing

Psychological testing has not been found useful in aiding the diagnosis of major depressive disorder.

Laboratory Findings & Imaging

No laboratory findings are diagnostic of a major depressive episode; however, several laboratory findings are abnormal in some patients with major depression, compared to the general population. It appears that most laboratory abnormalities are state dependent (i.e., they occur while patients are depressed), but some findings may precede the onset of an episode or persist after its remission.

HPA Axis

Dexamethasone Suppression Test

Although the dexamethasone suppression test (DST) has little use as a clinical marker for depression, it is worth mentioning some of the more pertinent DST findings: (1) The overall sensitivity (i.e., positive DST outcome) was 44% among all patients with major depression given 1 mg dexamethasone; sensitivity was significantly higher (65%) in elderly depressed patients but dropped to 31% in a smaller group of patients who received 2 mg of dexamethasone, (2) Within the depression spectrum, the rates of DST nonsuppression increased strikingly from grief reactions (10%) and dysthymic disorders (23%), to major depressive disorders (44%), major depressive disorders with melancholia (50%), psychotic affective disorders (69%), and depression with serious suicidality (78%), (3) In some depressed patients, the DST allowed researchers to predict or monitor long-term treatment outcome, (4) DST-positive patients appeared to respond more favorably to biological interventions such as antidepressants or ECT, (5) Among depressed patients, abnormal DST results correlated mostly with initial insomnia, weight loss, loss of sexual interest, ruminative thinking, and psychomotor retardation or agitation, and (6) HPA axis dysregulation contributed to cognitive dysfunction, although it is not known whether hypercortisolemia and cognitive impairment in depression are related indirectly or causally.

A hyperactive HPA axis has been observed in stroke patients with major depression; in pain patients with major depression; and in patients with anorexia nervosa, bulimia nervosa, alcoholism, obsessive–compulsive disorder, or anxiety disorders. It has not been observed in schizophrenic patients.

Corticotropin-Releasing Hormone Test

A blunted adrenocorticotropic hormone response after corticotropin-releasing hormone administration is another HPA axis abnormality commonly observed in major depression.

HPT Axis

T4 Concentrations

Most depressed patients appear to be euthyroid; however, longitudinal studies consistently found significant serum T4 reductions during a wide range of somatic treatments, including various antidepressants, lithium, sleep deprivation, or ECT. Evidence indicates that the T4 reduction was greater in treatment responders than in nonresponders. It is not known whether the initial T4 increase in depression is part of the pathophysiology of the illness, or whether it is a compensatory mechanism by which the organism delivers thyroid hormone to the brain.

Subclinical Hypothyroidism

Subtle thyroid dysfunctions are common in depression. Between 1% and 4% of patients show evidence of overt hypothyroidism, and between 4% and 40% show evidence of subclinical hypothyroidism. Comorbid subclinical hypothyroidism can be associated with cognitive dysfunction or with a diminished response to standard psychiatric treatments. Some depressed patients with subclinical hypothyroidism may respond behaviorally to thyroid hormone substitution.

TRH Test

The TRH test (i.e., measurement of serum TSH following TRH administration) has been used widely in psychiatry. More than 3000 patients have been studied, the majority of whom had major depressive disorder. Approximately 30% of patients had a blunted TSH response during depression, and a smaller number showed TSH blunting during remission; however, definitions of TSH blunting have varied among studies, different assays have been used, and a standard amount of TRH has not always been injected.

There appears to be no association between the TRH-induced TSH response and (1) the patient's body surface or age, (2) serum thyroid hormone or cortisol concentrations, (3) severity of depression, or (4) previous intake of antidepressant drugs (excluding long-term lithium administration). Further, the TRH test does not aid in the distinction between primary and secondary depression or between unipolar and bipolar subgroups. Preliminary evidence suggests that TSH blunting may be associated with a more prolonged course of depression and with a history of violent suicidal behavior.

Within psychiatric disorders, TSH blunting can also occur in some patients who have borderline personality disorder, anorexia nervosa, panic disorder, primary degenerative dementia, chronic pain, premenstrual syndrome, or alcoholism, both during acute withdrawal and after prolonged abstinence. The absence of TSH blunting in both schizophrenic patients and phobic patients during exposure therapy suggests that the abnormality is not a mere correlate of mental distress.

Sleep Electroencephalogram

Most depressed patients have insomnia. Sleep problems commonly reported include interruptions throughout the night, early morning awakenings, and less frequently, difficulty falling asleep. Sleep electroencephalogram (EEG) recordings reveal the following: (1) a shortened rapid eye movement (REM) latency (i.e., a shorter than normal interval between sleep onset and first REM period), more common in elderly depressed patients and often associated with unipolar depression; (2) a shift of slow-wave sleep (i.e., sleep stages 3 and 4), normally occurring during the first non-REM period, into the second non-REM period; and (3) an increased REM density (i.e., more frequent REM episodes) during the first few hours of sleep.

Because most of these sleep EEG abnormalities can be found in other illnesses, and some accompany normal aging, there is no agreement as to whether these abnormalities constitute diagnostic markers of depression or whether they reflect abnormal functioning in sleep-related processes but lack diagnostic specificity (see Chapter 27). There is evidence that EEG sleep variables are normal in depressed patients 6 months after an acute episode.

Brain Imaging

The rapidly increasing sophistication of neuroimaging techniques has improved our understanding of the neural substrates of emotion and its disorders. Neuroimaging studies have been particularly useful in characterizing the circuitry underlying emotional disorders.

Mood disorders may be associated with global and regional changes in cerebral blood flow and metabolism. Global cerebral blood flow and glucose metabolism appear normal, but may be decreased in late life depression. Decreased prefrontal cortex (PFC), especially left PFC, blood flow and metabolism in depressed unipolar patients are the most consistently replicated findings; there is preliminary evidence that they may correlate with severity of illness and cognitive impairment. Basal ganglia abnormalities have also been found in depressed unipolar patients, involving decreased blood flow and metabolism. Increased activity of amygdala is also observed. Other neuroimaging studies in major depression point to abnormalities in the hippocampus, cingulate, and related parts of the striatum and thalamus. These data suggest a neural model in which dysfunction of limbic, striatal, and PFC structures impair the modulation of the amygdala/hippocampus complex, leading to abnormal processing of emotional stimuli.

In unipolar depression, little is known about the state or trait characteristics of these findings, as the exact relationships between functional neuroimaging findings and clinical course has not been systematically investigated. Depression also tends to be associated with lesions in the left frontotemporal or right parieto-occipital regions. This concept is consistent with neuroanatomical and behavioral findings in stroke patients. Patients with dominant anterior or nondominant posterior strokes are especially vulnerable to secondary depressions, whereas patients with nondominant anterior or dominant posterior strokes are especially vulnerable to mania or hypomania.

Course of Illness

Major depressive disorder may be preceded by dysthymic disorder (10% in community samples, and 15–25% in clinical samples). Table 18–14 summarizes the course and prognosis of both non-bipolar disorders and bipolar disorders.

Major depressive disorder may begin at any age, with an average age at onset in the mid-teens to late 20s. Symptoms typically develop over days to weeks, and prodromal symptoms and preexisting comorbid conditions (e.g., generalized anxiety, panic attacks, phobias) are common. Although some patients have only a single episode, with full return to premorbid functioning, approximately 50% of patients with such episodes will eventually have another episode, at which time they will meet criteria for recurrent depression.

The course of recurrent depression is variable. Some patients have a few isolated episodes separated by stable intervals (years) of normal functioning. Others have clusters of episodes, and still others have increasingly frequent episodes with shortening of the interepisode interval and generally increased disease severity. Some patients experience a major depression without full remission before the next; this is referred to as major depression without interepisode recovery. About 50% of patients with one depressive episode will have a recurrence, and about 90% of patients who have had three episodes can be expected to have a fourth. Thus the number of past episodes can serve as a predictor of the future (i.e., as the number of recurrences increases, the episodes lengthen in duration and increase in both frequency and intensity). The average number of lifetime episodes is around five. About 5–10% of patients with an initial diagnosis of major depressive disorder subsequently develop a manic episode.

Depressive episodes may remit completely, partially, or not at all. The patient's functioning usually returns to the premorbid level between episodes, but 20–35% of patients show persistent residual symptoms and social or occupational impairment. Data from the prepsychopharmacology era (i.e., before 1960) suggest that, if untreated, a depressive episode may last about 12 months. Relapse is common. Almost 25% of patients relapse within the first 6 months of remission, especially if they have discontinued their antidepressant medications; 30–50% relapse in the first 2 years; and 50–75% relapse within the first 5 years. The risk of relapse during early remission can be reduced significantly by maintaining patients on antidepressants for at least 6 months—a regimen now generally viewed as important in managing the recurring nature of the illness.

A major depressive episode often follows acute psychosocial stressors, such as death, loss of a loved one, divorce, or acute medical illness. There is evidence that psychosocial stressors are more important in triggering or facilitating the first two depressive episodes and that their influence becomes less important in subsequent episodes.

Differential Diagnosis

A diagnosis of depression is made if the individual is significantly impaired by the depressive symptoms outlined in the preceding section, and if three exclusion criteria are met: (1) the illness is not due to the effects of a substance (e.g., drug of abuse or medication) or a general medical condition, (2) the illness is not part of a mixed episode (see section on Bipolar Disorders later in this chapter), and (3) the symptoms are not better accounted for by bereavement.

Medical Conditions

Most patients with depression will initially present not to a psychiatrist but to their general medical practitioner and often with a somatic complaint (i.e., “I can't sleep,” or “I have no energy”) rather than a psychiatric complaint (i.e., “I’m depressed”). This is especially true for elderly patients. It is also true that many medications and medical disorders commonly produce symptoms of depression (Table 18–3). Most of these sources of depression can be detected by a thorough review of a patient's history, a complete physical and neurologic examination, and standard laboratory tests. If an etiologic relationship exists between one of these causes and a mood disturbance, the diagnosis will be mood disorder due to a general medical condition (see section on Mood Disorder Due to a General Medical Condition later in this chapter).

Other Psychiatric Disorders

Depression can be a feature of almost all other psychiatric disorders; however, the diagnosis should not be made if the episode is part of a bipolar disorder or schizoaffective disorder.

Uncomplicated Bereavement

Bereavement generally refers to the grief symptoms experienced after the loss of a loved one. Bereavement is not considered a mental disorder even though it may have symptoms characteristic of a major depressive episode (e.g., sadness, insomnia, loss of appetite, weight loss, feelings of guilt or hopelessness). Data indicate that almost 25% of bereaved individuals meet criteria for major depression at 2 months and again at 7 months and that many of these people continue to do so at 13 months. Patients with more prolonged symptoms tend to be younger and tend to have a history of major depression. Antidepressant treatment is justified in bereavement when the behavioral symptoms are prolonged or if they are associated with continued functional impairment.

Treatment

An important aspect of the treatment of depression is for the physician to discuss with patients and their families the illness, its symptoms and course, and particularly its recurrent nature. Patients should recognize that the illness may recur; they—and their close family members—should be cognizant of the premonitory signs and symptoms of an impending episode (i.e., insomnia, especially early morning awakening, loss of energy, loss of appetite and libido, and diurnal changes in well-being); and they should be told to return to their physician as soon as they have noticed any combination of the premonitory signs for, say, longer than 1 week. This, of course, is best achieved if patients are followed up closely in and out of episodes, if the communication between the physician and the patient or patient's family is open and continuous, and if all parties are thoroughly aware that the nature of the illness necessitates both close, continuous observation of its course and early, decisive intervention to prevent relapse. It is also important to recognize that clinical features often have significant treatment implications. A summary of these features and their suggested solutions is provided in Table 18–4.

Psychopharmacologic Interventions

Clinical Pharmacology

Principles of Use

The following general principles provide a useful framework for the clinical use of antidepressants in major depressive disorder: (1) diagnose properly; (2) avoid treatment of symptoms (e.g., agitation, insomnia, memory disturbances) if possible, because most symptoms fit into specific diagnostic entities; and (3) be aware of the cycling course of the disease, because it necessitates different treatment approaches: Acute treatment for florid symptoms; continuation therapy to prevent early relapse; and maintenance therapy to make relapse (i.e., recurrence) less likely or, if it occurs, less severe.

Indications

Antidepressant drugs are used successfully to treat a variety of psychiatric and other conditions (19.1–4). In mood disorders, they are used most widely in the treatment of major depressive disorder, but they also have some (albeit reduced) therapeutic activity in treating dysthymic disorder and bipolar disorder (see sections on Dysthymic Disorders and Bipolar Disorders later in this chapter).

The effectiveness of antidepressants in young patients remains controversial, because some controlled clinical trials failed to show consistent beneficial effects in children and adolescents. Even though mitigating factors such as different study designs, medication compliance, and the impact of exogenous factors (e.g., family discord) on symptom manifestation limit the interpretation of these studies, antidepressant drugs are used widely in depressed children and adolescents. Moreover, antidepressants may increase risk of self-injury transiently. However, the best data suggest that the overall risk of suicide in children and adolescents is reduced by antidepressant treatment.

In contrast, antidepressants are very useful in elderly depressed patients, in whom daily dosage requirements are normally reduced because of pharmacokinetic changes associated with aging (i.e., reductions in both hepatic clearance and protein binding).

Antidepressants are usually initiated at a low dosage and increased over a 7–10-day period to achieve the initial target dosage. The dosage may have to be increased further in some patients in order to achieve the best results. With suicidal patients, the physician must take extra care early in the treatment because behavioral activation can precede observable mood effects, providing patients with sufficient energy to act on suicidal impulses.

Response to treatment should be evaluated every 3–4 weeks. If the patient is not fully recovered, and if there are no dose-limiting side effects, the dosage should be increased to the next incremental level. This plan should be followed until remission is achieved, dose-limiting side effects occur, or the upper end of the therapeutic range is reached. Absent full remission, the treatment should then be switched to a drug of a different therapeutic class. Augmentation or combination therapy should be considered if there is only partial response.

Once a therapeutic effect is achieved, the antidepressant medication should be continued through the period of high vulnerability for relapse (i.e., at least 6 months for continuation therapy). Because more than 60% of depressed patients will eventually relapse, especially if unprotected by medication, and because future episodes are more severe, it has been proposed that some depressed patients be placed on long-term treatment. Such maintenance therapy for extended periods of time (even years) should be considered if (1) the patient is older than 40 years and had two or more prior episodes of illness, (2) the first episode occurred at age 50 years or older, (3) the patient has a history of three or more depressive episodes, or (4) the patient has been depressed or dysthymic for 2 or more years before treatment.

Slow tapering of the antidepressant medication can be considered after at least 5 years of treatment if the patient is completely asymptomatic and is not experiencing or anticipating significant stressors. Some patients prefer lifetime treatment rather than risking a return of depression.

Adverse Effects

Table 18–5 lists common adverse effects associated with antidepressant drugs, their possible mechanisms, and their management. Table 18–6 indicates the relative potency of antidepressants at producing these effects (Table 18–7).

Monoamine Oxidase Inhibitors

Common side effects of MAOIs include those associated with α1-adrenergic or muscarinic/cholinergic antagonism (see Table 18–5). They are generally manageable by slowing the dose titration or by adjusting the maximum dosage. Less common reactions include ataxia, color blindness, hepatotoxicity, or induction of mania in bipolar patients. Withdrawal reactions—including anxiety, restlessness, insomnia, nausea, agitation, myoclonus, and in extreme cases, the induction of mania—may occur after abrupt discontinuation.

The most serious side effect associated with MAOI use is the development of an acute hypertensive crisis. The metabolism of certain dietary amino acids, especially tyramine, is blocked by MAOIs. The resulting increase in neurotransmitter availability can lead to an acute hypertensive crisis, with patients complaining about pounding headaches and presenting with flushing and blood vessel distention. The reaction must be considered a medical emergency to be treated with slow intravenous administration of the α-adrenergic antagonist phentolamine, 5 mg (which may be repeated hourly as needed).

MAOIs may also produce acute toxicity by interfering with the metabolic clearance of other drugs, including general anesthetics, barbiturates and other sedatives, antihistamines, alcohol, narcotics (particularly meperidine), anticholinergic agents, TCAs, and sympathomimetic amines (e.g., pseudoephedrine used commonly in decongestants).

Coadministration of MAOIs with either SSRIs or l-tryptophan can provoke a central serotonin syndrome, characterized by acute mental status changes (e.g., confusion, hypomania), restlessness, myoclonus, diaphoresis, tremor, diarrhea, hyperreflexia, and occasionally, seizures, coma, and death. The syndrome is usually mild in nature and resolves within 24 hours after drug discontinuation. MAOIs should therefore not be started within 2 weeks of discontinuation of most SSRIs (5 weeks in the case of fluoxetine, because of its long half-life time [HLT] [see next section]).

Coadministration of TCAs and MAOIs (used occasionally for the management of treatment-refractory depression) can produce serious side effects, including delirium and hypertension. It is therefore advisable to wait at least 2 weeks after discontinuing TCAs before initiating treatment with MAOIs. Table 18–8 lists foods and drugs that can produce serious adverse effects. Patients should be given a copy of this list before beginning treatment with MAOIs.

TCAs & Related Antidepressants

Tricyclic-induced side effects are the result of their binding to both specific (e.g., norepinephrine or serotonin) and nonspecific (e.g., histamine, muscarinic) sites. Table 18–6 summarizes these actions and their consequences.

The tertiary amine TCAs—amitriptyline, imipramine, and doxepin—tend to be more potent at binding sites unrelated to the mechanism of action and, therefore, tend to produce more severe side effects than the secondary amine TCAs nortriptyline and desipramine (see Tables 18–6 and 18–7).

TCAs and trazodone can affect the heart in the same way as the class 1 antiarrhythmics quinidine or procainamide, inducing atrioventricular (A-V) conduction delays. Special care should be taken with patients who have a first-degree A-V block (a relative contraindication) or a second-degree A-V block (an absolute contraindication) (see Table 18–6). As type 1 antiarrhythmics, these antidepressants may pose greater long-term risk than previously thought. The primary antiarrhythmics can increase the risk of sudden death from a presumed arrhythmic source. This may explain the long-standing observation of increased risk of sudden death in patients treated chronically with TCAs.

Maprotiline is associated with an increased risk for seizures at higher dosages and plasma levels and has never gained wide use in the United States. Venlafaxine, a mixed norepinephrine and serotonin reuptake inhibitor, can produce mild hypertensive reactions, especially at dosages greater than 300 mg/day. Risk factors for hypertension, such as age, race, or preexisting renal or hypertensive disorder, are usually not associated with this effect.

Selective Serotonin Reuptake Inhibitors

SSRIs potently and selectively block the uptake of serotonin, thereby producing characteristic side effects such as nausea, diarrhea, anorexia, anxiety, headache, insomnia, and orgasmic dysfunction (see Table 18–6). Sexual dysfunction is troublesome and can be persistent. It can be managed by reducing the dosage or by switching to bupropion or nefazodone. Cyproheptadine, yohimbine, or methylphenidate are also occasionally useful in reversing these effects. Finally, sildenafil has been shown in controlled trials to be beneficial in both men and women.

Pharmacokinetics

Absorption

All antidepressant drugs are highly lipophilic and are absorbed readily from the gastrointestinal tract. Plasma peaks typically occur within 30–60 minutes and remain unbound for 30 minutes.

Distribution

Most (80–90%) antidepressants are highly protein bound, although individual differences in binding can produce a fourfold variation in the amount of free drug. Tissue distribution and demethylation, the first step in metabolism of tertiary heterocyclics, occur rapidly. All antidepressants can be displaced by other drugs with similar protein-binding characteristics. Similarly, antidepressants will displace other compounds that are highly protein bound (e.g., the anticoagulant warfarin); this can increase the free fraction of these drugs and, therefore, both its therapeutic and its adverse effects.

Metabolism

Antidepressants are metabolized mainly in the liver. Metabolism includes demethylation, hydroxylation, and glucuronide conjugation. Individual differences in a patient's microsomal enzyme activity produce steady-state plasma concentrations (bound and unbound) that may vary as much as 40-fold. During the first pass through the liver, about 80% of the drug is degraded (the so-called first-pass effect). The first demethylation of imipramine and amitriptyline produces (the clinically highly active) desipramine and nortriptyline, respectively; these and other drugs have active metabolites that significantly prolong their biological half-lives. Glucuronide conjugation occurs after hydroxylation and makes the derivative water soluble; approximately 65% of the drug is eventually eliminated in the urine. Plasma clearance is slow.

Impairment of hepatic or renal function will reduce plasma clearance of antidepressants and thus potentiate their effects. Antipsychotics, by competing for metabolism, can have a similar effect. Conversely carbamazepine and barbiturates, by inducing metabolism, can reduce their therapeutic efficacy.

Half-Life Times

After a single oral dose, distribution is most important in regard to clinical efficacy; after chronic dosing, HLT becomes more important. The HLTs of the most commonly used antidepressants are given in Table 18–9.

Predicting Treatment Response & Managing Treatment Failure

Several factors are useful in predicting a patient's clinical response (positive and negative) to an antidepressant (Table 18–10).

Between 60% and 70% of persons with major depressive disorder respond to an adequate antidepressant drug trial. However, only about one-third of patients fully recover. The reasons for treatment failure include inadequate dose and/or serum level, inadequate duration of treatment (minimum 6 weeks), prominent side effects, noncompliance, and incorrect diagnosis.

Because single doses of antidepressants can produce widely varying drug serum levels, the patient's plasma levels should be measured in order to ensure optimal dosing when available. Unfortunately, only plasma levels of tricyclics are widely available. Associations between plasma levels and clinical response are known for several antidepressants (Table 18–11). The monitoring of the plasma level of nortriptyline is particularly important because there appears to be a therapeutic window in which the drug is most effective. Plasma levels can also be assessed for other antidepressants, including fluoxetine, but the relationship between plasma levels and clinical response is less well understood for these agents. Indications for monitoring drug levels of antidepressants are (1) failure to respond to a presumably adequate trial in terms of dose and duration; (2) concurrent physical illness, especially cardiovascular disease; (3) use of higher than usual doses (i.e., greater than 300 mg/day of imipramine); (4) treatment in the very young or old (or in any patient with decreased protein binding); (5) the presence of side effects uncertainly related to drug treatment; and (6) drug overdose.

When a true treatment failure has occurred, the physician needs to consider an alternative approach. This may involve trying another treatment approach (e.g., ECT or combining TCAs with an MAOI) or using one of several augmentation strategies (Table 18–12). A large body of evidence supports these strategies, and particularly in tertiary care centers (to which most nonresponders normally are referred) they are widely and safely used.

Lithium carbonate, thyroid hormone (preferably T3), or sleep deprivation are effective in augmenting the therapeutic effects of TCAs. Among these strategies, sleep deprivation is the least invasive and most easily conducted technique. Its antidepressant effects, observed in about 65% of patients after only one night of wakefulness, are immediate; it is entirely free of side effects; and patients can repeat it easily and safely even after they have been discharged from the hospital. Care must be taken, however, since patients may relapse quickly after recovery sleep. Lithium can be titrated to standard serum levels and T3 started at 25 μg/day; both augmentation strategies should be maintained for a minimum of 3 weeks in the absence of therapeutic response. About half of patients who have not responded to the standard antidepressants will experience sudden clinical improvement. It is not known whether lithium or T3 should be continued after a clinical response has occurred. T3 appears to be effective in the augmentation of SSRI's and related newer medications. However, the effectiveness of lithium has been called into question.

Alternative methods for treating refractory depression include the use of a combination of a noradrenergic antidepressant such as desipramine or bupropion with a serotonergic one such as fluoxetine. If TCA's are used, care must be taken to measure the plasma level since SSRIs are competitive inhibitors of metabolism by liver cytochrome enzymes, especially CYP2D6. This can lead to a dangerous increase in plasma levels. Hence TCA plasma levels should be measured with concomitant use. Psychostimulants such as amphetamine or methylphenidate may be useful in some patients but in a more limited way.

The addition of cognitive–behavioral psychotherapy (CBT) to antidepressant treatment also appears to be beneficial in many patients. In particular, a version of CBT specifically designed to aid in chronic depression, referred to as cognitive behavioral analysis system of psychotherapy (CBASP) has been shown to be effective in one large scale clinical trial. The effects of other psychotherapies have not been established.

If these techniques are ineffective, two other options remain. The first is to use a TCA and an MAOI together, but this strategy bears some risk of acute and potentially serious side effects. As well, the data supporting the effectiveness of this combination is very limited. Imipramine and phenelzine are commonly employed. Typically, the MAOI is started first. Once a stable dose is achieved, a tricyclic may be started at low dose. Patients must be cautioned regarding potential negative consequences, including hypertensive reactions. ECT is another available treatment that can be used in the face of nonresponse. A more extensive discussion of ECT is found below.

Other Interventions

Electroconvulsive Therapy

Between 80% and 90% of all ECT treatments in the United States are performed for the treatment of major depressive disorder. Although ECT appears to be most effective in the most severely depressed patients, attempts to use clinical symptoms, patient history, demographics, or other factors as predictors of clinical response have been largely unsuccessful.

ECT has shown efficacy in all types of major depressive disorder. Recent evidence suggests that it is less useful in patients whose depressive episodes occur in the context of a concurrent mental or medical disease (i.e., secondary depression) or in the treatment of depression that has been refractory to medications during the present episode.

When to Use ECT

The decision to refer a patient for treatment with ECT is made only after careful considerations of its risks and benefits. ECT is used as primary treatment when (1) an urgent need (either psychiatrically or medically) for a rapid response exists, (2) there is less risk with ECT than with other treatment alternatives, (3) there is a patient history of better response to ECT, or (4) there is a strong patient preference for its use. A majority of patients referred for ECT do not meet these criteria. ECT is used as a secondary treatment when (1) the patient has responded poorly to or is intolerant of alternative treatments or (2) the patient has deteriorated to the point at which a rapid response is needed urgently.

Contraindications

There are no absolute contraindications to the use of ECT, but some conditions are relative contraindications (Table 18–13). If treatment with ECT becomes necessary on a life- saving basis, these risks can generally be minimized to some extent using appropriate pharmacologic interventions. With appropriate preparation, ECT can be used effectively and safely in both pregnant and elderly patients.

Adverse Outcomes of Treatment

Mortality

The overall mortality rate of ECT is extremely low; it approximates that of brief general anesthesia. Estimates range between 1 in 10,000 (0.01%) and 1 in 1000 (0.1%) patients.

Cognitive Changes

There is no relationship between ECT and brain damage; however, temporary cognitive changes do occur and are often the most notable and most distressing side effects. Three types of cognitive change may be observed:

Postictal confusion. Because ECT induces seizure activity, all patients experience some transient confusion, lasting from a few minutes to a few hours, after they awaken from the ECT treatment. Most patients usually will not remember the immediate postictal period and will not experience the cognitive changes following a seizure as a significant disturbance. Reassurance, support, and avoidance of cognitive demands during the acute postictal period are typically all that is necessary for treatment. Postictal sedation with a short-acting benzodiazepine such as midazolam may be necessary if the patient becomes agitated.

Interictal confusion. Occasionally postictal confusion may not disappear fully, and when severe, it may develop into an interictal confusional state or delirium. This phenomenon, which is uncommon, may be cumulative over the ECT course but disappears rapidly over a period of days after the conclusion of treatments.

Memory impairment. Amnesia often occurs with ECT and varies considerably in both severity and persistence. Memory disturbances consist of both retrograde amnesia (i.e., difficulty in recalling information learned prior to the ECT course) and anterograde amnesia (i.e., difficulty in retaining newly learned information). When present, anterograde and retrograde amnesia disappear rapidly over a period of days to weeks after completion of the ECT course.

A small group of patients report that their memory never returned to normal after ECT. The etiology for this phenomenon is unclear, and objective memory testing generally does not substantiate the subjective complaint. Possible explanations for this phenomenon may include (1) patients incorrectly attribute memory problems associated with depression to prior ECT, (2) anesthetic misadventure (e.g., hypoxia), or (3) other, occult memory impairment (e.g., evolving dementia).

Cardiovascular Complications

During the seizure and acute postictal period, the sympathetic and parasympathetic autonomic systems are stimulated sequentially. Activation of the sympathetic system increases heart rate, blood pressure, and myocardial oxygen consumption, placing an increased demand on the cardiovascular system. Activation of the parasympathetic system causes a transient reduction in cardiac rate. These changes in heart rate and cardiac output challenge the cardiovascular system, occasionally giving rise to transient arrhythmias and, in susceptible individuals, transient ischemic changes. Most cardiovascular changes are minor, and the risk of complications can be diminished greatly by the use of oxygen and appropriate medications in susceptible patients.

Other Adverse Effects

General somatic complaints (i.e., headaches, nausea, muscle soreness) are common after ECT. Analgesics may be given prophylactically prior to ECT to patients who routinely have post-ECT headaches. The dosage of muscle relaxant should be increased if muscle soreness and headaches are due to inadequate relaxation.

About 7% of depressed bipolar patients switch into a manic or mixed state after ECT. This state can be managed by either continuing ECT or stopping ECT and administering an antimanic agent.

Course of Treatment

Number of Treatments

A typical ECT course involves 6–12 treatments, but the number required may be as low as 3 or as high as 20. Alterations in the ECT course should be considered for nonresponders or for patients whose clinical progress is minimal after approximately six treatments. Alterations may include changing from unilateral to bilateral electrode placement, increasing stimulus intensity, or potentiating the seizure pharmacologically. If there is still no response after three or four additional treatments, or if the patient's response has reached a plateau below a full level of remission, the ECT course should be terminated.

Multiple Monitored ECT

Multiple monitored electroconvulsive therapy (MMECT) was developed in an attempt to decrease the average length of the treatment course by inducing multiple seizures (usually 2–10) during a single treatment. MMECT appears to be associated with a more rapid clinical response, but the total number of seizures required is greater. There is also evidence that MMECT is associated with a higher frequency of prolonged seizure activity, exaggerated cardiovascular responses, and increased cognitive side effects.

Frequency of Treatments

In the United States, most ECT treatments are given three times a week (e.g., Monday, Wednesday, and Friday). An increase in the frequency is associated with a more rapid response but also increased cognitive side effects.

Continuation Treatment

Most psychiatrists agree that after the remission of a depressive or manic episode, ECT treatment should be continued for at least 6–12 months. After the conclusion of a course of ECT, there are three options for continued treatment: (1) administration of appropriate psychotropic medications (e.g., antidepressants); (2) continuation of ECT; or (3) psychotherapy, combined with pharmacotherapy or ECT.

Maintenance Treatment

Because of the severity and chronicity of some patients’ illnesses, many practitioners now prophylactically continue pharmacotherapy or ECT for longer time periods, particularly when attempts to discontinue treatment have resulted in a recurrence of the illness. Maintenance ECT treatments may be given monthly, although some individuals require more frequent administration (e.g., weekly). There is no lifetime maximum number of treatments that a patient may have. Cumulative amnestic effects are not usually seen.

Vagus Nerve Stimulation

A number of clinical studies have shown that vagus nerve stimulation (VNS) has antidepressant effects in patients with depression resistant to four or more treatments. Improvement with VNS result in enhanced neurocognitive function in many patients. VNS appears to be most effective in patients with low to moderate antidepressant resistance. Response rates usually range from 30–40%, and long-term VNS treatment appears to be associated with sustained symptomatic improvement. VNS is also known to improve mood in depressed patients with epilepsy.

Transcranial Magnetic Stimulation

Repetitive transcranial magnetic stimulation (rTMS) is a noninvasive and easily tolerated method of altering cortical physiology. Clinical studies support an antidepressant effect of high-frequency rTMS administered to the left PFC; however, antidepressive efficacy is not consistent, and where efficacy is demonstrated, degree of clinical improvement appears to be small. The absence of psychosis, and younger age may be predictors of treatment success. Low frequency TMS to the right PFC also has shown promise. Repetitive TMS may be useful in augmenting or hastening the response of antidepressant drugs in patients with major depressive disorder.

Psychotherapeutic Interventions

The establishment and maintenance of a supportive therapeutic relationship, wherein the therapist gains the patient's confidence and is available in times of crisis, is crucial in the treatment of depression. Beyond pure psychotherapeutic management the therapeutic relationship contains other important factors. These include observing emerging destructive impulses toward self or others; providing ongoing education, knowledge, and feedback about the patient's illness, its prognosis, and treatment; discouraging the patient from making major life changes while he or she is depressed; setting realistic, attainable, and tangible goals; and enlisting the support of others in the patient's social network.

Depression is often treated with psychotherapy, either alone or in combination with antidepressants. This is particularly true for unipolar depression. The psychosocial interventions that have fared well in controlled trials relative to antidepressants include interpersonal psychotherapy, cognitive–behavioral therapy, and some of the marital and family interventions. Traditional dynamic psychotherapy has not proved effective but may not have been evaluated adequately.

Interpersonal Psychotherapy

Interpersonal psychotherapy seeks to recognize and explore depressive precipitants that involve interpersonal losses, role disputes and transitions, social isolation, or deficits in social skills. It focuses on current interpersonal problems and deemphasizes childhood antecedents or extensive attention to transference in the therapeutic relationship. Interpersonal psychotherapy can be effective in reducing depressive symptoms in the acute phase of nonmelancholic major depression of lesser severity. It seems to produce a greater change in social functioning than antidepressants; this change may not appear until after several months of treatment (or until several months after treatment is over).

Cognitive–Behavioral Therapy

Cognitive–behavioral therapy is based on the premise that helping patients recognize and correct erroneous beliefs and maladaptive behaviors can relieve their affective distress. It also can be effective in reducing depressive symptoms in the acute phase of major depression. Cognitive–behavioral therapy may have an enduring effect that protects the patient against subsequent relapse following treatment termination. A recent large-scale clinical trial has shown that CBT is about as effective as the antidepressant paroxetine in moderate to severe major depression. Moreover, CBT reduced the likelihood of relapse over continuation placebo or medication over the subsequent year. Comorbid generalized anxiety disorder favored treatment with paroxetine. In the treatment of very severe depression in outpatients neither interpersonal therapy nor cognitive–behavioral therapy are quite as effective as antidepressant drugs.

A recent multi-center trial of patients with chronic depression compared the antidepressant effects of an antidepressant (Nefazodone), Cognitive Behavioral Analysis System of Psychopathology (CBASP), a psychotherapy method designed for chronic depression, or the combination. Overall, the effects of the antidepressant alone and psychotherapy alone were equal and significantly less effective than combination treatment. Among those with a history of early childhood trauma or abuse, psychotherapy alone was superior to antidepressant monotherapy, suggesting an important role of psychotherapy in the management of patients with chronic forms of major depression and a history of childhood trauma.

Marital Therapy & Family Therapy

Marital and family problems are common in the course of mood disorders. They may be a consequence of depression but may also increase the patient's vulnerability to depression and in some instances retard recovery. Research suggests that marital therapy and family therapy may reduce depressive symptoms and the risk of relapse in patients with marital or family problems.

Selection of Specific Therapies

Patient preference plays a large role in the selection of a particular form of psychotherapy. In choosing the most appropriate form, the psychiatrist should consider that an interpersonal approach may be more useful for patients who are in the midst of conflicts with significant others and for those having difficulty adjusting to an altered career or social role or other life transition. A cognitive approach can be helpful for patients who seek and are able to tolerate explicit, structured guidance from another party.

Complications/Adverse Outcomes of Treatment

Major depressive disorder is associated with considerable morbidity, disability, and risk for suicide. Up to 15% of patients with severe major depressive disorder commit suicide (among patients with recurrent depression, death by suicide occurs at the rate of about 1% per year). Among patients with major depressive disorder who are older than 55 years, the death rate is increased fourfold. Compared to nondepressive subjects, patients with major depressive disorder admitted to a nursing home are more likely to die within the first 5 years. Among patients in general medical settings, physically ill patients with depression have increased pain and physical illness and decreased social and role functioning. In fact, deterioration of functioning associated with depression or depressive symptoms can be equal to or worse than that associated with severe medical conditions. The combination of advanced coronary artery disease (CAD) and depressive symptoms was associated with roughly twice the reduction in social functioning as with either condition alone, suggesting that the effects of depressive symptoms and chronic medical conditions on functioning can be additive.

Major depression and depressive symptoms, although commonly observed in medical populations, are often underdiagnosed and undertreated. This is particularly relevant in cardiovascular disease where depression and its associated symptoms have been shown to be a major risk factor for both development of cardiovascular disease and death after myocardial infarction. In patients with cardiovascular disease, treatment of depression may thus stabilize mood imbalance, improve quality of life, and perhaps positively affect such outcome measures as longevity.

Major depressive disorder exacts a large toll on patients, family members, and society, because of its high level of morbidity and mortality. It may account for more bed days than any other physical disorder except cardiovascular disease, and it may be more costly to the economy than chronic respiratory illness, diabetes, arthritis, or hypertension. It has been estimated that depression and other mood disorders account for $43.7 billion in direct costs (i.e., treatment) and indirect costs (i.e., lost productivity through illness, absenteeism, or suicide). Before modern treatments, depressed patients typically spent one-fourth of their adult lives in the hospital and fully one-half of their lives disabled.

Prognosis

Table 18–14 summarizes the relevant prognosis factors. Major depressive disorder is not a benign condition, but careful management, including early, aggressive therapeutic intervention can largely improve its otherwise poor prognosis. Although the literature on long-term outcome has produced variable results because of study design (i.e., short duration of observation, focus on hospitalized episodes, exclusion of episodes preceding the index episode, and retrospective rather than prospective analysis) and different case definitions, the IOWA 500 study, involving 35 years of retrospective follow-up of patients initially presenting with a manic episode, found that outcome was good in 64% of patients, fair in 14%, and poor in 22%. Compared to well-matched patients with minor surgery or schizophrenia, the depressed patients’ outcome was intermediate (i.e., better than that of schizophrenic patients but worse than that of surgery patients).

Positive prognostic indicators include an absence of psychotic symptoms, a short hospitalization or duration of depression, and good family functioning. Poor prognostic indicators include a comorbid psychiatric disorder, substance abuse, early age at onset, long duration of the index episode, and inpatient hospitalization. A patient who has major depression with comorbid addiction is more likely to require hospitalization, more likely to attempt suicide, and less likely to comply with treatment than is a patient with depression of similar severity not complicated by comorbid addiction.

Clinicians have noted substantially increased mortality rates in major depressive disorder and bipolar disorder for more than 60 years. They recognized suicide as the most significant factor contributing to mortality, although other factors (e.g., malnutrition, comorbid medical illness) are likely to contribute. Early studies reported up to six times the mortality rates expected for a normal population of the same age, but later studies found a less striking yet still significant increase in mortality, averaging about two times the expected rates.

Essentials of Diagnosis

DSM-IV-TR Diagnostic Criteria

1. Depressed mood for most of the day, for more days than not, as indicated either by subjective account or observation made by others, for at least 2 years.

   Note: In children and adolescents, mood can be irritable and duration must be at least 1 year.

2. Presence, while depressed, of two (or more) of the following:

   1. poor appetite or overeating

   2. insomnia or hypersomnia

   3. low energy or fatigue

   4. low self-esteem

   5. poor concentration or difficulty making decisions

   6. feelings of hopelessness

3. During the 2-year period (1 year for children or adolescents) of the disturbance, the person has never been without the symptoms in Criteria A and B for more than 2 months at a time.

4. No major depressive episode has been present during the first 2 years of the disturbance (1 year for children and adolescents); i.e., the disturbance is not better accounted for by chronic major depressive disorder, or major depressive disorder, in partial remission.

   Note: There may have been a previous major depressive episode provided there was a full remission (no significant signs or symptoms for 2 months) before development of the dysthymic disorder. In addition, after the initial 2 years (1 year in children and adolescents) of dysthymic disorder, there may be superimposed episodes of major depressive disorder, in which case both diagnoses may be given when the criteria are met for a major depressive episode.

5. There has never been a manic episode, a mixed episode, or a hypomanic episode, and criteria have never been met for cyclothymic disorder.

6. The disturbance does not occur exclusively during the course of a chronic psychotic disorder, such as schizophrenia or delusional disorder.

7. The symptoms are not due to the direct physiologic effects of a substance (e.g., a drug of abuse, a medication) or a general medical condition (e.g., hypothyroidism).

8. The symptoms cause clinically significant distress or impairment in social, occupational, or other important areas of functioning.

Specify if:

Early Onset: if onset is before age 21 years

Late Onset: if onset is age 21 years or older

Specify (for most recent 2 years of Dysthymic Disorder With Atypical Features)

(Reprinted, with permission, from the Diagnostic and Statistical Manual of Mental Disorders, 4th edn., Text Revision. Washington DC: American Psychiatric Association, 2000.)

General Considerations

The term dysthymia is derived from the Greek word thymos, and its literal meaning is “disordered mood.” The classic dysthymic patient has been described by Akiskal as “ill-humored … (having the) … disposition to low-grade dysphoria … habitually brooding, overcontentious, incapable of fun, and preoccupied with personal inadequacy” (1995, p. 1073). The term dysthymic disorder was first used in the Diagnostic and Statistical Manual of Mental Disorders, third edition (DSM-III), which attempted to create a diagnostic concept under which a heterogeneous group of disorders previously referred to as depressive neurosis could be subsumed. The creation of this diagnostic entity as separate from major depressive disorder has been an impetus for increased research activity.

Despite our improved understanding of this condition, dysthymic disorder, as Max Hamilton stated just before his death in 1989, “lies on the border between the normal and pathological.” Because patients with dysthymic disorder experience symptoms that are of a lower intensity than those associated with major depression, the term subsyndromal mood disorder has been applied. In a rural primary care population, 3.8–6.0% of patients were classified as having low-grade or intermittent depressions according to research diagnostic criteria (which were used mainly for research purposes and preceded the development of the DSM-III), and almost twice as many patients (11.2%) had a psychiatric disorder with significant depressive symptomatology that did not fit into a specific depressive disorder category. (The diagnosis of intermittent depression using research diagnostic criteria can be seen as the precursor of the diagnosis of dysthymic disorder in DSM-III.) The data suggest that dysthymic disorder is the tip of a very large iceberg of subsyndromal mood states.

Epidemiology

Table 18–15 summarizes the relevant risk factors. The lifetime prevalence of dysthymic disorder in the general population (3.2%) is exceeded only by that for major depression (4.9%). Even more dramatic are the 1-month point prevalence rate estimates from the Epidemiologic Catchment Area (ECA) survey, which at 3.3% rank dysthymic disorder second only to phobia and more common than alcoholism and all other mental disorders assessed. The protracted course of dysthymic disorder makes it the most commonly encountered form of mood disorder. It affects women more often than men, with a ratio of nearly 2–1. Dysthymic disorder usually sets in before the age of 45 years. Although dysthymic disorder appears to be the least common among African Americans (2.5%), the ECA survey noted the largest number of depressive and dysthymic symptoms in that population. This finding may imply either denial of illness by some respondents or a higher rate of subsyndromal mood disorder (currently diagnosed as depressive disorder not otherwise specified or the proposed depressive personality disorder).

Etiology

Life Events

Personality features may be involved in the pathogenesis of dysthymic disorder. Psychoanalytic theorists have stressed the concept of inordinate interpersonal dependency, in which the patient's self-esteem is excessively reliant on approval, reassurance, attention, and love from others. Thus when such interpersonal concern from others wanes or terminates, depression ensues.

Cognitive theorists ascribe the symptoms of depression to disordered basic cognitive schema, or dysfunctional thoughts, that are developed early in life. The learned helplessness model holds that after repeated, inescapable shock, experimental animals will not initiate avoidance from aversive stimuli, even when avenues of escape are readily available. Even when faced with major traumatic events and losses, most people do not progress from grief to pathologic depression, suggesting that adverse life events and an underlying vulnerability interact to produce symptoms of chronic depression. It is also true that mood disorders create adverse conditions, such as separation, divorce, and suicide. When such conditions occur in parents, an already predisposed child is exposed to environmentally determined object losses that might precipitate more severe illness or earlier onset of depressive episodes.

Biological Theories

Family studies demonstrate that patients with dysthymic disorder and patients with major depression exhibit similar rates of mood disorders in first-degree relatives. Patients with dysthymic disorder also exhibit shortened REM latency during sleep onset, as do patients with major depression. Thus there may be a biological or familial relationship between these two disorders.

Personality inventories of patients with dysthymic disorder often reveal a disturbed personality structure; dependent, avoidant, borderline, and narcissistic personality disorders are the most common. Little is known about the exact biological determinants of dysthymic disorder. Many dysthymic individuals respond well to antidepressants, although the percentage of those who respond and the magnitude of their response are generally less than that observed in patients with major depression.

Genetics

Family studies support a genetic link between dysthymic disorder and major depression. Twin studies have not corroborated a strong relationship between dysthymic disorder and other personality disorders, with the exception of the depressive personality disorder. Depressive personality disorder is a construct noted in the DSM-IV, in which individuals demonstrate a pervasive pattern of depressive cognitions and behaviors beginning in early adulthood but do not develop dysthymic disorder or major depressive episodes. Depressive personality disorder may be a condition that subsumes the persistent trait-like qualities of the depressive disorders. Preliminary evidence from family studies indicates that this disorder is related to the major mood disorders such as dysthymic disorder. Thus far, research has not identified any specific gene or set of genes associated with dysthymic disorder.

Clinical Findings

Signs & Symptoms

Although there has been some controversy concerning the symptoms that best define dysthymic disorder, recent studies suggest that the symptoms most commonly encountered in dysthymic disorder may be those found in the alternative DSM-IV research criterion B for dysthymic disorder (Table 18–16).

In children, dysthymic disorder often results in impaired social interaction and school performance. Children and adolescents with dysthymic disorder are usually irritable, pessimistic, cranky, and depressed and have low self-esteem and poor social skills.

Psychological Testing

Patients with dysthymic disorder have increased scores for neuroticism and introversion on the Maudsley Personality Inventory; these abnormalities persist even after recovery.

Laboratory Findings & Neuroimaging

About 25–50% of adults with dysthymic disorder have EEG abnormalities similar to those found in major depressive disorder (e.g., reduced REM latency, increased REM density, reduced slow-wave sleep, impaired sleep continuity). Dysthymic patients with these abnormalities have a positive family history for major depressive disorder more often than do patients without these abnormalities; they also appear to respond better to antidepressant medications. It is not clear whether EEG abnormalities can also be observed in patients with pure dysthymic disorder (i.e., in those with no history of major depressive episodes). An abnormal DST is not common in dysthymic disorder, unless criteria are also met for a major depressive episode.

Differential Diagnosis

Major depressive disorder usually consists of one or more discrete episodes of depression that can be distinguished, by both onset and duration, from the person's usual functioning. During these episodes, vegetative symptoms such as insomnia, loss of appetite, loss of libido, weight loss, and psychomotor symptoms are common. In contrast, dysthymic disorder is characterized by chronic, less severe depressive symptoms that can persist for 2 or more years; vegetative symptoms are less common in dysthymic disorder than in major depressive disorder.

Depressive symptoms may be associated with chronic psychotic disorders and are not considered to indicate dysthymic disorder if they occur only during the course of the psychotic disorder (including residual phases). If mood disturbance is judged to be the direct physiologic consequence of a specific, usually chronic, general medical condition (e.g., multiple sclerosis), then the diagnosis is mood disorder due to a general medical condition. Similarly, substance-induced mood disorder is diagnosed when the depressive symptoms are thought to result from the ingestion of a drug of abuse or a medication or from exposure to a toxin.

Treatment

Psychopharmacologic Interventions

Substantial evidence from controlled treatment trials indicates that antidepressants are useful in the treatment of dysthymic disorder, with or without comorbid major depression (Table 18–17). TCAs, MAOIs, and SSRIs all appear to be effective treatments. Rates of response have varied, and there is currently no information assessing the rates of response to a second antidepressant if the first trial was unsuccessful. Controlled maintenance treatment studies are under way. Most dysthymic patients, if they respond to an antidepressant, require continued treatment for at least 6–12 months.

Because dysthymic patients with mild symptoms are not likely to tolerate side effects, the clinician should initiate treatment with an SSRI, because it produces less anticholinergic and sedating side effects and less weight gain. This is particularly important in women receiving long-term treatment. SSRIs can produce a number of mild and time-limited side effects, including gastrointestinal symptoms, insomnia, nervousness, sweating, tremor, dizziness, occasional somnolence, and sexual dysfunction. In patients who do not respond to SSRIs, nortriptyline and desipramine are effective. Among nonselective MAOIs, phenelzine is particularly beneficial for those with atypical reversed vegetative symptoms (e.g., hypersomnia, hyperphagia).

Many of the controlled medication trials reported only partial responses in some subjects with dysthymic disorder. Such patients might benefit from augmentation with a second antidepressant, lithium, thyroid hormone, or combined treatment with psychotherapy. However, there have been controlled studies neither of pharmacologic augmentation strategies, nor of combined pharmacotherapy and psychotherapy approaches.

Psychotherapeutic Interventions

Although antidepressants are effective in treating dysthymic disorder, a substantial proportion of patients either fail to respond or are unable to tolerate the side effects. Psychotherapy should be used to address impairments in social and occupational functioning, chronic pessimism and hopelessness, and lack of assertiveness (see Table 18–17). Interpersonal psychotherapy and cognitive–behavioral therapy are useful in treating major depression. Recent modifications of these approaches for dysthymic disorder have been investigated in small naturalistic studies, with successful results. Both therapies are typically time-limited (i.e., lasting less than 6 months).

Interpersonal psychotherapy emphasizes the interaction between the individual and his or her psychosocial environment. Interpersonal psychotherapy has the dual therapeutic goals of reducing depressive symptoms and developing more effective strategies for dealing with social and interpersonal relations. As applied to dysthymic disorder, interpersonal psychotherapy has a major focus on the therapeutic relationship, which is used as a model for other interpersonal interactions. The therapist encourages the patient to engage in occupational and social activities and helps the patient to identify personal needs, to assert those needs, and to set limits.

Cognitive–behavioral therapy is based on the assertion that depression is associated with negative thought patterns, cognitive errors, and faulty information processing that can be altered by specific psychotherapeutic strategies. These strategies help the patient to identify and test negative cognitions and to replace them with alternative, more flexible schemata, which are subsequently rehearsed and studied. As applied to dysthymic disorder, chronic beliefs of helplessness and lack of control are challenged and modified by the experience of mastery in producing desired outcomes, especially in the area of interactions with others.

Other Interventions

Interventions based on other than psychopharmacological or psychotherapeutic strategies have not been shown to be of value in the management of dysthymic disorder.

Complications/Adverse Outcomes of Treatment

Pure dysthymic disorder without prior major depressive disorder is a risk factor for developing major depressive disorder. Ten percent of patients with pure dysthymic disorder will develop major depressive disorder over the ensuing year. In clinical settings, individuals with dysthymic disorder frequently have superimposed major depressive disorder, which is often the reason for seeking treatment.

The condition that co-occurs most frequently with dysthymic disorder is major depressive disorder. The ECA survey estimated that 3.5% of the population had major depression only, 1.8% had dysthymic disorder only, and 1.4% of the population had both disorders, or double depression, at some point in their lives. Moreover, major depressive disorder occurred in 42% of patients with dysthymic disorder, and dysthymic disorders occurred in 28% of patients with major depressive disorder.

The National Comorbidity Survey (NCS) has confirmed that patients with dysthymic disorder have high rates of coexisting psychopathology. In that sample, only 8% of dysthymic patients were unaffected by another Axis I disorder. Besides major depression, the NCS revealed high rates of social phobia, posttraumatic stress disorder, and generalized anxiety disorder. Eating disorders, especially bulimia nervosa, are also associated with nonendogenous depression of only moderate severity. Furthermore, subjective observations and objective investigations have found that depressive symptoms in bulimia are highly reactive, unstable, short lived, and temporally related to bingeing and purging episodes.

Chronic mood symptoms can contribute to interpersonal problems or be associated with distorted self-perception. Thus the assessment of features of a personality disorder is difficult in such individuals. Nevertheless, concomitant personality disorders—especially cluster C personality diagnosis (i.e., avoidant, obsessive-compulsive, and dependent)—can complicate dysthymic disorder. The disorder can also be associated with the cluster B personality diagnoses (i.e., borderline, histrionic, and narcissistic).

Other chronic Axis I disorders (e.g., substance dependence) or chronic psychosocial stressors can be associated with dysthymic disorder in adults. In children, dysthymic disorder can be comorbid with attention-deficit/hyperactivity disorder (ADHD), conduct disorder, anxiety disorders, learning disorders, and mental retardation.

Prognosis

Dysthymic disorder often has an early and insidious onset (i.e., in childhood, adolescence, or early adult life) and a chronic course. Disease onset is usually before age 45, but a sizable subgroup of patients develop the disorder in late life (45% after age 45, and 19% after age 65). If dysthymic disorder precedes the onset of major depressive disorder, there is less probability that there will be spontaneous full recovery between major depressive episodes and a greater likelihood of having more frequent subsequent major depressive episodes.

The outcome literature of dysthymic disorder is not encouraging: Of dysthymic inpatients, only about 40% recovered after 2 years of follow-up, and only about 30% after 5 years of follow-up. Thus one can conclude that the available treatments for dysthymic disorder, as currently delivered, are only modestly effective and that relapse is quite common.

Essentials of Diagnosis

DSM-IV-TR Diagnostic Criteria

Manic Episode

1. A distinct period of abnormally and persistently elevated, expansive, or irritable mood lasting at least 1 week (or any duration if hospitalization is necessary).

2. During the period of mood disturbance, three (or more) of the following symptoms have persisted (four if the mood is only irritable) and have been present to a significant degree:

   1. inflated self-esteem or grandiosity

   2. decreased need for sleep (e.g., feels rested after only 3 hours of sleep)

   3. more talkative than usual or pressure to keep talking

   4. flight of ideas or subjective experience that thoughts are racing

   5. distractibility (i.e., attention too easily drawn to unimportant or irrelevant external stimuli)

   6. increase in goal-directed activity (either socially, at work or school, or sexually) or psychomotor agitation

   7. excessive involvement in pleasurable activities that have a high potential for painful consequences (e.g., engaging in unrestrained buying sprees, sexual indiscretions, or foolish business investments)

3. The symptoms do not meet the criteria for a mixed episode.

4. The mood disturbance is sufficiently severe to cause marked impairment in occupational functioning or in usual social activities or relationships with others, or to necessitate hospitalization to prevent harm to self or others, or there are psychotic features.

5. The symptoms are not due to the direct physiological effects of a substance (e.g., a drug of abuse, a medication, or other treatment) or a general medical condition (e.g., hyperthyroidism).

Note: Manic-like episodes that are clearly caused by somatic antidepressant treatment (e.g., medication, electroconvulsive treatment, light therapy) should not count toward a diagnosis of bipolar I disorder.

(Reprinted, with permission, from the Diagnostic and Statistical Manual of Mental Disorders, 4th edn., Text Revision. Washington DC: American Psychiatric Association, 2000.)

General Considerations

Bipolar disorders can be conceptualized into three distinct entities: Bipolar I disorder, consisting of episodes of mania cycling with depressive episodes; bipolar II disorder, consisting of episodes of hypomania cycling with depressive episodes; and cyclothymic disorder, consisting of hypomania and less severe episodes of depression (see section on Cyclothymic Disorder later in this chapter). Very few patients have only manic episodes.

Epidemiology

Table 18–18 summarizes the relevant risk factors. Among the key risk factors for bipolar disorder are being female, having a family history of bipolar disorder, and coming from an upper socioeconomic class. Data also suggest that people under age 50 years are at higher risk of a first attack of bipolar disorder, whereas someone who already has the disorder faces an increasing risk of a recurrent manic or depressive episode, as he or she grows older.

ECA studies support a lifetime prevalence of bipolar disorder ranging from 0.6% to 1.1% (between 0.8% and 1.1% in men, and between 0.5% and 1.3% in women). Estimates from community samples range from 0.4% to 1.6%, suggesting that the disorder affects over 3 million persons in the United States. Bipolar disorder accounts for one quarter of all mood disorders. Prevalence rates of bipolar disorders are usually underestimated, because of problems in identifying manic and, more so, hypomanic episodes.

Bipolar disorders are about equally distributed among males and females, but there are more females with serious bipolar disorder, especially rapid-cycling bipolar disorder. Higher rates of hypothyroidism, greater use of antidepressant medication, and changes in reproductive status and gonadal steroids may account for the greater prevalence of rapid-cycling bipolar disorder among women.

Approximately 10–15% of adolescents with the diagnosis of recurrent major depression will later develop bipolar I disorder. Mixed episodes are more likely in adolescents and young adults than in older adults. The first occurrence of a manic episode after the age of 50 years should alert the clinician to the possibility that the episode may be due either to a general medical condition or to substance use. Bipolar disorder is often misdiagnosed in adolescents (e.g., as ADHD). Patients with early-onset bipolar disorder are more likely to display psychotic symptoms and to have a poorer prognosis in terms of lifetime outcome.

Mood disorders often have seasonal patterns. Acute episodes of depression are common in spring and fall, whereas mania appears to cluster in the summer months. Corresponding data on suicides also show a peak in the spring and a (smaller) peak in the late fall.

Etiology

Despite intensive attempts to establish its etiologic or pathophysiologic basis, the precise cause of bipolar disorder is not known. As in major depressive disorder, there is consensus that multiple etiologic factors—genetic, biochemical, and socioenvironmental—may interact in complex ways. Bipolar disorder has a fairly strong genetic association, two-thirds or more of patients showing family history. Association studies have demonstrated linkage at several loci, most consistently at 18p/q and 21q, with others demonstrated as well (5p, 6p, 10q, 12q, 16p, 22p).

Life Events

Although psychosocial stressors can occasionally precede the onset of bipolar disorder, there is no clear association between life events and the onset of manic or hypomanic episodes. In fact, given the disruptive nature of mood cycling, bipolar disorder may be more likely to induce negative life events. For example, poor judgment during a manic episode may expose people to greater risk for trauma, sexual abuse, or debt.

Biological Theories

Neurotransmitters

Neurotransmitter theories initially conceptualized that depression and mania are on the opposite ends of the same continuum. For example, the norepinephrine hypothesis of affective illness centered around the availability of norepinephrine at synaptic sites, with less norepinephrine being available in depression, and more in mania. This theory, and later modifications of it, especially as they pertain to the separation of unipolar and bipolar disorders, are discussed earlier in this chapter. Simple models of variable transmitter levels have given way to more complex conceptualizations of the illness.

Neuroendocrine Factors

Abnormalities in the HPA axis and, more so, in the HPT axis are common in bipolar disorder; they are discussed later in this section.

Psychosocial Theories

Psychosocial theories pertaining to the etiology of bipolar disorders are the same as those described for major depressive disorder (see section on Major Depressive Disorder earlier in this chapter).

Environmental conditions contribute more to the timing of a bipolar episode than to the patient's inherent underlying vulnerability. In other words, more stressful life events appear to precede early episodes than later ones, and there is a pattern of increased frequency over time. It is possible that early precipitating events merely activate preexisting vulnerability, thereby making an individual more vulnerable for future episodes.

This clinical observation is supported conceptually by the kindling model. In laboratory animals, repeated electrical stimulation of the hippocampus will lead to the development of a spontaneous seizure disorder, in which seizures occur without external stimulation. Applied to bipolar disorder, this model provides a better understanding of several phenomena inherent in the illness, including the above-mentioned effects of repeated episodes on disease severity and outcome (e.g., rapid-cycling bipolar disorder usually develops late in the course of the illness) and the acute and prophylactic treatment effects of ECT and anticonvulsants such as carbamazepine and valproic acid. Because these therapies inhibit kindling in laboratory animals, their therapeutic effect in bipolar disorder may involve interruption of kindling.

The kindling model also has enormous treatment implications. Because it suggests that preventing or treating early episodes will favorably affect outcome, it stresses the need for early identification and intervention techniques. This pertains particularly to younger patients, because the risk of kindling is highest during adolescence.

Genetics

Twin and family studies provide strong evidence for a genetic component in bipolar disorder, but the precise mechanisms of inheritance are not known. In monozygotic twins, the concordance rate is higher for bipolar disorder (80%) than for unipolar disorder (54%). In dizygotic twins, concordance rates are 24% for bipolar disorder and 19% for unipolar disorder. Adoption studies have shown that the biological children of affected parents have an increased risk for developing a mood disorder, even when reared by unaffected parents, although the data are not uniform. Finally, first-degree biological relatives of bipolar I patients have elevated rates of bipolar I disorder (4–20%), bipolar II disorder (1–5%), and major depressive disorder (4–24%).

Clinical Findings

Signs & Symptoms

Bipolar disorders can be subdivided into two diagnostic entities: Bipolar I disorder (recurrent major depressive episodes with manic episodes) and bipolar II disorder (recurrent major depressive episodes with hypomanic episodes). The symptoms of both bipolar disorders involve changes in mood, cognition, and behavior (Table 18–19).

Bipolar I Disorder

Episodes typically begin relatively quickly, with a rapid escalation through the stages summarized in Table 18–19. In 50–60% of cases, a depressive episode immediately precedes or follows a manic episode. Mania without history of depression is very rare. The hallmark of a manic episode is abnormally and persistently elevated, expansive, or irritable mood. The elevated mood can be described as euphoric, cheerful, and with indiscriminate enthusiasm and optimism; therefore, others often perceive it as infectious. Although the patient's mood may be predominantly elevated, it may quickly become irritable, especially after demands are not satisfied.

Somatic Features

The sleep of bipolar patients varies with their clinical state. When depressed, bipolar patients may sleep too much; when manic, they sleep little or not at all and typically report feeling fully rested nevertheless. As the manic episode intensifies, patients may go without sleep for nights, their insomnia further intensifying the manic syndrome. There is speculation as to whether the insomnia precedes, and perhaps triggers or fuels, the manic episode. It is difficult to control an acute manic episode clinically if one cannot control the associated insomnia. Conversely, some bipolar patients may experience a manic or hypomanic episode after a single night of sleep deprivation.

Behavioral Features

Bipolar patients are initially social, outgoing, self-confident, and talkative and can be difficult to interrupt. Their speech is full of puns, jokes, and irrelevancies. Patients are often hypersexual, promiscuous, disinhibited, and seductive; they may present to an emergency room setting dressed in colorful, flamboyant, and inappropriate clothing. As the manic episode intensifies, their speech becomes loud, intrusive, rapid, and difficult to follow, and they can become irritable, assaultive, and threatening.

Cognitive Features

Manic patients are easily distracted. Their thought processes are difficult to follow because of racing thoughts and flight of ideas. They appear to have an unrestrained and accelerated flow of thoughts and ideas, which are often unrelated. Patients can be overly self-confident and become preoccupied with political, personal, religious, and sexual themes. They may exhibit an inappropriate increase in self-esteem and may have grandiose beliefs (e.g., of brilliance, success, wealth, etc.). Their judgment is impaired significantly, resulting in buying sprees, sexual indiscretions, and unwise business investments. Psychotic features such as paranoia, delusions, and hallucinations are sometimes present, not unlike those seen in patients with schizophrenia.

Risk of Suicide

Bipolar patients are at a substantial risk for suicide, with a mortality rate estimated to be 2–3 times higher than that of the general population. There is burgeoning evidence that lithium maintenance treatment attenuates the risk of suicide attempts and completions. When 16,800 bipolar patients were followed for a 2-year period, the annual risk of suicide or suicide attempts was 0.26–0.4 in patients taking lithium, and 1.68–1.5 in patients not taking lithium. Risk factors for suicide in bipolar patients include a history of previous suicide attempts, a history of rapid-cycling bipolar disorder, comorbid substance abuse, a mixed episode, and a current depressive episode.

Bipolar II Disorder

Characterized by recurrent episodes of major depression and hypomania, bipolar II disorder has been identified as a distinct disorder only in DSM-IV. Hypomanic symptoms are similar to manic symptoms but do not reach the same level of symptom severity or social impairment (i.e., the patient's capacity to function vocationally is seldom compromised). Hypomania, though associated with elated mood and increased self-assuredness, does not present with psychotic symptoms, or may not show severe racing thoughts, or marked psychomotor agitation. Hypomanic patients thus do not perceive themselves as being ill and are likely to minimize their symptoms and to resist treatment.

Rapid-Cycling Bipolar Disorder

By definition, patients with rapid-cycling bipolar disorder experience four or more affective episodes per year. Approximately 10–15% of bipolar patients experience rapid cycling. Although similar to other bipolar patients nosologically and demographically, patients with rapid-cycling bipolar disorder tend to have a longer duration and a more refractory course of the illness. Women are represented disproportionately, making up 80–95% of rapid-cycling patients, compared to about 50% of non-rapid-cycling patients. A variety of factors may predispose bipolar illness to a rapid-cycling course, including treatment with antidepressants. The development of clinical or subclinical hypothyroidism (spontaneously or during lithium treatment) in a manic patient predisposes to a more rapidly cycling course.

Psychological Testing

Psychological testing has not been shown to be of value in the management of bipolar disorder.

Laboratory Findings & Neuroimaging

No laboratory findings are diagnostic of a bipolar disorder or of a manic or hypomanic episode; however, several laboratory findings have been noted to be abnormal in groups of individuals with bipolar disorder. Compared to the wealth of findings in major depressive disorder, research findings in bipolar disorder are in general scarce and inconclusive. For one thing, bipolar disorder is not as common as major depressive disorder. Beyond that, patients with acute manic states are not as likely to collaborate in research studies; they experience lack of insight, agitation, irritability, and racing thoughts, among other symptoms; and they are normally in need of rapid containment of symptoms.

HPA Axis

The few available cross-sectional and longitudinal studies reveal increased plasma cortisol levels in some depressed bipolar patients. Abnormalities in the DST have also been noted. DST nonsuppression occurs more frequently in the mixed phase of the illness (78%), but also occurs in bipolar depression (38%) and mania (49%). Both hypercortisolemia and DST results normalize after the acute episode subsides, suggesting that these abnormalities are state and not trait dependent. Although these abnormalities are not specific for bipolar disorder or even major depression, their pathophysiology is suggestive of central (i.e., hypothalamic) rather than peripheral dysregulation of cortisol.

HPT Axis

HPT axis abnormalities are rather common in bipolar disorder, especially in rapid-cycling bipolar disorder, but the precise relationship of these abnormalities with the illness and its various clinical presentations is not known. Among these abnormalities are an attenuated nocturnal TSH peak, a blunted TSH response to TRH administration, and a high prevalence of various degrees of hypothyroidism.

There are intriguing associations among thyroid function, bipolar disorder, and gender. Hypothyroidism is very common in patients with bipolar disorder; it is especially common in female patients who present with a rapid-cycling course. Treatment with hypermetabolic doses of T4 has shown some promise in that it may reduce acute symptoms, number of relapses, and duration of hospitalizations.

Two main conclusions can be drawn from these observations. Clinically, comorbid hypothyroidism seems to negatively affect disease outcome by predisposing the individual patient to a rapid-cycling course. Substitution with T4 has proven useful in some patients, but often, high doses are necessary to induce clinical response. Conceptually these findings support the hypothesis that a relative central thyroid hormone deficit may predispose to the marked and frequent mood swings that characterize rapid-cycling bipolar disorder. We may ask whether the central thyroid hormone deficit serves only as a risk factor for the development of rapid cycling in a known bipolar patient, or whether it can predispose most affectively ill patients to any major behavioral change (i.e., the switch from depression into recovery, from recovery into depression, or from depression into mania).

Sleep EEG

Sleep EEG recordings have revealed normal results in acutely ill bipolar patients, including normal REM latencies, but not all studies agree. These data, of course, are in stark contrast to those reported in major depressive disorder, in which shortened REM latencies are common. However, bipolar patients in full clinical remission can exhibit an increased density and percentage of REM and a sleep architecture more sensitive to arecoline (an acetylcholine agonist known to produce a shortened REM latency). The discrepancies between unipolar and bipolar patients are thought to be the result of different levels of clinical severity, variable proportions of bipolar I and bipolar II patients, and age differences.

Brain Imaging

The most consistent neuroimaging findings in depressed (unipolar and) bipolar patients are decreased PFC blood flow and metabolism. Basal ganglia abnormalities have also been found in depressed bipolar patients, involving decreased blood flow and metabolism. In bipolar disorders, structural imaging studies suggest an increased number of white matter hyperintensities, which may be unique to bipolar disorder and its treatment, or related to cardiovascular factors. Metabolic and blood flow studies provide evidence for decreased activity of the PFC in depressed bipolar patients. It is presently not known whether these findings are state or trait related.

Course of Illness

The natural course of bipolar disorders is variable. Patients usually experience their first manic episode in their early 20s, but bipolar disorders sometimes start in adolescence or after age 40 years. Manic episodes typically begin suddenly, with a rapid escalation of symptoms over a few days, often, they are triggered by psychosocial stressors. In about 50% of patients a depressive episode immediately precedes a manic episode.

Although medications are effective in treating acute episodes and in maintaining patients in remission, they are not completely effective in preventing future episodes. Studies have shown that patients maintained with lithium levels of 0.8–1.0 ng/mL are less likely to have a recurrence than are patients maintained on lower lithium levels (0.4–0.6 ng/mL). Combinations of lithium and anticonvulsants have shown promise in maintaining euthymia and increasing cycle intervals.

Differential Diagnosis

Medical Disorders

Numerous medical disorders and medications can induce or mimic the clinical picture of bipolar disorder, exacerbate its course and severity, or complicate its treatment. It has been widely noted that psychopharmacologic interventions are less successful if the bipolar disorder is due to a primary medical condition; whenever possible, it is advisable to attempt to correct the underlying medical disorder before beginning psychopharmacologic treatment.

DSM-IV criteria specify that to make a diagnosis of bipolar disorder, the symptoms cannot be the direct result of a substance or a general medical condition. A late onset of the first manic episode (over age 50) should prompt the clinician to carefully exclude a possible medical cause. Table 18–20 lists the many organic causes of mania and hypomania.

Psychiatric Disorders

The Axis I and Axis II disorders associated with hypomanic or manic symptoms need to be included in the differential diagnosis of bipolar disorders (Table 18–21). First, the psychotic features associated with schizophrenia or schizoaffective disorder are often indistinguishable from those associated with acute mania. Second, major depressive episodes may be associated prominently with irritable mood and may thus be difficult to distinguish from a mixed bipolar episode. Third, in children and adolescents, ADHD and mania are both characterized by overactivity, impulsive behavior, poor judgment and academic performance, and psychological denial. Fourth, patients with certain personality disorders (e.g., borderline or histrionic personality disorders) can exhibit impulsivity, affective instability, and paranoid ideations, as do manic patients.

Finally, substance abuse is exceedingly common in patients with bipolar disorder. As many as 41% of bipolar patients abuse or are dependent on drugs, 46% abuse or are dependent on alcohol, and as many as 61% abuse or are dependent on any substance. Careful etiologic evaluation and a drug-free washout period after intoxication are often necessary to distinguish whether the mood disturbance is the consequence of substance abuse or whether the substance abuse is the consequence of a mood disturbance. Such differentiation is important because psychiatric comorbidity complicates the acute manic state and negatively affects disease outcome. The mood disorders complicated by substance abuse are therefore treated in units especially designed to treat aspects of dual diagnosis.

Treatment

Psychopharmacologic Interventions

The clinical management of bipolar disorder involves treatment of the acute episodes and maintenance therapy. Acute bipolar episodes usually demand immediate symptom containment, which is achieved most easily using pharmacologic means (Table 18–22). TCAs typically should be avoided in depressive episodes, because of the potential for the induction of mania, mixed state, or rapid cycling. This appears to occur with other antidepressants as well. A limited amount of data suggests that lamotrigine may be effective in bipolar depression, in doses in the range of 200–400 mg/day. It also has mood-stabilizing effects, making it a reasonable medication to use during depressive episodes.

All atypical antipsychotics appear to have acute anti-manic effects, and most have been shown to be effective as maintenance treatment. One drug, quetiapine, recently received US Food and Drug Administration (USFDA) approval for bipolar depression as a monotherapy agent. Olanzapine also has been shown to reduce depressive symptoms, although this appears primarily to be vegetative symptoms such as sleep and appetite disturbance, more than so-called “core” depressive symptoms such as down mood or negative thinking. However, the combination of olanzapine with the SSRI fluoxetine has been shown to be quite efficacious, and appears to produce its effects very rapidly. This combination treatment has received approval by the USFDA. However, the use of the combination has been limited. In part, this may be the result of significant issues with weight gain and the potential for metabolic syndrome associated with olanzapine. Acute manic episodes can be managed with lithium, divalproex, or an atypical antipsychotic. If delusional symptoms and agitation are present, antipsychotics (particularly atypical antipsychotics such as olanzapine or risperidone) or benzodiazepines (e.g., clonazepam) need to be added.

After the resolution of depressive or manic episodes, the maintenance treatment phase is aimed at prevention of future episodes. Drugs shown to be effective for maintenance management include lithium, divalproex, lamotrigine, or atypical antipsychotics. Often, treatments will need to be combined for optimal stabilization. (see Table 18–22).

Mood charting, though cumbersome and time consuming, allows careful delineation of the frequency and severity of episodes and of the drug's efficacy, including evidence of partial response patterns or loss of efficacy. Benzodiazepines (e.g., clonazepam or lorazepam) are often useful in relieving the insomnia and agitation associated with acute illness. Clinicians agree that control of insomnia is paramount in treating acute bipolar illness, and there is speculation that continued insomnia may fuel the illness. Finally, agents with the propensity to destabilize mood (e.g., antidepressants, steroids, alcohol) need to be identified and discontinued if possible (although continuation antidepressant treatment may be needed).

Lithium

Indications

There are two main indications for lithium in patients with bipolar disorder: (1) for the management of the acute manic or hypomanic episode, and (2) for the prevention of further episodes of both mania and depression. Lithium is also useful in potentiating or augmenting the effects of antidepressants (mainly in refractory depression, as noted earlier in this chapter); in managing impulse-control disorders (especially episodic violence); in the long-term prophylactic treatment of cluster headaches; and in treating schizophrenia in a limited number of patients, particularly if they experience prominent affective symptoms (in which case a neuroleptic is also needed to treat the intrinsic schizophrenic symptoms).

Acute Management of (Hypo)Mania

Lithium is useful in acute manic episodes. Approximately 80% of manic patients show at least a partial response to lithium monotherapy. Lithium also has some acute antidepressant effects, but these actions tend to be incomplete and require 3–4 weeks of treatment to achieve maximal effect. Often antidepressants are needed to effectively treat or prevent depressive episodes, but their use is always associated with the risk of precipitating (hypo)manic episodes or of destabilizing the illness course. Bupropion or SSRIs are less likely to produce these unwanted side effects relative to TCAs. In acute manic states, lithium monotherapy can be beneficial in a matter of days, but usually there is a 5–10-day latency of response. Because of this delayed response lithium has not become the sole agent in everyday clinical practice for acute mania. Most clinicians now use antipsychotics or benzodiazepines in addition to lithium in the acute phase of mania to control psychosis, hyperactivity, insomnia, and agitation. Antipsychotics appear to be superior to lithium only in the initial management of the acute state (i.e., in the first few days of treatment).

Patients are often concerned about continuation of treatment indefinitely. The clinician should carefully explain to the patient that he or she may have a chronic condition, that a manic or depressive episode is likely to recur in the future, that it is necessary to be cognizant of the early symptoms of a developing episode (e.g., insomnia, elation or depression of mood, increased or decreased activity), and that treatment should be sought as soon as such symptoms occur. It is also important to point out, ideally to both the patient and family members, that patients are not likely to seek treatment in the early hypomanic states because the associated self-assuredness, mood elation, and boundless energy are generally perceived as pleasant. If lithium is withdrawn abruptly, approximately 50% of patients relapse within 5 months.

Prevention of Relapse

The efficacy of lithium in the long-term prophylactic treatment of bipolar disorder has been well documented; it appears to involve a reduction in both the number of episodes and their intensity. The preventive effects of lithium are most robust when there is evidence that the interval between episodes has shortened, that is, when the course of bipolar disorder has intensified in both severity of symptoms and frequency of episodes.

Management of Breakthrough Depression

A depressive episode sometimes “breaks through” lithium maintenance treatment. If this occurs, the following steps should be taken: (1) Increase lithium dose; (2) maximize thyroid function; (3) add an SSRI (the drug of choice), bupropion, or an MAOI; (4) consider treatment alternatives (e.g., valproic acid, carbamazepine); and (5) consider sleep deprivation or ECT for severe depression. Because acute side effects and toxicity are common with lithium treatment, a thorough pretreatment and follow-up evaluation are necessary (Table 18–23).

In healthy patients, lithium carbonate can be initiated at 300 mg twice daily and titrated with frequent blood level monitoring. The blood level should be adjusted and maintained in a range of 0.6–1.2 mmol/L. This will require widely varying doses (600–1800 mg/day). However, most patients are maintained effectively at 900–1200 mg/day.

Occasionally treatment calls for the use of medication in liquid form. Lithium citrate, a suspension form of the drug is available. Not only does this preparation allow for finer gradations of titration, but it reduces the likelihood of nausea and vomiting, which are associated with lithium treatment.

Adverse Effects

Lithium is well tolerated by most patients. However, careful management of lithium plasma concentrations is required because of its narrow therapeutic window and because of the close association between plasma levels and toxicity (Table 18–24). Within the normal range of lithium plasma levels, one can commonly observe persistent but benign side effects, including increased thirst or urination, fine tremor, weight gain, and edema. Above the normal range of lithium plasma levels, serious side effects can occur rapidly; they include (with increasing plasma concentrations and symptom severity) nausea, vomiting, diarrhea, drowsiness and mental dullness, slurred speech, confusion, coarse tremor and twitching, muscle weakness, and above levels of 3.0 mmol/L, seizures, coma and death.

The major side effects of lithium affect the endocrine, renal, hematologic, cardiovascular, cutaneous gastrointestinal, ocular, and nervous systems (Table 18–25). Side effects are usually dosage dependent and transient in nature. Lithium has teratogenic effects, particularly when taken during the first trimester of pregnancy, it has been linked to the serious cardiac birth defect Ebstein's Anomaly and should be avoided in pregnant women. The risk factors predisposing to lithium side effects and toxicity include renal disease or reduced renal clearance with age; organic brain disorder; dehydration after vomiting, diarrhea, increased perspiration, and strenuous exercise; low sodium intake or high sodium excretion; prolonged dieting, especially salt-restriction diets; and early pregnancy.

Management of Lithium Side Effects

Because most side effects are dosage dependent, reduction of lithium intake will quickly ameliorate the acute symptoms, but this may increase the risk of relapse (see Table 18–25). Nausea, vomiting, and diarrhea can be reduced in some patients by switching from the carbonate to the citrate salt of lithium. Polydipsia and polyuria can be managed by giving the entire daily dosage of lithium at bedtime. Tremor is often responsive to β-blockers such as atenolol or propranolol. These and other common side effects (i.e., memory problems, weight gain, and tremor)—although not immediately harmful and dangerous—can be quite troublesome, may be intolerable to patients, and often negatively affect compliance. Goodwin and Jamison (1990), pooling percentages from 12 studies including 1094 patients, showed that subjective complaints were common. Among the most frequent were thirst (36%), polyuria (30%), memory problems (28%), tremor (27%), weight gain (19%), drowsiness (12%), and diarrhea (9%). Only 26% of patients had no complaints. Memory problems were most likely to cause noncompliance, followed by weight gain, tremor, polyuria, and drowsiness.

Thyroid dysfunction can be associated with lithium treatment. A small proportion of patients receiving chronic lithium treatment will develop thyroid enlargement with elevations in plasma TSH concentrations. Few patients, however, develop frank hypothyroidism. When this occurs, lithium may be discontinued, if possible, or thyroid hormone supplementation may be initiated.

The most common renal problems in patients taking lithium are polydipsia and polyuria, both of which are usually reversible. Rarely, patients can develop diabetes insipidus or an interstitial nephritis-like syndrome (see Table 18–25), both of which require regular monitoring of kidney functioning (see Table 18–23). We now use lower plasma concentrations during maintenance therapy (i.e., 0.6–0.8 nmol/L) than, say, two decades ago; this reduction in lithium dosing has resulted in a lower side-effect profile, especially of renal side effects.

Serious cardiac side effects are uncommon (see Table 18–25). T-wave flattening or inversion occur often and are not associated with negative treatment outcome. Some patients taking lithium over the long-term may experience sudden death of presumed cardiac origin. In particular, sinoatrial node dysfunction (sick sinus syndrome) can occur with increased frequency in these patients. Routine monitoring of the patient's electrocardiogram and pulse is necessary in order to minimize cardiac risk.

The use of lithium during pregnancy is controversial (see Table 18–25). Mild transient hypothyroidism and somnolence are common in newborns exposed in utero. The concentration of lithium in breast milk may also adversely affect nursing infants. The possibility of cardiovascular abnormalities in some infants exposed to lithium during the first trimester in utero necessitates both a careful initial risk-benefit analysis and close monitoring.

Contraindications to Lithium

Relative or absolute contraindications to lithium are severe renal disease, acute myocardial infarction (in which complications may occur owing to arrhythmias, use of diuretics and digoxin, reduced fluid or salt intake, cardiac failure, and reduced renal function), myasthenia gravis (in which lithium interferes with the release of acetylcholine and the depolarization and repolarization of the motor endplate), first trimester of pregnancy, and breast-feeding mothers. As well, nonsteroidal anti-inflammatory drugs can markedly elevate plasma levels of lithium and should be avoided.

Management of Lithium Treatment Failure

About 60% of bipolar patients respond to lithium treatment alone, although breakthrough of mania or depression is common. If patients do not respond to lithium treatment, the clinician has several alternative options. The first is to change the medication to divalproex. Carbamazepine also has been used as a mood stabilizer, although there are very limited controlled clinical trial data. Lamotrigine also has been shown to have some mood-stabilizing effects and is approved for maintenance use in bipolar disorder. These can be administered with or without continuing lithium. Table 18–26 summarizes other, less common options.

Preliminary evidence indicates that maintenance treatment with anticonvulsants may carry a higher risk of suicide than maintenance treatment with lithium, but no firm conclusions can be drawn yet. The following are predictors of a good response to anticonvulsants: Rapid-cycling course of the illness, mixed episode, previous poor response to lithium, secondary mania, and recurrent substance abuse.

Carbamazepine

Of the two anticonvulsants, carbamazepine has been used clinically for a longer period of time. Doses are typically started at 200–400 mg/day and titrated to achieve a plasma concentration of 3–14 μg/mL. Common side effects include somnolence, dizziness, nausea and vomiting, and mild ataxia; these effects can be managed effectively with dosage reduction or slow dosage titration. Mild elevations of liver enzymes and mildly reduced white blood counts are also common. These problems typically improve or resolve with time. Finally, rash is common and usually requires discontinuation.

Two potentially serious adverse reactions associated with carbamazepine treatment require close and regular monitoring during treatment (Table 18–27). Carbamazepine can acutely damage the liver, resulting in marked increases in liver enzymes and sometimes frank jaundice. Hepatic failure, however, is rare.

Blood dyscrasias, including granulocytopenia, agranulocytosis, or aplastic anemia, pose a second category of problems. Although rare (the incidence of agranulocytosis is approximately six per million population per year, whereas aplastic anemia occurs in two per million per year), the possibility of these adverse reactions necessitates that the patient have a complete blood count taken regularly (see Table 18–27). If the patient's absolute neutrophil count falls below 2000, the count should be monitored at least weekly until a rebound is observed. At below 1500, the carbamazepine dosage should be lowered quickly to about 50% of previous dosage and the count monitored weekly until it rebounds. If the count falls below 1000, carbamazepine should be discontinued and the patient monitored under hospital conditions until the count rebounds.

The teratogenic effects of carbamazepine are well known. Carbamazepine appears to induce neural tube defects such as spina bifida at high rates (about 1% of exposures); therefore, it should not be used in pregnant women. Human chorionic gonadotropin testing is indicated in women of childbearing potential who are to be started on this agent.

Valproic Acid (Divalproex Sodium)

Clinical trials have established the effectiveness of valproic acid in the acute management of mania, but long-term maintenance research is lacking. The agent is available as regular sodium valproate or as a dimer molecule divalproex sodium, which is most commonly used. The latter dissociates in the gastrointestinal tract to form free valproate, which is then absorbed. This slower release reduces gastrointestinal side effects that are common with valproic acid.

Divalproex is started in the 500–750 mg/day range and titrated in 250 mg increments to achieve an adequate plasma level. The therapeutic plasma-level range is 50–125 μg/mL. The higher end of the plasma-level range is indicated for patients whose illness is poorly controlled on lower levels.

Common side effects associated with sodium valproate include nausea and vomiting, diarrhea or constipation, mild elevations of liver enzymes, mild drowsiness or fatigue, and skin rash. More serious reactions are rare. Severe hepatitis that can result in hepatic failure has been reported and has resulted in fatalities; it is especially prevalent in young children. Liver function studies are required at baseline and within the first month of treatment. They should be repeated at 3 and 6 months, then at least annually thereafter. Mild elevations of liver enzymes are an indication for more frequent monitoring. Clotting abnormalities also have been reported. Finally, like carbamazepine, valproate may produce spina bifida in infants exposed in utero. Use in pregnancy must be avoided.

Atypical Antipsychotics

All of the currently available atypical antipsychotics (except for clozapine and the recently approved drug paliperidone) have been described to have acute anti-manic effects. Several appear beneficial in relapse prevention, although only olanzapine has a maintenance indication approved by the USFDA. Nonetheless, all atypicals are used for both acute and maintenance treatment. Side effects of atypicals include extrapyramidal effects, akathisia, somnolence, activation (particularly for ziprasidone and aripiprazole), and weight gain with a risk for the metabolic syndrome. The latter issue is a particular problem for olanzapine, while ziprasidone and aripiprazole produce little weight gain. One concern with the use of antipsychotics of any type (with the possible exception of clozapine) is tardive dyskinesia (TD). TD is a syndrome of abnormal involuntary movements typically involving the face (grimacing), tongue (writhing movements and protrusion), and lips (including lip smacking), but which can involve the trunk and extremities as well. The latter symptoms are typically choreiform in nature. TD may be permanent, so clinicians need to monitor patients for the emergence of symptoms carefully. The risk of TD is much lower with the atypical antipsychotics than with older, typical agents such as haloperidol, with rates under 1% in longer-term use. However, vigilance is needed and other treatments should be used first before considering atypicals. One exception to this rule is the presence of psychotic mania or depression.

Psychotherapeutic Interventions

Little is known about the role of psychotherapy in the treatment of bipolar disorder. Efforts are currently under way to evaluate whether treatments such as interpersonal psychotherapy or cognitive therapy work as well in bipolar patients as they do in unipolar patients. Preliminary evidence indicates that family education may reduce the risk for relapse and that cognitive therapy may enhance compliance with medications. On the whole, some of the newer psychosocial interventions appear to represent a viable alternative or valuable adjunct to pharmacotherapy in mild to moderate unipolar disorder. Whether those advantages apply to more severely depressed unipolar patients or to patients with bipolar disorders remains to be determined.

Other Interventions

The management of bipolar disorders heavily depends on the utilization of psychopharmacological agents and, if necessary, ECT. Other strategies have not been shown to be therapeutically useful.

Electroconvulsive Therapy

Approximately 80% of manic patients show substantial improvement after ECT, but the widespread (and successful) use of lithium and other antimanic agents, often in combination with antipsychotic medications, has limited ECT to patients who are intolerant of medications or whose illness is refractory to medications. ECT has proven especially useful for those manic patients who did not respond to medication and for those in mixed states who present with a high risk for suicide and are thus in need of acute symptom containment.

Complications/Adverse Oucomes of Treatment

Complications and adverse outcomes related to the various biological interventions have been detailed in the above paragraphs where emphasis was given to complications specific to the various therapeutic intervention strategies.

Prognosis

It appears that the availability of treatment for bipolar disorder has significantly affected the length of an average episode. Before 1960 (i.e., before the introduction of psychopharmacologic medications such as lithium), episode length averaged 7–13 months; thereafter, it averaged 2–4 months. With the emergence of early intervention and successful treatment of the illness, the effects of kindling on outcome (described earlier in this section) may have been interrupted, resulting in fewer episodes over time and reduced cycle length. Early disease onset suggests poor prognosis; patients who experience their initial episode in their late teens are likely to have a less favorable outcome than patients who experience their initial episode in their early 30s. Patients with early-onset bipolar disorder should be quickly identified and targeted for aggressive treatment intervention.

As discussed earlier in this section, somatic interventions can have a marked effect on the prognosis of bipolar disorder. Comorbidity with substance abuse, antisocial behavior, and personality disorders often complicates the clinical outcome. Substantial psychosocial morbidity can affect marriage, children, occupation, and other aspects of the patient's life. Divorce rates are two to three times higher in bipolar patients than in the general population, and occupational status is twice as likely to deteriorate. Completed suicide rates approximate 15% in untreated bipolar patients and is usually associated with a depressive or mixed episode.

Essentials of Diagnosis

DSM-IV-TR Diagnostic Criteria

1. For at least 2 years, the presence of numerous periods with hypomanic symptoms and numerous periods with depressive symptoms that do not meet criteria for a major depressive episode.

   Note: In children and adolescents, the duration must be at least 1 year.

2. During the above 2-year period (1 year in children and adolescents), the person has not been without the symptoms in Criterion A for more than 2 months at a time.

3. No major depressive episode, manic episode, or mixed episode has been present during the first 2 years of the disturbance.

   Note: After the initial 2 years (1 year in children and adolescents) of cyclothymic disorder, there may be superimposed manic or mixed episodes (in which case both bipolar I disorder and cyclothymic disorder may be diagnosed) or major depressive episodes (in which case both bipolar II disorder and cyclothymic disorder may be diagnosed).

4. The symptoms in Criterion A are not better accounted for by schizoaffective disorder, and are not superimposed on schizophrenia, schizophreniform disorder, delusional disorder, or psychotic disorder not otherwise specified.

5. The symptoms are not due to the direct physiological effects of a substance (e.g., drugs of abuse, a medication) or a general medical condition (e.g., hyperthyroidism).

6. The symptoms cause clinically significant distress or impairment in social, occupational, or other important areas of functioning.

(Reprinted, with permission, from the Diagnostic and Statistical Manual of Mental Disorders, 4th edn., Text Revision. Washington DC: American Psychiatric Association, 2000.)

General Considerations

Epidemiology

Cyclothymic disorder often begins early in life. Lifetime prevalence rates for this disorder range between 0.4% and 1.0% in the general population, and between 3% and 5% in mood disorder clinics. In community samples, cyclothymic disorder is equally common in men and women, but women are more prevalent in clinical settings (in a ratio of approximately 3–2).

Etiology

Cyclothymic individuals alternate between the extremes of dysthymia (gloomy and depressed) and hyperthymia (cheerful and uninhibited). They can be moody, impulsive, erratic, and volatile but usually do not meet the full syndromal criteria for bipolar disorder. Repeated romantic or conjugal failures occurring in the context of interpersonal and social disturbances are common. Cyclothymia first received an operational definition in the Diagnostic and Statistical Manual of Mental Disorders, third edition.

There is still considerable controversy concerning the association between cyclothymic disorder and both other mood disorders and cluster B personality disorders. However, current research favors the conceptualization of cyclothymic disorder as a bona fide mood disorder.

Genetics

Among first-degree relatives of cyclothymic patients, the diagnoses of major depressive disorder and bipolar I and II disorders are more common than in the general population (Table 18–28). There may also be an increased familial risk of substance-related disorders.

Clinical Findings

Signs and Symptoms

Cyclothymic disorder is characterized by periods of depression alternating with periods of hypomania, which are generally of less severity or shorter duration than those associated with bipolar I disorder. The changes in mood tend to be irregular and abrupt, sometimes occurring within hours of each other. Approximately 50% of patients with cyclothymic disorder are predominantly depressed; a minority has primarily hypomanic symptoms and is unlikely to consult a mental health practitioner. Almost all individuals with cyclothymic disorder have periods of mixed symptoms with marked irritability, leading to unprovoked disagreements with friends, family, and coworkers. Cyclothymic individuals may have a history of multiple geographical moves, involvement in religious cults, or inability to maintain a career pathway.

Psychological Testing

Psychological testing has not been shown to be useful in aiding the diagnosis of cyclothymic disorder.

Laboratory Findings

There are no laboratory findings suggesting or supporting the diagnosis of cyclothymic disorder.

Neuroimaging

There are no neuroimaging findings specific for cyclothymic disorder.

Course of Illness

Hypomania can be triggered by interpersonal stressors or by losses. Although most individuals with cyclothymic disorder seek psychiatric help for depression, their problems are often related to the interpersonal and behavioral crises (e.g., romantic failures) caused by their hypomanic episodes. The unpredictable nature of the mood changes causes a great deal of stress, and patients will often feel as if their moods are out of control. Patients will also present with disorganization and work productivity problems following hypomanic episodes.

Differential Diagnosis

When the mood disturbance is judged to be the direct physiologic consequence of a specific, usually chronic general medical condition (e.g., hypothyroidism), then the diagnosis should be mood disorder due to a general medical condition (see below). If substances, such as stimulants, are judged to be etiologically related to the mood disturbance, then a substance-induced mood disorder should be diagnosed. The mood swings suggestive of substance-induced mood disorder usually dissipate following cessation of drug use.

Although cyclothymic disorder may resemble bipolar I or II disorders with rapid cycling, the mood states in cyclothymic disorder do not meet the criteria for a manic, major depressive, or mixed episode. Borderline personality disorder is frequently confused with cyclothymic disorder, because of its strong affective component with marked shifts in mood. If criteria are met for each disorder, both may be diagnosed. In children and adolescents, ADHD can be clinically difficult to differentiate from cyclothymic disorder, but a pharmacologic trial of stimulants might be diagnostic, as stimulants tend to improve the symptoms of ADHD but may worsen the mood swings of cyclothymic disorder.

Treatment

Psychopharmacologic Interventions

There are no systematic data on the treatment of cyclothymic disorder. Lithium has been used to suppress the hypomanic cycles, but most patients seek treatment for their labile, irritable moods and their anxious, dysphoric depressions. Some of these patients may shift into acute manic states or may develop rapid cycling if their depressions are treated solely with TCAs. In this case, bupropion, MAOIs, and low-dose SSRIs, in conjunction with lithium or other mood stabilizers, may be appropriate.

Psychotherapeutic Interventions

Although traditional insight-oriented psychotherapeutic approaches have been used to treat the so-called temperamental and Axis II comorbid features of cyclothymic disorder, experts now also advocate psychoeducational and interpersonal strategies as adjuncts to pharmacotherapy directed at the affective instability.

Other Interventions

Interventions based on other than psychopharmacological or psychotherapeutic strategies have not been shown to be of value in the management of cyclothymic disorder.

Complications/Adverse Outcomes of Treatment

There are several sources of comorbidity for cyclothymic disorder. For example, mood disorders, especially major depressive disorder and bipolar II disorder, may follow the onset of cyclothymic disorder. Other comorbid disorders include personality disorders, especially borderline personality disorder (in 10–20% of patients), impulse-control disorders such as intermittent explosive disorder, and substance- related disorders (in 5–10% of patients). Patients may use substances such as alcohol and sedatives to self-medicate the fluctuating moods, or they may ingest stimulants in order to achieve even further stimulation. Sleep disorders (such as problems initiating and maintaining sleep) are occasionally present.

Prognosis

Cyclothymic disorder usually appears in adolescence or early adult life; its onset is insidious and its course chronic. The disorder is considered by some to reflect a temperamental predisposition to other mood disorders, and there is a 15–50% risk that a bipolar I or II disorder will develop subsequently.

Essentials of Diagnosis

DSM-IV-TR Diagnostic Criteria

1. A prominent and persistent disturbance in mood predominates in the clinical picture and is characterized by either (or both) of the following:

   1. depressed mood or markedly diminished interest or pleasure in all, or almost all, activities

   2. elevated, expansive, or irritable mood

2. There is evidence from the history, physical examination, or laboratory findings that the disturbance is the direct physiological consequence of a general medical condition.

3. The disturbance is not better accounted for by another mental disorder (e.g., adjustment disorder with depressed mood in response to the stress of having a general medical condition).

4. The disturbance does not occur exclusively during the course of a delirium.

5. The symptoms cause clinically significant distress or impairment in social, occupational, or other important areas of functioning.

Specify type:

• With depressive features: if the predominant mood is depressed but the full criteria are not met for a Major Depressive Episode
• With major depressive-like episode: if the full criteria are met (except Criterion D) for a major depressive episode
• With manic features: if the predominant mood is elevated, euphoric, or irritable
• With mixed features: if the symptoms of both mania and depression are present but neither predominates
• Coding note: Include the name of the general medical condition on Axis I, e.g., mood disorder due to hypothyroidism, with depressive features
• Coding note: If depressive symptoms occur as part of a preexisting vascular dementia, indicate the depressive symptoms by coding the appropriate subtype, i.e., vascular dementia, with depressed mood

(Reprinted, with permission, from the Diagnostic and Statistical Manual of Mental Disorders, 4th edn., Text Revision. Washington DC: American Psychiatric Association, 2000.)

General Considerations

Epidemiology

Table 18–29 summarizes the prevalence rates of major depression in a variety of medical conditions. As shown there, depression is particularly common in certain neurological disorders, endocrine disorders, and cardiovascular disorders. Approximately 27–57% of patients with certain neurologic conditions (e.g., Parkinson's disease, multiple sclerosis, Huntington's disease, and epilepsy) develop symptoms of severe depression at some point during their illness. For medical conditions that do not directly affect the brain, prevalence rates appear to be more variable, ranging from less than 8% (for chronic renal disease), to 19% (for coronary artery disease [CAD]), to 40% (for primary hypothyroidism).

Etiology

Diagnostic criteria require that the disturbance—based on history, physical examination, or laboratory findings—(i) is the direct physiological consequence of a general medical condition, (ii) is not accounted for by another mental disorder, and (iii) does not occur exclusively during the course of a delirium.

Genetics

Little is presently known about the genetics of mood disorders due to a general medical condition.

Underdiagnosis

Although symptoms of depression are rather common in medically ill patients, depression is usually underdiagnosed and undertreated in medicine clinics or primary care settings. It has been estimated that fewer than 50% of all depressed patients are identified and adequately treated in primary care settings. The primary focus on physical signs and symptoms may only partly explain this problem.

The problem of underdiagnosis appears especially problematic in patients with cancer or cardiovascular disease where treatment of depression can improve quality of life, and perhaps contribute to rising life expectancy.

Morbidity and Mortality

There is a burgeoning body of evidence suggesting that depression co-occuring with specific medical disorders can negatively affect morbidity and mortality. For example, the presence of either major or minor depression has been shown to (a) predict myocardial infarction, angioplasty and death during the 12 months following catheterization, (b) be a risk factor for mortality in patients with myocardial infarction at 6-month follow-up, (c) increase the risk for death 3.4 times in stroke patients during a 10-year period following the stroke (patients with poststroke depression and few social contacts are especially vulnerable—about 90% died), (d) increase—when co-occurring with a life-threatening illness—the risk for death or further life-threatening illnesses, and (e) increase the likelihood of death in patients with cancer. Although the precise factors that mediate the increased risk for morbidity and mortality in medically ill depressed patients remain obscure, there is preliminary evidence that changes in the neuroendocrine and/or immune system may play a critical role.

Depression as Initial Manifestation of Physical Illness

Depression may precede the clinical signs and symptoms of some medical illnesses by months and sometimes years. Depressive symptoms are particularly common in the early stages of cancer of the pancreas, cancer of the breast, Huntington disease, or such endocrine disorders as Cushing's disease, Addison's disease, and hypothyroidism. In the endocrine conditions depression can sometimes be seen even before the classic signs and symptoms are present.

Characteristics of Depression Secondary to Medical Illness

Table 18–30 details some of the characteristic differences between major depression and depression secondary to medical illness.

Clinical Findings

Signs and Symptoms

The diagnostic criteria for mood disorder due to a general medical condition are less vigorous than those for the primarily psychiatric mood disorders and require only the presence of depressed mood/diminished enjoyment or elevated, expansive, or irritable mood (manic symptoms). Regarding the medical illness, the required signs, symptoms, and laboratory findings are simply those that, in conjunction with the clinical history, yield the medical diagnosis. Furthermore, history, physical examination, or laboratory results must suggest that the mood disturbance is a physiologic consequence of the medical illness. Causality is usually established if the clinician demonstrates the presence of a medical condition known to cause depression and if the symptoms improve as the medical condition is treated. The many medical conditions and medications that are associated with or can cause mood problems are delineated in Tables 18–29 and 18–30.

Specific Medical Conditions Associated with Depression

Cardiovascular Disorders

Depression is common in CAD (Table 18–29). It has been shown that 18% of patients with coronary angiograms that proved CAD met criteria for major depression. Similarly, 16–19% of patients with acute myocardial infarction met criteria for major depression. In CAD, the depressed group was characterized by a positive history of depression, female gender, large infarcts, and poor social relationships.

Comorbid depression in cardiac disease cannot be considered a benign condition, as depression and its associated symptoms have been shown to be a major risk factor for both development of cardiovascular disease and death after myocardial infarction. In patients with cardiovascular disease, depression needs to be identified early and treated aggressively; such treatment may improve quality of life by stabilizing mood imbalance, and perhaps positively affect longevity.

Neurological Disorders

Among neurological disorders, depression following stroke is the most extensively studied. Among poststroke patients, prevalence rates for depression range from 30% to 50%; these rates include major depression (26%) as well as minor depression (20%). Six months following the stroke, 34% still fulfilled criteria for major depression, and 26% for minor depression. There is evidence that patients with left hemisphere lesions show a higher frequency and severity of depression than patients with right hemisphere lesions; the highest frequency of major depression was seen following left anterior lesions. Psychopharmacologic management using nortriptyline or trazodone has proved beneficial. The high period for developing depression extends to about two years after the stroke.

Other neurological disorder with high prevalence rates for depression are epilepsy, multiple sclerosis, Parkinson's disease, and Huntington's disease (Table 18–29).

Endocrine Disorder

Endocrine diseases commonly associated with depression are Cushing's disease, primary hypothyroidism (occuring in 1–6% of middle-aged females), diabetes mellitus, and Addison's disease (Table 18–29).

Depression may be among the earliest symptoms of endogenous Cushing's disease, sometimes preceding its final diagnosis by years. Patients with endogenous Cushing's disease usually present with depression (prevalence rates are as high as 65%), whereas patients with iatrogenic Cushing's disease (e.g., resulting after high doses of glucocorticoids for allergic or autoimmune conditions) are more likely to present with mood elevation. It is noteworthy that depressive symptoms not always remit when cortisol levels normalize, suggesting that in Cushing's disease psychological remission does not necessarily parallel endocrine remission or cure. The same phenomenon has been observed in patients whose primary hypothyroidism was treated with thyroid hormone substitution.

About one quarter of patients with diabetes mellitus complain about symptoms of depression. For long period of time, it has been known by primary care physicians and endocrinologists that depression can complicate both course and treatment of diabetes mellitus. During episodes of depression, it may be difficult to appropriately control the endocrine aspects of diabetes mellitus. A burgeoning number of clinical studies shows that comorbid depression can be associated with a less predictable course and wider fluctuations of blood glucose levels, often necessitating a more intensive treatment approach.

Medications and Substances Associated with Depression

Table 18–3 details the list of drugs that are known to cause or exacerbate depressive symptoms. Among them, the most common are analgesics (indomethacin, opiates), antihypertensives (propranol, reserpin, methyl-dopa, clonidine), and steroids.

Psychological Testing

Psychological testing has not been shown to be useful in aiding the diagnosis of mood disorder due to a general medical condition.

Laboratory Findings

There are no laboratory findings characteristic of, or specific for, mood disorders due to a general medical condition; this, of course, excepts the laboratory findings characteristic for the primary medical condition.

Neuroimaging

Little is known about neuroimaging findings in mood disorders due to a general medical condition.

Course of Illnesss

In most cases, successful treatment of the (primary) medical condition will beneficially affect the (secondary) mood disorder. However, behavioral remission does not always parallel medical remission. For example, patients with Cushing's syndrome who have been cured by surgical removal of the pituitary microadenoma, may continue to fulfill criteria for depression for several months after successful surgery.

Differential Diagnosis

The differential diagnosis of a mood disorder due to a general medical condition must include delirium (a separate diagnosis of mood disorder due to a general medical condition is not given if the mood disturbance is part of a delirium), dementia of the Alzheimer's type, vascular dementia, substance- induced mood disorder, and the other major mood disorders. Table 18–30 compares the characteristics of depression secondary to medical illness with those of (primary) major depression.

Treatment

Treatment approaches are directed primarily at the medical condition. It is noteworthy that particularly in endocrine disorders behavioral remission often does not parallel medical remission, that is, patients may continue to be depressed even after they have been cured endocrinologically. It is possible that in some patients (i.e., in those with a genetic vulnerability for developing depression) the mood disorder was simply triggered by the general medical condition, whereas in others it was more causally linked.

Psychopharmacologic Interventions

The psychopharmacologic treatment of comorbid mood disorders in medically ill patients requires a careful risk-benefit assessment. The considerations necessary for such assessment include the following: (1) side effects of psychotropic agents that may complicate existing medical illness (examples include the use of TCAs in patients with hypothyroidism, orthostatic hypotension, chronic heart disease including myocardial infarction and congestive heart failure, prostatic hypertrophy, or narrow-angle glaucoma); (2) potential interactions of psychotropic agents (e.g., antidepressants) with drugs used to treat medical disorders; and (3) the effects of impaired renal, hepatic, or gastrointestinal functioning on psychotropic drug absorption and metabolism. Medically ill patients are often older and more sensitive to drugs, they require lower dosages, and they metabolize drugs slower than do younger individuals. Therefore, when treating mood disorders in elderly, medically ill patients, clinicians should use low dosages of antidepressants at the beginning of treatment and raise the dosages gradually depending on the patient's toleration of side effects and response to treatment. The following sections summarize the management of comorbid depression in specific medical conditions:

Asthma

MAOIs should be administered with caution to patients who have asthma because of interactions with sympathomimetic bronchodilators. Other antidepressants appear to be safe.

Cardiac Disease

Because of their quinidine-like effects, the TCAs have antiarrhythmic properties and prolong the Q-T interval. A prolonged Q-T interval (i.e., >0.440 seconds) presents a relative contraindication for the use of TCAs. Among patients with preexisting but asymptomatic conduction defects, TCAs may induce symptomatic conduction defects and orthostatic hypotension. TCAs may also provoke arrhythmia in patients with subclinical sinus node dysfunction. SSRIs, trazodone, and bupropion present safer alternatives for treating depression in patients with cardiac disease. It is also advisable to monitor carefully those patients who already take another class I antiarrhythmic agent; to evaluate patients for preexisting but asymptomatic conduction defects such as interventricular conduction delay and bundle branch block; to be aware that patients with prolonged Q-T intervals, whether preexisting or drug induced, are predisposed to the development of ventricular tachycardia; and to obtain an EKG for all patients over age 40 years before initiating treatment.

Dementia

Patients with dementia are particularly sensitive to the negative effects of anticholinergic agents on memory and attention. Antidepressants with low anticholinergic profiles are the drugs of choice (e.g., SSRIs, bupropion, trazodone, desipramine, nortriptyline). Stimulants can be useful occasionally. ECT can be used if medications are either contraindicated or not tolerated, and if immediate resolution of depressive symptoms is medically indicated.

Hypertension

Antihypertensive medications and antidepressants may interact in various ways, either intensifying or weakening the effects of the antihypertensive therapy. The antihypertensive effect of prazosin is intensified by TCAs, because both block the α-receptors. In contrast, TCAs antagonize the therapeutic actions of guanethidine, clonidine, or α-methyldopa. Orthostatic hypotension can occur when antihypertensives, especially diuretics, are combined with TCAs, trazodone, or MAOIs. β-Blockers, especially propranolol, are known to cause depression in some individuals; the situation is remedied by changing to another antihypertensive medication.

Narrow-Angle Glaucoma

The anticholinergic side effects of TCAs can acutely exacerbate narrow-angle glaucoma in susceptible individuals (i.e., in those with shallow anterior chambers). Patients who receive miotics for their glaucoma may take antidepressants, including those with strong anticholinergic effects, provided that their intraocular pressure is monitored; SSRIs, trazodone, and bupropion present safer alternatives. Trazodone, however, can produce priapism in some patients (about 1 in 7000 men treated); priapism may necessitate surgical intervention, which is often associated with impotence or erectile problems.

Obstructive Uropathy