Mood Disorders

From WiserWiki

[edit] Mood Disorders

Steven A. Cole

Five mood disorders are of importance to primary care physicians: major depression, dysthymia, adjustment disorder with depressed mood, depression caused by a general medical condition (or substance), and bipolar disorder. Depression is one of the most painful conditions that can afflict an individual. Depressed persons generally report that their depression has caused more suffering than any other problem. Depression is common, disabling, and often unrecognized and untreated in primary care practices. This chapter reviews the epidemiology, diagnosis, and treatment of major depression in detail and briefly discusses other mood disorders.

Major depression (MD) is the most severe form of depression and is associated with considerable disability, morbidity, and mortality. In several careful epidemiologic studies, MD has been associated with as much disability (days in bed; days absent from work; impaired social, role, and interpersonal functioning) as other chronic illnesses, including arthritis and coronary artery disease. In addition, MD is also associated with considerable physical morbidity and mortality, most specifically related to the risk of suicide, but also as a risk factor for poor outcome and death in nursing home patients and those with general medical conditions (e.g., coronary artery disease, traumatic brain injury).

Dysthymia, or "chronic" depression, represents a milder form of depressive illness that does not meet the severity criteria of MD but is marked by considerable chronicity. Dysthymia is diagnosed when a depressed mood and at least two other symptoms of depression are present "most days" during the previous 2 years. Dysthymia can be considered a form of "minor depression" but is itself associated with significant pain and impaired functioning and may not remit spontaneously.

Adjustment disorder with depressed mood describes a psychiatric condition resulting from an identifiable stressor (e.g., divorce, job loss) that represents a level of impairment greater than what could normally be expected for most individuals. It can only be diagnosed within the first 6 months after a stressor has occurred. If the condition lasts longer than 6 months, the diagnosis would change to depressive disorder, NOS (not otherwise specified). A "normal" reaction to a distressing life event should not be diagnosed as an adjustment disorder. If an identifiable stressor precipitates a depressive syndrome that meets the severity criteria for MD, the diagnosis of MD is made (and not adjustment disorder).

Mood disorder caused by a general medical condition (or substance) refers to a psychiatric syndrome that the physician judges to result from the direct physiologic consequence of a general medical condition or medication (e.g., hypothyroidism or reserpine). The treatment should focus on resolution of the underlying general medical problem or withdrawal of the causative medication, but specific psychiatric treatment is also usually needed and recommended.

Bipolar disorder, previously called "manic-depressive" illness, occurs in about 1% of the population and represents a common and quite severe form of mental illness that has a very strong biologic and genetic substrate. About 10% of children who have one parent with bipolar disorder will develop the illness.

[edit] EPIDEMIOLOGY AND ETIOLOGY

Most epidemiologic studies indicate that the lifetime prevalence of MD is 5% to 10% in men and 10% to 20% in women. The reasons for these gender differences are unclear; numerous factors point to endocrine and biologic factors, whereas others attribute the differences to sociocultural factors. The 1-year prevalence of MD in the community is 2% to 4% for men and 4% to 8% for women. Studies of medical outpatients demonstrate similar rates, whereas studies of medical inpatients and studies in some diseases thought to predispose biologically (e.g., cerebrovascular accident [CVA, stroke], Parkinson's disease, traumatic brain injury) or psychologically (cancer) indicate much higher rates (10% to 50%). Although clinical lore has long suggested depression is more common in elderly persons, recent epidemiologic findings demonstrate lower 1-year prevalence rates of depression in elderly persons (1% to 2%).

Many patients and physicians consider depression to be "expected" in the face of significant stress. Stressful life events certainly predispose to major depression, but MD is not the uniform outcome (either clinically or statistically) of any stressful event. In fact, studies of individuals under stress (e.g., from terminal cancer, natural disaster) show rates of MD greater than the general population rate, but almost always less than 50%. That is, despite the seriousness of the circumstance, most individuals do not develop an MD syndrome under stress. Sad or depressed affect is an expected accompaniment of a stressful event, but the full syndrome of MD does not uniformly emerge. In this sense, no "good reasons" exist for MD. When such a syndrome develops, it is best considered a major (clinical) depression and not a "reactive depression."

Historically, the term reactive depression suggested (1) a mild version of the syndrome; (2) one that resulted entirely from a psychologic precipitant; (3) one that did not have a biologic substrate; and (4) one that should be treated with psychotherapy. None of these four assumptions is true about depressive syndromes precipitated by life events or physical illness. A very severe depressive syndrome can result from a stressful event (just as a myocardial infarction [MI] can result from a stress); a biologically predisposed individual may have MD in response to a life event; a MD resulting from a life stress may develop a biologic substrate; and a MD resulting from a life stress may respond as well or better to biologic therapy than to psychotherapy.

Thus the presence or absence of identifiable precipitants (stressors) is irrelevant to the diagnosis of MD. MD may be precipitated by stressful life events, but the diagnosis should be based only on the presentation of the signs and symptoms of MD. When MD results from a life stress or general medical illness, it is best to consider the MD a dread complication of the stress (or illness) and to diagnose aggressively and treat the depression as a comorbid condition.

Many physical symptoms of MD (see following discussion), such as fatigue, anorexia, and psychomotor retardation, can often be attributed to a comorbid general medical condition such as cancer or Parkinson's disease. This contribution of a general medical illness to the symptom profile of depression often leads physicians to discount their relevance and thus overlook the possibility of a treatable depression. Emerging data and the revised Diagnostic and Statistical Manual of Mental Disorders, fourth edition (DSM-IV) criteria for MD point to the importance of including these symptoms in the diagnostic approach to depression in medically ill patients. Although this inclusive approach might tend to overdiagnose MD, studies of CVA, Parkinson's disease, and traumatic brain injury indicate that the problem of overdiagnosis, if it exists, is quite low (about 2%).

[edit] PATHOPHYSIOLOGY: BIOPSYCHOSOCIAL MODEL OF DEPRESSION

Because MD is still a syndromal diagnosis, the illness itself probably represents a heterogenous group of disorders, many of which include a variable array of etiologic determinants. No clear anatomic, physiologic, or biochemical lesion has been found that can explain MD. Most investigators agree that MD is a complex biopsychosocial syndrome that, at present, can only be diagnosed on clinical criteria. Many promising studies point to possible etiologic dimensions and suggest future therapeutic interventions.

Research to date indicates mixtures of environmental and biologic factors underlying severe mood disorders. Genetic and family experiences certainly play a role, but not a determining one. Even for bipolar disorder, monozygotic twins share less than a 50% concordance and dizygotic twins about 10% concordance (about the same as siblings). With respect to MD in women, recent data indicate that negative life events also play a significant etiologic role. Animal studies demonstrate that early environmental stress predisposes to biologic and behavioral abnormalities associated with depression that may not emerge until adult life.

Numerous biologic "markers" of depression may underlie MD but do not necessarily "cause" it. An environmental stress may lead to psychic distress, which then precipitates a biologic cascade leading to MD. Although no marker is specific enough for diagnostic use, several markers reliably and statistically differentiate groups with and without MD. Some of these markers include endocrine factors: elevated cortisol, failure to suppress cortisol after the administration of exogenous dexamethasone (DST), blunted response of thyroid-stimulating hormone (TSH) to challenge with thyroglobulin-releasing factor (TRF), and increased growth hormone (GH) response to prolactin. Neurotransmitter levels such as norepinephrine (NE) and serotonin (5-HT) may be altered, but more likely, NE or 5-HT receptor function or number is altered during depression. One marker of altered central nervous system (CNS) 5-HT receptor function is platelet imipramine or platelet paroxetine binding. Sleep physiology is also changed in MD, with an early induction of rapid eye movement (REM) sleep and an overall increase of REM density during sleep. Finally, neuroimaging studies using positron emission tomography (PET) have demonstrated anatomically specific metabolic differences between depressed individuals and control subjects.

Loss of a parent (especially mother) is weakly predictive of future depression, whereas a low level of perceived social support (especially a wife's perception of an unsupportive husband) is predictive of future depression. Recent interpersonal, behavioral, and cognitive approaches to understanding depressive etiology and pathology have also generated important and promising therapeutic approaches, some of which are as effective as medication for the treatment of mild but not severe depression.

[edit] PATIENT HISTORY

DSM-IV criteria for MD requires five of nine symptoms for a 2-week period (Box 48-1). One of the nine symptoms must be either a persistent depressed mood or a pervasive anhedonia (loss of interest or pleasure in living). "Persistent" is defined as "present most of the day, nearly every day." Four hallmarks of MD for the purpose of clinical evaluation are (1) depressed mood, (2) anhedonia, (3) physical symptoms (sleep disorder, appetite problem, fatigue, psychomotor changes), and (4) psychologic symptoms (trouble concentrating or indecisiveness, guilt or low self-esteem, and hopelessness). The physical symptoms are important because they are predictive of a good response to biologic treatments. In particular, when middle insomnia is present (awaking at 3 or 4 am with inability to return to sleep) and when a diurnal variation in mood is present (feeling more depressed in am), patients are more likely to respond to biologic intervention. The psychologic symptoms are helpful in recognizing MD in patients with general medical illnesses.

[edit] Medical Interview

The medical interview is the key to making the diagnosis of MD. Although MD is diagnosed by a positive response to five of nine symptoms, physicians often miss the diagnosis if they do not routinely screen for depression. About 50% of depressed patients are not recognized because of time limitations for the interview, lack of knowledge and skill, fear of "opening Pandora's box," and the stigma associated with psychiatric illness.

Often, nonverbal cues can suggest depression to the busy physician. Downcast eyes, slow speech, wrinkled brow, and tearful looks all express a sad mood. However, a depressed mood is not synonymous with MD. Physicians interested in screening for MD can focus on anhedonia ("What do you do for a good time?") and sleep ("How is your sleep?"). When MD is present, these questions often yield positive responses, despite the patient's focus on other complaints and a tendency to deny depressed mood.

When physicians suspect a psychiatric disorder underlying a presenting physical problem, an open-ended questioning style may help reveal important data early in the interview. When patients indicate emotional distress surrounding a stressful life situation or distress related to a physical symptom, the physician should investigate the emotional issue immediately. Recognizing and treating a major depression early in the course of a general medical evaluation can save much time and expense.

[edit] Somatic Presentations

The depressed patient in primary care does not say, "I have a major depression, doctor; please prescribe proper treatment." More often, the patient experiences a somatic problem such as pain (headache, backache), fatigue, insomnia, or spells. The physician should attempt to rule out significant general medical problems and assess depression simultaneously. Developing these skills to evaluate both general medical and psychiatric problems simultaneously can save much time and expense and decrease frustration of both physician and patient.

[edit] Resistance (Stigma)

Many patients are reluctant to accept the diagnosis of depression. They do not experience this illness as a psychologic problem, and they often resist their physicians' explanation of a psychiatric problem. There is a stigma to having depression, and patients often think they are to blame. Physicians can help overcome this stigma by explaining that depression is an illness with a biochemical derangement, as in diabetes, and that proper treatment is necessary to restore function. For severe depression, proper treatment is usually biologic. Patients should be told that depression is not their fault and is a common condition. Severe depression does not usually remit on its own, but proper treatment usually ensures a good outcome.

[edit] "Pandora's Box"

Many physicians often are reluctant to pursue psychiatric or emotional problems for fear this will open Pandora's box and take an unreasonable amount of time. In fact, when physicians are able to respond appropriately to patients' emotions and recognize psychiatric disorders, the overall amount of time can be decreased. Extended workups for nonspecific physical complaints can be avoided.

[edit] Suicide

Suicidal ideation needs to be evaluated in all patients with symptoms of depression. Suicide is one of the top 10 causes of death in all age groups and one of the top three causes in young adults and teenagers. Besides depression, risk factors for suicide include gender (elderly white males are the highest-risk group), alcoholism, psychosis, chronic physical illness, and lack of social support. Clinically, suicidal intent, hopelessness, and a well-formulated plan all indicate high risk. Many patients who eventually commit suicide visit a primary care physician in the months before they take their lives.

The topic of suicidal ideation can be approached gradually with nonspecific questions such as, "Do you ever feel so discouraged that life does not seem worth living?" (Box 48-2). Patients can be asked to elaborate on their answers. Ultimately the physician should ask a very direct question, "Do you ever feel like taking your own life?" Physicians need to be aware that asking about suicide will not increase a patient's risk. To the contrary, inquiries about suicide can reassure the patient and enable the physician and patient together to make a plan to prevent suicide.

Once a patient has admitted to any suicidal ideation, the primary care physician must decide whether emergency psychiatric consultation and hospitalization are necessary. This is a clinical judgment for which no absolute guidelines can be set. Risk factors should be kept in mind, but the clinical evaluation should be primary. If this assessment suggests outpatient management, physicians should consider using a "no suicide contract." Although data on the effectiveness of this technique are not available, it does represent a standard clinical practice that has a high degree of face validity. Patients are simply asked to promise the physician that they will contact the physician (or alternative covering caregiver) if they think they are losing control of a suicidal impulse.

[edit] PHYSICAL EXAMINATION

No specific signs on physical examinations indicate depression. Nonverbal cues can be helpful, however, such as furrowed brow, downcast eyes, slow speech, psychomotor retardation or agitation, hand wringing, frequent sighing, and frequent shoulder shrugging. Some general medical conditions may also present with depression or depressive symptoms, such as endocrine problems (hypothyroidism, Cushing's syndrome), Parkinson's disease, and cerebrovascular disease.

[edit] LABORATORY STUDIES AND DIAGNOSTIC PROCEDURES

A routine laboratory screen, including complete blood count (CBC), chemistry profile, and urinalysis, should be part of the workup of depression because many general medical illnesses present with the symptoms of depression (e.g., fatigue, poor sleep). Thyroid studies (including TSH) and vitamin B₁₂ levels should also be completed on depressed patients to ensure that a metabolic derangement is not leading to depressive symptoms. Unfortunately, these studies only rule out other conditions that may mimic or exacerbate depression. No laboratory studies are yet available to "rule in" MD. In treatment of refractory cases or when indicated by medical history or physical examination, computed tomography (CT), magnetic resonance imaging (MRI), electroencephalography (EEG), or lumbar puncture can be considered, but these studies do not need to be part of the standard workup. In patients over age 40, however, an electrocardiogram (ECG) is usually necessary to rule out conduction disturbances or bradycardia, although this does not usually add to the differential diagnosis.

Screening instruments for depressive symptoms (e.g., Zung Depression Scale, Beck Depression Scale) can be helpful to recognize potential patients and monitor changes. However, these tools rate the severity of symptoms and do not yield diagnoses. A new scale, PHQ-9, is the first paper-and-pencil instrument that yields valid diagnoses as well as provides quantitative severity measures to monitor response to treatment.

[edit] DIFFERENTIAL DIAGNOSIS

[edit] Psychiatric Disorders

Many other psychiatric disorders present with symptoms similar to depression and can lead to misdiagnosis. Furthermore, depression often presents in addition to another psychiatric disorder. Thus an awareness of the other psychiatric disorders common in primary care is essential. The best way to avoid problems is to evaluate and treat MD if it is present. When another psychiatric condition is also present, treatment must be adapted to account for this comorbidity.

[edit] Anxiety Disorders

Anxiety is a ubiquitous symptom in primary care practices. Anxiety is common in MD, and depressive symptoms are common in anxiety disorders. Primary care physicians should be especially aware of generalized anxiety disorder (GAD), panic disorder, and obsessive-compulsive disorder (OCD). Key symptoms of these disorders include long-lasting and pervasive anxiety (GAD), discrete panic attacks, and the presence of unreasonable and disabling behaviors or thoughts (OCD). When MD occurs with GAD, panic, or OCD, a psychiatric referral is generally indicated. Treatment of MD, however, often helps to resolve or improve these other conditions.

[edit] Somatoform Disorders

Some patients are "addicted" to their physical problems. For unclear reasons, they focus on body ailments, and reassurance from the physician often does not help. Because depression also presents frequently with unexplained body complaints, the differentiation between a depressive illness and a somatoform disorder (SD) can be quite difficult. Primary care physicians will do best by focusing on the symptoms of MD. When MD is present, despite symptoms of an SD, appropriate treatment of the MD often significantly improves the SD.

[edit] Substance Abuse

Alcoholism or other substance abuse problems are common and often disabling conditions that can present with MD. Unlike anxiety disorders or SD, treatment of the MD that is comorbid with alcoholism does not usually relieve the substance abuse problem. Physicians need to evaluate patients for substance abuse and design separate treatment programs when substance abuse is present, whether or not MD is present and treated. In general, physicians should be cautious about treating only MD in the patient with a substance abuse problem. Such an action can be enabling to the substance abuser. Rather, the substance abuse needs aggressive confrontation and treatment.

[edit] Personality Disorders

Personality disorders can complicate the diagnosis and treatment of a mood disorder. Because patients with personality disorders can be difficult and demanding, physicians (as with others) often minimize their contacts with such individuals. This may lead to avoiding emotional issues and missing the diagnosis of depression. When MD coexists with a personality disorder, however, effective treatment of the MD often improves functioning, in general, even if the underlying personality disorder is not fundamentally changed. Thus physicians should evaluate the basic symptoms of depression in all distressed individuals, whether or not they have a comorbid personality disorder.

[edit] Depression Caused by Condition or Substance

The DSM-IV diagnosis of "depression due to a general medical condition" implies a psychiatric condition that the physician considers to be the direct physiologic result of a condition such as hypothyroidism or hyperthyroidism, pancreatic cancer, Parkinson's disease, and left-sided CVAs (Table 48-1). Because the prevalence of MD is not 100% in any physical illness, the final diagnosis of "depression due to a general medical condition" ultimately must be made on clinical inference alone. Unfortunately, no clear criteria can help guide clinicians in this endeavor. When five of nine of the symptoms of MD are present, however, the diagnosis becomes "depression due to a general medical condition, with major depressive episode." Thus, when this diagnosis is made, physicians should indicate the presence of a major depressive episode, whether or not it is presumably caused by another general medical condition. Furthermore, when this severity criterion is reached, data seem to indicate that standard treatments for MD should be used and are effective.

Table 48-1 General Medical Conditions with High Prevalence Rates of Major Depression

Similarly, DSM-IV recognizes that depression can be caused by exogenous medications (Box 48-3). The prototype of this condition is reserpine, which has long been known to cause a severe depressive condition in 15% of patients. No medication has been noted to "cause" depression in all patients. The most important clinical factors are evaluating the history and linking the initiation of depressive symptoms to starting a new medication or changing the dosage.

[edit] MANAGEMENT

[edit] Patient/Family Education: Overcoming Stigma/Resistance

The stigma associated with depression is so common that primary care physicians should routinely assume that patients and families will resist the diagnosis and its treatment. Most resistance can be managed by clarifying that MD (1) is common, (2) is not the patient's fault, (3) reflects a biologic disorder, (4) causes great suffering, (5) can exacerbate other physical complaints or illnesses, and (6) is treatable. Also, the medication to treat MD corrects biologic disorders and is not addictive.

Enlistment of family or other supports can also be extremely important in gaining patient acceptance of the diagnosis and adherence to treatment. Patients and families can be educated about the relative benefits and costs of medication vs. psychotherapy. For mild MD, psychotherapy (including office counseling by the physician) may be effective. Mild MD may also remit on its own; if the MD has not improved in 2 to 3 months after "watchful waiting" or after formal psychotherapy, however, antidepressant medication is clearly indicated. On the other hand, severe MD responds much more effectively to medication than to psychotherapy. The distinction between mild and severe MD is relative, with no clear-cut criteria. The physician can make this clinical distinction, however, through clinical judgment based primarily on the patient's level of functioning. If functioning remains high and suffering mild, the MD can be considered relatively mild. If suffering is great and functioning impaired, the MD should be considered severe. Often, family, friends, or colleagues are needed to help the physician evaluate the extent of impairment. Patients are often reluctant to admit changes in functioning and may even be unaware of such changes.

[edit] Treatment Efficacy

Treatment is usually effective in more than 90% of patients. If the first treatment attempt (psychotherapy, medication, or combination) is not effective, switching medications, changing psychotherapy, or initiating another treatment (e.g., electroconvulsive therapy [ECT]) is almost always effective. Patients should be told that (1) they deserve to feel better; (2) they will probably continue to suffer the same symptoms without treatment; and (3) underlying conditions may also have a worse outcome.

[edit] Length of Treatment and Prophylaxis

Biologic treatment of an MD episode should last 6 to 9 months after it has fully remitted. Because approximately 50% of patients with one episode of MD have a recurrence, many should receive prophylactic medication probably should not decrease the chances of recurrence. Patients probably should not receive medication for life after a first depressive episode. Thus, 6 to 9 months after the first episode abates, the physician can begin gradually withdrawing the patient from the medication. Slow tapering over 6 to 8 weeks is preferred.

Once it is recognized that a patient clearly has recurrent major depression (i.e., at least three episodes), research now supports the concept of prophylactic antidepressant treatment to decrease the likelihood of recurrence. The data support chronic use of the full treatment dosage of antidepressant medication.

[edit] Role of Physician Support

The importance of the physician-patient relationship cannot be overemphasized. The physician must convey a feeling of concern for the patient to be willing to discuss personal and distressing life issues, including suicidal feelings. The effective physician also must be skilled at the recognition and management of emotional distress in patients. Many physicians possess some of these skills, and literature and workshops are available for further training.

Regular visits are essential for the proper care of the depressed patient. Brief weekly visits (or biweekly) are usually indicated at the start of treatment to evaluate dosage, side effects, and changes in condition. Instead of weekly visits, weekly phone contacts can be substituted. Patients should generally not receive more than 1 week's supply of a medication that can be lethal in overdose. Once the patient has stabilized on a medication, monthly visits are important for support. If chronic prophylactic treatment is necessary, quarterly visits are usually appropriate, if the depression itself has remitted.

[edit] Role of Psychotherapy

Short-term (12 to 16 weeks) interpersonal, cognitive, and behavioral therapies have demonstrated efficacy for the treatment and prophylaxis of depression. Short-term "problem-solving" therapy (four sessions of 30 minutes each) has shown promise for depressed patients in primary care settings. Such treatment is as effective as medication for patients with mild MD but generally not as effective as medication for severe MD. Psychotherapy may also help prevent recurrences.

Office counseling by the primary care physician may be helpful to many patients. The physician should clarify that the patient is not being offered formal psychotherapy (unless the physician has such training). Physicians who become involved in complex interpersonal issues or notice that strong feelings (positive or negative) emerge in either the patient or themselves during office counseling should consider supervision or consultation. Psychotherapeutic situations invariably arouse strong emotions in both patients and physicians, and sensitivity to these issues is often essential to good outcomes.

[edit] Antidepressant Medications

About two thirds of patients with MD respond to an antidepressant medication within 3 weeks after reaching a therapeutic plasma level. This two-thirds proportion of responders can be increased to 90% by switching initial nonresponders to another class of antidepressants or by using augmentation strategies (e.g., addition of lithium or triiodothyronine). About one third of patients with MD (usually milder forms) improve with a placebo or general support.

Patients treated for MD should keep taking medication for 4 to 9 months after full remission of the syndrome. Patients with recurrent MD should be considered for long-term prophylaxis. When stopping antidepressants, medications should be tapered over 1 to 2 months or more slowly and the dose raised if prodromal symptoms of depression reappear.

Some uncertainty surrounds antidepressant treatment of dysthymia or adjustment disorders (including grief). Data from randomized clinical trials increasingly point to the efficacy of antidepressant medication for the acute treatment and long-term prophylaxis of dysthymia. No good outcome studies, however, have evaluated the treatment of minor depressive episodes, such as adjustment disorders with depressed mood. Emotional support by the physician may be sufficient to resolve an adjustment disorder, but when a patient experiences significant impairment in function (e.g., poor work performance, poor sleep, distressed relationships), an antidepressant may be helpful.

Numerous medications are available to treat depression (Table 48-2). All the agents are equally efficacious, and no particular agent or class has been shown to be more effective in ameliorating certain symptoms of depression, such as agitation or insomnia. No agent has been shown to improve depressive symptoms at a faster rate than other medications. Thus the choice of antidepressant can only be made on issues other than efficacy (e.g., side effects, costs, compliance).

Table 48-2 Antidepressants: Side Effects, Mechanisms of Action, and Dosages

✢Blockade of α₂NE, 5-HT_2A, 5-HT_2C, and 5-HT₃ receptors.

†Blockade of 5-HT_2A receptors.

[edit] Heterocyclic Medications

The heterocyclic medications include the tricyclics, which have been available since the 1950s, and several other agents that are similar in structure, including maprotiline, amoxapine, and trazodone. The heterocyclic antidepressants are similar in side effects, dosing strategies, and efficacy. Their major advantage over newer agents is their lower cost per unit dosage. Recent pharmacoeconomic studies, however, indicate that total health system expenditures (e.g., total psychiatric and nonpsychiatric outpatient visits, laboratory expenditures) for patients taking tricyclics and for those taking the newer agents are about the same. The newer agents are more expensive, but patients taking tricyclics incur extra costs related to more office visits and more general medical expenditures.

Use of heterocyclic medications requires starting at low doses and gradually building up doses to a therapeutic level. Plasma levels can be followed, but with the exception of nortriptyline, which has a therapeutic window (i.e., levels below and above the window are less likely to lead to remission of depression), blood levels function primarily as crude indicators of whether or not the patient is taking the medication. More importantly, physicians should treat the patient and not the blood level. Many patients with high or low blood levels may do very well clinically. For problematic situations, consultation with psychiatrists expert in psychopharmacology may be advisable. Patients may require up to 3 weeks at a therapeutic blood level to respond fully, so they should be informed that it may take time before they begin to notice any positive effects.

The heterocyclic antidepressants are characterized by varying degrees of problematic side effects. The anticholinergic effects include dry mouth, constipation, urinary retention, tachycardia, increased ocular pressure, and confusion. The primary antihistaminic effects are sedation and inhibition of gastric acid secretion. These effects have led to use of some antidepressants for urticaria, especially doxepin, which seems to be the most potent. The antiadrenergic effects cause postural hypotension, which can be quite dangerous in medically ill or elderly patients. Nortriptyline seems to be the safest of the heterocyclics in this regard. All the heterocyclics (except trazodone) have quinidine-like effects, delaying conduction across the bundle of His and increasing the QT interval on the ECG. Patients with bundle branch block and increased conduction from other causes are at risk of higher degrees of heart block when given these medications. This side effect is probably responsible for the high lethality of these drugs in overdose. A corrected QT interval of 0.44 differentiates between low-risk and high-risk patients.

Among the tricyclics, desipramine (a metabolite of imipramine) and nortriptyline (a metabolite of amitriptyline) are the least anticholinergic, the least sedating, and the least likely to cause postural hypotension. Among the heterocyclic medications, trazodone does not cause anticholinergic or conduction problems and has very low lethality in overdose. Trazodone, however, is quite sedating and also causes postural hypotension.

[edit] Newer Agents

New agents have revolutionized psychiatric practice, especially for the treatment of depression in patients with comorbid general medical illnesses and in elderly persons. The side effects are so much less toxic and the medication so well tolerated that many patients who were too physically ill to be safely treated in the 1970s and 1980s can now receive antidepressant medication without fear of dangerous side effects. These newer agents seem safe from the standpoint of overdose, with a very low therapeutic index.

The newer agents include four selective serotonin reuptake inhibitors (SSRIs: citalopram, fluoxetine, paroxetine, and sertraline) and four others: bupropion, mirtazapine, nefazodone, and venlafaxine. These agents are remarkably safe for use in elderly patients and those with comorbid general medical illnesses. They do not cause postural hypotension or cardiac conduction delay; antihistaminic and anticholinergic side effects are minimal. Choosing among these agents can be difficult because they are all so well tolerated and effective (see Table 48-2).

Fluoxetine has the longest half-life (24 to 27 hours) and has a long-acting, active metabolite (half-life of 7 days). Doses can be given every other day, and when the drug is being withdrawn, the doses can eventually be given once or twice a week to allow for very smooth tapering. The other SSRIs have half-lives of about 24 hours and have no active metabolites with longer half-lives. This allows once-a-day dosing and rapid washout. The other new agents have shorter half-lives and must be given more than once a day, except for the slow-release form of venlafaxine.

The starting dose of the SSRIs may also be the effective treatment dose, as with fluoxetine (20 mg) and possibly with paroxetine (20 mg) and sertraline (50 mg). Patients not responding to the starting doses of SSRIs after 1 month should be given increased doses. Frail elderly persons and patients with liver disease or other general medical illness require smaller starting doses. One-half the recommended starting dose in this population is usually preferred. The SSRIs inhibit various isoenzymes of the cytochrome P-450 system in the liver, thus potentially leading to the buildup of other medications metabolized in the liver, such as anticonvulsants, digitalis, and coumadin. To avoid potentially dangerous pharmacokinetic interactions, it may be necessary to monitor blood levels or clinical indicators of toxicity.

The SSRIs sometimes cause side effects such as anxiety, insomnia, gastrointestinal distress, agitation, and sexual difficulties. With the exception of sexual difficulties, which may occur in 30% to 50% of patients, these side effects occur in less than 20% of patients, are usually mild, and do not lead to discontinuation of medication. Adjunctive use of a sedating antihistamine (diphenhydramine, hydroxyzine), a sedating antidepressant (trazodone, doxepin), or an anxiolytic (lorazepam, clonazepam) can be helpful. These adjuncts can usually be discontinued after a short time, although continued treatment with two agents may be indicated. If a long-acting benzodiazepine (clonazepam) is used in elderly persons, care must be exercised to avoid buildup of medication over time, which can lead to confusion, sedation, or falls after several weeks of treatment.

Among the newer agents, venlafaxine and bupropion have effects similar to those of the SSRIs. In doses greater than 300 mg, venlafaxine can cause persistent blood pressure elevation in about 10% of patients. Bupropion also has been associated with a 1% to 4% prevalence of seizures, which is a very low overall risk but probably slightly more common than with other antidepressant medications. The medication must be given in divided doses because of its short half-life, should not be given to individuals at risk for seizures (e.g., head trauma), and should never be administered in any single dose greater than 150 mg. The starting dose (75 mg two or three times a day) should be increased slowly (once a week) to therapeutic levels of 300 to 450 mg/day to minimize the risk of seizures. The medication should also not be given to bulimic patients (patients who gorge food and often induce vomiting) because of a possible increased seizure risk.

Nefazodone is a new serotonergic agent that blocks a postsynaptic serotonin receptor, in addition to being a weak inhibitor of serotonin reuptake. It is more sedating than pure SSRIs, but it does not cause sexual side effects and may be more effective than SSRIs and other newer agents in preserving normal sleep architecture. Because of its anxiolytic properties, nefazodone can often be used as monotherapy for patients with depression and coexisting anxiety. Mirtazapine blocks several serotonin receptions and adrenergic receptors. It shares the benefits of a favorable side effect profile, with the exception of causing sedation and weight gain. All four newer non-SSRI agents can be used as alternatives to the SSRIs for patients who develop sexual side effects. Nefazodone and mirtazapine can be useful for patients with anxiety.

[edit] Elderly and Medically Ill Patients

As noted, the newer agents have generally become the drugs of choice in elderly and medically ill patients. Among the heterocyclics, the safest agents are nortriptyline and desipramine, which are still widely used in these patients. They also seem to have a role in the treatment of refractory depression.

Dosing strategies in elderly and medically ill patients need to "start low and go slow." From the pharmacokinetic point of view (i.e., absorption, metabolism, and elimination; the effect of the body on medication), these agents are metabolized more slowly in these two groups. Accumulation and toxicity can become a problem. Protein binding is also lower in these patients because albumin levels are lower; this can lead to toxic effects. Similarly, lower doses may lead to efficacious treatment. Pharmacodynamic differences (i.e., impact of the medication on the body) in elderly persons can also lead to toxicity or efficacy at lower-than-expected doses.

[edit] Electroconvulsive Therapy

ECT is still the most effective treatment available for the treatment of depression. Although many patients and physicians have prejudices and fears about ECT, new methodologies of administration have shown ECT to be a safe and effective treatment modality. In frail elderly persons, ECT can be safer than antidepressants. Some reversible short-term memory loss is a common side effect, but this reverts to normal in virtually all patients. ECT can be lifesaving and should not be denied to patients because of lack of understanding or unrealistic fear.

ECT does not lead to permanent remission of depression in patients susceptible to recurrence. Patients with recurrent depression who receive ECT should receive either prophylactic medication or maintenance ECT, about once a month on an outpatient basis.

[edit] Referral

The clinical criteria for referral to a psychiatrist or other mental health specialist depend greatly on the experience and expertise of the primary care physician. In general the most common reasons for referral are lack of response to initial treatment, suicidal ideation, or psychosis. Physicians can inform patients early that referral to a mental health specialist is sometimes needed; this can make referral at a later stage much more acceptable. Because partial remission often occurs, physicians should establish clear indications of predepression functioning, and the patient who does not return to baseline functioning should be referred to a specialist.

[edit] ADDITIONAL READINGS

• S Cole, J Bird: The medical interview: the three function approach ed 2. St Louis: Mosby; 2000:
• SA Cohen-Cole, K Kaufman: Major depression in physical illness: diagnosis, prevalence, and antidepressant treatment (a ten-year review: 1982-1992). Depression 1993; 1:181.
• Depression Guideline Panel: Depression in primary care. Detection and diagnosis, Clinical practice guideline no 5 1993; vol 1: Rockville, Md: US Department of Health and Human Services; 1993:
• Depression Guideline Panel: Depression in primary care. Treatment of major depression, Clinical practice guideline no 5 1993; vol 2: Rockville, Md: US Department of Health and Human Services; 1993:
• RL Spitzer, K Kroenke, JBW Williams: Validation and utility of a self-report version of PRIME-MD: the PHQ primary care study. JAMA 1999; 282:1737 - 1744.
• S Stahl: Selecting an antidepressant by using mechanism of action to enhance efficacy and avoid side effects. J Clin Psychiatry 1998; 59 (suppl 18):23.
• LR Wulson, GE Vaillant, VE Wells: A systematic review of the mortality of depression. Psychosom Med 1999; 61:6.