Anxiety
Robert G. Robinson
Jess G. Fiedorowicz
I. Background
A. Definitions
Anxiety refers to a vague, often unpleasant, emotion associated with a sense of apprehension, arousal, or tension. Karl Jaspers described the experience of abnormal anxiety as “free-floating and unattached” and associated with a “feeling of restlessness.” Anxiety symptoms may be consciously experienced as psychological symptoms (anticipation, apprehension, arousal, dread, fear, worry, irritability, insomnia, inattention) or as physical (somatic) symptoms (nausea/dyspepsia, dry mouth, racing heart, palpitations, shortness of breath, sweating, tremor, muscular tension).
Anxiety is universally experienced and normal anxiety is thought to be adaptive as illustrated by the Yerkes–Dodson law demonstrating an “inverted U” relationship between arousal and performance. Anxiety may also be a manifestation of a neuropsychiatric disorder.
Anxiety is commonly divided into state and trait anxiety. State anxiety refers to anxiety in the setting of a specific stressor, whereas trait anxiety refers to an enduring pattern of high anxiety at baseline or a tendency to respond with exaggerated anxiety in mildly stressful situations. Clinically, the boundary between state and trait anxiety is often blurred.
Phenomenologically, the construct of anxiety assumes a breadth of symptoms including situational anxiety, general anxiety, panic attacks, phobic states, obsessions, and compulsions. General anxiety is distinguished from situational anxiety by the perceived lack of a sufficient stressor. Panic attacks are discrete episodes of limited duration with prominent psychological and physical anxiety symptoms in conjunction with a sense of impending doom. Phobias consist of involuntary fears out of proportion to the object of fear leading to avoidance of that object. Obsessions represent recurring, intrusive thoughts, or images. Obsessions may be linked to compulsions, which are
repeated actions driven by a desire to relieve an anxiety preceding the action and compelling the completion of the action.
TABLE 6.1 Classification of Anxiety Disorders of Particular Relevance to Neuropsychiatry
Generalized anxiety disorder
Persistent anxiety and worry out of proportion to circumstances
Obsessive-compulsive disorder
Obsessions and compulsions of clinically sufficient severity and duration
Panic disorder
Recurrent, spontaneous panic attacks accompanied by fear of repeated attacks
B. Classification and epidemiology of anxiety syndromes
This chapter will focus on state anxiety syndromes associated with neurologic conditions. The anxiety syndromes with the strongest association are detailed in Table 6.1.
C. Normal anxiety contrasted with generalized anxiety disorder
Anxiety is ubiquitous in human mental life and plays a role in maintaining arousal and response to environmental stress. Individuals differ dimensionally on baseline levels of anxiety as well as in their ability to mount an anxious response to a stressor. The ability to be aroused by threatening aspects in the environment condones clear survival advantages.
Individuals who have excessive anxiety at baseline or display exaggerated responses to relatively unthreatening or only moderately arousing events may be categorized as suffering from a generalized anxiety disorder (GAD). Criteria for GAD include persistent, excessive, and uncontrollable worries over a period of at least 6 months and encompassing multiple aspects of one’s life, not limited to a single activity or event and not confined to a specific aspect of another psychiatric disorder such as, worrying about having a serious illness in hypochondriasis. This anxiety is clinically significant and associated with symptoms such as edginess, fatigue, impaired concentration, irritability, restlessness or sleep disturbance, and particularly, initial insomnia. Physical symptoms associated with generalized anxiety may include muscular tension, a sense of a racing pulse, sweating, headaches, and nausea.
The diagnosis of anxiety disorder due to a general medical condition with generalized anxiety is applied by convention when a clinically significant syndrome of excessive general anxiety is deemed to be the result of a medical condition. The presence of a temporal relationship with the condition, atypical features, and an evidence base of an association between the specific condition and generalized anxiety support the diagnosis.
The lifetime prevalence of GAD in the general population is approximately 2% to 5%. The point prevalence is approximately 1% to 2%.
D. Obsessive-compulsive disorder
Significant obsessions and compulsions are not commonly experienced in the general population. In addition to occurring in those with obsessive-compulsive disorder (OCD), they may arise in the context of affective syndromes, psychotic disorders, and in those with trait obsessionality such as in obsessive-compulsive personality disorder, in which case these symptoms are less likely to be perceived as intrusive or unreasonable.
OCD is characterized by either obsessions or compulsions that are recognized as excessive or unreasonable. The symptoms are associated with significant clinical impairment through psychological distress, consumption of time, or impairment in function. Common obsessions include fear of contamination, fear of harming oneself or others, overwhelming concern with neatness or order, fear of making a mistake, fear of losing something, fear of embarrassing oneself, and persistent thinking about a specific image, number, sound, or word. Common compulsions include washing, checking, counting, repeating words or actions, arranging items in a specific order, and collecting or hoarding items no longer needed. Compulsive symptoms are often recognized as voluntary acts arising from oneself although they are typically not perceived as willful. Individuals often feel a desire to resist compulsions, albeit a seemingly futile desire at times.
The diagnosis of anxiety disorder due to a general medical condition with obsessive-compulsive symptoms is applied by convention when a preponderance of clinically significant obsessions and compulsions are deemed to be the result of a medical condition. The diagnosis is supported by the presence of a temporal relationship with the condition, atypical features, and an evidence base of an association between the specific condition and a syndrome of obsessions and compulsions.
The lifetime prevalence of OCD in the general population is approximately 2% to 2.5%. Peak incidence occurs around 20 years of age with more than half developing the illness before 25 years of age.
E. Panic disorder
Panic attacks may occur in the setting of a depressive syndrome, schizophrenia, phobias, GAD, OCD, or with separation anxiety. Panic attacks elicited by a specific situation are typically differentiated from spontaneous panic attacks.
Panic disorder is defined by convention as the presence of spontaneous, recurrent panic attacks followed by a fear of repeated attacks for at least 1 month. The panic attack is classically sudden in onset and typically reaches maximum intensity within 10 minutes. The duration of a panic attack can vary from minutes to hours although panic attacks typically last 10 to 20 minutes. Given the intensity of the event, the reported duration may overestimate the actual duration. In addition to intense fear or anxiety, patients may experience a rapid even pounding heart rate, sweating, shaking, tremor, shortness of breath,
hyperventilation, a choking sensation, chest discomfort, nausea, dyspepsia, lightheadedness, unsteadiness, derealization, depersonalization, sense of losing control, sense of going crazy, sense of impending doom or death, paresthesia, or chills. Panic disorder may be accompanied by agoraphobia, a fear of situations where escape may be difficult should a panic attack occur.
The diagnosis of anxiety disorder due to a general medical condition with panic attacks is applied by convention when panic attacks predominate and are deemed to be the result of a medical condition. The diagnosis is supported by the presence of a temporal relationship between panic attacks and the medical condition, atypical features, and an evidence base of an association between the specific condition and panic attacks.
The lifetime prevalence of panic disorder in the general population is approximately 1% to 2%. Rates have been found to be as high as 1% to 6% in primary care clinics and 17% to 25% in those presenting to an emergency department with chest pain. Rates are also thought to be elevated in neurology clinics, particularly among those referred for evaluation of dizziness and vestibular function.
F. Post-traumatic stress disorder
By current criteria, post-traumatic stress disorder (PTSD) follows the experiencing or witnessing of a traumatic event where threatened death, serious injury, or the physical integrity of self or others evoked fear, helplessness, or horror.
PTSD then involves persistently experiencing that traumatic event, avoiding stimuli associated with the traumatic event, and symptoms of hyperarousal such as hypervigilance or an exaggerated startle response. The symptoms must result in significant distress or impairment and must last for more than 1 month (otherwise operationalized as acute stress disorder). PTSD is considered chronic if symptoms persist for more than 3 months.
References
1. American Psychiatric Association. Diagnostic and statistical manual of mental disorders, 4th ed. Text Revision. Washington, DC: American Psychiatric Association; 2000.
2. Sadock BJ, Sadock VA. Kaplan & Sadock’s comprehensive textbook of psychiatry, 7th ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2000.
3. Sims A. Symptoms in the mind: An introduction to descriptive psychopathology, 3rd ed. London: WB Saunders; 2003.
II. Neurobiology and Pathophysiology of Anxiety
A. Normal anxiety
The neurobiology of normal anxiety remains to be fully elucidated. There are two systems of anxiety that probably overlap: A “defense system” directed to making an immediate response to a threat and a “behavioral inhibition” system, which suppresses
potentially dangerous behaviors. Sensory input is relayed by the dorsal thalamus to the modality-appropriate cortical areas. Upon reaching secondary association cortex for a particular modality, information is relayed to emotion and memory centers in the amygdala, hippocampus, entorhinal cortex, orbitofrontal cortex, and anterior cingulate. Further, neural circuits orchestrating anxiety are organized at various levels throughout the brain. Brainstem regions appear to mediate somatic anxiety symptoms. Midbrain regions such as the periaqueductal gray and locus coeruleus are prominently involved in simple reflexive responses. Intermediate regions such as the amygdala and septohippocampal systems manage practiced responses. Cortical regions such as the paralimbic cortex appear to address more cognitively demanding situations.
The clinical efficacy of pharmaceuticals targeting γ-amino-butyric acid (GABA), serotonin (5-HT), and norepinephrine (NE) systems has promulgated interest in the role of these neurotransmitter systems in anxiety. Generally speaking, increased GABA release attenuates anxiety and increased NE release augments anxiety. The role of 5-HT is less straightforward and likely involves an adaptive or modulating role.
B. Generalized anxiety
GAD is thought to result from a dysfunction or change in the set point of the aforementioned anxiety systems.
Acute increases in 5-HT levels which are presumed to be treatment induced have been associated with increased anxiety; however, chronic increases are believed to be associated with diminished anxiety.
C. Obsessions and compulsions
Cortico-subcortical-thalamic-cortical circuits, particularly the orbitofrontal circuit appears to be dysfunctional in OCD. It has been demonstrated that patients with OCD have significantly more gray and less white matter than those without OCD. The onset of OCD may be immediately preceded by streptococcal infection in children (Pediatric Autoimmune Neuropsychiatric Disorders Associated with Streptococcal infections, [PANDAS]), encephalitis, head injury, or damage to the striatum.
OCD has been linked to disruption in serotonergic pathways involving the basal ganglia and orbitofrontal cortex.
D. Panic attacks
The locus coeruleus, a noradrenergic nucleus, has been identified in playing a prominent role in the etiology of panic. Positron emission tomography (PET) evidence has further identified involvement of the hippocampal and parahippocampal regions.
A dysregulation of noradrenergic systems, particularly the locus coeruleus, has been implicated in the pathogenesis of panic attacks. A shift in the “set point” of the benzodiazepine receptor has been proposed as a contributing factor in panic disorder.
Sodium lactate, hypercapnia, and cholecystokinin may provoke panic.
References
1. Sandford JJ, Argyropoulos SV, Nutt DJ. The psychobiology of anxiolytic drugs. Part 1: Basic neurobiology. Pharmacol Ther. 2000;88(3):197–212.
2. Argyropoulos SV, Sandford JJ, Nutt DJ. The psychobiology of anxiolytic drugs. Part 2: Pharmacological treatments of anxiety. Pharmacol Ther. 2000;88(3):213–227.
III. Management
A. Principles of management
The evidence base for the specific treatment of anxiety in patients with neurologic illness is limited. In the absence of compelling data from a study specific to a particular neurologic condition, it is generally advisable to follow the treatment guidelines for patients with primary anxiety disorders when appropriate. The American Psychiatric Association has published practice guidelines in panic disorder (1998) and for acute stress disorder and PTSD (2004). Cochrane reviews exist for “Antidepressants for generalized anxiety disorder (2003)” and “Pharmacologic management for agitation and aggression in people with acquired brain injury (2003).”
B. Performing a comprehensive diagnostic evaluation
The importance of a comprehensive examination (psychiatric, medical, and neurologic) of patients with neuropsychiatric disorders cannot be overemphasized. This examination should include a history, mental status examination, and physical examination. Although an interdisciplinary approach is advantageous, the primary responsibility for the complete examination should not be delegated. Information from outside informants such as family members or primary care providers is essential, particularly to ascertain the patient’s premorbid function, premorbid personality, and the temporal relationship between the onset of anxiety symptoms and the neurologic illness. Given the often insidious onset of anxiety symptoms and frequent impairments in the patient’s historical abilities secondary to neuropsychiatric illness, it may be difficult to delineate the relationship between anxiety symptoms and the illness. A prudent approach is to clearly separate observations from interpretations. One must avoid assuming “who wouldn’t be anxious with condition X” and postpone such explanation until formulating the case. Although often tempting, it is inadvisable to reflexively disregard a patient’s psychological suffering as a normal reaction. Such an approach need not underestimate the tragedy of the neurologic illness, but may rather call attention to the comorbid anxiety. A rigorous and thorough elucidation of phenomenology is vital for an accurate assessment while further fostering the development of an empathic therapeutic relationship.
Medical and psychiatric comorbidity should be thoroughly assessed, including alcohol and other substance abuse, personality
vulnerabilities, and frank personality disorders. Mood disorders are a common cause of anxiety and occur with increased frequency in those with anxiety disorders. Anxiety may occur as a consequence of asthma, central nervous system (CNS) infection, dehydration, fever, hyperventilation, metabolic abnormalities, mitral valve prolapse, myocardial infarction, neurosyphilis, pheochromocytoma, respiratory illness, thyroid disease, and other endocrine conditions. Alcohol, amphetamines, anabolic steroids, caffeine, cocaine, ecstasy, γ-hydroxybutyrate, inhalants, and phencyclidine (PCP) commonly produce clinically significant anxiety symptoms with intoxication and habitual use. Withdrawal from alcohol, barbiturates, benzodiazepines, cocaine, sedative/hypnotics and opiates may be associated with significant anxiety and agitation. Medications may be associated with anxiety and a thorough review of current and recent medications should be performed. Anxiety may also occur as an adverse reaction to medications including but not limited to acyclovir, antiretrovirals (esp. efavirenz), bronchodilators, corticosteroids, interferons, interleukin-2, isoniazid, lidocaine, pentamidine, procaine, sympathomimetics, and stimulants.
Laboratory tests may be required to follow up on any concerns elucidated during the comprehensive evaluation. The laboratory workup is, of course, guided by clinical suspicion as well as the estimated prevalence rate of the condition screened and will therefore appropriately vary from one patient or one setting to another. Because of the relatively high prevalence of thyroid abnormalities and a strong association with anxiety, a thyroid-stimulating hormone (TSH) test should be obtained as a screening test in patients with GAD or panic disorder if thyroid abnormalities have not been ruled out because of the onset of anxiety symptoms. Additionally, any patient who is neuropsychiatric, with a constellation of symptoms suggestive of thyroid disease including a change in the frequency of bowel movements, hair changes, skin changes, soft or brittle nails, sweating, temperature intolerance, tremor, or weight change should be screened for thyroid abnormalities with a minimum TSH. A complete medical panel may reveal hypercalcemia associated with hyperparathyroidism, liver abnormalities suggestive of alcoholism (i.e., elevated aspartate aminotransferase [AST] alanine aminotransferase [ALT] ratio), or reflections of other endocrine disorders. Other tests should include a complete blood count, erythrocyte sedimentation rate (ESR), vitamin B12 level, folate level, and a rapid plasma reagin (RPR). The rate of medical comorbidities with neuropsychiatric illness is high enough to warrant such screening if not already performed.
No definitive practice guidelines exist for the use of neuroimaging in the evaluation of patients with anxiety in the setting of a neuropsychiatric condition. Structural neuroimaging (i.e., computed tomography [CT] scan or magnetic resonance
imaging [MRI]) should be obtained when the onset of anxiety symptoms begins after age 50, as follow-up of any abnormalities from the neurologic exam suggestive of intracranial pathology, or in the setting of a temporal association between symptoms and head trauma.
Performing a risk assessment, which includes a suicide risk assessment as well as assessment of general dangerousness to self or others is important.
C. Formulating the case
A comprehensive formulation should summarize the salient aspects of the history and physical examination. It may be helpful to organize one’s diagnostic thinking into a biopsychosocial or an etiopathic model. Aspects of the patient’s presentation can be organized into those that are likely the result of brain disease, the provocation of psychological vulnerabilities, the product of maladaptive behaviors, or demoralization from a tragic illness or other life events.
The neuropsychiatric diagnosis is a product of the formulation. It may be difficult to judge whether the anxiety disorder is primary or occurs secondary to the neurologic illness. The diagnosis of anxiety disorder secondary to a general medical condition is supported by the presence of a temporal relationship between the anxiety syndrome and the neurologic condition, atypical features, and an evidence base of an association between the specific neurologic condition and an anxiety disorder. Neurologic conditions associated with specific anxiety syndromes are detailed in Table 6.2.
D. Implementing a treatment plan
The determination of treatment setting (outpatient, partial hospitalization, or inpatient) is based on multiple factors, including but not limited to severity of illness, level of function, strength of family and social supports, and treatment compliance. The
decision regarding treatment setting is a fluid one and must be reassessed on a regular basis. In the outpatient setting, the frequency of outpatient visits should be titrated to patient need.
TABLE 6.2 Neurologic Conditions Associated with Specified Anxiety Syndromes
Generalized anxiety disorder
Dementia, epilepsy, Huntington disease, multiple sclerosis?, Parkinson disease, restless legs syndrome, stroke, traumatic brain injury
Obsessive-compulsive disorder
Huntington disease?, Parkinson disease?, stroke (involving basal ganglia), traumatic brain injury, Tourette syndrome, Sydenham’s chorea
Panic disorder (with or without agoraphobia)
Migraine headache, epilepsy, Lyme disease?, Parkinson disease, restless legs syndrome, traumatic brain injury
Post-traumatic stress disorder
Traumatic brain injury
?, association not fully established.
Given the subjective nature of anxiety symptoms, having a benchmark to monitor progress and treatment effectiveness is necessary. This may include a scale such as the Yale-Brown Obsessive Compulsive Scale (Y-BOCS), a target symptom or symptoms, a functional assessment, or an outside informant’s impression. Titration of psychiatric medications for anxiety disorders can be a prolonged process and it can be difficult for patients or providers to recall how current symptoms and function compare to prior months or previous treatment. The assessment of treatment effects is further complicated in neurodegenerative disorders where treatment effects must be inferred from a projected course of worsening illness.
The patient’s personality and cognitive strengths and vulnerabilities should be assessed. Counseling and education may help patients capitalize on strengths and attend to vulnerabilities. Function-oriented guidance, delineation of prognosis and expectations, and exploration of means of coping with illness and handicap serves to combat demoralization for the patient and family.
Address avoidance behaviors, substance use (including caffeine), and other maladaptive behaviors proactively. Substance abuse or dependence is optimally managed in a structured substance abuse program.
Implementing a biopsychosocial treatment plan.
Patients should be instructed to eliminate or at least limit caffeine and alcohol use.
Psychotherapeutic approaches include cognitive-behavioral, supportive, and psychodynamic.
Medications commonly prescribed for anxiety include selective serotonin reuptake inhibitors (SSRIs), venlafaxine, tricyclic antidepressants (TCAs), and benzodiazepines.
Treatment outcomes and compliance should be routinely monitored. Any worsening in illness or decline in adherence should be dealt with proactively.
References
1. McHugh PR, Slavney PR. The perspectives of psychiatry, 2nd ed. Baltimore and London: The Johns Hopkins University Press; 1998.
IV. Generalized Anxiety in Patients with Neurologic Illness
A. Generalized anxiety disorder general treatment pearls
Psychotherapeutic approaches to GAD include cognitive-behavioral, supportive, and psychodynamic. Cognitive-behavioral approaches have been well studied and should be employed when expertise is available. Specific behavioral approaches include relaxation techniques and biofeedback.
TABLE 6.3 Selected Psychiatric Medications for Treatment of Generalized Anxiety Disorder
Medication
Starting Dose
Usual Dosing Range
Tricyclic Antidepressants
Nortriptyline (Pamelor, Aventyl HCl)
25 mg/d
Typically 50–150 mg qhs (dosed to 50–150 ng/mL)
Selective Serotonin Reuptake Inhibitors
iCtalopram (Celexa)
10–20 mg/d
20–60 mg/d
Escitaloprama (Lexapro)
10 mg/d
10–20 mg/d
Paroxetinea (Paxil)
10 mg/d
40 mg/d
Sertraline (Zoloft)
25–50 mg/d
100–200 mg/d
Other Antidepressants
Mirtazapine (Remeron, Remeron SolTab)
15 mg/d
15–45 mg qhs
Venlafaxinea (Effexor XR)
75 mg/d
150–300 mg qd
Antipsychotics
Olanzapine (Zyprexa)
2.5–5 mg/d
2.5–10 mg/d
Other
Diphenhydramine (Benadryl)
25 mg
25–50 mg q4 hours as needed
Pindolol (Visken)
5–10 mg/d
10–60 mg/d
Propranolol (Inderal)
60–80 mg/d
80–520 mg/d
aU.S. Food and Drug Administration indication for generalized anxiety disorder.
Patients should be instructed to eliminate or at least limit caffeine and alcohol use.
Medications commonly prescribed for primary GAD include SSRIs, venlafaxine, TCAs, buspirone, and benzodiazepines. Starting and target doses are detailed in Table 6.3. Long-term treatment is required for many patients. To limit adverse reactions, including a possible initial worsening in anxiety, antidepressants must often be initiated at low doses and titrated slowly.
B. Poststroke anxiety
1. Clinical background
Approximately one fourth of patients, after acute stroke, meet criteria for GAD. Nearly another one fourth of patients will go on to develop clinically significant generalized anxiety. Approximately half of all poststroke patients with anxiety will have a clinically significant depressive syndrome (major or minor).
The occurrence of poststroke anxiety is not associated with acute social, cognitive, or physical impairment, but it is associated with later impairment in activities of daily living and a more limited social support network. Early onset poststroke anxiety, as defined by occurring within
the first 3 months, is associated with a history of psychiatric disorder. Late-onset poststroke anxiety, arising more than 3 months after stroke, has not been associated with a previous psychiatric history. The use of stimulants to treat poststroke apathy may precipitate or contribute to anxiety.
The phenomenology of poststroke anxiety is similar to that of a primary anxiety disorder. The phenomenology of early-onset and late-onset poststroke anxiety are also similar.
Poststroke anxiety may last several months to years. When comorbid with depression, poststroke anxiety is associated with a longer duration of depression and greater functional and social impairment.
2. Pathogenesis of poststroke anxiety
The pathophysiology of poststroke anxiety remains unclear. The combination of depression and anxiety has been associated with left frontal cortical lesions. Poststroke anxiety without depression has been associated with right hemisphere lesions.
3. Diagnosis
As in primary anxiety disorders, the diagnosis is clinical and based on the presence of consensus criteria. The diagnosis may be impeded by the presence of a comorbid depression, aphasia, anosognosia, or dementia. Observations of family and other caretakers may assist in making the diagnosis.
4. Treatment
The clinician should optimize treatment and rehabilitate the patient, provide adequate pain management, and treat any underlying medical conditions.
Environmental modifications should be implemented, including a daily routine with structured activities and good sleep hygiene. Behavioral management techniques should be employed by staff and taught to the family for any agitation secondary to anxiety.
Although psychotherapy has not been explicitly studied in poststroke anxiety, cognitive behavior therapy (CBT) should be considered given its efficacy in patients with primary anxiety disorders.
Somatic therapy should be implemented if anxiety is severe or persists despite the steps listed in the preceding text. There are no U.S. Food and Drug Administration (FDA)-approved somatic treatments for poststroke anxiety. Therapeutic options are the same as for primary anxiety disorders.
Nortriptyline titrated slowly from 25 to 75 mg per day should be administered based on data from the treatment of poststroke depression comorbid with GAD. Tolerance and subsequent titration of nortriptyline may be limited by sedation, orthostatic hypotension, and anticholinergic side effects such as dry mouth, constipation, urinary retention, and blurry vision. In patients where nortriptyline is either
not desired or contraindicated such as during acute recovery after myocardial infarction, escitalopram 10 mg per day or venlafaxine up to 300 mg per day are reasonable alternatives based on safety profile and efficacy in primary anxiety disorders.
Benzodiazepines are commonly prescribed in the treatment of primary generalized anxiety. However, benzodiazepines have not been studied in patients with stroke and carry significant risk of side effects and adverse events such as falls. In addition, tolerance is developed to benzodiazepines, which may result in rebound anxiety and withdrawal symptoms when prescribed for more than 4 weeks. Given these safety concerns and the high comorbidity of depression with poststroke anxiety, a course of treatment with an antidepressant is preferable.
Only one in four patients with early onset generalized anxiety recover by 1 year following the stroke. Most of the remaining experience a chronic course defined as at least 2 years’ duration. Therefore, long-term treatment is required for many patients with poststroke anxiety similar to a primary GAD. We recommend continuing effective acute therapy for at least 1 year in responders. Partial responders or those whose symptoms return during subsequent taper may require chronic antidepressant treatment.
