Articles

Print Friendly 

Dan J. Stein, MD, PhD

Primary Psychiatry. 2004;11(6):55-78
 

Dr. Stein is professor of psychiatry at the University of Stellenbosch, CapeTown, in Tygerberg, South Africa, and the University of Florida in Gainesville.

Disclosure: Dr. Stein has received grants and/or honoraria from AstraZeneca, Eli Lilly, GlaxoSmithKline, Lundbeck, Orion, Pfizer, Pharmacia, Roche, Servier, Solvay, Sumitomo, and Wyeth.

Please direct all correspondence to: Dan J. Stein, MD, PhD, University of Stellenbosch, PO Box 19063, Tygerberg 7505, South Africa; Tel: 27-21-938-9228; Fax: 27-21-933-5790; E-mail: danheath@comcast.net.
 

Focus Points

• Algorithms can provide a useful scaffolding for the practice of
evidence-based pharmacotherapy of depression and anxiety disorders in primary care.
• Algorithms for the pharmacotherapy of mood and anxiety disorders must begin with careful diagnosis and assessment, and then proceed to the choice of a first-line medication.
• A step-by-step process can then be formulated toward issues, such as duration and dosage of an adequate trial and the approach to the treatment of the refractory patient.

Abstract

Depression and anxiety disorders are the most prevalent psychiatric disorders seen in the primary care setting. This volume provides a series of algorithms, or step-by-step guidelines, on the evidence-based pharmacotherapy of these conditions. It aims to provide a concise, logical, and user-friendly approach to the pharmacotherapy of depression and anxiety disorders in the primary care context.

 

Introduction

 

Mood and anxiety disorders are the most common psychiatric disorders in the community and in primary care settings.1 This educational review is designed to be an evidence-based reference for primary care clinicians using psychiatric medication to treat patients with these conditions. The algorithms included here cover the major mood and anxiety disorders, namely major depressive disorder (MDD), generalized anxiety disorder (GAD), obsessive-compulsive disorder (OCD), panic disorder, posttraumatic stress disorder (PTSD), and social anxiety disorder.

 

There are good reasons for formulating evidence-based approaches in medicine and psychiatry.2 Given the ever increasing volume of data and with multiple randomized controlled trials (RCTs) being undertaken annually, clinical practice guidelines that incorporate evidence on efficacy, safety, and effectiveness3 potentially allow integration of information and facilitate decision making.4 Nevertheless, current evidence-based approaches also have limitations. First, such approaches are only as good as the evidence base, which is often limited. In addition, many expert consensus guidelines are too complex to be appealing to clinicians and remain unread. The algorithms in this review provide a concise, logical, and user-friendly guideline to the pharmacotherapy of psychiatric disorders in the primary care context.

 

A medical algorithm is a step-by-step approach to describing diagnosis or treatment. The psychopharmacology algorithms presented invariably begin with diagnosis as step 1, an initial medication as step 2, and then proceed to step 7 for patients that prove particularly refractory to standard treatment approaches. Before proceeding to the algorithms themselves, some of their pros and cons are briefly considered.

 

Pros and Cons of Evidence-Based Algorithms

Why Evidence-Based Algorithms?

 

There are many good reasons for the recent proliferation of evidence-based guidelines for psychiatric disorders. First, the complexity of decision making in psychiatry has steadily increased in recent years as a result of the introduction of new medications, the completion of hundreds of clinical trials, and the growth in systematic and anecdotal information about the management of treatment-resistant conditions.4 Guidelines and algorithms may provide a concise and logical “scaffolding” with which to approach the practice of evidence-based medicine.

 

Second, there is increasing emphasis by government and private funding of health care on resource allocation and cost effectiveness in the practice of medicine. The field of pharmacoeconomics has grown rapidly and increasingly plays a role in the development of clinical practice guidelines.5 Once again, guidelines and algorithms provide a way of summarizing currently available information and specifying appropriate clinical options to minimize wasteful and suboptimal practices. They may be able to provide a pithy, step-by-step approach to cost-effective practices.

 

Third, the development of computer-based decision tools for psychiatric assessment and treatment encourages the formulation of algorithms. Many such programs are based on logical rules and these are often most usefully articulated with the aid of flow charts or algorithms.6 Guidelines can in turn be used not only to encourage best clinical practices, but also as a teaching tool and as a pointer to future research. As new clinical data is gathered, such rules should be revised in a timely way.

 

Limitations of Evidence-Based Algorithms

 

There are important limitations to evidence-based medicine that need to be borne in mind.6,7 First, evidence-based guidelines and algorithms are only as good as the evidence. Despite the important advances in the past several decades in medicine in general and psychiatry in particular, there are many important gaps in our knowledge about both treatment efficacy and effectiveness.8

 

Second, the application of guidelines and algorithms is dependent on the skills of the treating clinician. Thus, for example, accurate diagnosis is often required before a treatment guideline can be applied; inaccurate diagnosis will lead to the use of the wrong guideline. Guidelines are intended to provide a “scaffolding,” rather than to automate the practice of medicine.

 

Third, guideline and algorithm formulation is not the same as guideline and algorithm dissemination and application.9-11 There is, however, growing awareness of the factors that facilitate the closing of the gap between development and application.12-15 Not the least of these may be the complexity and unwieldliness of specialized clinical practice guidelines, particularly from the perspective of users in a primary care setting.

 

Applying Algorithms in the Clinic

 

Development of evidence-based approaches encourages a rigorous review of the literature. For example, the development of an algorithm requires the clinician to carefully justify clinical decisions in terms of the existing data. Furthermore, algorithms point to the future. The development of an algorithm provides an important focus on those areas where clinical practice, even when expert, is not supported by research and where further investigation is required.

 

Ideally, such algorithms have several characteristics, including reliability, flexibility, clarity, and multidisciplinary input.16 The algorithms here are not based on formal expert consenses or systematic meta-analyses, but rather on a summary of the previous literature. Their strength lies not so much in consensus and complexity, but rather, in their concision, logic, and user-friendliness.

 

The development of algorithms and their revision after critical assessment and after the gathering of new data is arguably a useful didactic and research tool. Nevertheless, all algorithms need to be applied with good clinical judgement. While algorithms can provide a useful overview, they may also run the risk of oversimplification. Simple algorithms may, for example, be irrelevant in patients with complex comorbid psychiatric and medical conditions. Algorithms are intended to provide a “scaffolding,” rather than to regulate or automate the practice of medicine. Certainly, algorithms will never replace the individual clinician; only an experienced and competent decision-maker is able to apply an algorithm in the clinic.

 

Major Depressive Disorder

 

An algorithm for the pharmacotherapy of depression is presented in Algorithm 1. Each line of the algorithm is briefly explained below. However, it is crucial to state at the outset that psychotherapy is often a useful treatment of depression, either alone in less severe cases or in conjunction with medication.17,18 Similarly, psychoeducation about depression may be an important adjunct to both pharmacotherapy and psychotherapy.19 Organizational strategies within primary care settings can also contribute to the improved diagnosis and treatment of depression.20

 

The algorithm here focuses on major depression. Nevertheless, there is increasing recognition that dysthymia also responds to standard antidepressant treatments.21,22 This disorder, which is characterized by chronic depressive symptoms, has significant associated morbidity when left untreated. It therefore deserves additional attention from both clinicians and researchers.

 

The algorithm here also assumes that the patient is an adult. The tricyclic antidepressants (TCAs) may offer some benefit for adolescents but are ineffective and often unsafe in children,23 and some of the newer antidepressants (eg paroxetine, venlafaxine) should be used only with a great deal of caution.24 In addition, differences in dosing and risk-benefit determination need to be considered (eg, clinicians are less likely to use untested augmentation strategies in children than in adults).25,26 Indeed, specialist consultation would seem advisable in younger patients.

 

Step 1

 

Depression should be diagnosed using specific diagnostic criteria, such as those outlined in the Diagnostic and Statisical Manual of Mental Disorders, Fourth Edition, Text-Revision, (DSM-IV-TR) (Table 1). A range of practical questionnaires that incorporate these criteria are available to help clinicians identify and diagnose patients with major depression.27,28 Particular attention should be paid to symptoms that are chosen as targets for pharmacotherapy, which include mood symptoms, associated symptoms (such as pain), and disability. It should be stressed that current evidence indicates that depression remains significantly underrecognized in primary care practice.29

 

By definition, the diagnosis of MDD is not made when there is a history of manic or hypomanic episodes. Rather, such patients are diagnosed with a bipolar disorder. For example, postpartum depressions are often in the bipolar spectrum. It is particularly important to exclude a history of mania or hypomania in patients with a family history of bipolar disorder. The treatment of depression in bipolar disorder is not discussed further here, but it can be noted that mood-stabilizing agents (rather than antidepressants alone) are the essential ingredients of its treatment.30,31

 

The differential diagnosis of depression includes not only other psychiatric disorders, but also mood disorders due to general medical conditions and substance-induced mood disorders. Most of these disorders can readily be excluded by means of a thorough medical history, physical examination, and routine blood and urine tests.32 At the same time, it is important to be aware of the comorbidity between depression and general medical disorders, with increasing evidence that depression contributes to the morbidity and mortality of illnesses such as diabetes and heart disease.33

 

Step 2

 

Depression may be complicated in several ways, impacting decisions about pharmacotherapy and other interventions. Brief explanations of these various complicating features and their implications for pharmacotherapy follow.

 

Melancholia

 

Melancholic features of depression include loss of pleasure in activities, lack of reactivity to pleasurable stimuli, and various neurovegetative symptoms such as exacerbation of depression in the morning, early-morning awakening, and significant weight loss. There is some evidence that TCAs and venlafaxine may be more effective than selective serotonin reuptake inhibitors (SSRIs) in patients with depression accompanied by melancholic features34 and in possibly related subgroups, such as in-patients with depression,35,36 although not all data is consistent.37 Certainly the presence of melancholic features in depression predicts a poor response to nonsomatic interventions.

 

Psychotic Features

 

Clinicians should be alert for psychotic symptoms in depression, as these are sometimes subtle. Depression with psychotic features is associated with increased risk for suicide and recurrent depression. Most authorities hold that it is important to use an antipsychotic agent in addition to an antidepressant in order to achieve a response.38 New-generation antipsychotics have a superior side-effect profile and are therefore a good choice in this context. Electroconvulsive treatment (ECT) is highly effective, and may also be considered as a first-line treatment for depressed patients with psychotic features.39

 

Atypical Features

 

Atypical features of depression include mood reactivity, as well as neurovegetative symptoms of reversed polarity (ie, increased rather than decreased sleep and appetite), severe lack of energy or leaden feelings in the limbs (leaden paralysis), and pathologic sensitivity to interpersonal rejection. Patients may present with only some of these symptoms. Classical monoamine oxidase inhibitors (MAOIs) are more effective than TCAs in atypical depression.40 However, in view of the inconvenience caused by dietary and other restrictions when using these MAOIs, SSRIs may be considered a first-line class of medication41 (despite the fact that not all studies of these agents in this subgroup have been positive).

 

Seasonal Affective Disorder

 

Seasonal affective disorder (SAD) is a subtype of depression where there is a regular temporal relationship between season (usually autumn/winter) and depression, which cannot be accounted for by seasonal stressors. SAD has been shown to respond to light therapy. Symptoms of depression are often atypical rather than typical and SSRIs may also be considered as a first-line treatment for SAD.42,43

 

Comorbid Anxiety or Anxiety Disorders

 

Comorbidity of depression and anxiety is extremely common. Given that short-term augmentation of an antidepressant with a benzodiazepine may reduce suffering and increase compliance,44,45 this combination seems reasonable to consider in depressed patients with high levels of anxiety. In patients with depression and comorbid panic disorder, it is often useful to begin at much lower doses of antidepressants than usual because anxiety symptoms are often initially exacerbated at ordinary doses. In patients with depression and comorbid OCD, it is advisable to use a serotonergic medication (ie, clomipramine or an SSRI). In patients with depression and comorbid social anxiety disorder, SSRIs and venlafaxine, rather than TCAs, would be favored as first-line agents.

 

Alcohol and/or Substance Use

 

A detailed history of substance abuse is essential in patients with depression because of the frequent comorbidity of these disorders. Depressed patients with comorbid substance use disorders are more likely to require hospitalization, more likely to attempt suicide, and less likely to comply with treatment. However, abstinence may itself lead to improvement in symptoms of depression. It may therefore be advisable to detoxify patients prior to beginning antidepressant treatment. On the other hand, a positive family history of depression, a history of depression preceding alcohol or other substance use, or a history of major depression during periods of sobriety raises the possibility that early intervention with antidepressants may be useful. There is some evidence that the SSRIs in particular may be useful as a primary treatment of alcohol dependence.46

 

Suicide

 

In patients with suicidal ideation, it is important to consider whether pharmacotherapy should be initiated in hospital or under supervision. The risk of suicide in some patients recovering from depression may increase transiently as they develop the energy and capacity to act on self-destructive plans made earlier in the course of their illness. If suicidal patients are treated as outpatients, it may be advisable to favor medications that are safe in overdose (such as the SSRIs). If TCAs are used in this setting, it may be advisable to prescribe only limited amounts of medication.

 

Pregnancy, Lactation, Menopause

 

Depression during pregnancy should be treated with nonpharmacotherapeutic interventions when possible. Where clinical considerations outweigh the risk of medication,47 one of the SSRIs or perhaps one of the TCAs may be considered48 in consultation with a specialist. In particular, there is a growing literature pointing toward relative safety of fluoxetine in pregnancy. ECT may be used in selected cases, or in patients in whom depression is complicated by psychotic features, as it is both safe and effective in depression during pregnancy. There is some evidence for the efficacy of antidepressants in postpartum depression49—they are often used prophylactically in patients with a history of postpartum depression, despite conflicting data on their value.48

 

Depression during lactation should be treated with nonpharmacotherapeutic interventions when possible. There is some evidence for the safety of SSRIs during lactation,50 although there is some passage of drug and metabolites to the breastfeeding infant. Sertraline has the lowest relative dose passed on to the infant, but expert recommendation is that the same SSRI to which the patient has responded should be continued.48 Specialist consultation should again be sought prior to using medication in this context. Depression may be exacerbated in susceptible patients during menopause and hormone replacement therapy may be considered as an adjunct to standard pharmacotherapy.48,51

 

Comorbid Medical Disorders and Medications

 

Antidepressants are effective in a wide range of general medical disorders with comorbid MDD.52 However, clinicians need to be aware of the multiple interactions between antidepressants and other medications.53 In addition, specific medical disorders in patients with depression may impact choice of antidepressant medication.

 

Cardiac Disease. TCAs are contraindicated in several cardiac conditions including ventricular arrhythmia, subclinical sinus node dysfunction, conduction defects (including asymptomatic conduction defects), prolonged QT intervals, and a recent history of myocardial infarction.54 Several SSRIs and ECT appear to be safe for patients with pre-existing and existing cardiac disease. A depressed patient with a history of any cardiac disorder should be monitored for the emergence of cardiac symptoms, electrocardiogram (EG) changes, and orthostatic blood-pressure decrements. Consultation with a cardiologist before and during antidepressant treatment may be advisable.

 

Stroke. It has been suggested that in the acute stroke period depression is associated with left-frontal and left-basal lesions, while in the subacute period depression is associated with right-frontal and right-parietal lesions.55 There is also some evidence that the TCA nortriptyline may be superior to an SSRI in poststroke depression.55

 

Epilepsy. Many antidepressants lower the seizure threshold and theoretically exert an adverse effect on seizure control in depressed patients with epilepsy. Although TCAs can still be used in such patients, the initial dosages should be lower than usual and subsequent dosage increases should be gradual. Epilepsy is not a contraindication to ECT.

 

Glaucoma, Obstructive Uropathy, etc. Anticholinergic effects of the TCAs are undesirable in a number of settings. TCAs may, for example, precipitate acute narrow-angle glaucoma in susceptible individuals (ie, those with shallow anterior chambers). Prostatism and other forms of bladder outlet obstruction are relative contraindications to the use of TCAs. Anticholinergic antidepressants may also exacerbate memory problems in patients with central nervous disorders such as Parkinson’s disease or dementia.56,57

 

Step 3

 

Any one of the different antidepressants can be used as a first-line treatment in uncomplicated patients with MDD. There is a good deal of experience not only with older medications, such as the TCAs and SSRIs, but also with newer agents, such as the serotonin norepinephrine reuptake inhibitors (SNRIs) and noradrenergic and specific serotonergic antidepressants (NaSSAs). Despite some suggestive findings from meta-analyses of depression trials,58 it is probably premature to conclude that any of the currently available antidepressants has a more rapid onset of action59 or is more effective.60 Thus, for now, the choice between antidepressants is primarily made on the basis of tolerability and side effects.

 

SSRIs are particularly useful as a first-line treatment in view of their established efficacy and tolerable side-effect profile.61 Prescription of therapeutic dosages is typically straightforward and most SSRIs appear relatively safe in overdose. Although some of the SSRIs remain more expensive than many of the TCAs, enhanced compliance resulting from ease of use and increased tolerability may in fact translate into greater cost effectiveness. SSRIs may be particularly helpful in younger patients62 and perhaps the elderly, where side effects of older antidepressants can be problematic.25

 

Venlafaxine (an SNRI), reboxetine (a noradrenergic reuptake inhibitor [NARI] not currently available in the US), and mirtazapine (a NaSSA) are also useful first-line options for the treatment of MDD. Venlafaxine is a predominant serotonin reuptake blocker at lower doses and a combined serotonin norepinephrine reuptake blocker at higher doses; this profile may contribute to its apparently pronounced effect on remission of symptoms.

 

Mirtazapine is an α2-noradrenergic antagonist (increasing noradrenergic and serotonergic transmission) and a 5-HT2 and 5-HT3 antagonist. This unique mode of action is associated with a useful range of clinical effects and side effects (eg, increased sedation, few sexual side effects). Bupropion, a norepinephrine and dopamine reuptake inhibitor (NDRI), is an antidepressant that is also marketed for decreasing the craving associated with smoking cessation and that is relatively free of sexual side effects.

 

TCAs have long been used in the treatment of depression. They work as quickly and are as effective as any of the more recently introduced antidepressants. Clinicians often prescribe subtherapeutic dosages of these agents, partly because of less tolerable side-effects at higher doses.63 The TCAs are also unsafe in overdose. On the other hand, low doses can be therapeutic for some patients64 and some TCAs (eg, the secondary tricyclics) have relatively favorable side-effect profiles,35,65 so that these agents remain an option even in the elderly.66,67

 

St. John’s wort may be a useful option in patients who refuse conventional medication, particularly when depression levels are only mild to moderate.68 A range of other herbals or nutriceuticals (eg, tryptophan, 5-hydroxytryptophan) have some evidence for efficacy, but cannot be recommended as first-line agents.69 The older, irreversible MAOIs, which require specific dietary restrictions, are no longer used as first-line agents.

 

Several factors not mentioned thus far may also influence choice of agent. Of course, agents that have been proven effective and tolerable in a particular individual in the past should be favored in the present, and vice versa. Similarly, a family history of response to a particular antidepressant may favor the choice of that agent.70 Patient preference for a particular agent is often a deciding factor.

 

An interesting area of current research focuses on interventions to decrease the response time of antidepressants. There are theoretical reasons for suggesting that certain combinations of agents, by combined receptor action, may be able to achieve this result. To date, however, such work has not translated into clinical recommendations.59

 

Important aspects of psychoeducation regarding the antidepressants include the facts that antidepressants are not addictive, side effects are typically transient, therapeutic response is relatively slow in onset, and if one agent does not work another should be tried.

 

Step 4

 

To determine response to medication, it is important to ask about changes in those symptoms initially targeted for treatment. Side effects of the medication should also be determined, with particular attention to those that patients may be reluctant to disclose (eg, sexual dysfunction). It may be useful to have the patient complete a rating scale of depression (Table 2) to help quantify response to medication. In addition to monitoring depression symptoms, it is important to ascertain overall change in objective disability and subject well-being (ie, quality of life).

 

Patients who are intolerant of a particular medication can of course be switched to another agent. When TCA side effects prove intolerable, it may well be useful to switch to an SSRI. Within the SSRIs, adverse effects may not be seen when an alternative SSRI is used. For example, some studies have found fluvoxamine to have a lower risk of sexual side effects.71 Mirtazapine and bupropion may be useful for patients who have specific SSRI side effects such as sexual dysfunction.

 

When there is a poor response to medication, it is important to optimize dosage and duration of the antidepressant. For many of the antidepressants, there is a linear relationship between dose and response and between dose and side effects. The steepness of the dose-curve relationship varies from drug to drug (it is relatively more flat for the SSRIs) and in some cases there is a therapeutic window beyond which increased doses are less effective (eg, nortriptyline). However, in general, optimal dosage is as close to maximum recommended doses that the patient can tolerate. Elderly patients generally require lower doses than younger adults.

 

As noted earlier, clinicians often prescribe suboptimal doses of antidepressants. This is particularly so in the case of the TCAs, where medications such as imipramine and amitriptyline are not typically raised to suggested average doses of 150–200 mg/day. Even in the case of the SSRIs, some patients may fail to respond to the standard initial starting dose, but do well at higher doses.72

 

Furthermore, there is increased awareness that some patients may be rapid metabolizers of medication and therefore require significantly higher doses than usual.73 Thus, for example, when patients on TCAs have little response and few anticholinergic side effects on average doses of medication (eg, imipramine 150–200 mg), it may be useful to further increase dosage while monitoring EG and perhaps drug levels. Response to medication may take 6–8 weeks or longer.25,74,75

 

While there is some evidence that patients who do not begin to respond within the first 3–4 weeks of treatment are more likely to have a relatively poor outcome at 8 weeks, it is important to give each patient a trial of medication that is of adequate duration.76

 

Step 5

 

Patients should be reassessed at the end of a clinical trial of optimal dosage and duration. Increasingly, the literature is emphasizing not only significant reduction in depressive symptoms, but also the value of aiming for near-complete remission of symptoms. Patients with reduced but not remitted symptoms continue to experience significant disability and are at increased risk for relapse.

 

When the patient has a good response to medication, it is important to reinforce the necessity for continuing the medication at the therapeutic dose despite this improvement. Guidelines for maintenance therapy of depression have become increasingly conservative, favoring longer courses of medication, in view of the safety of modern antidepressants and the likelihood of additional episodes of depression in patients with repeated episodes.77,78 It is reasonable to continue medication for at least 1 year before gradual tapering, and to continue for even longer in patients with a risk of recurrence. Risk factors for recurrence include history of multiple episodes, persistent dysthymic symptoms after recovery from MDD, and comorbid psychiatric and medical disorders. There is also increased awareness of the potential long-term adverse effects of antidepressants (ie, insomnia, weight gain, sexual problems), and these require careful monitoring. A combination of pharmacotherapy and psychotherapy may the most effective form of maintenance therapy.79

 

When there is a partial response despite an optimum trial of medication, it may be useful to consider an augmenting agent. Even among responders, residual symptoms of depression are common, and, as noted earlier, are associated with greater likelihood of relapse. There are few controlled studies directly comparing augmentation strategies with switching medications,80 although both seem effective in approximately 50% of cases.81,82 Augmentation offers the advantage of retaining any possible gains from the first agent, but the potential disadvantages of polypharmacy (more side effects, drug interactions).81

 

There is good evidence from controlled trials that lithium (at serum levels >0.4 mEq/L) and triiodothyronine (T3) up to 50 μg/day are effective in enhancing the response of TCAs,83 although there is less evidence for augmentation of other antidepressants including the SSRIs.84 Improvement tends to occur within 3–4 weeks, and, when successful, the augmenting agent is usually continued for 6–9 months before an attempt is made at a gradual taper. Augmentation with estrogen and atypical antipsychotics may be considered in certain treatment-refractory patients. Other agents (buspirone, pindolol, dopamine agonists and pyschostimulants, anticonvulsants, inositol, opiates, dehydroepiandosterone, folate S-adenosyl-methionine [SAMe], and omega-3 fatty acids) have also been suggested, but the data to date remains anecdotal or conflicting.81,85

 

In recent years there has also been a trend for specialists to combine antidepressants that affect different neurotransmitters sytems; for example, combing an SSRI with a noradrenergic agent.86 There are, however, few controlled studies that support augmentation of an antidepressant with a second antidepressant agent, and practitioners should be aware of the significant drug interactions that may ensue from such combinations.81 There is some controlled evidence, however, for adding mirtazapine to an SSRI.

 

Step 6

 

When MDD does not respond to a clinical trial of optimal dose and duration, it is useful to reassess a number of factors.87 The presence of certain features may impact the choice of the subsequent intervention.

 

Severity

 

In patients with a severe MDD, the need for hospitalization should be carefully monitored on a continuous basis. In addition, in such patients the need for ECT should be considered. Referral to a specialist may be advisable.

 

Compliance

 

It is not unlikely that clinicians often overestimate the compliance of their patients. Many patients worry that medication is addictive or is “a crutch.” It is well worth checking with the patient and family whether medication is in fact being taken on a daily basis regularly.

 

Comorbid Substance Use

 

In patients who fail to respond to pharmacotherapy of MDD, the possibility of comorbid substance use should be considered. There may be a need to detoxify the patient before tackling the depression.

 

Comorbid Personality Disorders

 

Although antidepressants may be useful,88 additional interventions, such as psychotherapy, may be crucial in patients with depression and comorbid personality disorders. While improvement in depressive symptoms may reduce maladaptive behavior in patients with comorbid personality disorder, there are other patients (eg, those with borderline personality disorder) in whom the personality disorder itself may need to be a major target of treatment.

 

Underlying Medical Disorder

 

Depressed patients who fail to respond to medication should be thoroughly reassessed for an underlying medical disorder. Apathy has been reported as an adverse event of some antidepressants, and may masquerade as depression.

 

Pharmacokinetic Issues

 

Drug-drug interactions may result in a subtherapeutic dose of the prescribed antidepressant.

 

Psychosocial Issues

 

Psychosocial circumstances that continue to complicate the course of a depression need to be assessed, as these may necessitate appropriate intervention.89

 

Step 7

 

After the failure of an adequate clinical trial of medication in a patient where reassessment sheds no light on any further unresolved factors, a different antidepressant should be used. Switching between classes of antidepressants has long been considered a useful strategy in nonresponders, and there is controlled data to support this idea for certain agents.82,90 There is also some evidence that switching from one SSRI to another may be useful.80 In most cases cross-tapering is possible, but half-life (eg, fluoxetine has the longest half-life of the SSRIs) and drug-interactions (eg, MAOIs should not be overlapped with other antidepressants) should be kept in mind.

 

Some medications may be particularly useful in cases of depression that do not respond to one or more antidepressants. Although some clinicians no longer use TCAs due to their side-effect profile, these agents continue to have a role, perhaps particularly in patients with melancholic features34 and in possibly related subgroups, such as in-patients with depression,35,36 although not all evidence is consistent.37 High doses of the SNRI venlafaxine (up to 375 mg) have been demonstrated to be effective in patients with treatment-resistant depression.91 Furthermore, the classic MAOIs remain a useful resource for patients that have not responded to other more commonly used classes of medication.92 ECT can be considered, and in the future other nonpharmacologic somatic interventions (transcranial magnetic resonance imaging, vagus-nerve stimulation) may also have a role.93,94

 

Estrogen augmentation may have a role in perimenopausal/postmenopausal treatment-refractory depression95,96 and in postpartum depression.96,97 The role of the atypical antipsychotics as augmenting agents in refractory depression also appears promising.98 Indeed, as response to augmentation may not be related to the degree of nonresponse to the preceeding monotherapy trial, augmentation strategies can sometimes be useful in patients who have not responded to multiple single agents.99

 

Generalized Anxiety Disorder

 

Recently, there have been important advances in the nosology and treatment of GAD. In particular, there is increasing evidence that patients with GAD and mixed anxiety-depression frequently present in primary care settings,100 and the DSM-IV-TR now provides fairly user-friendly criteria for the diagnosis of GAD (Table 3).

 

A few simple points can perhaps be made to help conceptualize GAD. In primary care practice, GAD can perhaps be viewed as a “tension disorder.” This is a useful term insofar as it crosses the boundary between psychic symptoms (worries, feeling keyed up, irritability) and somatic complaints (muscle tension, restlessness, insomnia). While GAD patients may not present with “worries,” they often describe themselves as “worriers”: worry may represent an avoidance behavior that is used to diminish tension (analogous to the way that agoraphobia may develop after panic attacks).

 

The algorithm presented here (Algorithm 2) provides a step-by-step approach to the pharmacotherapy of GAD, based on reading the research literature. It is important to mention at the outset, however, that psychotherapy approaches may be a first-line intervention in some GAD patients and should be considered in all patients with this disorder. In addition, psychoeducation is of the utmost importance, particularly in the initial stages of treatment, and should address the direct effects of anxiety on the life of the patient, as well as possible effects on family members.

 

Step 1

 

GAD is characterized by chronic, uncontrollable, and excessive worry, accompanied by a range of somatic symptoms. Making the correct diagnosis is essential. Given that GAD often presents with somatic symptoms and comorbid psychiatric disorders, the diagnosis is frequently overlooked. On the other hand, improvements in the nosology and treatment of GAD now make it useful to establish whether diagnostic criteria for this disorder are met.

 

Particular attention should be paid to the evaluation of symptoms that are chosen as targets for pharmacotherapy and to symptoms that may point to the presence of other psychiatric disorders. It is also useful to determine the severity of GAD symptoms using a scale, such as the Hamilton Rating Scale for Anxiety. It is possible that the situation in GAD mirrors that in depression, where less severe forms of the disorder respond equally well to pharmacotherapy and to psychotherapy.

 

It is also necessary to rule out the presence of comorbid psychiatric and medical disorders. Mood disorders, such as depression and dysthymia, and other anxiety disorders are common in patients with GAD. In addition, attention should be paid to the possibility of comorbid somatization disorder or substance abuse, dependence, or withdrawal. In particular, excessive alcohol or caffeine use may contribute to chronic anxiety symptoms and should be excluded. Children with pervasive anxiety likely deserve evaluation by a specialist before a diagnosis of GAD is made.

 

Step 2

 

Several factors may complicate GAD, thus impacting decisions about pharmacotherapy and other interventions. The most important of these factors, along with their treatment implications, are listed below.

 

Geriatric Patients

 

Research indicates that GAD in the elderly is not uncommon and is often accompanied by depression.101 Given the possibility of accumulation of the drug and consequent adverse effects, such as motor vehicle accidents, falls, and fractures, benzodiazepines (particularly in high doses or those with long half-lives) should be prescribed only with great caution in this population. In addition, dosages of many other psychotropics require adjustment in the elderly.

 

Alcohol and/or Substance Use

 

When the diagnosis of GAD predates the onset of substance abuse, treatment may be initiated relatively soon after abstinence. However, when symptoms of anxiety have their onset during substance abuse or withdrawal, it is likely that a longer period of abstinence is indicated prior to re-evaluation of the need for treatment. In addition, given the risk of dependence, benzodiazepines should be used with caution in patients with a history of substance abuse.102

 

Other Comorbid Disorders

 

As noted earlier, there is a high rate of comorbidity among GAD, other anxiety disorders, and mood disorders. GAD will often respond to the antidepressants that are used as first-line medication in these disorders and these agents should therefore be initially prescribed. Similarly, in patients with chronic anxiety and comorbid personality disorder (eg, borderline personality disorder), antidepressants may be particularly useful.

 

Pregnancy, Lactation, Menopause

 

Pharmacotherapy should ideally be avoided during pregnancy and lactation. Nevertheless, where clinical considerations outweigh the risk of medication, such intervention should be considered after consultation with a specialist. In particular, there is a growing literature pointing toward relative safety of fluoxetine in pregnancy.103 Certain benzodiazepines (eg, chlordiazepoxide) may be safer, while others (eg, alprazolam) should be avoided during pregnancy and lactation; the lowest effective dose should be prescribed for the shortest possible duration, and high-peak concentrations should be avoided by dividing the daily dosage into two or three doses.104 Anxiety symptoms may be exacerbated in susceptible patients during menopause, and hormone replacement therapy may be considered as an adjunct to standard pharmacotherapy.

 

Comorbid Medical Disorders and Medications

 

Clinicians need to be aware of the multiple interactions between medications used in the treatment of GAD and the treatment of other disorders, as well as of the impact of the medication’s adverse effects on medical disorders.

 

Step 3

 

The first-line treatment of uncomplicated GAD is somewhat controversial. At this point, given the substantial comorbidity of GAD with depression and other disorders for which antidepressants are effective, expert consensus favors the use of one of these agents (specifically, TCAs, venlafaxine, or SSRIs).105,106 However, a number of other agents may also be useful in patients with GAD.106-108

 

The TCAs have been shown effective in GAD in several controlled trials, although many of these agents have troublesome side-effect profiles. Venlafaxine is an SNRI that is effective in both short- and long-term studies of GAD and depression. There is growing evidence that the SSRIs are useful in the treatment of GAD, and a number of these agents (ie, escitalopram and paroxetine) are approved by the Food and Drug Administration for this indication. Other classes of antidepressants, such as the classic MAOIs (eg, phenelzine), may also be effective in the treatment of GAD, although relatively less has been published on their use in this disorder.109

 

There is a good deal of evidence for the efficacy of the benzodiazepines in GAD as well as for their safety during long-term use, and other authors have recommended them as first-line agents. Nevertheless, the high comorbidity of symptoms of depression in GAD, and the significant difficulties experienced by many patients during benzodiazepine withdrawal, constitute a strong argument against their use.

 

Benzodiazepines with longer half-lives or slow-release preparations may, however, be associated with fewer withdrawal problems. Furthermore, hydroxyzine, a tranquilizer that is an antagonist at both the H1 and the 5-HT2 receptor, has long been available, and has recently received renewed attention with studies demonstrating not only efficacy in GAD, but also a lack of withdrawal symptoms after discontinuation.110

 

Buspirone, a 5-HT1A agonist, takes 2–4 weeks or longer to begin working, and appears to be experienced as less helpful in patients recently treated with benzodiazepines.111 Its advantages lie in its benign side-effect profile, the lack of dependence, and its proven efficacy in GAD. Disadvantages include a lack of efficacy against the depressive symptoms often found in GAD and a lack of efficacy in some trials. Whereas some SSRIs have been shown useful in children and adolescents with GAD, a controlled study of buspirone in this population was negative.

 

Although β-blockers are often prescribed by general practitioners for anxiety symptoms, there is not sufficient evidence to include them as a first-line medication for GAD. Kava extract is an herbal that showed some promise for the treatment of anxiety,112 but it has not been studied rigorously enough in GAD and its safety remains unclear. Similarly, given their disadvantageous side-effect profiles (including tardive dyskinesia), one should be extremely cautious about the use of low-dose antipsychotic medications in the treatment of chronic anxiety, despite the anecdotal impression of many clinicians that these agents can be useful for this indication.

 

Step 4

 

The next step is to determine response to the medication. This is achieved by careful evaluation of change in symptoms initially targeted for treatment. These are typically the excessive worry, various somatic symptoms, and the consequent functional impairment. Determining the side effects of the medication is also important, as these may influence compliance. Gathering collateral information from the family of the patient may be useful in making these evaluations.

 

Patients who are intolerant of a particular medication can be switched to another agent or to another class of agents. For example, when TCA side effects prove intolerable, it may well be useful to switch to an SSRI. Within the SSRIs, adverse effects may not be seen when an alternative SSRI is used.

 

When there is a poor response to medication, the first course of action is to optimize dose and duration of the treatment. For many of the antidepressants, there is a linear relationship between dose and response and between dose and side effects. Thus, optimal dosage is as close to maximum recommended doses that the patient can tolerate. Elderly patients generally require lower doses than younger adults. Particularly in the case of the TCAs, clinicians often prescribe suboptimal doses, rather than using doses of 150 mg or more of medications such as imipramine. Even in the case of the SSRIs, some patients may fail to respond to the standard initial starting dose, but do well at higher doses.

 

Furthermore, there is increasing awareness that some patients may be rapid metabolizers of antidepressant medication and, therefore, require significantly higher doses than usual. Thus, for example, when patients on TCAs have little response and few anticholinergic side effects on average doses of medication (eg, imipramine 150 mg), it may be useful to further increase dosage while monitoring EG and perhaps drug levels. In addition, response to antidepressant medication may take 6–8 weeks or longer.

 

Buspirone treatment usually begins at 5 mg TID. This dose may be increased by 5 mg every 2–3 days. Therapeutic doses of buspirone range from 30–60 mg QD, typically given in divided doses. Buspirone has at least a 2–4 week time lag from initiation to clinical onset; optimum duration of a trial of treatment should thus be no less.

 

Although benzodiazepines are not suggested as a first-line treatment, should these be used, it is important to administer them in an optimal fashion. In particular, it may be useful to replace short-acting agents with slow-release compounds or long-acting agents. All too frequently, patients on short-acting compounds have intermittent increases of anxiety before the next dose of medication is to be taken.

 

Step 5

 

At the end of a clinical trial of optimal dose and duration, patients should be thoroughly reassessed. There is growing recognition of the importance of residual anxiety symptoms in causing disability and predicting relapse, and of the consequent necessity of aiming for remission of symptoms as the endpoint of treatment.113

 

When the patient has a good response to medication, it is important to reinforce the necessity for continuing the medication at the therapeutic dose despite this improvement.114 It has been said that more than half of anxious patients treated for <6 months with buspirone or benzodiazepines relapse at the end of 1 year. Indeed, guidelines for maintenance therapy of GAD emphasize the safety of modern agents, the likelihood of additional episodes of illness in patients with repeated past episodes, and the theoretical possibility that appropriate treatment may prevent the onset of secondary disorders.106 Such guidelines have become increasingly conservative, favoring longer courses of medication.

 

Neither augmentation nor switching strategies in GAD have been well researched. Augmentation offers the advantage of retaining any possible gains from the first agent, but the potential disadvantages of polypharmacy (more side effects and drug interactions).81 Given the literature on combining SSRIs with buspirone in the treatment of refractory depression, this strategy perhaps deserves consideration in GAD patients with partial response.

 

Step 6

 

When GAD does not respond to a clinical trial of adequate dose and duration, it may be useful to reassess a number of important factors that may influence choice of further interventions.

 

Comorbidity

 

It is important to establish whether comorbid mood or other anxiety disorders are present. For example, comorbid dysthymia may not respond to buspirone alone, comorbid social anxiety disorder is unlikely to respond to a TCA (other than clomipramine), and comorbid hypochondriasis may require high doses of serotonin reuptake inhibitors. Excluding important comorbid psychiatric disorders is perhaps the most important step in the evaluation and management of refractory GAD.

 

Compliance

 

Many patients with GAD suffer from extreme anxiety and are in fact compliant with their medication. Nevertheless, there is perhaps a tendency for clinicians to overestimate patient compliance. Patients are particularly likely to be concerned about physical or psychological dependence on medication. It is well worth checking not only with the patient, but also with the family, whether medication is in fact taken as prescribed.

 

Comorbid Substance Use

 

In the presence of active alcohol or substance use, it may be necessary to shift the emphasis of treatment toward a substance use disorder as the primary diagnosis, with the anxiety as a secondary problem. Detoxification is typically a first step in the management of these patients.115,116

 

Comorbid Personality Disorders

 

Although antidepressants may be useful, additional interventions, such as psychotherapy, may be helpful in patients with chronic anxiety and comorbid personality disorder. While improvement in anxiety symptoms may reduce maladaptive behavior in patients with comorbid personality disorder, there are other patients (eg, those with borderline personality disorder) in whom the personality disorder itself may need to be a major target of treatment.

 

Underlying Medical Disorder

 

Patients with GAD who fail to show any noticeable response to treatment should be thoroughly reassessed for the possibility of an underlying medical condition. A range of different medical disorders may lead to chronic anxiety, including endocrine disorders (eg, hyperthyroidism), respiratory disorders (eg, chronic obstructive pulmonary disorders), cardiac disorders (eg, congestive heart failure) and others. If present, such disorders naturally require specific intervention. Note that when using a benzodiazepine in patients with liver dysfunction, consider using those metabolized only by conjugation (eg, lorazepam, oxazepam).

 

Pharmacokinetic Issues

 

Drug-drug interactions may result in a subtherapeutic dose of the prescribed antidepressant.

 

Psychosocial Issues

 

In some cases, a diagnosis of an adjustment disorder with anxious features may be more accurate than that of GAD, and a psychotherapeutic approach therefore indicated. (This factor may partially explain high rates of placebo response in some clinical trials in GAD). In other cases of chronic anxiety, psychosocial factors may be enduring and therefore continuously complicate treatment of GAD until given independent attention.

 

Step 7

 

Following failure of an adequate clinical trial of medication, and in the event of a thorough reassessment providing no further important clues, the medication should be changed. The research literature on approaches to resistant GAD is unfortunately extremely scant.117 Referral to a specialist should be considered.

 

In such cases, following some of the strategies that have proven useful in the treatment of resistant depression is suggested. In particular, switching to a different class of antidepressants is recommended. Venlafaxine should be considered if it has not already been employed. In particularly resistant cases, one should consider the classical MAOIs.

 

One study has argued that GAD patients with more cognitive symptoms respond better to antidepressants, while those with more somatic symptoms respond better to a benzodiazepine. A minority of patients may, indeed, ultimately require long-term treatment with benzodiazepines after consideration of their benefits and risks.

 

Panic Disorder, Social Anxiety Disorder, and PTSD

 

Panic disorder, social anxiety disorder (social phobia), and PTSD are all anxiety disorders and are among the most common of the psychiatric disorders.1 Each is characterized by panic attacks or hyperarousal and by underdiagnosis or misdiagnosis as physical illness in primary care settings. In addition, each may be associated with significant morbidity and functional impairment. Indeed, it has been estimated that anxiety disorders account for one third of the costs of all mental illness.118

 

The role of psychotherapy in the treatment of panic disorder, social anxiety disorder, and PTSD must be emphasized.119 Cognitive-behavioral therapy can augment the anxiolytic effects of medication, is essential in reducing avoidance behaviors (eg, agoraphobia, avoidance of social situations), and may be important in ultimately allowing medication discontinuation. In addition, psychoeducation of both patient and family is crucial in the treatment of anxiety disorders.

 

Algorithm 3 assumes that the patient is an adult. Although there is less data on the treatment of children and adolescents with these disorders, what does exist suggests that a similar approach may be useful in some younger patients.120,121 Specialist consultation may, however, be indicated in such cases.

 

Step 1

 

The first step in any treatment algorithm involves correct diagnosis. Panic attacks (Table 4) may be present in panic disorder, social anxiety disorder, and PTSD, although shortness of breath may be more common in the panic attacks of panic disorder; blushing and stuttering may be more common in those of social anxiety disorder; and hyperarousal including a startle response is characteristic of PTSD.

 

Panic attacks in panic disorder and PTSD (Tables 5, 6) may be spontaneous or may be cued by exposure to stimuli previously associated with a panic attack (eg, a confined space), while panic attacks in social phobia (Table 7) are cued by feared social situations (eg, public speaking).

 

The initial evaluation should include assessment of the severity and frequency of panic attacks and hyperarousal, the degree of avoidance behavior, and the extent of functional impairment. Panic disorder, social anxiety disorder, and PTSD may all be associated with other psychiatric symptoms, particularly depressive and substance abuse symptoms, which therefore also deserve particular attention.

 

Panic attacks may occur in other psychiatric disorders, such as depression, specific phobias, OCD, and substance use disorders. In addition, certain general medical conditions (eg, hyperthyroidism) may present with panic attacks. Caffeine may exacerbate panic attacks as well. A thorough medical history, physical examination, and routine blood and urine tests, are therefore useful in patients presenting with panic attacks.

 

Step 2

 

Panic disorder, social anxiety disorder, and PTSD may be complicated in several ways, impacting decisions about pharmacotherapy. Brief explanations of these complicating factors and their implications for pharmacotherapy follow.

 

Severity

 

Patients with severe symptoms may require brief hospitalization to help contain symptoms. The possible association between panic disorder and increased risk of suicide, for example, needs to be taken seriously. In addition, acute administration of high-potency benzodiazepines (eg, alprazolam, clonazepam) may be necessary. Slow-release preparations or benzodiazepines with longer half-lives (eg, clonazepam) have the advantage of avoiding rebound anxiety between doses. While high-potency benzodiazepines have been shown to be effective in panic disorder and in social anxiety disorder in controlled trials, in view of their dependence potential, these agents should be reserved for acute rapid anxiety reduction in patients with severe symptoms.122

 

Melancholia

 

Patients with anxiety disorders often have comorbid depression, but this usually also responds to first-line anti-anxiety treatments such as the SSRIs. However, there is some evidence (albeit controversial) that TCAs and venlafaxine may be more effective than SSRIs in patients with depression accompanied by melancholic features34 and in possibly related subgroups, such as in-patients with depression,35,36 although not all data is consistent.37 Interestingly, clomipramine is a TCA with properties that overlap with the SSRIs in that it is relatively selective for the serotonin system. Melancholic features of depression include loss of pleasure in activities, lack of reactivity to pleasurable stimuli, and various neurovegetative symptoms, such as exacerbation of depression in the morning, early-morning awakening, and significant weight loss.

 

Alcohol and/or Substance Use

 

Alcohol and other substance use disorders are associated with exacerbation of anxiety disorders. Anxiety disorders may lead to use of alcohol and other substances in order to self-medicate anxiety. Thus, although it is generally advisable to detoxify patients prior to beginning pharmacotherapy, in some cases such pharmacotherapy is an integral part of the treatment of the secondary substance use disorder.102 There is some evidence that the SSRIs in particular may be useful as a primary treatment of alcohol dependence. Benzodiazepines are relatively contraindicated in these patients.46

 

Pregnancy, Lactation, Menopause

 

Pharmacotherapy should ideally be avoided during pregnancy and lactation. Furthermore, panic attacks may diminish during pregnancy. Nevertheless, where clinical considerations outweigh the risk of medication, such intervention should be considered after consultation with a specialist. In particular, there is a growing literature pointing toward relative safety of fluoxetine in pregnancy,103 and certain benzodiazepines may be safe if used circumspectly.104 Panic attacks may be exacerbated in susceptible patients during menopause, and hormone replacement therapy may be considered as an adjunct to standard pharmacotherapy.

 

Comorbid Medical Disorders and Medications

 

Clinicians need to be aware of the multiple interactions between medications used in the treatment of the anxiety disorders and other medications, as well as of the impact of medication adverse effects on medical disorders. Fortunately, certain SSRIs have relatively few interactions with other medications, and the SSRIs as a class are well tolerated in most medical disorders.

 

Step 3

 

In Algorithm 3, SSRIs or venlafaxine are suggested as the first-line treatment of panic disorder, social anxiety disorder, and PTSD. There is now a great deal of evidence that SSRIs are both well tolerated and effective in these disorders.105,123-128 More recently, venlafaxine has been shown to be useful in panic disorder and social anxiety disorder.

 

Panic disorder also responds to the TCAs and to certain other antidepressants (but not all). The relatively poor tolerability and safety profile of the TCAs and other older antidepressants, such as the MAOIs,129 means that these agents should not be used as first-line medications. No matter which antidepressant is used in the treatment of panic disorder, however, it is crucial to initiate treatment with very low doses (eg, fluoxetine 5 mg or imipramine 10 mg) in order to prevent early exacerbation of panic attacks.

 

Social anxiety disorder does not respond to the TCAs (with the possible exception of clomipramine, a predominantly serotonergic TCA), and these agents should not be used as first-line treatments of this disorder. When social anxiety symptoms are limited only to specific performance situations (“performance anxiety”), β-blockers can be prescribed on an as-needed basis. These agents appear particularly useful for reducing “peripheral” symptoms of anxiety such as tremor and palpitations.130 However, there is growing data that the SSRIs are useful both in more generalized social anxiety disorder (when symptoms extend to most social situations), and less generalized social anxiety disorder.131

 

PTSD also responds to TCAs and certain other antidepressants. The relatively poor tolerability and safety profile of TCAs and other older antidepressants, such as the MAOIs, means that these agents should not be used as first-line medications. Treatment response rate is possibly lower in some combat veterans than in civilians with PTSD, so that any pessimism about prognosis which derives from studies of PTSD in veterans should not apply to all patients.125

 

There are few studies showing that any one SSRI is superior to another in the treatment of anxiety disorders. Of course, agents that have been proven effective and tolerable in a particular individual in the past should be favored in the present, and vice versa. Similarly, despite an absence of supporting empirical data, many clinicians suggest that family history of response to a particular antidepressant favors the choice of that agent.

 

Important aspects of psychoeducation regarding the SSRIs and SNRIs include the fact that these are not addictive, that despite being termed antidepressants they are highly effective in anxiety disorders, that side effects are typically transient, that therapeutic response is relatively slow in onset, and that if one agent does not work another should be tried.

 

High-potency benzodiazepines (alprazolam, clonazepam) have also been shown to be effective in the treatment of panic disorder and social anxiety disorder. In PTSD these agents do not appear effective. Some authors have suggested that high-potency benzodiazepines qualify as first-line monotherapies in panic disorder and social anxiety disorder, but others have emphasized the potential problems of long-term treatment. A compromise position is that short-term augmentation of antidepressant agents with benzodiazepines may be useful in rapidly stabilizing symptoms, and should therefore be considered in panic disorder132 and perhaps social anxiety disorder, particularly when high levels of anxiety threaten to disrupt ongoing pharmacotherapy.

 

Although β-blockers are often prescribed by primary care practitioners for anxiety symptoms, there is not sufficient evidence to include them as a first-line medication for panic disorder, generalized social anxiety disorder, or PTSD. Similarly, given their disadvantageous side-effect profiles (including tardive dyskinesia), one should be extremely cautious about the use of low-dose antipsychotic medications in the treatment of anxiety symptoms, despite the anecdotal impression of many clinicians that these agents can be useful for this indication.

 

Step 4

 

To determine response to medication, it is important to ask about change in those symptoms initially targeted for treatment. Side effects of the medication should also be determined, with particular attention to those that patients may be reluctant to disclose (eg, sexual dysfunction). It may be useful to complete symptom rating scales (Tables 8, 9) in order to help quantify response to medication.

 

Patients who are intolerant of a particular medication can of course be switched to another agent. Within the SSRIs, adverse effects may not be seen when an alternative SSRI or clomipramine is used.

 

When there is a poor response to medication, it is important to optimize dosage and duration of the medication. Dose-response relationship of the SSRIs has not been as well studied in the anxiety disorders as in depression. Optimal dosage may be as close to maximum recommended doses that the patient can tolerate (usually 2–3 times starting dose). Certainly, the lowest effective antidepressant dose of an SSRI may not be sufficient for some patients with panic disorder, social anxiety disorder, and PTSD. Elderly patients generally require lower doses than younger adults do.

 

Furthermore, response to SSRIs in these disorders may take 6–8 weeks or longer. PTSD in particular may be associated with a relatively slow response to antidepressant treatment. It is important to give each patient a trial of medication that is of adequate duration.

 

Step 5

 

At the end of a clinical trial of optimal dose and duration, the patient should be thoroughly reassessed. There is growing recognition of the importance of residual anxiety symptoms in causing disability and predicting relapse, and of the consequent necessity of aiming for remission of symptoms as the endpoint of treatment.113

 

When the patient has a good response to medication it is important to reinforce the necessity for continuing the medication at the therapeutic dose despite this improvement. Guidelines for maintenance therapy of anxiety disorders have become increasingly conservative, favoring longer courses of medication, in view of the safety of modern antidepressants and the likelihood of relapse in patients with an untreated chronic illness. In panic disorder, social anxiety disorder, and PTSD it is not unreasonable to continue medication for at least a year before gradual tapering.126-128,133 Cognitive-behavioral therapy may be useful prior to beginning and during medication withdrawal in order to maintain gains, although more research on the optimal combination and sequencing of pharmacotherapy and psychotherapy in these disorders is needed.

 

When there is a partial response despite an optimum trial of medication, it may be useful to consider an augmenting agent. Neither augmentation nor switching strategies in these conditions have been well researched. Augmentation offers the advantage of retaining any possible gains from the first agent, but the potential disadvantages of polypharmacy (more side effects, drug interactions).81 Of all the augmentation strategies in the treatment of the anxiety disorders, however, arguably the most important is augmentation of pharmacotherapy with additional psychotherapy.134,135

 

In panic disorder, high-potency benzodiazepines have been anecdotally described as useful in treatment-resistant panic disorder,136 and these agents (particularly slow-release agents or those with a longer half-life) may be expected to reduce anxiety secondary to antidepressants and to combat the primary disorder. However, the potential for dependence must be borne in mind. The anticonvulsant gabapentin has shown some efficacy as monotherapy in panic disorder,137 does not have dependence potential, and is another theoretical possibility for use as an augmenting agent. An alternative strategy noted in uncontrolled reports, is to augment an SSRI with low doses of a predominantly noradrenergic TCA.138,139 However, such strategies demand caution as very high levels of a TCA may result if combined with certain SSRIs. Another possibility is the use of pindolol to augment an SSRI—a strategy found useful in a preliminary controlled trial in panic disorder.140 Specialist consultation may be advisable when augmentation seems indicated.

 

In social anxiety disorder, given its efficacy as monotherapy, augmentation with clonazepam is a theoretical consideration.141 Nevertheless, this again runs the risk of benzodiazepine dependence. Buspirone augmentation has been noted to be useful in an open-label trial. The anticonvulsant gabapentin also showed some efficacy as monotherapy in social anxiety disorder,142 does not have dependence potential, and is another theoretical possibility for use as an augmenting agent. Nevertheless, there is a paucity of research on augmentation strategies in social anxiety disorder. Pindolol augmentation of a SSRI was ineffective in a controlled trial of social anxiety disorder.143

 

In PTSD, a range of augmentation strategies have been suggested, but few have been studied in controlled trials.144 The anticonvulsant lamotrigine showed some efficacy as monotherapy in PTSD,145 and is a theoretical possibility for use as an augmenting agent. There is a growing literature on the use of atypical antipsychotics as monotherapy for PTSD, and these medications therefore also have a possible role as augmentating agents.144 Again, however, further research in this area is needed before definitive recommendations can be made. In the interim, specialist consultation should be sought in such cases.

 

Step 6

 

When anxiety disorders do not respond to a clinical trial of optimal dose and duration, it is useful to reassess a number of factors. The presence of certain features may impact on the choice of the subsequent intervention.

 

Compliance

 

Clinicians often overestimate the compliance of their patients and it is often useful to check with patients and their families whether medication is being taken as prescribed. Many patients worry that medication is addictive or is a “crutch.”

 

Comorbid Substance Use

 

In patients who fail to respond to pharmacotherapy, the possibility of comorbid substance use should again be considered. Self-medication with alcohol and other substances is particularly common in social anxiety disorder and PTSD. There may be a need to withdraw the patient before tackling the anxiety disorder per se.115,116

 

Comorbid Personality Disorders

 

Although SSRIs may be useful, additional interventions, such as psychotherapy, may be crucial in patients with anxiety disorder and comorbid personality disorder. While improvement in anxiety symptoms may reduce maladaptive behavior in comorbid personality disorder, the personality disorder itself may need to be a major target of treatment.

 

Underlying Medical Disorder

 

Anxious patients who fail to respond to medication should be thoroughly reassessed for an underlying medical disorder.

 

Pharmacokinetic Issues

 

Drug-drug interactions may result in a subtherapeutic dose of the prescribed antidepressant.

 

Psychosocial Issues

 

Psychosocial circumstances that continue to complicate the course of an anxiety disorder need to be assessed, as these may necessitate appropriate intervention. Loss or separation may precipitate panic disorder, and this may need to be addressed. Ongoing avoidance patterns, such as avoidance of feared social situations in social anxiety disorder, may need to be specifically addressed. PTSD may be associated with disruptions in interpersonal relationships, altered perceptions of the self and others, survivor guilt, etc, which require specific intervention.

 

Step 7

 

After the failure of an adequate clinical trial of medication in a patient where reassessment sheds no light on any further unresolved factors, a different agent should be used. It is often advisable to switch classes of medication. It is also possible that switching from one SSRI to another may be useful.

 

Some medications may be particularly useful in cases of treatment-resistant anxiety disorders. Venlafaxine has been suggested to be useful in patients with social anxiety disorder who have failed to respond to SSRIs.146 The classical irreversible MAOIs, despite necessitating measures, such as a specialized diet, remain a useful resource for patients with panic disorder, social anxiety disorder, or PTSD who have not responded to other more commonly used classes of medication.147 Finally, augmentation strategies can sometimes be useful in patients who have not responded to multiple single agents.

 

If all avenues have been thoroughly explored, a second opinion by an expert is indicated. Medication trials that ended prematurely or that did not use optimal dosing can be repeated. It may ultimately be necessary to revert to the regimen on which the patient demonstrated the best response.

 

Obsessive-Compulsive and Related Disorders

 

OCD is characterized by obsessions and compulsions and is classified as an anxiety disorder. It is interesting to note, however, that a number of other disorders are also characterized by repetitive thoughts and rituals, and though not classified as anxiety disorders, they may also respond to standard OCD treatment. These so-called OCD spectrum disorders include body dysmorphic disorder (characterized by recurrent concerns with imagined ugliness), hypochondriasis (characterized by recurrent concerns with imagined illness), and trichotillomania (characterized by recurrent hair-pulling).148

 

Step 1

 

Current evidence indicates that OCD is commonly underdiagnosed and undertreated.149 There is also the converse possibility that various disorders with intrusive symptoms, such as PTSD or GAD, can be misdiagnosed as OCD. Diagnostic criteria for OCD are provided in Table 10.

 

Most patients with OCD have both obsessions (which increase anxiety) and compulsions (which aim to decrease anxiety), particularly now that the DSM-IV-TR definition of compulsion includes mental rituals. Evaluation should include assessment of symptom pattern, severity, and functional impairment. Comorbid axis I and II disorders, including tic disorders, as well as medical conditions (including pregnancy) and disorders need to be accurately identified. There is growing evidence that OCD and/or tics in some patients, particularly children, are precipitated or exacerbated by streptococcal throat infections.150

 

Evaluation of the OCD patient also requires attention to psychosocial factors that may have precipitated or exacerbated OCD symptoms. For example, are family members involved in the patient’s rituals? What is the patient’s explanatory model of OCD—does he or she regard it as a sign of weakness or as evidence of brain dysfunction? Certainly, psychoeducation as part of the management of OCD is crucial. Similarly, cognitive-behavioral techniques are an important aspect of OCD treatment, whether used alone or in combination with medication.151

 

In the discussion here, we assume that the patient is an adult. Nevertheless, there is increasing data on the pharmacotherapy of OCD in children.152,153 Indeed, Algorithm 4 here can readily be adapted for children, bearing in mind considerations such as differences in dosing and differences in risk-benefit determination (eg, clinicians are less likely to use untested augmentation strategies in children). Specialist consultation may, however, be indicated in such cases.

 

Step 2

 

Possible complications in OCD that may impact on pharmacotherapy include the following.

 

Severity

 

Patients with severe symptoms may require brief hospitalization to help contain symptoms. In general, however, the principles of behavior therapy suggest that patients should attempt to continue with their ordinary daily routines where possible.

 

Melancholia

 

There is some evidence (albeit controversial) that TCAs may be more effective than SSRIs in patients with depression accompanied by melancholic features34 and in possibly related subgroups, such as in-patients with depression,35,36 although not all data is consistent.37 Melancholic features of depression include loss of pleasure in activities, lack of reactivity to pleasurable stimuli, and various neurovegetative symptoms such as exacerbation of depression in the morning, early-morning awakening, and significant weight loss. The only TCA that is effective in OCD is clomipramine.

 

Tourette’s Disorder

 

Tourette’s disorder (TD) is characterized by both motor and vocal tics. Many patients with TD have comorbid OCD. Although this OCD may respond to standard OCD treatments, additional medication that targets the tics (eg, dopamine blockers such as haloperidol, pimozide, or risperidone) may be necessary for resolution of the range of symptoms that characterize the disorder.154

 

Pregnancy, Lactation, Menopause

 

Pharmacotherapy should ideally be avoided during pregnancy and lactation. Nevertheless, where clinical considerations outweigh the risk of medication, such intervention should be considered after consultation with a specialist. In particular, there is a growing literature pointing toward the relative safety of fluoxetine in pregnancy.103

 

Comorbid Medical Disorders and Medications

 

Clinicians need to be aware of the multiple interactions between medications used in the treatment of OCD and other medications, as well as the impact of a medication’s adverse effects on medical disorders. Fortunately, certain SSRIs have relatively few interactions with other medications, and the SSRIs as a class are well tolerated in most medical disorders.

 

Step 3

 

The first line of medication in the treatment of OCD should comprise a serotonin reuptake inhibitor (SRI). Consistent with growing evidence for the importance of serotonin in OCD, both clomipramine and the SSRIs appear to be more effective than desipramine in the treatment of OCD.155,156 Furthermore, the efficacy and safety of clomipramine and the SSRIs in the treatment of OCD have been well researched, with studies indicating that at least half of patients will respond to one of these agents. The SRIs are also useful for body dysmorphic disorder, hypochondriasis, obsessive-compulsive symptoms in TD, and possibly in trichotillomania (albeit with relatively less robust responses), compulsive skin-picking and other stereotypic movements, so-called compulsive sexual behavior, and pathological gambling.148,157

 

An immediate question, however, is which SRI to use first. Given the apparent lack of differences in efficacy between the SRIs, the side-effect profile of these agents may be an important question in considering which agent to use first. Certainly, there are invariably fewer side effects during treatment with the SSRIs than during treatment with clomipramine. Thus, it seems reasonable to suggest that treatment of OCD be initiated with a SSRI. While any SSRI is acceptable, choice of a SSRI may be influenced by characteristic side-effect profiles of the different SSRIs, previous individual and familial responses to pharmacotherapy, etc. Low doses should initially be used in patients with comorbid panic disorder.

 

Step 4

 

To determine response to medication, it is important to ask about change in those symptoms initially targeted for treatment. Side effects of the medication should also be determined, with particular attention to those that patients may be reluctant to disclose (eg, sexual dysfunction). It may be useful to complete a symptom rating scale (Table 11) to help quantify response to medication.

 

Patients who are intolerant of a particular medication can of course be switched to another agent. Within the SRIs, adverse effects may not be seen when an alternative SSRI or clomipramine is used.

 

When there is a poor response to medication, it is important to optimize dosage and duration of the medication. Although some patients with OCD respond to standard doses of SRIs, others require doses that are much higher than in depression. In adults, clomipramine should be increased to approximately 250 mg, and the SSRIs should be increased to maximal dosages (eg, fluoxetine 60–80 mg). Unfortunately, the likelihood of side effects increases at these high doses. Electrocardiogram monitoring isnecessary when children and adolescents, or patients with pre-existing heart disease, are treated with clomipramine.

 

Furthermore, response to SSRIs in OCD may take rather longer than in many other disorders—up to 12 weeks. It is obviously important to give each patient a trial of medication that is of adequate duration. Therefore, patients need to be educated that response may take a significant length of time and that they need to remain optimistic even when no change is seen at first.

 

Step 5

 

At the end of a clinical trial of optimal dose and duration, patients should be thoroughly reassessed. There is growing recognition of the importance of residual anxiety symptoms in causing disability and predicting relapse, and of the consequent necessity of aiming for remission of symptoms as the endpoint of treatment.113 Nevertheless, many OCD patients who are judged “responders” to medication therapy may continue to experience obsessions and compulsions, albeit with less intensity. In clinical trials, a decrease of 25% to 35% on the Yale-Brown Obsessive-Compulsive Scale typically corresponds to a categorical treatment response.

 

In patients where an SRI is effective, maintenance pharmacotherapy should be instituted. Rapid discontinuation of these agents risks the return of symptoms. Nevertheless, a maintenance dose of SSRIs in OCD may be lower than the dose initially required during acute treatment.158 At least 1 year of maintenance pharmacotherapy is reasonable. When a decision is made to attempt discontinuation of medication, it is advisable to taper medication off slowly (eg, by 25% every 2 months). Concomitant behavioral treatment during pharmacotherapy may well increase chances of being able to discontinue medication without relapse.

 

Comparison of augmentation with switching strategies in OCD has not been well researched. Augmentation offers the advantage of retaining any possible gains from the first agent, but the potential disadvantages of polypharmacy (more side effects, drug interactions).81 Of all the augmentation strategies in the treatment of OCD, perhaps the most important is augmentation of pharmacotherapy with additional psychotherapy.159 However, when there is a partial response despite an optimum trial of medication, or when there are comorbid tics, it may be useful to consider an augmenting medication. Certainly, in patients with comorbid tics, there is good evidence that augmentation of a SRI with a dopamine blocker can be effective.160 The introduction of the new-generation antipsychotics has led to increased use of these agents in the augmentation therapy of OCD, and they appear useful even treatment-refractory patients even in the absence of comorbid tics.161 Another possible strategy is to supplement an SSRI with a low dose of clomipramine,162 although careful monitoring of adverse effects and electrocardiographs may be warranted with such a combination. Other augmentation strategies have been suggested, but there are few positive controlled trials. There is also relatively little work on augmentation strategies in OCD-related disorders, although addition of a dopamine blocker may also be useful in some of these patients.163

 

Step 6

 

When OCD does not respond to a clinical trial of optimal dose and duration, it is useful to reassess a number of factors. The presence of certain features may impact on the choice of the subsequent intervention.

 

Compliance

 

Clinicians often overestimate the compliance of their patients and it is often useful to check with patients and their families whether medication is being taken as prescribed. Many patients worry that medication is addictive or is a “crutch.”

 

Comorbid Substance Use

 

In patients who fail to respond to pharmacotherapy, the possibility of comorbid substance use should again be considered. There may be a need to withdraw the patient before tackling the anxiety disorder per se.115,116

 

Comorbid Personality Disorders

 

Although SSRIs may be useful, additional interventions, such as psychotherapy, may be crucial in patients with OCD and comorbid personality disorder. While improvement in OCD symptoms may reduce maladaptive behavior in comorbid personality disorder, the personality disorder itself may need to be a major target of treatment.

 

Underlying Medical Disorders

 

Anxious patients who fail to respond to medication should be thoroughly reassessed for an underlying medical disorder. In OCD in children, for example, the role of streptococcal throat infection may be of particular importance.

 

Pharmacokinetic Issues

 

Drug-drug interactions may result in a subtherapeutic dose of the prescribed antidepressant.

 

Psychosocial Issues

 

Psychosocial circumstances that continue to complicate the course of OCD need to be assessed, as these may necessitate appropriate intervention. In particular, the participation of friends and family in rituals may serve to derail treatment.

 

Step 7

 

After the failure of an adequate clinical trial of medication in a patient where reassessment sheds no light on any further unresolved factors, a different agent should be used. Although an SRI has less chance of being effective in patients who have already failed a number of trials of other SRIs, some of these patients will in fact ultimately respond to a new SRI.164 Given the possible superiority of clomipramine in certain cases of OCD and depression, it may be argued that all OCD patients who have failed to respond to one or more of the SSRIs deserve a trial of clomipramine.165 While results of studies correlating plasma drug levels and therapeutic response in OCD have been mixed, in the case of clomipramine, obtaining drug levels at high doses may be useful. Anecdotal experience suggests that certain non-SRI agents, such as the classical MAOIs and venlafaxine, may on occasion be effective in treatment-resistant OCD.166 Recent trials of intravenous clomipramine also show efficacy in treatment-resistant OCD.167

 

For patients who have failed multiple medication and behavioural treatments (including intensive partial or full hospitalization programs),168 and where severity of the disorder is marked, neurosurgery should also be considered.169 Several studies have suggested that specific lesions to the corticostriatal pathways may lead to significant reduction in OCD symptoms in treatment-refractory patients. Patients can be referred to specialized centers in the United States, Sweden, and elsewhere for such treatments. ECT may anecdotally be helpful for some OCD patients,170 and research on transcranial magnetic stimulation or deep-brain stimulation in OCD may ultimately provide new methods of treatment.171 PP

 

References

1. Kessler RC, McGonagle KC, Zhao S, et al. Lifetime and 12-month prevalence of DSM-III-R psychiatric disorders in the United States: results from the National Comorbidity Survey. Arch Gen Psychiatry. 1994;51:8-19.
2. Rush AJ. Clinical practice guidelines: Good news, bad news, or no news? Arch Gen Psychiatry. 2001;50;483-490.
3. Wells KB. Treatment research at the crossroads: the scientific interface of clinical trials and effectiveness research. Am J Psychiatry. 1999;156:5-10.
4. Eddy DM. Clinical Decision-Making: From Theory to Practice. Sudbury, MA: Jones and Bartless; 1996.
5. Gold MR, Siegel JE, Russell LB, Weinstein MC. Cost-Effectiveness in Health and Medicine. New York, NY: Oxford University Press; 1996.
6. Stein DJ, Patterson B, Hollander E. Expert systems for psychiatric pharmacotherapy. Psych Annals. 1994;24:37-41.
7. Fawcett J, Stein DJ, Jobson KO. Textbook of Treatment Algorithms in Psychopharmacology. Chichester, UK: John Wiley & Sons; 1999.
8. Klein DF. Clinical psychopharmacological practice: the need for a developing research base. Arch Gen Psychiatry. 1993;50:491-494.
9. Daniels A, Biesma R, Otten J, et al. Ambivalence of primary health care professionals towards the South African guidelines for hypertension and diabetes. S Afr Med J. 2000;90:1206-1211.
10. Peveler R, Kendrick T. Treatment delivery and guidelines in primary care. Br Med Bull. 2001;57:193-206.
11. Bauer MS. A review of quantitative studies of adherence to mental health clinical practice guidelines. Harv Rev Psychiatry. 2002;10:138-153.
12. Rush AJ, Crismon ML, Toprac MG, et al. Implementing guidelines and systems of care: experiences with the Texas Medication Algorithm Project (TMAP). J Pract Psychiatry Behav Health. 1999;5:75-86.
13. Freeman AC, Sweeney K. Why general practitioners do not implement evidence: qualitative study. BMJ. 2001;323:1100.
14. Stein DJ, Xin Y, Osser D, Li X, Jobson K. Clinical psychopharmacology guidelines: different strokes for different folks. World J Biol Psychiatry. 2002;3:64-67.
15. Bero L, Grilli R, Grimshaw JM, Harvey E, Oxman AD, Thomson MA. Closing the gap between research and practice: an overview of systematic reviews of interventions to promote implementation of research findings by health care professionals. BMJ. 1998;317:465-468.
16. Field MJ, Lohr KN. Clinical Practice Guidelines: Directions for a New Program. Washington, DC: National Academy Press; 1990.
17. Persons JB, Thase ME, Crits-Christoph, P. The role of psychotherapy in the treatment of depression: review of two practice guidelines. Arch Gen Psychiatry. 1996;53:283-290.
18. Parker G, Roy K, Eyers K. Cognitive behavior therapy for depression? Choose horses for courses. Am J Psychiatry. 2003;160:825-834.
19. Grilli R, Freemantle N, Minozzi S, Domenighetti G, Finer D. Mass media interventions: effects on health services utilization. Cochrane Database Syst Rev. 2002;1:CD000389.
20. Gilbody S, Whitty P, Grimshaw J, Thomas, R. Educational and organizational interventions to improve the management of depression in primary care: a systematic review. JAMA. 2003;289:3145-3151.
21. de Lima S, Hotopf M. A comparison of active drugs for the treatment of dysthymia. Cochrane Database Syst Rev. 2003;3:CD004047.
22. Nobile M, Cataldo G, Marino C, Molteni M. Diagnosis and treatment of dysthymia in children and adolescents. CNS Drugs. 2003;17:927-946.
23. Hazell P, O’Connell P, Heathcote D, Henry, D. Tricyclic drugs for depression in child and adolescents. Cochrane Database Syst Rev. 2003;4:CD002317.
24. Ramchandani P. Treatment of major depressive disorder in children and adolescents. BMJ. 2004;328:3-4.
25. Ambrosini PJ. A review of pharmacotherapy of major depression in children and adolescents. Psychiatr Services. 2000;51:627-633.
26. Birmaher B, Brent DA, Benson RS. Summary of the practice parameters for the assessment and treatment of children and adolescents with depressive disorders: American Academy of Child and Adolescent Psychiatry. J Am Acad Child Adolesc Psychiatry. 1998;37:1234-1238.
27. Gilbody SM, House AO, Sheldon TA. Routinely administered questionnaires for depression and anxiety: systematic review. BMJ. 2001;322:406-409.
28. Williams JW Jr, Noël PH, Cordes JA, Ramirez G, Pignone M. Is this patient clinically depressed? JAMA. 2002;287:1160-1170.
29. Montano CB. Recognition and treatment of depression in a primary care setting. J Clin Psychiatry. 1994;55(suppl):18-34.
30. Srisurapanont M, Yatham LN, Zis AP. Treatment of acute bipolar depression: a review of the literature. Can J Psychiatry. 2002;40:533-544.
31. Zornberg GL, Pope HG Jr. Treatment of depression in bipolar disorder: new directions for research. J Clin Psychopharmacol. 1993;13:397-408.
32. Hall RC, Gardner ER, Popkin MK, et al. Unrecognized physical illness prompting psychiatric admission: a prospective study. Am J Psychiatry. 1981;183:629-635.
33. Katon W. Clinical and health services relationships between major depression, depressive symptoms, and general medical illness. Biol Psychiatry. 2003;54:216-226.
34. Roose SP, Glassman AH, Attia E, et al. Comparative efficacy of selective serotonin reuptake inhibitors and tricyclics in the treatment of melancholia. Am J Psychiatry. 1994;151:1735-1739.
35. Anderson IM. Selective serotonin reuptake inhibitors versus tricyclic antidepressants: a meta-analysis of efficacy and tolerability. J Affect Disord. 2000;58:19-36.
36. Clerc GE, Ruimy P, Verdeau-Palles J. A double-blind comparison of venlafaxine and fluoxetine in patients hospitalized for major depression and melancholia. The Venlafaxine French Inpatient Study Group. Int Clin Psychopharmacol. 1994;9:139-143.
37. Joyce PR, Mulder RT, Luty SE, McKenzie JM, Rae AM. A differential response to nortriptyline and fluoxetine in melancholic depression: the importance of age and gender. Acta Psychiatr Scand. 2003;108:20-23.
38. Rothschild AJ. Challenges in the treatment of depression with psychotic features. Biol Psychiatry. 2003;53:680-690.
39. Kantor SJ, Glassman AH. Delusional depression: natural history and response to treatment. Br J Psychiatry. 1977;131:351-360.
40. Liebowitz MR, Quitkin FM, Stewart JW, et al. Antidepressant specificity in atypical depression. Arch Gen Psychiatry. 1988;45:129-137.
41. Lonnqvis J, Shivo S, Syvalahti E, et al. Moclobemide and fluoxetine in atypical depression: a double-blind trial. J Affect Disord. 1994:169-177.
42. Lam RW, Gorman CP, Michalon M, et al. Multicentre, placebo-controlled study of fluoxetine in seasonal affective disorder. Am J Psychiatry. 1995;152:1765-1770.
43. Partonen T, Lonnqvist J. Moclobemide and fluoxetine in treatment of seasonal affective disorder. J Affect Disord. 1996;41:93-99.
44. Smith WT, Longboorg PD, Glaudin V, et al. Short-term augmentation of fluoxetine with clonazepam in the treatment of depression: a double-blind study. Am J Psychiatry. 1998;155:1339-1345.
45. Furukawa TA, Streiner DL, Young LT. Antidepressant plus benzodiazepine for major depression. Cochrane Database Syst Rev. 2001;2:CD001026.
46. Naranjo CA, Knoke DM. The role of selective serotonin reuptake inhibitors in reducing alcohol consumption. J Clin Psychiatry. 2001;62(suppl 20):18-25.
47. Wisner KL, Zarin DA, Holmbroe ES, et al. Risk-benefit decision making for treatment of depression during pregnancy. Am J Psychiatry. 2000;157:1933-1940.
48. Altshuler LL, Cohen LS, Moline ML, et al. The expert consensus guideline series: treatment of depression in woman. Postgrad Med. 2001;3(spec no):1-116.
49. Hoffbrand S, Howard L, Crawley H. Antidepressant drug treatment for postnatal depression. Cochrane Database Syst Rev. 2001;2:CD002018.
50. Miller LJ. Postpartum depression. JAMA. 2002;287:762-765.
51. Schneider LS, Small GW, Hamilton SH, et al. Estrogen replacement and response to fluoxetine in a multicenter geriatric depression trial: Fluoxetine Collaborative Study Group. Am J Geriatr Psychiatry. 1997;5:97-106.
52. Gill D, Hatcher S. Antidepressants for depression in medical illness. Cochrane Database Syst Rev. 2000;4:CD001312.
53. Nemeroff CB, DeVane CL, Pollock BG. Newer antidepressants and the cytochrome P450 system. Am J Psychiatry. 1996;153:311-320.
54. Roose SP, Glassman AH, Giardina EG, et al. Tricyclic antidepressants in depressed patients with cardiac conduction disease. Arch Gen Psychiatry. 1974;44:273-275.
55. Robinson RG, Schultz SK, Castillo C, et al. Nortriptyline versus fluoxetine in the treatment of depression and in short-term recovery after stroke: a placebo-controlled, double-blind study. Am J Psychiatry. 2000;157:351-359.
56. Shabnam GN, Th C, Kho DH, Ce C. Therapies for depression in Parkinson’s disease. Cochrane Database Syst Rev. 2003;3:CD003465.
57. Bains J, Birks JS, Dening TR. The efficacy of antidepressants in the treatment of depression in dementia. Cochrane Database Syst Rev. 2002;4:CD003944.
58. Anderson IM. Meta-analytic studies on new antidepressants. Br Med Bull. 2001;57:161-178.
59. Gelenberg AJ, Chesen CL. How fast are antidepressants. J Clin Psychiatry. 2000;61:712-721.
60. Freemantle N, Anderson IM, Young P. Predictive value of pharmacological activity for the relative efficacy of antidepressant drugs. Meta-regression analysis. Br J Psychiatry. 2000;177:292-302.
61. MacGillivray S, Arroll B, Hatcher S, et al. Efficacy and tolerability of selective serotonin reuptake inhibitors compared with tricyclic antidepressants in depression treated in primary care: systematic review and meta-analysis. BMJ. 2003;326:1014.
62. Birmaher B. Should we use antidepressant medications for children and adolescents with depressive disorders? Psychopharmacol Bull. 1998;34:35-39.
63. Guaiana G, Barbui C, Hotopf M. Amitriptyline versus other types of pharmacotherapy for depression. Cochrane Database Syst Rev. 2003;2:CD004186.
64. Furukawa T, McGuire H, Barbui C. Low dosage tricyclic antidepressants for depression. Cochrane Database Syst Rev. 2003;3:CD003197.
65. Barbui C, Hotopf M, Freemantle N, et al. Selective serotonin reuptake inhibitors versus tricyclic and heterocyclic antidepressants: comparison of drug adherence. Cochrane Database Syst Rev. 2000;4:CD002791.
66. Mittmann N, Herrmann N, Einarson TR, et al. The efficacy, safety and tolerability of antidepressants in late life depression: a meta-analysis. J Affect Disord. 1997;46:191-217.
67. Wilson K, Mottram P, Sivanranthan A, Nightingale A. Antidepressants versus placebo for depressed elderly. Cochrane Database Syst Rev. 2001;2:CD000561.
68. Linde K, Mulrow CD. St John’s wort for depression. Cochrane Database Syst Rev. 2000;2:CD000448.
69. Shaw K, Turner K, Del Mar C. Tryptophan and 5-hydroxytryptophan for depression. Cochrane Database Syst Rev. 2002;1:CD003198.
70. Franchini L, Serretti A, Gasperini M, et al. Familial concordance of fluvoxamine response as a tool for differentiating mood disorder pedigrees. J Psychiatr Res. 1998;32:255-259.
71. Lane RM. A critical review of selective serotonin reuptake inhibitor-related sexual dysfunction; incidence, possible aetiology and implications for management. J Psychopharmacol. 1997;11:72-82.
72. Fava M, Rosenbaum JF, McGrath PJ, et al. Lithium and tricyclic augmentation of fluoxetine treatment for resistant major depression: a double-blind, controlled study. Am J Psychiatry. 1994;151:1372-1374.
73. Kraus RP, Diaz P, McEachran, A. Managing rapid metabolizers of antidepressants. Depress Anxiety. 1996;4:320-327.
74. Donovan SJ, Quitkin FM, Stewart JW, et al. Duration of antidepressant trials: clinical and research implications. J Clin Psychopharmacol. 1994;14:64-66.
75. Quitkin FM, Petkova EP, McGrath PJ, et al. When should a trial of fluoxetine for major depression be declared failed? Am J Psychiatry. 2003;160:734-740.
76. Nierenberg AA, McLean NE, Alpert JE, et al. Early nonresponse to fluoxetine as a predictor of poor 8-week outcome. Am J Psychiatry. 1995;152:1500-1503.
77. Reimherr FW, Amsterdam JD, Quitkin FM, et al. Optimal length of continuation therapy in depression: a prospective assessment during long-term fluoxetine treatment. Am J Psychiatry. 1998;155:1247-1253.
78. Geddes JR, Carney SM, Davies S, et al. Relapse prevention with antidepressant drug treatment in depressive disorders: a systematic review. Lancet. 2003;361:653-661.
79. Reynolds CF 3rd, Frank E, Perel JM, et al. Nortriptyline and interpersonal psychotherapy as maintenance therapies for recurrent major depression: a randomized controlled trial in patients older than 59 years. JAMA. 1999;281:39-45.
80. Nelson JC. Treatment of antidepressant nonresponders: augmentation or switch. J Clin Psychiatry. 1998;59(suppl):35-41.
81. Fava M. Augmentation and combination strategies in treatment-resistant depression. J Clin Psychiatry. 2001;62(suppl 18):4-11.
82. Marangell LB. Switching antidepressants for treatment-resistant major depression. J Clin Psychiatry. 2001;62(suppl 18):12-17.
83. Joffe RT, Singer W, Levitt AJ, et al. A placebo-controlled trial comparison of lithium and triiodothyronine augmentation of tricyclic antidepressants in unipolar refractory depression. Arch Gen Psychiatry. 1993;50:387-393.
84. Joffe RT, Sokolov STH. Thyroid hormone treatment of primary unipolar depression: a review. Int J Neuropsychopharmacol. 2000;3:143-147.
85. Taylor MJ, Carney S, Geddes J, Goodwin G. Folate for depressive disorders. Cochrane Database Syst Rev. 2003;2:CD003390.
86. Nelson JC. Augmentation strategies with serotonergic-noradrenergic combinations. J Clin Psychiatry. 1998;59(suppl):65-68.
87. Fava M. Diagnosis and definition of treatment-resistant depression. Biol Psychiatry. 2003;53:649-659.
88. Krishnan KRR. Comorbidity and depression treatment. Biol Psychiatry. 2003;53:701-706.
89. Grote NK, Frank E. Difficult-to-treat depression: the role of contexts and comorbidities. Biol Psychiatry. 2003;53:660-670
90. Thase ME, Rush J, Howland RH, et al. Double-blind switch study of imipramine or sertraline treatment of antidepressant-resistant chronic depression. Arch Gen Psychiatry. 2002;59:233-239.
91. Nierenberg AA, Feighner JP, Rudoph R, et al. Venlafaxine for treatment-resistant unipolar depression. J Clin Psychopharmacol. 1994;14:419-423.
92. McGrath PJ, Stewart JW, Nunes EV, et al. A double-blind crossover trial of imipramine and phenelzine for outpatients with treatment-refractory depression. Am J Psychiatry. 1993;150:118-123.
93. Martin JL, Barbanoj MJ, Schlaepfer TE, et al. Transcranial magnetic resonance stimulation for treating depression. Cochrane Database Syst Rev. 2002;2:CD003493.
94. George MS, Rush AJ, Sackheim HA, Marangell LB. Vagus nerve stimulation (VNS): utility in neuropsychiatric disorders. Int J Neuropsychopharmacol. 2003;6:73-83.
95. Klaiber EL, Broverman DM, Vogel W, et al. Estrogen therapy for severe persistent depression in women. Arch Gen Psychiatry. 1979;36:550-557.
96. Stahl SM. Augmentation of antidepressants by estrogen. Psychopharmacol Bull. 1998;34:319-321.
97. Lawrie, T A, Herxheimer, A, and Dalton, K. Oestrogens and progestogens for preventing and treating postnatal depression. Cochrane Database Syst Rev. 2000;2:CD001690.
98. Tohen M, Shelton R, Tollefson GD, et al. Olanzapine plus fluoxetine: double-blind and open-label results in treatment-resistant major depressive disorder. Eur Neuropsychopharmacology. 1999;9(suppl 5):S246.
99. Joffe RT, Levitt AJ. Relationship between antidepressant partial and nonresponse and subsequent response to antidepressant augmentation. J Affect Disord. 1999;52:257-259.
100. Roy-Byrne PP, Katon W. Generalized anxiety disorder in primary care: the precursor/modifier pathway to increased health care utilization. J Clin Psychiatry. 1997;58S:34-38.
101. Schneider F, Weiss U, Kessler C, et al. Subcortical correlates of differential classical conditioning of aversive emotional reactions in social phobia. Biol Psychiatry. 1999;45:863-871.
102. Posternak MA, Mueller TI. Assessing the risks and benefits of benzodiazepines for anxiety disorders in patients with a history of substance abuse or dependence. Am J Addict. 2000;10:48-68.
103. Nulman I, Rovet J, Stewart DE, et al. Neurodevelopment of children exposed in utero to antidepressant drugs. N Eng J Med. 1997;336:258-262.
104. Iqbal MM, Sobhan T, Ryals T. Effects of commonly used benzodiazepines on the fetus, the neonate, and the nursing infant. Psychiatr Serv. 2002;53:39-49.
105. Bandelow B, Zohar J, Hollander E, Kasper S, Moller HJ, for the WFSBP Task Force on Treatment Guidelines for Anxiety, Obsessive-Compulsive, and Posttraumatic Stress Disorders. World Federation of Societies of Biological Psychiatry (WFSBP) guidelines for the pharmacological treatment of anxiety, obsessive-compulsive, and posttraumatic stress disorders. World J Biol Psychiatry. 2002;3:171-199.
106. Ballenger JC, Davidson JR, Lecrubier Y, et al. Consensus statement on generalized anxiety disorder from the International Consensus Group on Depression and Anxiety. J Clin Psychiatry. 2001;62S11:53-58.
107. Gould R, Otto MW, Pollack MH, et al. Cognitive behavioral and pharmacological treatment of generalized anxiety disorder: a preliminary meta-analysis. Behavior Therapy. 1997;28:285-305.
108. Gale C, Oakley-Brown M. Anxiety disorder: clinical review. Br Med J. 2000;321:1204-1207.
109. Kapczinski F, Lima MS, Souza JS, Schmitt R. Antidepressants for generalized anxiety disorder. Cochrane Database Syst Rev. 2003;2: CD003592.
110. Ferreri M, Hantouche EG. Recent clinical trials of hydroxyzine in generalized anxiety disorder. Acta Psychiatr Scand. 1998;393(suppl):102-108.
111. DeMartinis N, Rynn M, Rickels K, et al. Prior benzodiazepine use and buspirone response in the treatment of generalized anxiety disorder. J Clin Psychiatry. 2000;61:91-94.
112. Pittler MH, Ernst E. Kava extract for treating anxiety. Cochrane Database Syst Rev. 2003;1:CD003383.
113. Ballenger JC. Treatment of anxiety disorders to remission. J Clin Psychiatry. 2001;62(suppl 12):5-9.
114. Schweizer W, Rickels K, Uhlenhuth EH. Issues in the long-term treatment of anxiety disordersychopharmacology. In: Bloom FE, Kupfer DJ, eds. Psychopharmacology: The Fourth Generation. New York, NY: Raven Press; 1995.
115. Castenada R, Sussman N, Westreich L, et al. A review of the effects of moderate alcohol intake on the treatment of anxiety and mood disorders. J Clin Psychiatry. 1996;57:207-212.
116. Schadé A, Marquenie LA, van Balkom AJ, de Beurs E, van Dyck R, van den Brink W. Do comorbid anxiety disorders in alcohol-dependent patients need specific treatment to prevent relapse? Alcohol Alcohol. 2003;38:255-262.
117. Thompson PM. Generalized anxiety disorder treatment algorithm. Psych Annals. 1996;26:227-232.
118. Dupont RL, Rice DP, Shiraki S, et al. Economic costs of obsessive-compulsive disorder. Med Interface. 1995;8:102-109.
119. Stein DJ, Hollander E. Textbook of Anxiety Disorders. Washington, DC: American Psychiatric Press; 2002.
120. Hawkridge S, Stein DJ. Risk-benefit assessment of drug therapies for anxiety disorders in children and adolescents. Drug Saf. 1998;19:283-297.
121. Bernstein GA, Shaw K, for the Work Group on Quality Issues. Practice parameters for the assessment and treatment of children and adolescents with anxiety disorders. J Am Acad Child Adolesc Psychiatry. 1997;36(suppl 10):69-84.
122. Shader RI, Greenblatt DJ. Use of benzodiazepines in anxiety disorders. New Eng J Med. 1993;328:1398-1405.
123. Boyer W. Serotonin uptake inhibitors are superior to imipramine and alprazolam in alleviating panic attacks: a meta-analysis. Int Clin Psychopharmacol. 1995;10:45-49.
124. van der Linden GJ, Stein DJ, van Balkom AJ. The efficacy of the selective serotonin reuptake inhibitors for social anxiety disorder (social phobia): a meta-analysis of randomized controlled trials. Int Clin Psychopharmacol. 2000;15(suppl 2):15-24.
125. Stein DJ, Zungu-Dirwayi N, van der Linden GJ, Seedat S. Pharmacotherapy for posttraumatic stress disorder. Cochrane Database of Systematic Rev. 2000;4:CD002795.
126. Ballenger JC, Davidson JR, Lecrubier Y, et al. Consensus statement on panic disorder from the International Consensus Group on Depression and Anxiety. J Clin Psychiatry. 1998;59S:47-54.
127. Ballenger JC, Davidson JR, Lecrubier Y, et al. Consensus statement on posttraumatic stress disorder from the International Consensus Group on Depression and Anxiety. J Clin Psychiatry. 2000;61(suppl 5):60-66.
128. Ballenger JC, Davidson JA, Lecrubier Y, et al. Consensus statement on social anxiety disorder from the international consensus group on depression and anxiety. J Clin Psychiatry. 1998;59:54-60.
129. Bakker A, van Balkom AJ, Spinhoven P. SSRIs vs TCAs in the treatment of panic disorder: a meta-analysis. Acta Psychiatr Scand. 2002;106:163-167.
130. Liebowitz MR, Schneier F, Campeas R, et al. Phenelzine vs atenolol in social phobia: a placebo-controlled comparison. Arch Gen Psychiatry. 1992;49:290-300.
131. Stein DJ, Stein MB, Goodwin W, Hunter B. The selective serotonin reuptake inhibitor paroxetine is effective in more generalized and less generalized social anxiety disorder. Psychopharmacology. 2001;158:267-272.
132. Goddard AW, Brouette T, Almai A, Jetty P, Woods SW, Charney D. Early coadministration of clonazepam with sertraline for panic disorder. Arch Gen Psychiatry. 2001;58:681-686
133. Noyes R, Perry P. Maintenance treatment with antidepressants in panic disorder. J Clin Psychiatry. 1990;51(suppl):24-30.
134. Southwick SM, Yehuda R. The interaction between pharmacotherapy and psychotherapy in the treatment of posttraumatic stress disorder. Am J Psychotherapy. 1993;47:404-411.
135. Pollack MH, Otto MW, Kaspi S, et al. Cognitive behavior therapy for treatment-refractory panic disorder. J Clin Psychiatry. 1994;55:200-205.
136. Tesar GE, Rosenbaum JF. Successful use of clonazepam in patients with treatment-resistant panic disorder. J Nerv Ment Dis. 1986;174:477-482.
137. Pande AC, Pollack MH, Crockatt J, et al. Placebo-controlled study of gabapentin treatment of panic disorder. J Clin Psychopharmacol. 2000;20:467-471.
138. Tiffon L, Coplan JD, Papp LA, et al. Augmentation strategies with tricyclic or fluoxetine treatment in seven partially responsive panic disorder patients. J Clin Psychiatry. 1994;55:66-69.
139. Zamorski MA, Albucher RC. What to do when SSRIs fail: eight strategies for optimizing treatment of panic disorder. Am Fam Physician. 2002;66:1477-1484.
140. Hirschmann S, Dannon PN, Iancu I, et al. Pindolol augmentation in patients with treatment-resistant panic disorder: a double-blind, placebo-controlled trial. J Clin Psychopharmacol. 2000;20:556-559.
141. Davidson JRT, Potts N, Richichi E, et al. Treatment of social phobia with clonazepam and placebo. J Clin Psychopharmacol. 1993;13:423-428.
142. Pande AC, Davidson JR, Jefferson JW, et al. Treatment of social phobia with gabapentin: a placebo-controlled study. J Clin Psychopharmacol. 1999;19:341-348.
143. Stein MB, Sareen J, Hami S, Chao J. Pindolol potentiation of paroxetine for generalized social phobia: a double-blind, placebo-controllled, crossover study. Am J Psychiatry. 2001;158:1725-1727.
144. Stein DJ, Seedat S, van der Linden G, Kaminer D. Pharmacotherapy of posttraumatic stress disorder. In: Davidson J, Nutt D, Zohar J, eds. Posttraumatic Stress Disorder: Assessment and Management. London, UK: Martin Dunitz; 2000.
145. Hertzberg MA, Butterfield MI, Feldman ME, et al. A preliminary study of lamotrigine for the treatment of posttraumatic stress disorder. Biol Psychiatry. 1999;45:1226-1229.
146. Altamura AC, Pioli R, Vitto M, et al. Venlafaxine in social phobia: a study in selective serotonin reuptake non-responders. Int Clin Psychopharmacol. 1999;14:239-245.
147. Aarre TF. Phenelzine efficacy in refractory social anxety disorder: a case series. Nord J Psychiatry. 2003;57:313-315.
148. Stein DJ, Hollander E. Serotonin reuptake inhibitors in obsessive-compulsive disorder and related disorders. In: Feighner J, Boyer WF, eds. Selective Serotonin Reuptake Inhibitors: Advances in Basic Research and Clinical Practice. 2nd ed. New York, NY: John Wiley: 1996.
149. Stein DJ, Hollander E, Rowland C, Serebro P, Roberts M. Quality of life and pharmacoeconomic aspects of obsessive-compulsive disorder: a South African survey. S Afr Med J. 1996;86:1579-1585.
150. Swedo SE, Leonard HL, Garvey M, et al. Pediatric autoimmune neuropsychiatric disorders associated with streptococcal infections: clinical description of the first 50 cases. Am J Psychiatry. 1998;155:264-271.
151. Broocks A, Hohagen, F. Psychotherapy in OCD. In: Fineberg N, Marazziti D, Stein, DJ, eds. Obsessive-Compulsive Disorder: A Practical Guide. London, UK: Martin Dunitz; 2001.
152. Geller DA, Biederman J, Stewart SE, et al. Which SSRI? A meta-analysis of pharmacotherapy trials in pediatric obsessive-compulsive disorder. Am J Psychiatry. 2003;160:1919-1928.
153. King RA, Leonard H, March J, for the Work Group on Quality Issues. Practice parameters for the assessment and treatment of children and adolescents with obsessive-compulsive disorder. J Am Acad Child Adolesc Psychiatry. 1998;37(suppl 10):27-45.
154. Hawkridge S, Stein DJ, Bouwer C. Combining neuroleptics with serotonin specific reuptake inhibitors in Tourette’s syndrome. J Am Acad Child Adolesc Psychiatry. 1996;35:703-704.
155. Stein DJ, Spadaccini E, Hollander E. Meta-analysis of pharmacotherapy trials for obsessive compulsive disorder. Int Clin Psychopharmacol. 1995;10:11-18.
156. Ackerman DL, Greenland S. Multivariate meta-analysis of controlled drug studies for obsessive-compulsive disorder. J Clin Psychopharmacol. 2002;22:309-317.
157. Stein DJ. Neurobiology of the obsessive-compulsive spectrum of disorders. Biol Psychiatry. 2001;47:296-304.
158. Pato MT, Hill JL, Murphy DL. A clomipramine dosage reduction study in the course of long-term treatment of obsessive compulsive disorder patients. Psychopharmacol Bull. 1990;26:211-214.
159. Stein DJ, Fineberg N, Seedat S. An integrated approach to the treatment of OCD. In: Fineberg N, Marazziti D, Stein DJ. Obsessive-Compulsive Disorder: A Practical Guide. London, UK: Martin Dunitz; 2001.
160. McDougle CJ, Goodman WK, Leckman JF. Haloperidol addition in fluvoxamine-refractory obsessive-compulsive disorder: a double-blind placebo-controlled study in patients with and without tics. Arch Gen Psychiatry. 1994;51:302-308.
161. McDougle CJ, Epperson CN, Pelton GH, et al. A double-blind, placebo-controlled study of risperidone addition in serotonin reuptake inhibitor-refractory obsessive-compulsive disorder. Arch Gen Psychiatry. 2000;57:794-802.
162. Ravizza L, Barzega G, Bellino S, et al. Therapeutic effect and safety of adjunctive risperidone in refractory obsessive-compulsive disorder (OCD). Psychopharmacol Bull. 1996;32:677-682.
163. Stein DJ, Bouwer C, Hawkridge S, et al. Risperidone augmentation of serotonin reuptake inhibitors in obsessive-compulsive and related disorders. J Clin Psychiatry. 1997;58:119-122.
164. Marrazitti D, Dell’Osso L, Gemignani A, et al. Citalopram in refractory obsessive-compulsive disorder: an open study. Int Clin Psychopharmacol. 2001;16:215-219.
165. Hollander E, Mullen L, DeCaria CM, et al. Obsessive compulsive disorder, depression, and fluoxetine. J Clin Psychiatry. 1991;52:418-422.
166. Ananth J, Burgoyne K, Smith M, et al. Venlafaxine for treatment of obsessive-compulsive disorder. Am J Psychiatry. 1995;152:1832.
167. Koran LM, Sallee, FR, Pallanti S. Rapid benefit of intravenous pulse loading of clomipramine in obsessive-compulsive disorder. Am J Psychiatry. 1997;154:396-401.
168. Bystritsky A, Munford PR, Rosen RM, et al. A preliminary study of partial hospital management of severe obsessive-compulsive disorder. Psychiatr Serv. 1996;47:170-174.
169. Martuza RL, Chiocca EA, Jenike MA, et al. Stereotactic radiofrequency thermal cingulotomy for obsessive-compulsive disorder. J Neuropsychiatry. 1990;2:331-336.
170. Maletzky B, McFarland B, Burt A. Refractory obsessive compulsive disorder and ECT. Convuls Ther. 1994;10:34-42.
171. Martin JL, Barbanoj MJ, Perez V, Sacristan M. Transcranial magnetic stimulation for the treatment of obsessive-compulsive disorder. Cochrane Database Syst Rev. 2003;3:CD003387.