ANNOTATED

ANXIETY DISORDERS24 P S Y C H I AT R I C T I M E S F E B R UA RY 2 0 0 9w w w. p s y c h i a t r i c t i m e s . c o m

by Kristalyn Salters-Pedneault, PhD

T he 2 most common anxiety dis- orders are generalized anxiety disorder (GAD) and panic dis-

order. Approximately 5.7% of people in community samples will meet di- agnostic criteria for GAD in their life- time; the rate is about 4.7% for panic disorder (with or without agorapho- bia).1 GAD—which is characterized by excessive and uncontrollable wor- ry about a variety of topics (along with associated features such as trou- ble sleeping and impaired concen- tration)—is often chronic and is as- sociated with significant costs to the individual and to society.2,3

Panic disorder—characterized by recurrent, unexpected panic attacks— can be similarly intractable (particu- larly when is it accompanied by ago- raphobia) and costly.4,5 Panic attacks are discrete periods of intense fear or discomfort that manifest with sweat- ing, trembling, accelerated heart rate, and concern about having another panic attack. Many people experience panic attacks without meeting full diagnostic criteria for panic disorder (about 28.3% lifetime prevalence).6

Furthermore, panic attacks have been identified as a risk factor for various other forms of psychopathology, in- cluding GAD.7

Given the widespread occurrence of GAD, panic disorder, and panic at- tacks, it is not surprising that these conditions are frequently comorbid. An international study of lifetime co- morbidities found a high rate (21.8%) of panic disorder and GAD; most people (55.8%) reported that the symptoms of GAD and panic disor- der began within 1 year of each oth- er.8 Unfortunately, while effective treatment strategies are available for both panic disorder and GAD, little is known about how to best treat these disorders when they are comorbid.

Diagnostic issues The first task to effectively address comorbid GAD and panic disorder is to make the appropriate diagnosis. A variety of general medical conditions that mimic features of panic disorder and/or GAD (eg, hyperthyroidism, pheochromocytoma) need to be con- sidered. In addition, somatic and as- sociated symptoms may be present in

both GAD and panic disorder; thus it is important to understand the context of these symptoms.

In GAD, somatic symptoms such as muscle tension or feeling “keyed up” or “on edge” may occur, but they present differently than those that arise during a panic attack. Panic at- tack–related somatic symptoms tend to develop abruptly, then peak and subside relatively quickly, whereas GAD-related somatic symptoms tend to come on more gradually, and are present at a lower level for longer periods. GAD-related somatic symp- toms may be experienced as aversive by the individual but generally are not catastrophically misinterpreted (eg, “I am dying”).

In addition to the symptom over-

lap between GAD and panic attacks, there is overlap between GAD and panic disorder; worry is a component of both disorders. Panic disorder is associated with recurrent panic at- tacks and worries about future attacks and their consequences or meaning.

With GAD, worry is much more pervasive and tends to cover a variety of life’s domains (eg, finances, work, family). Thus, if the worry is con- fined to apprehension about having a panic attack or about the implications or consequences of a panic attack, the diagnosis of GAD is excluded. How- ever, a diagnosis of panic disorder is ruled out if a person has GAD and panic attacks, but does not exhibit a month or more of fear of having another attack, worry about the im- plications of the attack, or significant behavior change (ie, behavioral avoidance). In this case, only GAD is diagnosed, although the presence of panic attacks should be noted as a

clinically significant feature. It is also important to establish

whether GAD or panic disorder is primary, because this will probably affect the treatment plan. This assess- ment can be difficult given the symp- tom overlap that exists. However, it is often particularly helpful to consider which disorder is having the greatest functional impact or the most poten- tial to impede treatment compliance and outcome. Many patients with moderate to severe panic disorder re- port that the symptoms are more in- terfering than GAD symptoms. These patients tend to be more motivated to seek treatment because of the intense fear produced by attacks. They may also be more likely to be referred to treatment because of frequent emer-

gency department (ED) visits. How- ever, there are persons for whom GAD is far more interfering, so this issue warrants consideration.

Treatment There is little empirical guidance in the area of treatment planning for comorbid GAD, panic disorder, and panic attacks, but more work is emerging in this area. For example, psychosocial approaches, such as cognitive-behavioral therapy (CBT), are being tailored specifically for treating comorbid GAD and panic disorder.9

CBT is the most researched and effective psychotherapy for both pan- ic disorder and GAD. Findings from CBT trials that target GAD or panic disorder suggest that CBT should be the first-line psychological treat- ment—regardless of whether panic disorder or GAD is primary.10,11 The elements of empirically supported

CBTs for panic disorder and GAD generally include similar techniques, and treatment for the comorbid con- ditions can include all or some of these elements: psychoeducation, ex- posure, cognitive restructuring, and relaxation.

Psychoeducation is critically im- portant for individuals with comorbid panic disorder and GAD. The pa- tient’s expectations about treatment are set and treatment goals are devel- oped collaboratively. The next focus of treatment depends on the goals arrived at with the patient. However, keep in mind that research has demonstrated that panic disorder may interfere with GAD treatment more than other comorbid disorders.12 This may be because the intense nature of panic attacks leads to increased func- tional impairment and obscures the significance of GAD treatment. Fre- quent and severe panic attacks have the potential to interfere with treat- ment compliance, particularly if ag- oraphobia develops. Therefore, it is often prudent to address panic attacks early in treatment.

One way to address panic attacks directly is through interoceptive ex- posure, which draws on principles of habituation and extinction learning to reduce the occurrence of panic at- tacks. Generally, during interoceptive exposure, the patient engages in a variety of exercises that induce the feared physical sensations that often precede or accompany a panic attack. Exercises might include spinning in a chair, breathing through a straw, hy- perventilating, or walking up stairs. The patient is instructed to continue the exposure until he or she experi- ences a reduction in anxiety (ie, ha- bituation has occurred).

While interoceptive exposure is intended to reduce panic, it may also help reduce symptoms of GAD. There are models of GAD and a growing research literature, which suggest that worry and GAD are characterized by a tendency to avoid a variety of emotional experiences, including fear, sadness, anger, and positive emotions.13 There is a distinct possibility that exposure to the un- comfortable emotions and sensations of a panic attack will also reduce the emotional avoidance typical of GAD by introducing or reinforcing the no-

Strategies for Assessing and Treating Comorbid Panic and Generalized Anxiety Disorder Understanding the Differences Between GAD, Panic Disorders, and Panic Attacks

CHECK POINTS

üThe elements of empirically supported cognitive-behavioral therapies (CBTs) for both panic disorder and generalized anxiety disorder (GAD) generally include similar techniques, and treatment for the comorbid conditions can include all or some of these elements: psychoeducation, exposure, cognitive restructuring, and relaxation.

ü It is often prudent to address panic attacks early in treatment. üRandomized controlled trials have generally found that combination

treatments have limited value over monotherapies for panic disorder, and studies of naturalistic combination treatments for panic disorder have demonstrated poorer outcomes than CBT alone.

Although combination treatment is common, this practice raises con- cerns that patients may experience symptom relapse if the medication is withdrawn at some point in the future. During or immediately after SSRI discontinuation, a short booster course of CBT may be indicated to prevent relapse and reinforce previ- ously learned anxiety management skills.

Dr Salters-Pedneault is a research associate

in the National Center for Posttraumatic

Stress Disorder Behavioral Science Division at

the VA Boston Healthcare System and an in-

structor of psychiatry at Boston University

School of Medicine. Her research is funded by

the Department of Veterans Affairs. She re-

ports no conflicts of interest concerning the

subject matter of this article.

Drug Mentioned in This Article

Alprazolam (Xanax)

References

1. Kessler RC, Berglund P, Demler O, et al. Lifetime prevalence and age-of-onset distributions of DSM-IV disorders in the National Comorbidity Survey Repli- cation. Arch Gen Psychiatry. 2005;62:593-602. 2. Noyes R, Woodman C, Garvey MJ, Cook BL. Gen- eralized anxiety disorder vs. panic disorder: distin- guishing characteristics and patterns of comorbidity. J Nerv Ment Dis. 1992;180:369-379. 3. Wittchen HU. Generalized anxiety disorder: preva- lence, burden, and cost to society. Depress Anxiety. 2002;16:162-171. 4. Yonkers KA, Bruce SE, Dyck IR, Keller MB. Chronic- ity, relapse, and illness—course of panic disorder, social phobia, and generalized anxiety disorder: find- ings in men and women from 8 years of follow-up. Depress Anxiety. 2003;17:173-179. 5. Batelaan N, Smit F, de Graaf R, et al. Economic costs of full-blown and subthreshold panic disorder. J Affect Disord. 2007;104:127-136. 6. Kessler RC, Chiu WT, Jin R, et al. The epidemiology of panic attacks, panic disorder, and agoraphobia in the National Comorbidity Survey Replication. Arch Gen Psychiatry. 2006;63:415-424. 7. Goodwin RD, Lieb R, Hoefler M, et al. Panic attack as a risk factor for severe psychopathology. Am J Psy- chiatry. 2004;161:2207-2214. 8. Kessler RC, Andrade LH, Bijl RV, et al. The effects of co-morbidity on the onset and persistence of gen- eralized anxiety disorder in the ICPE surveys: Interna- tional Consortium in Psychiatric Epidemiology. Psy- chol Med. 2002;32:1213-1225. 9. Labrecque J, Dugas MJ, Marchand A, Letarte A. Cognitive-behavioral therapy for comorbid general- ized anxiety disorder and panic disorder with agora- phobia. Behav Modif. 2006;30:383-410. 10. Roemer L, Orsillo SM, Salters-Pedneault K. Effi- cacy of an acceptance-based behavior therapy for generalized anxiety disorder: evaluation in a random- ized controlled trial. J Consult Clin Psychol. 2008;76: 1083-1089. 11. Tsao JC, Mystkowski JL, Zucker BG, Craske MG. Impact of cognitive-behavioral therapy for panic disorder on comorbidity: a controlled investigation. Behav Res Ther. 2005;43:959-970. 12. Provencher MD, Ladouceur R, Dugas MJ. Comor- bidity in generalized anxiety disorder: prevalence and course after cognitive-behavior therapy [in French]. Can J Psychiatry. 2006;51:91-99. 13. Roemer L, Orsillo SM. Mindfulness- and Accep- tance-Based Behavioral Therapies in Practice. New York: Guilford Press; 2008. 14. Roy-Byrne P, Wingerson D, Cowley D, Dager S. Psychopharmacologic treatment of panic, generalized

2 weeks because it was making her feel jittery. After she switched to a benzodi- azepine, the number of panic attacks was significantly reduced, as were the somat- ic symptoms of GAD. Nevertheless, she continued to worry excessively.

After 1 year of therapy, Julia noted that her life circumstances had become less stressful. The benzodiazepine dosage was slowly tapered, but within 1 week af- ter the medication had been discontinued, Julia had another panic attack. Her psy- chiatrist decided to try another SSRI but this time at a much lower initial dosage. She was also referred to a CBT therapist for treatment of both panic disorder and GAD.

Julia was quite motivated for CBT treatment and complied with all therapy elements. While she was initially appre- hensive about interoceptive exposure ex- ercises, she was willing to try them and did report less fear of her physical sensa- tions after completing these exercises. She also reported that the cognitive re- structuring and relaxation skills helped her feel greater mastery over her anxiety and worry. At the conclusion of the 16 sessions of CBT, Julia no longer met diag- nostic criteria for panic disorder but still met criteria for GAD (although she report- ed reduced severity of symptoms). One year after CBT treatment, she was contin- uing to take the SSRI and was still expe- riencing some symptoms of GAD but had not had a panic attack since completing CBT.

Conclusions This hypothetical vignette demon- strates several important and com- mon issues in the treatment of co- morbid panic disorder and GAD. Julia experienced a typical course of panic disorder and GAD, including onset in the late teens, with a relaps- ing/remitting course of panic disorder and chronic GAD (with relatively less GAD response to pharmacother- apy). Her initial treatment with an SS- RI was unsuccessful because of non- compliance with the treatment regiment secondary to discomfort with the physical sensations associat- ed with medication initiation, a prob- lem that is common in persons with panic disorder.20

Although the panic attacks remit- ted with the benzodiazepine, there was little effect on the GAD symp- toms, and the panic disorder returned on discontinuation (despite slow ta- per), possibly due to rebound anxiety. A trial of a slowly titrated SSRI helped reduce both symptoms of pan- ic disorder and GAD. This, combined with a course of CBT that focused on reducing symptoms of both disorders, was most effective.

tients with symptoms severe enough to interfere with adherence to psycho- therapy. However, there is reason to use this combination approach judi- ciously. Research has shown that con- current use of pharmacotherapy and CBT may not be helpful or may be problematic in many individuals with anxiety symptoms. Randomized con- trolled trials have generally found that combination treatments have limited value over monotherapies for panic disorder, and studies of natura- listic combination treatments for pan- ic disorder have demonstrated poorer outcomes than CBT alone.17,18

There are theoretical reasons to avoid combination therapy for panic disorder and GAD.15 First, it is possi- ble that the phenomenon of state- dependent learning may be implicat- ed in relapse following combination treatment. The extinction learning that occurs during CBT in combina- tion with a medication may not be re- trieved as readily from memory once the medication is discontinued. In ad- dition, medication may reinforce a patient’s belief that his internal ex- periences should be avoided, which may detract from the CBT message that emotions, physical sensations, and thoughts are not harmful and do not need to be avoided. Furthermore, combination therapy may cause pa- tients to attribute symptom reduction to the medication and thereby in- crease the risk of relapse following medication withdrawal.19

CASE VIGNETTE

Julia is a 19-year-old who referred herself for psychiatric treatment following 3 vis- its to the ED for heart palpitations, short- ness of breath, dizziness, and trembling. These attacks began a few months after her father learned he had cancer. All phys- ical causes of the attacks were ruled out. In addition to the panic attacks, Julia re- ported that she was very worried that she would have another attack in public. She mentioned (when asked) that she worried a lot about many issues in her day-to-day life, including paying her bills, the health of family members, and school. She also reported that for much of her life she had been “an anxious person.” After thorough assessment, panic disorder without ago- raphobia and GAD was diagnosed.

Julia was offered pharmacotherapy and CBT but felt that her life circum- stances would prevent her from attending therapy appointments regularly. She opt- ed for pharmacotherapy and an SSRI was prescribed for her. At her follow-up ap- pointment she reported that she had de- cided to stop taking the medication after

tion that fear and fear-related sensa- tions and thoughts are not harmful.

Another CBT element that may be applied to both panic disorder and GAD symptoms is cognitive restruc- turing. In panic disorder, cognitive restructuring is used to address cata- strophic beliefs about one’s physical sensations and/or panic attacks. For example, many people believe they are dying during a panic attack. In cognitive restructuring, this thought is identified and the patient is encour- aged to examine the evidence that this thought is true. The process is much the same when doing cognitive re- structuring of worry; an anxious thought is identified (eg, “What will happen if I lose my job?”) and the evidence of the likelihood and/or cat- astrophic outcome of the thought is examined.

Finally, many CBTs for panic dis- order and GAD include a relaxation component that may include skills such as diaphragmatic breathing or progressive muscle relaxation. These skills are popular with clinicians and patients because they are easy to learn (and teach), and often have signifi- cant short-term impact on panic symptoms, worry, and GAD somatic symptoms.

Both panic disorder and GAD re- spond to pharmacological manage- ment. Some medications have dem- onstrated efficacy for both conditions (although they may be FDA-ap- proved for only 1 condition), includ- ing some benzodiazepines, tricyclic antidepressants, SSRIs, and serotonin noradrenaline reuptake inhibitors. Other medications, such as azapir- ones and b-blockers, may alleviate symptoms of GAD but are less effec- tive for panic.14

High-potency benzodiazepines, such as alprazolam, have shown su- perior efficacy for panic symptoms. However, the disadvantages of ben- zodiazepines, such as abuse potential and rebound anxiety after discontin- uation, detract from their overall val- ue in treating comorbid panic disor- der and GAD.15 (A recent effect-size analysis suggests that GAD is gener- ally less responsive than panic disor- der to pharmacotherapy; also, resid- ual symptoms of GAD may need to be addressed by CBT.16)

Studies that compared treatment responses to medication or CBT have found that while medication may ini- tially produce greater symptom re- duction, CBT may produce a more durable response.17 Many practitio- ners have adopted combination (phar- macotherapy plus CBT) treatment approaches, particularly for those pa-

ANXIETY DISORDERS 25P S Y C H I AT R I C T I M E S F E B R UA RY 2 0 0 9 w w w. p s y c h i a t r i c t i m e s . c o m

(Please see Comorbid Disorders, page 26)

anxiety disorder, and social phobia. Psychiatr Clin North Am. 1993;16:719-735. 15. Westra HA, Stewart SH. Cognitive behavioural therapy and pharmacotherapy: complementary or contradictory approaches to the treatment of anxiety? Clin Psychol Rev. 1998;18:307-340. 16. Hidalgo RB, Tupler LA, Davidson JR. An effect-size analysis of pharmacologic treatments for generalized anxiety disorder. J Psychopharmacol. 2007;21:864- 872. 17. Barlow DH, Gorman JM, Shear MK, Woods SW. Cognitive-behavioral therapy, imipramine, or their combination for panic disorder: a randomized con- trolled trial [published corrections appear in JAMA. 2000;284:2450; JAMA. 2001;284:2597]. JAMA. 2000;283:2529-2536. 18. Arch JJ, Craske MG. Implications of naturalistic use of pharmacotherapy in CBT treatment for panic disorder. Behav Res Ther. 2007;45:1435-1447. 19. Basoglu M, Marks IM, Kiliç C, et al. Alprazolam and exposure for panic disorder with agoraphobia: attri- bution of improvement to medication predicts subse- quent relapse. Br J Psychiatry. 1994;164:652-659. 20. Mathew SJ, Coplan JD, Gorman JM. Management of treatment-refractory panic disorder. Psychophar- macol Bull. 2001;35:97-110.

Evidence-Based References

Arch JJ, Craske MG. Implications of naturalistic use of pharmacotherapy in CBT treatment for panic dis- order. Behav Res Ther. 2007;45:1435-1447. Labrecque J, Dugas MJ, Marchand A, Letarte A. Cog- nitive-behavioral therapy for comorbid generalized anxiety disorder and panic disorder with agoraphobia. Behav Modif. 2006;30:383-410. r

Comorbid Disorders Continued from page 25

ANXIETY DISORDERS26 P S Y C H I AT R I C T I M E S F E B R UA RY 2 0 0 9w w w. p s y c h i a t r i c t i m e s . c o m

by Marco Mula, MD, PhD

A nxiety disorders are chronic conditions that follow a re- lapsing/remitting course.1 The

evidence to support this view comes primarily from cross-sectional and retrospective assessments of duration of illness and, in part, from prospec- tive studies. The waxing and waning nature of panic disorder and general- ized anxiety disorder (GAD), for ex- ample, has been clearly demonstrat- ed. Much less information is available about the course of illness of social phobia. However, both com- munity studies and patient samples suggest an age of onset of social pho- bia in mid to late teens with a chronic- ity that is equal to or greater than that of panic disorder.2 Nevertheless, this recognition has not reshaped our ba- sic treatment approach, which focus- es almost entirely on the acute control of symptoms and only secondarily acknowledges relapse prevention.

In addition, the natural history of anxiety disorders is frequently com- plicated by Axis I and Axis II comor- bidity that seems to be significantly higher among patients who seek treatment than in persons in the com- munity who are not in treatment.1 In fact, it has been estimated that 73% of patients with panic disorder had oth- er comorbid conditions that ranged from major depression to substance abuse until the onset of the Axis II disorders, mostly cluster C type 1 to 2. It is, therefore, evident that any long-term anxiolytic treatment strat- egy must take account of these high rates of comorbidity that appear to develop during the longitudinal phase of the anxiety disorder.

A variety of drug classes have been shown to be effective in treating anxiety disorders. SSRIs are the cur- rent gold standard for anxiety disor- ders. Effective in about 50% to 60% of patients, serotonin noradrenalin reuptake inhibitors are now consid- ered the gold standard specifically for panic disorder and GAD. Benzodi- azepines have a rapid onset of action, but their long-term use may lead to complications, such as abuse liabili- ty, dependence, and withdrawal risk if the dosage is not tapered properly. Moreover, in some patients, benzo-

diazepines may cause sedative effects and cognitive deficits that signifi- cantly affect quality of life and social functioning.

The basis for the use of anticon- vulsant drugs in treating anxiety dis- orders can be found in the main cere- bral structures involved in fear circuits. Although numerous brain re- gions are likely to be involved, the

amygdala and the hippocampus play a key role.3 The amygdala is im- portant in experiencing fear and its autonomic and endocrine response through the output to the hypothala- mus, while the output to periaque- ductal gray matter is mainly impli- cated in avoidance behavior, which is also typical of fear response.4 In ad- dition, the hippocampus is important

Can Anticonvulsants Help Patients With Anxiety Disorders? What Does the Evidence Show?

CHECK POINTS

üThe rationale for the use of anticonvulsants in anxiety disorders is supported by neurobiological underpinnings that make these compounds a likely alternative for short-term treatment in patients who do not respond to benzodiazepines or who have a contraindication.

üPregabalin and gabapentin have been shown to be promising in social phobia. Data about gabapentin in panic disorder suggest a possible role only in patients who are moderately to severely affected. These compounds need to be considered second- or even third-line treatment, however.

üWe are still far from having straightforward guidelines on the use of anticonvulsants in patients with anxiety disorders. Apart from the direct comparison between pregabalin and venlafaxine for generalized anxiety disorder, there are no head-to-head comparisons with first-line agents for the long-term treatment of anxiety symptoms.

Principal mechanisms of action of anticonvulsant drugs

Calcium channel GABA block (type) potentiation

Barbiturate NK + (A)

Benzodiazepine NA ++ (A)

Carbamazepine + (L) NK

Ethosuximide ++ (T) NA

Felbamate + (L) + (A)

Gabapentin ++ (N, P/Q) NK

Lamotrigine + (N, P/Q, R) +

Levetiracetam + (N) NK

Oxcarbazepine + (N, P) NK

Phenytoin NK NA

Pregabalin ++ (N, P/Q) NA

Tiagabine NA ++

Topiramate + (L) + (A)

Valproate NK +

Vigabatrin NA ++

Zonisamide ++ (N, P, T) NK

GABA, g-aminobutyric acid; ++, primary action; +, secondary action; NA, no activity; NK, controversial findings; A, GABA potentiation through GABAA receptors.

Table 1

Residents, Do You Like to Write?

Do you have an experience you have learned from during your res- idency that you would like to share with your colleagues?

Is there a patient interaction that has helped you to become a better doctor? A meaningful lesson you learned from a patient, nurse, or supervisor? A difficulty you en- countered that raised questions about your career path or your ap- proach to patients?

Is there an aspect of your career development you have been think- ing about that may interest other residents as well as practicing psychiatrists?

Psychiatric Times invites con- tributions from residents concern- ing their experiences, challenges, frustrations, and triumphs.

Residents are encouraged to send a brief description of the topic they would like to write about, a sample of their writing, and their curriculum vitae to: Editor, Psychiatric Times, 535 Connecticut Avenue, Suite 300, Norwalk, CT 06854 or by e-mail to [email protected]. r

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.