The interest in effective psychotherapeutic and pharmacologic treatments for panic disorder continues to grow, as the prevalence, cost, and impact on quality of life of the disorder have been recognized. The National Comorbidity Survey reported a 3.5% prevalence of panic disorder in the U.S. population,1 and later assessments of these data showed significant underdiagnosis of the condition and a high associated economic burden based largely on excess nonpsychiatric medical costs.2 A recent study of primary care services found that patients with panic disorder had significant disability and an increased number of medical and mental health visits; however, only 64% received any treatment for panic disorder, with 42% being given medication and 36% being given psychotherapy.3 Further, many patients who do receive treatment nonetheless remain symptomatic when broad measures of assessment, including panic attacks, anticipatory anxiety, agoraphobic avoidance, and functional impairment, are considered.4
Thus, despite the availability of effective cognitive-behavioral and pharmacologic treatments, further efforts to improve the recognition of panic disorder and the effectiveness, tolerability, and availability of treatment are warranted. In addition, what to do next for patients who do not respond to a first-line intervention is a critical area of clinical concern requiring further research, In this article, we review the pharmacologic and cognitive-behavioral therapies for panic disorder, as well as novel strategies for the treatment of patients who have refractory symptoms.
Selective Serotonin Reuptake Inhibitors (SSRIs)
SSRIs have become first-line pharmacotherapy for panic disorder due to their relatively favorable side effect profile and broad spectrum of efficacy against many mood and anxiety disorders. Given the frequent comorbidity of depression and other anxiety disorders with panic, the SSRIs offer the possibility of monotherapy for a range of symptomatology. Further, the issue of potential abuse, and the physiologic dependence associated with benzodiazepines, may be avoided; this issue is particularly relevant for patients who have past or present comorbid alcohol or other substance abuse. However, although efficacy rates based on panic-free status in short-term studies have been reported to range from 50% to 70%, rates of full remission appear to be lower (20% to 40%) when broader assessments are applied.45 In addition, some patients are unable to tolerate the initiation of treatment because of side effects such as anxiety, agitation, and insomnia, whereas others are unable to tolerate side effects such as sexual dysfunction or gastrointestinal distress that emerge over time.
Generally, SSRIs should be initiated at low doses (eg, 5 to 10 mg/d for fluoxetine or paroxetine, 12.5 to 25 mg/d for sertraline) and titrated up slowly. Therapeutic dose levels are generally comparable to and sometimes higher than those for depression. The therapeutic lag in efficacy and the anxiety associated with initiation have driven the frequent strategy of combining SSRIs with benzodiazepines.6 The choice of SSRI is often guided by side effect profile, patient preference, and past personal or family history of response to a particular agent.
Tricyclic Antidepressants (TCAs) and Monoamine Oxidase Inhibitors (MAOIs)
Although there are abundant data supporting the efficacy of TCAs, particularly Imipramine and clomipramine, for panic disorder, they are no longer considered first-line therapy due to their more aversive side effect profile and more constricted spectrum of efficacy. Further, TCAs may be fatal in overdose and can be of concern for patients who have comorbid medical illnesses because of potential drug interactions and cardiac effects. Rates of panic-free status achieved in TCA trials are similar to those found with high-potency benzodiazepines and SSRIs, but the TCAs may be somewhat less effective than the SSRIs.7 Nonetheless, TCAs do remain a treatment option for patients who fail to respond to or who remain symptomatic after other pharmacologic interventions, and they may be used to augment the response to SSRIs. At least two reports suggest that clomipramine may be the most effective TCA for panic,8,9 perhaps due to its potent serotoninergic effects. TCAs do offer the benefit of effectiveness for comorbid depression, but are not efficacious for social anxiety disorder,10 which occurs in 20% to 50% of patients who have panic disorder.11
MAOIs (ie, phenelzine, franylcypromine, and isocarboxazid) also have proven efficacy for panic disorder, and are effective for comorbid depression and social phobia as well.12 However, the aversive side effect profile of the MAOIs and the necessity for strict dietary proscriptions with their administration have resulted in diminished use since the advent of safer and more tolerable antidepressants. MAOIs are now largely reserved for patients who are refractory to other safer pharmacologic and cognitive-behavioral interventions and who are responsible enough to manage the strict dietary regimen.
High-potency benzodiazepines (ie, alprazolam and clonazepam) have proven efficacy in panic disorder; they have a rapid onset of effect without the initial increase in anxiety and agitation that may occur with antidepressants. Of the high-potency benzodiazepines, clonazepam has a pharmacokinetic advantage over alprazolam. Alprazolam's shorter half-life usually results in a need for at least 4 times per day dosing, may induce interdose rebound anxiety or withdrawal after a missed dose, and heightens a patient's sense of dependency on the medication.
In contrast, clonazepam may be dosed 2 to 3 times per day and its longer half-life allows for more steady plasma levels, reducing the likelihood of interdose anxiety and withdrawal symptoms with abrupt discontinuation. Clonazepam is typically effective at 1 to 3 mg/d (administered on a twice a day schedule), with anti-panic efficacy maintained over time without dose escalation.13-15
Benzodiazepines do not provide antidepressant efficacy and should be avoided, if possible, for patients with a history of drug or alcohol abuse. However, they remain a useful, frequently employed option, particularly for patients who require rapid anxiolysis or situational dosing, who fail to tolerate antidepressants, or who remain symptomatic despite adequate therapy with antidepressants.
Pharmacologic Treatment of Panic Disorder
There are little systematic data available supporting the common practice of combining pharmacologic agents for the treatment of panic disorder. However, the combination of benzodiazepines and antidepressants is frequently employed in clinical practice because of the potential for rapid anxiolysis from the benzodiazepines and protection against emergent anxiety secondary to initiation of antidepressants, and the benefit of the broad spectrum of efficacy of antidepressants. Only one study has been published to date on this combined treatment strategy for panic disorder: patients were randomized to Imipramine alone or imipramine plus alprazolam, with the high-potency benzodiazepine tapered during weeks 4 to 6. The study demonstrated initial benefit for the combined treatment group, but emergent distress associated with discontinuation of the high-potency benzodiazepine.16
Our group is currently examining the efficacy and safety of paroxetine alone versus paroxetine plus clonazepam, with the high-potency benzodiazepine either gradually tapered after week 6 to minimize the effects of discontinuation or maintained over the course of the trial. A preliminary analysis suggests benefit for both combined treatment groups over paroxetine alone,17 with a final analysis of the differential effect of the combined treatment cells awaiting study completion. Further study of combined pharmacologic treatment strategies is clearly warranted.
Table 1 presents the dosages and limitations of pharmacologic agents commonly prescribed for the treatment of panic disorder.
The role of other newer, non-SSRI antidepressants such as venlafaxine, nefazodone, bupropion, and mirtazapine for panic disorder is still emerging, with minimal controlled data available to date. Venlafaxine, a serotonin-norepinephrine reuptake inhibitor with a more tolerable side effect profile than the TCAs, was effective in a small double-blind trial.18 Venlafaxine, similar to other antidepressants employed for anxious patients, should be started at low doses (ie, 18.75 to 37.5 mg/d) and titrated slowly to avoid initial increased anxiety. Therapeutic doses for most patients with the extended release formulation are in the range of 75 to 225 mg/d, although occasionally patients may require higher doses to achieve efficacy. Anecdotal reports suggest potential efficacy of mirtazapine or nefazodone for panic disorder, but no controlled data have yet been published.29,20 The use of bupropion for panic disorder has perhaps been prematurely limited by one small single-blind study that failed to show efficacy for this agent.21 Anecdotal experience suggests potential efficacy for bupropion, and further examination of this agent is warranted.
Recently, there has been substantial interest in the use of anticonvulsants for anxiety disorders, and specifically valproate and gabapentin for panic disorder. One advantage of anticonvulsants is the relatively low likelihood of emergent anxiety with initiation of treatment. In addition, the anticonvulsants may provide an advantage for patients with comorbid bipolar illness or mood lability. The use of valproate in panic disorder is supported by studies of typical and treatment-refractory patients, with dosing guided by blood levels and treatment response.22-25
The anxiolytic effects of gabapentin have emerged clinically in panic and other anxiety disorders. The use of gabapentin in panic disorder is further supported by a randomized, double-blind, placebo-controlled trial in which gabapentin demonstrated efficacy compared with placebo for patients with a moderate to high level of symptomatology (score of s* 20 on the Panic and Agoraphobia Scale).26 Dosing with gabapentin is generally initiated at 100 to 300 mg/d, with therapeutic dosing typically in the range of 900 to 2,700 mg/d (divided in twice a day or three times a day dosing). The dose is gradually titrated against side effects such as lightheadedness and sedation. Anecdotal experience suggests that some patients may benefit from higher dosing of up to 3,600 to 5,400 mg/d.
COGNITIVE-BEHAVIORAL THERAPY (CBT)
CBT has been clearly shown to be a highly efficacious and well-tolerated treatment, and should be considered a first-line option for panic disorder. Rates of remission and treatment tolerability appear to be at least as good as those for pharmacotherapy. A meta-analysis of 43 controlled trials published from 1974 to 1994 reported a 70% panic-free response rate for patients treated with CBT compared with 57% for pharmacotherapy in acute treatment, with a lower attrition rate for CBT (6%) than for pharmacotherapy.27 However, the medication studies included largely focused on TCAs and benzodiazepines and did not include more typically employed current pharmacotherapy (eg, SSRIs).
Early reports from a recent large multicenter trial comparing Imipramine with CBT also support the efficacy and tolerability of CBT compared with pharmacotherapy.28 As with pharmacotherapy, the acute response to CBT is better than the long-term outcome, particularly when strict remission criteria are applied. One long-term outcome study of CBT for 63 patients with panic disorder reported 75% of the patients as being panic free at 2 years of followup, with 57% meeting criteria for high end-state functioning when assessed cross-sectionally. However, only 1 in 5 patients maintained complete remission during the 2-year follow-up period.29
One of the issues limiting the applicability of CBT to panic populations has been the relative paucity of therapists trained in this highly specialized treatment modality. To help address this issue, recent studies have examined the efficacy of CBT administered in a group format; positive results were obtained both acutely30 and in naturalistic follow-up at 1 year.31 One current focus in CBT research is the possibility of transporting the treatment from specialized to more generalized treatment settings. One study demonstrated the feasibility of administering CBT in a community mental health center.32 Relatively brief and inexpensive versions of CBT programs have also demonstrated efficacy.33 Self-help CBT programs may provide an additional option for patients with panic disorder who do not have access to behavioral therapists.
COMBINED CBT AND PHARMACOTHERAPY
A frequent treatment approach in clinical practice has been to combine pharmacotherapy with CBT. There are limited data assessing the benefit of this combined approach for panic disorder with or without agoraphobia. The greatest support for this approach is for patients with agoraphobia or for patients attempting withdrawal from benzodiazepines. Currently, however, no definitive conclusion may be drawn about the additive benefit of CBT and pharmacotherapy in panic disorder.34
The recently completed multicenter trial comparing irnipramine, CBT, and their combination28 has shed additional light on this issue. In this trial, patients were randomly assigned to receive Imipramine alone (up to 300 mg), CBT alone, placebo alone, CBT plus irnipramine, or CBT plus a pill placebo for 3 months, with responders receiving 6 months of maintenance treatment. Patients were then observed for 6 months after discontinuation of treatment.
In the acute and maintenance phases, response rates were equivalent for the CBT and the irnipramine groups, with both superior to placebo, although the inüpramine group had significantly fewer panic symptoms, reflecting a fuller response. CBT plus irnipramine was equivalent to CBT plus placebo acutely, but gained superiority in the 6-month maintenance phase. In the 6 months following the discontinuation of treatment, patients who had responded to CBT alone or to CBT with placebo did significantly better than patients who had responded to irmprarnine with or without CBT.
Thus, this trial suggests that patients respond to both CBT and irnipramine, although they have a more complete response with irnipramine, and that combined treatment confers an advantage in maintenance treatment. However, the effects of CBT without medication may be more durable after the discontinuation of treatment.
Some investigators have questioned whether adding medication to CBT may hinder the effects of CBT. One report suggested a higher rate of relapse in patients receiving combined treatment compared with those receiving CBT alone.35 Nonetheless, clinical experience supports the theoretical rationale that providing patients with the skills learned in CBT may help them manage episodes of anxiety and decrease rates of relapse both during and after pharmacotherapy.
TREATMENT-REFRACTORY PANIC DISORDER
Despite the plethora of research in recent years examining first-line treatment interventions for panic disorder, there has been little systematic research on what to do next for patients who remain symptomatic despite initial treatment interventions. The need for such research has been heightened as the goals of treatment for panic disorder have shifted from acute panic-free status to persistent remission with inclusion of broader measures of outcome (eg, anticipatory anxiety, agoraphobic avoidance, and functional impairment). However, there are limited data, and no systematic study, regarding what to do next for patients who do not respond or who respond only partially to initial treatment. Current research in our program is aimed at addressing this issue.
The first step in determining what to do for a treatment-refractory patient is to assess the reason for treatment failure (Table 2). Cowley et al.36 retrospectively assessed 106 patients with 252 past medication trials for panic disorder. Intolerance of side effects was the most common reason for treatment failure, occurring in 27% (51 of 190) of the trials with anti-panic agents. However, treatment was largely with TCAs, with few patients having been treated with SSRIs at the time of the review. Other considerations for lack of or partial response to treatment include misdiagnosis, medical illness, substance abuse, psychiatric comorbidity, noncompliance, selection of inappropriate medication, inadequate dosing or duration of treatment, and complicating psychosocial factors.4
CBT FOR TREATMENT-REFRACTORY PANIC DISORDER
At least two small open case series have examined the effectiveness of adding CBT for patients who remain symptomatic despite medication. In one, 15 patients referred for CBT were receiving high-potency benzodiazepines with or without antidepressants; only 7 had received an adequate trial of pharmacotherapy due to a number of factors, including side effects and their unwillingness to take higher doses of benzodiazepines because of fears of dependency. Nonetheless, 40% of the patients achieved remission after 12 weeks of CBT.37 A second case series reported on group CBT for 10 patients with panic disorder (9 of whom had agoraphobia) who remained refractory despite an adequate trial of medication. Sixty percent of the patients were in remission at ike end of 12 weeks.38
Reasons for the Failure of Treatment of Panic Disorder
Thus, preliminary evidence suggests that CBT in an individual or a group format may provide an effective treatment option for patients with panic disorder that is refractory to medication, although additional controlled data are needed. Currently, no conclusive data are available regarding the efficacy of adding pharmacotherapy for patients with panic disorder who remain symptomatic despite an adequate trial of CBT, although clinical experience suggests that this can be effective.
PHARMACOTHERAPY FOR TREATMENT-REFRACTORY PANIC DISORDER
There are several open case series and case reports describing pharmacologic approaches for patients with panic disorder who remain refractory after initial pharmacotherapy. These open reports provide preUminary support for techniques such as adding high-potency benzodiazepines, valproate, gabapentin, or buspirone to standard pharmacologic or behavioral interventions.4 Clinical experience and case series suggest that the addition of a TCA to an SSRI may provide an alternate combination strategy for patients who do not respond to an SSRI alone, although monitoring of plasma levels of the TCA is recommended.39 Many clinicians believe that the MAOIs may be effective for some patients who are refractory to other interventions, but this has not been subjected to systematic study.
Options for the treatment of panic disorder are rapidly expanding, with a focus on achieving full remission of symptoms and return to a high level of functioning and a favorable quality of life. Current recommendations for first-line therapy include SSRIs and CBT, and treatment selection is largely determined by patient and physician preference, treatment availability, comorbidity, and side effect profile. Despite an improved treatment armamentarium, no one treatment has emerged with clearly superior efficacy and many patients remain symptomatic despite initial treatment. There are relatively little data to guide "nextstep" interventions for patients who have treatment-refractory panic disorder, although combining pharmacotherapies, optimizing medication doses, adding CBT, and switching between medications are all viable options. The challenges that remain for the field in treating panic disorder include the need for additional strategies to optimize outcome, increase access to effective treatments, and develop a data-based algorithm to guide the treatment of patients with panic disorder who remain symptomatic despite initial treatment intervention.
1. Kessler RC, McGonagle KA, 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. Greenburg PE, Sisitsky T, Kessler RC, et al. The economic burden of anxiety disorders in the 1990s. J Clin Psychiatry. 1999;60:427-435.
3. Roy-Byrne PP, Stein MB, Russo J, et al. Panic disorder in the primary care setting: comorbidity, disability, service utilization, and treatment. J Clin Psychiatry. 1999;60.492-499.
4. Simon NM, Pollack MH. Treatment-refractory panic disorder. Psychiatr Clin North Am. 1999;6:115-140.
5. Roy-Byrne P, Cowley D. Clinical approach to treatmentresistant panic. In: Rosenbaum JF, Pollack MH, eds. Panic Disorder and Us Treatment. New York: Dekker; 1998:205-228.
6. Uhlenhuth EH, Baiter MB, Ban TA, Yang K. International study of expert judgment on therapeutic use of benzodiazepines and other psychotherapeutic medications: VI. Trends in recommendations for the pharmacotherapy of anxiety disorders. Depress Anxiety. 1999;9:107-116.
7. Boyer W. Serotonin uptake inhibitors are superior to irnipramine and alprazolam in alleviating panic attacks: a meta-analysis. Int Clin Psychopharmacol. 1995;10:45-49.
8. den Boer JA, Westenberg HG, Kamerbeek WD, Verhoeven WM, Kahn RS. Effect of serotonin uptake inhibitors in anxiety disorders: a double-blind comparison of clomipramine and fluvoxamine. Int Clin Psycliopharmacol. 1987;2:21-32.
9. Modigh K, Westberg P, Eriksson E. Superiority of domipramine over irnipramine in the treatment of panic disorder: a placebo-controlled trial. J Clin Psychopharmacol. 1992;12:251-261.
10. Simpson HB, Schneier FR, Campeas R, et al. Irnipramine in rne treatment of social phobia. J Clin Psychopharmacol. 1998;18:132-135.
11. Stein MB, Shea CA, Uhde TW. Social phobic symptoms in patients with panic disorder: practical and theoretical implications. Am J Psychiatry. 1989;146:235-238.
12. Rosenbaum JF, Pollack MH, Fredman SJ. The pharmacotherapy of panic disorder. In: Rosenbaum JF, Pollack MH, eds. Panic Disorder and Us Treatment. New York: Dekker; 1998:153-180.
13. Moroz G, Rosenbaum JF. Efficacy, safety, and gradual discontinuation of clonazepam in panic disorder: a placebocontrolled, multicenter study using optimized doses. J Clin Psychiatry. 1999;60:604-612.
14. Rosenbaum JF, Moroz G, Bowden CL, Clonazepam Panic Disorder Dose-Response Study Group. Clonazepam in the treatment of panic disorder with or without agoraphobia: a dose-response study of efficacy, safety, and discontinuance. J Clin Psychopharmacol. 1997;17:390-400.
15. Worthington JJ HL Pollack MH, Otto MW, McLean RY, Moroz G, Rosenbaum JF. Long-term experience with clonazepam in patients with a primary diagnosis of panic disorder. Psychopharmacol Bull. 1998;34:199-205.
16. Woods SW, Nagy LM, Koleszar AS, Krystal JH, Heninger GR, Charney DS. Controlled trial of alprazolam supplementation during irnipramine treatment of panic disorder. J Clin Psychopliarmacol. 1992;12:32-38.
17. Simon N, Pollack M, Worthington J, Scott E, Otto M. The effectiveness and safety of combined treatment with paroxetine and clonazepam for panic disorder: interim analysis. Presented as a poster at the New Clinical Drug Evaluation Unit Program (NCDEU); Boca Raton, FL; 1998.
18. Pollack MH, Worthington JJ HT, Otto MW, et al. Venlafaxine for panic disorder: results from a doubleblind, placebo-controlled study. Psychopharmacol Bull. 1996;32:667-670.
19. Carpenter LL, Leon Z, Yasmin S, Price LH. Clinical experience with mirtazapine in the treatment of panic disorder. Ann Clin Psychiatry. 1999;11:81-86.
20. DeMartinis NA, Schweizer E, Rickets K. An open-label trial of nefazodone in high comorbidity panic disorder. J Clin Psychiatry. 1996;57:245-248.
21. Sheehan DV, Davidson J, Manschreck T, VanWyck Fleet J. Lack of efficacy of a new antidepressant (bupropion) in the treatment of panic disorder with phobias. Clin Psychopharmacol. 1983;3:28-31.
22. Baetz M, Bowen R. Efficacy of divalproex sodium in patients with panic disorder and mood instability who have not responded to conventional therapy. Can J Psychiatry. 1998;43:73-77.
23. McElxoy SL, Keck PE, Lawrence JM. Treatment of panic disorder and benzodiazepine withdrawal with valproate. J Neuropsychiatry Clin Neurosci. 1991;3:232-233.
24. Ontiveros A, Fontaine R. Sodium valproate and clonazepam for treatment-resistant panic disorder. J Psychiatry Neurosci. 1992;17:78-80.
25. Woodman C Noyes R Jr, Baüenger JC, Lydiard RB, Sievers G, Mihalko D. Predictors of response to alprazolam and placebo in patients with panic disorder. J Affect Disord. 1994;30:5-13.
26. Pande AC, Pollack MH, Crockatt J, et al. Placebo-controlled study of gabapentin treatment of panic disorder. J Clin Psychopharmacol. 2000;20:467-471.
27. Gould R, Otto M, Pollack M. A meta-analysis of treatment outcome for panic disorder. Clin Psychol Rev. 1995,15:819-844.
28. Barlow DH, Gorman JM, Shear MK, Woods SW. Cognitive-behavioral therapy, imipramine, or their combination for panic disorder: a randomized controlled trial. JAMA 2000;283:2529-2536.
29. Brown T, Antony MM Barlow DH. Diagnostic comorbidity in panic disorder: effect on treatment outcome and course of comorbid diagnoses following treatment. J Consult Clin Psychol, 1995;63:408-418.
30. Penava SJ, Otto MW, Maki KM, Pollack MH. Rate of improvement during cognitive behavioral group treatment for panic disorder. Behav Res Ther. 1998;36:665-673.
31. Martinsen EW, Olsen T, Tonset E, Ny land KE, Aarre TF. Cognitive-behavioral group therapy for panic disorder in the general clinical setting: a naturalistic study with 1year follow up. J Clin Psychiatry. 1998;59:437-442.
32. Wade WA, Treat TA, Stuart GL. Transporting an empirically supported treatment for panic disorder to a service clinic setting: a benchmarking strategy. J Consult Clin Psychol. 1998;66:231-239.
33. Clark DM, Salkovskis PM, Hackmann A, Wells A, Ludgate J, Gelder M. Brief cognitive therapy for panic disorder: a randomized controlled trial. J Consult Clin Psychol. 1999;67:583-589.
34. Gelder MG, Combined pharmacotherapy and cognitive behavior therapy in the treatment of panic disorder. J Clin Psychopharmacol. 1998;18(suppl 2):2S-5S.
35. Otto MW, Pollack MH, Sabatino SA. Maintenance of remission following cognitive behavior therapy for panic disorder: possible deleterious effects of concurrent medication treatment. Behavior Therapy. 1996;27:473-482.
36. Cowley DS, Ha EH, Roy-Byrne PP. Determinants of pharmacologic treatment failure in panic disorder. J Clin Psychiatry. 1997;58:555-561.
37. Pollack MH, Otto MW, Kaspi SP, Hammerness PG, Rosenbaum JF. Cognitive behavior therapy for treatmentrefractory panic disorder. J Clin Psychiatry. 1994;55:200-205.
38. Otto MW, Pollack MH, Penava SJ, Zucker BG. Group cognitive-behavior therapy for patients failing to respond to pharmacotherapy for panic disorder: a clinical case series. Behav Res Ther. 1999;37:763-770.
39. Tiffon L, Copian JD, Papp LA, Gorman JM. Augmentation strategies with tricyclic or fluoxetine treatment in seven partially responsive panic disorder patients. J Clin Psychiatry. 1994;55:66-69.
Pharmacologic Treatment of Panic Disorder
Reasons for the Failure of Treatment of Panic Disorder