Treatment Planning for Panic Disorder

Panic disorder with or without agoraphobia is a chronic, debilitating psychiatric illness that affects about 4.7% of the general US population.¹ Kessler and colleagues² report that close to one third of the general population has met criteria for panic disorder within the past year.² The mean age at onset is in one's 20s, and women are twice as likely as men to present with panic disorder.³

Panic disorder is associated with poor quality of life^4,5 and with substantial and moderately severe functional impairment in 45% and 30% of persons, respectively.² Many patients have at least one other psychiatric diagnosis, most commonly substance use disorder, mood disorder, or another anxiety disorder.³ Panic disorder is associated with a 2-fold increased risk of coronary heart disease⁶ and frequent use of emergency and medical services.^7,8

Despite its chronic nature and severity of illness, many patients do not receive therapy that meets standard treatment guidelines.³ In the primary care setting, only 22% of patients with panic disorder receive adequate medication and only12% receive adequate psychotherapy.⁸ The purpose of this brief report is to increase the awareness of evidence-based treatments for panic disorder.

Patients should be thoroughly evaluated to eliminate any conditions that may mimic symptoms of panic disorder, including medical illnesses (eg, cardiac, respiratory, or thyroid diseases) and substance use (eg, marijuana, cocaine, methamphetamines). One of the core symptoms of panic disorder is recurrent, unexpected panic attacks. A full-symptom panic attack requires the abrupt onset of at least 4 physical symptoms (eg, palpitations or dizziness) or cognitive symptoms (eg, fear of going crazy or losing control) that reach a peak within 10 minutes. Another core feature is anticipatory anxiety about future panic attacks or the implications of these attacks. This often leads to agoraphobia in which the person avoids situations where escape or help is not readily available if a panic attack occurs. Thus, expected, or cued, panic attacks can be triggered by the anticipation or presence of a phobic situation.

Panic attacks and phobic avoidance are also present in other anxiety disorders, and examining the focus of the patient's fear helps in the differential diagnosis. According to the cognitive model of panic disorder, panic attacks are the result of catastrophic misappraisals of bodily sensations.⁹ The focus of fear is the perceived harmful consequence of panicking (eg, losing control, going crazy, and so forth) and the associated physical sensations (eg, having a heart attack or stroke). In other anxiety disorders, the fear is not of the panic attacks themselves. In generalized anxiety disorder, for example, panic attacks are the exacerbation of persistent worries and anxiety about daily life events and the anticipation of future bad events. Panic attacks in social anxiety disorder are limited to the anticipation or presence of social or performance situations caused by fear of embarrassment or humiliation. Panic attacks in posttraumatic stress disorder are triggered by reminders of the traumatic event. In specific phobia, the focus of fear is often the perceived danger of the object or situation.¹⁰

Cognitive-behavioral therapy (CBT) and pharmacotherapy are among the empirically supported treatments for panic disorder.

Treatment response is usually defined by a "panic-free" status that is the absence of full-symptom panic attacks. Other measures include clinician's ratings of overall improvement and severity on the Clinical Global Impression (CGI) scale¹¹ and changes in panic symptoms in the Panic Disorder Severity Scale (PDSS).¹² Remission has been defined in one of several ways. The international consensus group considers remission as the complete resolution of symptoms for at least 3 months.¹³ However, most pharmacotherapy studies have defined remission by using a combination of panic-free status and improved CGI and/or PDSS scores. CBT trials have focused on rates of "clinically significant improvement," such as achieving high end-state functioning¹⁴ or meeting the criteria developed by Jacobson and colleagues.¹⁵

Treatment components of CBT include education, breathing retraining, cognitive restructuring, interoceptive exposure, and in vivoexposure.^16,17 Education helps establish treatment alliance and teach key concepts, including the course of panic disorder, the nature of panic attacks, behaviors that maintain the panic cycle, the cognitive model, and the treatment rationale. Breathing retraining was initially thought to reduce physical symptoms of panic, but it was later conceptualized as a way to demonstrate how hyperventilation can exacerbate physical symptoms.¹⁸

The goal of cognitive restructuring is to identify and challenge anxious thoughts regarding a panic attack. In interoceptive exposure, patients perform various exercises (eg, hyperventilation, spinning in a chair) in a controlled setting in order to repeatedly confront anxiety-provoking physical symptoms. With exposure, patients learn that symptoms are not as dangerous as originally perceived. In vivo exposure traditionally has been used in the treatment of agoraphobia. During in vivo exposure, patients confront the actual phobic situation, such as traveling alone or being in an airplane or elevator. These treatment components are based on 2 CBT approaches to panic disorder: panic control treatment (PCT) developed by Barlow and colleagues,¹⁶ which introduced interoceptive exposure as a key element; and the cognitive therapy (CT) program developed by Clark,⁹ which emphasized cognitive restructuring.

In the first controlled trial of PCT (N = 56), 3 active treatments—PCT, muscle relaxation, and PCT plus muscle relaxation—were compared with wait-list controls.¹⁶ PCT outperformed the other treatment conditions. Approximately 87% of patients in the PCT groups, 60% of patients who received muscle relaxation, and 36% of controls were panic-free at the end of their respective treatments. There was a greater distinction when patients who had dropped out were included in the analysis: 74% to 79% of PCT patients, 40% of patients who received muscle relaxation, and 33% of controls were panic-free. However, only 46% of patients who were treated with PCT achieved high-end functioning, despite the high panic-free status associated with this group. This suggests that many patients were still symptomatic at the end of treatment.

A controlled study of 2 versions of CT (N = 42) showed favorable results. A full 12-session and a brief 5-session course of CT were compared with wait-list controls.¹⁹ A greater proportion of patients who received CT had panic-free status relative to controls; 79% in the 12-session CT group, 71% in the 5-session CT group, and 8% in the control group had panic-free status. Thus, the investigators concluded that brief intervention was as effective as full 12-session treatment.

Other randomized controlled studies have confirmed the efficacy of CBT for panic disorder, with findings of panic-free status achieved in approximately 70% to 90% of patients.^17,20-22 Rates of clinically significant improvement were found to be 38% to 79%, depending on the criteria used.^23-25

Dismantling studies have attempted to determine which treatment component is the "active" ingredient or the most effective intervention. With the exception of breathing retraining, which was not shown to be helpful and may possibly increase the risk of relapse,²⁴ the other 3 main interventions—cognitive restructuring, interoceptive exposure, and in vivoexposure—have been shown to be equally effective.^23,26-28

Several classes of psychotropics have been shown to be effective and have comparable efficacy in the treatment of panic disorder, including tricyclic antidepressants (TCAs), benzodiazepines, and SSRIs.^29-34 SSRIs are now considered first-line pharmacotherapy because of a more favorable adverse-effect profile than with TCAs³⁰ and less concern for withdrawal and dependence than with benzodiazepines.³⁵ Based on clinical practice, SSRIs are generally started at one half to one third of the typical starting dose for depression to limit the adverse effects of jitteriness and anxiety.³⁶ Benzodiazepines are better tolerated than antidepressants and may help with adverse effects.³³ Table 1 presents common adverse effects and suggested dosages.

On average, 55% to 60% of patients treated with pharmacotherapy achieve panic-free status.^30,33 Efficacy data are most extensive for imipramine(Drug information on imipramine), fluvoxamine(Drug information on fluvoxamine), paroxetine(Drug information on paroxetine), and alprazolam. There is relatively less controlled data for monoamine oxidase inhibitors, the newer SSRI escitalopram(Drug information on escitalopram), and the serotonin-norepinephrine reuptake inhibitor venlafaxine. Studies have shown either minimal or modest efficacy for the newer drugs.^37,38

A recent placebo-controlled trial compared the efficacy of venlafaxine extended release (ER) with that of paroxetine.³⁹ Venlafaxine ER and paroxetine achieved a greater panic-free status compared with placebo (70%, 58%, and 48%, respectively). Venlafaxine ER also resulted in greater improvement than paroxetine in the PDSS scale. Based on CGI-improvement responders, active treatments resulted in a response rate of 80% to 85%, compared with a placebo response of 60%.

In addition to monotherapy, there is some evidence that combining an antidepressant with a benzodiazepine may facilitate early treatment response. For example, one study randomized patients (N = 50) to clonazepam(Drug information on clonazepam) or placebo during the initial 4 weeks of treatment with sertraline(Drug information on sertraline).⁴⁰ Compared with sertraline alone, combination treatment resulted in a greater reduction of panic symptoms. Treatment gains were maintained even after clonazepam was discontinued.

Remission rates in pharmacotherapy trials are infrequently reported. A retrospective analysis of paroxetine studies found a rate of 24.6% to 35.5%, depending on the definition of remission.⁴¹ Remission was achieved in 50% of patients treated with venlafaxine ER based on CGI severity score.³⁹ These rates are probably an overestimation of true remission, because the more stringent definition proposed by the international consensus group was not used.

A large (N = 312) randomized, placebo-controlled, head-to-head comparison trial examined the efficacy of monotherapy (CBT alone or imipramine alone) with combination treatment (CBT and imipramine).⁴² All active treatments were superior to placebo during the acute treatment phase, but imipramine produced a higher quality of response than CBT. During maintenance, combination treatment fared better than monotherapy. However, imipramine appeared to decrease the long-term efficacy of CBT.

Once treatment was discontinued, patients treated with medication (alone or with CBT) were more vulnerable to relapse than those treated with CBT alone. One possible reason may be that medication was discontinued too rapidly (over 1 to 2 weeks). Based on this study, all 3 treatment modalities are effective for acute stabilization. Medication is helpful if a more potent response is desired, but the long-term effects of medication are limited. Discontinuation of medication over a longer period may reduce the chance of relapse. CBT alone is better tolerated and has a more lasting effect.

In addition to treatment efficacy, other factors should be considered when recommending treatment, including patient preference, treatment history, severity of illness, and presence of comorbid disorders. Table 2 summarizes the advantages and disadvantages of the different treatment modalities.