A middle-aged woman presents to a hospital emergency room complaining of severe chest pains, a pounding heart and dizziness. She is convinced she is dying. A differential diagnosis reveals she is having a panic attack rather than a heart attack.
To assist physicians in caring for patients with panic disorder, the American Psychiatric Association recently published its Practice Guideline for the Treatment of Patients with Panic Disorder (American Psychiatric Association, 1998).
Co-chairing the guideline work group were Katherine Shear, M.D., professor of psychiatry at the University of Pittsburgh and director of the Anxiety Disorders Prevention Program; and Jack Gorman, M.D., professor and vice chairman for research for the department of psychiatry at Columbia University. Other members were Deborah Cowley, M.D.; C. Deborah Cross, M.D.; John March, M.D.; Walton Roth, M.D.; and Michael Shehi, M.D.
"A typical panic attack is a sudden onset of...fear or apprehension which is accompanied by somatic symptoms, such as feeling unable to catch one's breath, heart palpitations, pounding heart, chest pain, dizziness or light-headedness, weakness, hot or cold flashes, shakiness and sometimes stomach upset, nausea or diarrhea," said Shear. "Generally, physical symptoms that are caused by what we think of as activation of the sympathetic or autonomic nervous system are very typical."
Because of the very prominent physical symptoms of panic disorder, those affected believe there is something wrong with their physical health and frequently see their primary care doctor or a specialist, such as a pulmonologist, first.
For any patients who present with chest pains and/or heart palpitations, Shear urged physicians to include panic attacks in the differential diagnosis, explaining that patients with panic disorder are often misdiagnosed, resulting in repeated visits to emergency rooms. The timing of the attack, Shear indicated, offers a clue to differential diagnosis.
"Panic attacks tend to occur at rest, and most anginal chest pain more likely occurs during exercise or some kind of activation state. Activation can relieve a panic attack," she said. "We think some of the symptoms of panic may relate to low blood bicarbonate. Panic patients show some evidence of chronic hyperventilation. Sometimes they hyperven-tilate during acute episodes."
Prior to the initiation of any treatment, the practice guideline recommends "a comprehensive general medical and psychiatric evaluation...to determine whether potential general medical and substance-induced conditions may be causing the panic symptoms, complicating treatment or requiring specific interventions, especially with a patient who has a new onset of symptoms. In addition, the assessment of developmental factors, psychosocial stressors and conflicts, social supports and general living situation will aid the treatment." Natural History and Epidemiology
Several different types of panic attacks may occur, according to the guideline. The most common is the unexpected attack, defined as one not associated with a known situational trigger. But individuals also may experience panic attacks linked to certain situations or particular emotional contexts, attacks involving limited symptoms and nocturnal attacks. Panic disorder also is related to agoraphobia.
"Agoraphobia technically is feeling trapped or unable to escape, or all alone or unable to get help in the event of having panic or panic-like symptoms. You don't have to meet the full criteria for panic disorder in order to have agoraphobia," Shear said. "Usually, you have some sort of panic or panic-like symptoms that are very frightening. And people who have it severely are very constricted in their activities."
Some agoraphobics refuse to leave their homes, often for years at a time, while others leave only when accompanied by a trusted friend. The lifetime prevalence for panic disorder is about 3.5%, according to the National Comorbidity Survey (Kessler et al., 1994), but Shear said there is a differential between men and women-5% of women and 2% of men.
"For whatever combination of psychosocial and possibly biological reasons, there is this gender difference in anxiety and depressive disorders across the board," she said, noting that the differences show up in household studies done across cultures (Weissman et al., 1997).
"We know that poverty and physical and sexual abuse in childhood predispose to these disorders to some degree. And then also, it is true that hormones have effects on neurotransmitters, and there may be some relationship there as well. Certainly, the postpartum period is a period of vulnerability," she said. "The gender difference does seem to start earlier in life for anxiety disorders; girls have more phobias and separation anxiety than boys."
Patients with panic disorder also have a higher prevalence of depression, according to Shear. People with depression have a higher prevalence of panic than a control group or other population.
"Sometimes people become very demoralized as a result of having the panic attacks, and the depression grows out of that demoralization," she said. "People who have both panic and depression are at higher risk for suicidal thoughts and behaviors than people with either of them alone. There also have been reports in the literature that panic attacks alone may be associated with suicidality."
Treatment ApproachesWith regard to treatment, Gorman said that after a comprehensive literature review, the Work Group on Panic Disorder concluded there are two types of treatment that have been rigorously shown to be effective for panic disorder.
"One is medication and one is cognitive-behavioral psychotherapy," he said, adding that no evidence exists "at the present time to indicate which one is better than the other for individual patients. Therefore, the decision is based on individual physician and patient characteristics.
"Although there is a not a substantial body of scientific literature, there is a long tradition of using psychoanalytic therapy to treat conditions like panic disorder, and many clinicians find these helpful."
Of the medications, the selective serotonin reuptake inhibitors (SSRIs) "should, for most patients, be considered first-line medications," Gorman said.
Two of the SSRIs, paroxetine(Drug information on paroxetine) (Paxil) and sertraline(Drug information on sertraline) (Zoloft) have been approved by the U.S. Food and Drug Administration for use in treating panic disorder. But there is evidence that all of the available SSRIs are effective for panic disorder, he added.
Acute treatment with medication also could include tricyclic antidepressants (TCAs), high-potency benzodiazepines and monoamine oxidase inhibitors (MAOIs).
"For many patients, SSRIs are likely to have the most favorable balance of efficacy and adverse effects," the guideline said. "Although SSRIs carry a risk of sexual side effects, they lack the cardiovascular side effects, anticholinergic side effects and toxicity associated with overdose that occur with TCAs and MAOIs. SSRIs also lack the potential for physiologic dependency associated with benzodiazepines."
Shear said treatment with medications is usually for six months to a year after the panic is under control. She noted that even when patients are on therapeutic doses of medications, they can occasionally have a panic attack. "That is something that is well-observed," she said. "That doesn't necessarily mean that they are going to have a recurrence of the whole full-blown disorder."
If patients have experienced significant or full improvement on maintenance therapy, physicians can attempt a trial of medication discontinuation, according to the guideline.
"Although data on the percentage of patients that remain well after medication discontinuation has been widely divergent," the guideline said, "evidence suggests that it is between 30% and 45% [Ballenger et al., 1993]."
Psychiatrists are generally less aware of the fact that cognitive-behavior treatments (CBT) have been found to be equally efficacious to medication in treating panic disorder, Shear said.
"What is important about that [fact] for psychiatrists is that for patients who don't want or cannot tolerate medication, there is an alternative," she said. "It is worthwhile for psychiatrists to pay attention to what the cognitive-behavioral strategies are and to begin to incorporate them into their treatment of patients with panic disorder even when they treat them with medications. It can help provide very clear psychoeducation. It is going to help to know that patients are reactive to bodily sensations.
The cognitive-behavioral model of panic attacks is that the person has an unexplained panic attack...then they begin to monitor their bodies for signs of these physical sensations that are frightening. Maybe they had a heart palpitation before they ever had panic disorder, but after they have had their first panic attack, that sporadic heart palpitation becomes a focus of attention and fear. They become more afraid and that creates more heart palpitations from the fear; it becomes a vicious cycle and that can contribute to onset of panic attacks."
Shearer continued: "It is important that anybody working with panic patients recognize that vulnerability. It helps for physicians to remember to prescribe extra low doses of medications for panic patients, since panic patients will be sensitive to any physical sensations from the medications."
CBT encompasses a range of treatments, each consisting of several elements, including psychoeducation, continuous panic monitoring, anxiety management techniques such as abdominal breathing to control the physiologic reactivity, cognitive restructuring to identify and counter fear of bodily sensations, and exposure to fear cues.
Shear added that CBT used in the treatment of panic disorder has been described in manuals which have both patient versions and therapists' guides: Mastery of Your Anxiety and Panic (Craske et al., 1994), Therapist's Guide for the Mastery of Your Anxiety and Panic (MAP) Program (Craske and Barlow, 1990) and Agoraphobia Supplement to Mastery of Your Anxiety and Panic (Craske and Barlow, 1994).
As to whether combining medication and psychotherapy is efficacious, Shear said "Jack Gorman and I, along with David Barlow, M.D., Ph.D., and Scott Woods, M.D., Ph.D., just completed an eight-year study [Multicenter Collaborative Treatment Study of Panic Disorder] looking at exactly that question. We compared CBT to imipramine(Drug information on imipramine) [Tofranil] and then the combination. We studied only panic disorder patients who had mild or no agoraphobia. We did not find an advantage of the combination over the CBT alone or the imipramine alone, and the CBT and imipramine were equivalent. Intuitively, you would think it would be better if you do both, but that didn't really seem to have a better outcome. Our study is under submission, and has not been published as yet. It was presented at NCDEU [New Clinical Drug Evaluation Unit]...It also was presented at the Anxiety Disorder Association meeting in Boston last March."
For severe agoraphobics, however, Shear said there is some indication in the literature that combining behavior therapy and medication may be best.
In her own treatment of patients, Shear often uses medications, but for patients who do not want to take medications and who are not urgently ill, she will often do cognitive-behavioral treatment. The other situation where CBT treatments are extremely helpful, she said, is when the physician wants to discontinue medications with patients.
"I, myself, generally combine the treatments, although not in a formal way," she said. "The other thing I need to mention is that the whole concept of exposure as a treatment for anxiety is very useful. Basically, the idea is that the more the patient can do things he or she is frightened of, as opposed to avoiding doing them, the better."
Research DirectionsBoth Gorman and Shear agree that serious questions about the treatment of panic disorder need to be addressed through research.
"First of all, we need to determine whether there are any characteristics that would guide physicians and patients to select one type of treatment over the other," Gorman said. "We also need more research to see if the combination of psychotherapy plus medication is more effective than either one alone, to see what the effectiveness of psychoanalytical psychotherapies are, to determine how long treatment should last before one can be reasonably certain that discontinuing the treatment will not lead to a rapid relapse and to find biological markers for treatment effects."
Gorman said his research team is conducting studies "where we are looking at various brain imaging measures to see if we can find exactly what it is that antidepressant medication and cognitive behavioral therapy make better in people with panic disorder when they respond to treatment. We also are conducting studies trying to find out how long cognitive behavioral therapy and medication therapy should last."
What is known from epidemiological and some long-term naturalistic follow-up studies, Shear said, is that there is a reasonably high likelihood of recurrence and chronicity with panic disorder.
"What we don't have is long-term follow-up of the best treatment that we have. Remember that a lot of panic patients don't get the best treatment-they don't get CBT or appropriate medications or the right doses of medications, or the right kind of medication management. So given that, we can't really say that if someone does get those things, that they would have recurrence," she said.
Another big area of needed research "is what predisposes to this illness," Shear said. "What kinds of early experiences or genetic vulnerabilities might there be that we could identify early, so as to work on primary prevention."
Shear is looking at separation anxiety as a possible precursor of panic disorder and specifically studying early infant separation. Some other areas requiring further research mentioned in the guideline include the relationship between panic disorder and other psychiatric disorders; long-term health risks, such as cardiovascular morbidity and mortality, associated with panic disorder; and panic disorder in childhood.
