General Hospital Psychiatry 24 (2002) 265–268
Allergy in panic disorder patients: a preliminary report Barbara L. Kennedy, M.D., Ph.D.*, Richard L. Morris, M.A., John J. Schwab, M.D. Department of Psychiatry, University of Louisville, Louisville, KY, USA
Abstract In this study of panic disorder patients, 71.8% were found to have allergies. Compared to the nonallergic subjects, the allergic subjects had more full-blown situational panic attacks and significantly higher scores on 7 of the 14 symptoms in the Hamilton Anxiety Scale. More subjects in the allergic group had comorbid anxiety disorders, but only the number with specific phobias was significantly higher than for the nonallergic. These results are discussed in terms of our present theories of panic disorder. © 2002 Elsevier Science Inc. All rights reserved. Keywords: Allergy; Panic Disorder; Comorbidity
1. Introduction In a recent preliminary study of depressed subjects’ responses to antidepressant medications, we found that about 73% of those with depression had a history of allergies . That percentage is about 2–3 times larger than reported for the U.S. population for food allergies (28%) or positive skin tests (20.2%) . Also, as a group, our allergic subjects did not respond as well as the nonallergic to antidepressant medications. Moreover, even though the allergic and nonallergic groups’ Hamilton Depression Scale (HAM-D)  scores were about the same prior to starting treatment, the allergic group’s mean Hamilton Anxiety Scale (HAM-A)  score was significantly higher (19.7) than the nonallergic group’s mean score (15.5) (unpublished data). Consequently, we became interested in further comparisons of allergic and nonallergic subjects with anxiety disorders, especially panic disorder (PD). There have been only a few scattered reports related to allergies and panic disorder in the literature. Jasnoski and colleagues  reported an increased incidence of panic attacks in individuals with a history of type I allergies. Schmidt-Traub and Bamler  reported that the frequency
* Corresponding author. Tel.: ⫹1-412-856-8770; fax: ⫹1-412-8568790. E-mail address: [email protected]
(B.L. Kennedy). This work was done at the University of Louisville, Kentucky. Dr. Kennedy is now in the Department of Psychiatry at the Western Psychiatric Institute and Clinic at the University of Pittsburgh Medical Center, Pittsburgh, PA. Mr. Morris is now in private practice in Alabama.
of type I allergies in panic patients was high, 74%, and Ramesh and colleagues  found that panic patients had increased levels of immunoglobulin A (IgA). Inasmuch as panic and fear are associated, Gerritsen and colleagues  found that a laboratory induced fear situation (public speaking) produced changes in the experimental subjects’ immune systems, specifically a decrease of T helper cells (CD4) and a decrease in T lymphocyte, but not B lymphocyte responsivity, suggestive of possible common immune mechanisms. In accord with these research reports indicating that there is an association between allergies and panic disorder, we formulated the following questions: 1. Are there differences between allergic and nonallergic PD patients’ frequency of panic attacks and/or the circumstances under which they are experienced? 2. Are there some differences between allergic and nonallergic PD patients’ panic and anxiety symptoms? 3. Are there differences between allergic and nonallergic panic patients’ diagnosed comorbid Axis I anxiety disorders?
2. Materials and methods Our data were obtained from the charts of 45 patients who entered a clinical drug trial. All the 45 patients met DSM-III-R criteria  for PD, and all gave written informed consent prior to participation. In addition to the psychiatric
0163-8343/02/$ – see front matter © 2002 Elsevier Science Inc. All rights reserved. PII: S 0 1 6 3 - 8 3 4 3 ( 0 2 ) 0 0 1 8 6 - X
B.L. Kennedy et al. / General Hospital Psychiatry 24 (2002) 265–268
evaluation, each received a physical examination, laboratory tests, and an electrocardiogram. The psychiatrist conducting the screening interview gathered the information about allergies prospectively. In addition to responses to more general questions about whether the person had allergies and the type, it included data from specific questions about drug sensitivity to prescription or over-the-counter medications and about whether the person had food or contact allergies. Also, each subject was asked to keep a daily diary that contained information about panic attacks and whether they were situational (bound or cued by a specific environment) or unexpected (those which occur “out of the blue”). In addition to the frequency and type of attacks experienced, the subjects were asked whether they had full blown panic attacks (four or more symptoms) and/or a limited symptom attack (fewer than four symptoms). We administered the Structured Clinical Interview for DSM-III-R (SCID-P)  to all the subjects. Those meeting criteria for current substance abuse/dependence were excluded, as were patients with any other significant primary Axis I diagnosis (e.g., depression) or any history of a psychotic disorder. At baseline, all subjects were administered the Hamilton anxiety rating scale (HAM-A) by a trained rater . The rater was not aware of the patient’s allergic status or diary data. There were no significant differences between the allergic and nonallergic groups’ mean age (allergic 29.8 years vs. the nonallergic 28.4 years), or gender (allergic; 17 (52%) females, 15 (48%) males vs. nonallergic; 7 (54%) females, 6 (46%) males). We did not include data on race inasmuch as almost all the subjects were Caucasians. Analyses were carried out using the SPSS PC statistical program. The first were independent groups’ “T’ tests for data considered to have a normal distribution, such as the HAM-A scores, or number of panic attacks experienced. The second were 2 analyses of independent groups for data that were proportional, such as the percentage of patients with comorbid disorders.
3. Results As shown in Table 1, the allergic patients reported a significantly higher number of total full-blown attacks (situational and unexpected) than the nonallergic. Also, the allergic reported almost 5 times more full-blown situational panic attacks than the nonallergic (2.97 vs. 0.62). In contrast, full-blown unexpected panic attacks were reported with almost the same frequency by both the allergic and nonallergic patients (mean 3.38 vs. 3.15). Also, there were no significant differences between the allergic and nonallergic subjects’ total number of limited symptom attacks. The allergic groups reported more limited symptom situational panic attacks than the nonallergic (mean 1.19 vs 0.62)
Table 1 Report of panic attacks for allergic and non-allergic panic patients for a two-week period (mean ⫾ standard deviation) Type of Attack Full blown attacks (4 or more symptoms) Situational Unexpected Total full blown attacks Limited symptom attacks (3 or fewer symptoms) Situational Unexpected Total limited symptom attacks
Allergic (N ⫽ 32)
Non-Allergic (N ⫽ 13)
2.97 ⫾ 3.1 3.38 ⫾ 3.5 6.35 ⫾ 4.8
0.62 ⫾ .96 3.15 ⫾ 2.2 3.77 ⫾ 1.8
⬍.0005 N.S. ⬍.012
1.19 ⫾ 1.2 1.59 ⫾ 2.2 2.78 ⫾ 3.6
0.62 ⫾ 1.2 1.00 ⫾ 1.6 1.62 ⫾ 2.1
N.S. N.S. N.S.
Results of “T” tests for independent groups.
and more limited symptom unexpected panic attacks (mean 1.59 vs 1.00), but the differences were not statistically significant. The allergic group’s HAM-A score was significantly higher than that of the nonallergic group (mean 13.96 vs 8.09). Table 2 shows the allergic and nonallergic subjects’ HAM-A total and item scores. The allergic subjects had higher scores on all items than the nonallergic and significantly higher scores on 7 of the 14 items: fears, insomnia, depressed mood, cardiovascular symptoms, respiratory symptoms, gastrointestinal symptoms and genitourinary symptoms. Table 3 presents the comorbidity data for the subjects’ DSM-III-R anxiety disorders. Having an anxiety disorder was reported by about twice as many of the allergic than the nonallergic subjects, but the difference did not quite reach significance (P⬍.06). However, Specific Phobia was signifTable 2 Allergic and non-allergic panic patients’ Hamilton Anxiety Scale Item scores Item
Allergic (N ⫽ 28)
Non-Allergic (N ⫽ 11)
Anxious mood Tension, restlessness Fears Insomnia Intellectual Depressed mood General somatic (muscular) General somatic (sensory) Cardiovascular symptoms Respiratory symptoms Gastrointestinal symptoms Genitourinary symptoms Autonomic symptoms Behavior at interview Total Score
1.64 ⫾ .87 1.71 ⫾ .76 0.36 ⫾ .73 1.25 ⫾ .93 0.96 ⫾ .84 1.04 ⫾ .64 0.68 ⫾ .72 0.79 ⫾ .74 1.11 ⫾ .79 0.68 ⫾ .55 1.07 ⫾ .90 0.68 ⫾ .77 1.36 ⫾ .78 0.64 ⫾ .68 13.96 ⫾ 5.45
1.27 ⫾ .79 1.36 ⫾ .67 0.00 0.64 ⫾ .67 0.64 ⫾ .67 0.55 ⫾ .52 0.55 ⫾ .93 0.36 ⫾ .51 0.55 ⫾ .69 0.18 ⫾ .41 0.45 ⫾ .69 0.18 ⫾ .60 0.82 ⫾ .87 0.55 ⫾ .52 8.09 ⫾ 4.25
N.S. N.S. .015 .05 N.S. .05 N.S. N.S. .05 .01 .05 .05 N.S. N.S. .005
* “T” tests for independent samples. 0⫽symptom not present; 1⫽mild; 2⫽moderate; 3⫽severe; 4⫽very severe.
B.L. Kennedy et al. / General Hospital Psychiatry 24 (2002) 265–268 Table 3 Allergic and non-allergic panic patients’ comorbidity Disorder
Allergic (N ⫽ 32)
Non-allergic (N ⫽ 13)
Any anxiety disorder Generalized anxiety disorder Social phobia Specific phobia Agoraphobia
40.6% 12.5% 12.5% 31.3% 56.3%
15.4% 0.00% 15.4% 7.7% 53.8%
33.3% 8.9% 13.3% 24.4% 55.6%
.06 N.S. N.S. .02 N.S.
* Results of Chi Square Analysis.
icantly more common in the allergic than the nonallergic group. Generalized Anxiety Disorder (GAD) was more frequent in the allergic than the nonallergic group, but the difference was not statistically significant. Social Phobia occurred with almost the same frequency in the allergic and nonallergic groups (12.5% vs. 15.4%), as did agoraphobia (about 55% each). We have no data on comorbidity with obsessive-compulsive disorder (OCD) because those with it were excluded for these clinical drug trials.
4. Discussion The percentage of PD subjects with allergies (71.8%) was remarkably close to the percentage (73.1%) of depressed patients who reported allergies in one of our studies . Both percentages are about the same as the 74% reported by Schmidt-Traub and Bamler  for panic patients. The high percentages of subjects with allergies in our depression study and this panic study probably are attributable to the well-known high frequency of allergic disorders, especially those affecting the upper respiratory tract, in the Ohio Valley Population. Finding that allergic panic patients reported more fullblown panic attacks and also more situationally related attacks than the comparable group of nonallergic panic patients is consistent with differences between medicated allergic and nonallergic depressed patients . In accord with the hypothesis advanced by Klein , it is possible that the allergic patients may be having more situational, but not unexpected panic attacks than the nonallergic because they may have a higher sensitivity to CO2. Also in this respect, we wonder whether the patients’ allergies affecting the respiratory tract and interfering with ventilation, led to relatively high CO2 concentrations that might be responsible for the situational attacks. The allergic subjects’ having more situational but not more unexpected panic attacks than the nonallergic may be attributable to the allergic having significantly more comorbid anxiety disorders. Also, we found that the allergic had significantly higher total HAM-A scores on 7 of the 14 HAM-A symptoms. The rates of comorbid anxiety disorders were: social phobia 13.3%, specific phobia 24.8%, and agoraphobia
55.6%, all of which are about the same as those listed in DSM-IV . The percentage of our subjects with GAD was low, only 8.9%, possible because of clinical difficulties differentiating between GAD and PD symptomatology. We have no OCD data because subjects with significant OCD symptoms were excluded from the research. The associations between allergies and emotional and nervous disorders have intrigued many great clinicians since the days of Hippocrates who warned asthmatic patients against anger. In the middle part of the last century, during the “hey-day” of psychosomatic medicine, the topic of syndrome shift, of patients having alternating conditions was popular in medicine and in psychiatry. In 1951, Walter Alvarez wrote about reinforcing associations between emotions and allergies, especially between anxiety or depression and bronchial asthma . In accord with classic views, he emphasized that both were attributable to heightened sensitivity to stimuli such as stress or allergens respectively. The evidence of significant anxiety that we found in our allergic subjects supports those views. In our personal clinical experience we have seen cases in which the patients alternated fairly dramatically between severe bronchia asthma, often requiring hospitalization and either GAD or bouts of major depression. We are acutely aware of the limitations of clinical research of this type. We recognize that the number of subjects was relatively small. However, perhaps the most worrisome problem was the persisting one of defining the term, allergy, with the precision and uniformity needed to meet research requirements. For example, we know that Jasnoski and colleagues  applied the term narrowly only when there were definite IgE mediated responses. In our strictly clinical research, we have used it in a more inclusive way to describe subjects’ symptomatic reactions to various medications and environmental stimuli, but know that it would have been desirable and wish it had been possible to conduct studies of CO2 sensitivity and immune [7,8] and/or neurotransmitter [14,15] responsivity in these allergic panic patients.
References  Kennedy B, Morris R, Schwab J. Responsivity of allergic depressed subjects to antidepressant medications: a preliminary study. Depression 1996;3:286 – 89.  Bierman C, Pearlman D, Shapiro G, et al. Allergy, asthma, and immunology from infancy to adulthood. 3rd. Ed. Philadelphia: W.B. Saunders and Co, 1996, pp. 81–5.  Hamilton M. A rating scale for depression. Jnl of Neurol Neurosurg Psychiatry 1960;23:56 – 62.  Hamilton M. The assessment of anxiety by rating. British Jnl of Medical Psych, 1959;32:50 –5.  Jasnoski M, Bell I, Peterson R. What connections exist between panic symptoms, shyness, type I hypersensitivity, anxiety, and anxiety sensitivity. Anxiety, Stress, and Coping 1994;7:19 –34.
B.L. Kennedy et al. / General Hospital Psychiatry 24 (2002) 265–268
 Schmidt-Traub S, Bamler K. Psychoimmunologischer zusammenhang zwischen allergien, panik unc agoraphobie. Qeitschrift Fur Klinische Psychologie, Psychopathologie Und psychotherapie, 1992; 40(4):325– 45.  Ramesh C, Yeragami V, Balon R, et al. A comparative study of immune status in panic disorder patients and controls. Acta Psychiatrica Scandinavica 1991;84:396 –7.  Gerritsen W, Heijnen J, Wiegant V, et al. Experimental social fear. Immunological, hormonal, and autonomic concomitants. Psychosomatic Medicine 1996;58:273– 86.  American Psychiatric Association. Diagnostic and statistical manual of mental disorders (3rd. Ed. rev.) Washington, DC, 1987.  Spitzer P, Williams J, Gibbon M, et al. Structured clinical interview
   
for DSM-III-R Patient edition (SCID-P Version 1.0). American Psychiatric Press, Washington DC, 1990. Klein D. Panic Disorder and Agoraphobia: Hypothesis Hothouse. J Clin Psychiatry 1996;57:21–7. American Psychiatric Association. Diagnostic and statistical manual of mental disorders (4th. ed.) Washington DC, 1994. Alvarez WC. The Neuroses. W B Saunders Co, Philadelphia, 1951, pp. 466 –9. Kaliner M, Shelhamer J, Davis P, et al. Autonomic nervous system abnormalities and allergy. Annals of Internal Medicine 1982;96:349 – 57. Nesse R, Cameron O, Curtis G, et al. Adrenergic function in patients with panic anxiety. Arch Gen Psychiatry 1984;41:771– 6.