Modulation by Muscarinic Antagonists of the Response to Carbon Dioxide Challenge in Panic Disorder

ORIGINAL ARTICLE Modulation by Muscarinic Antagonists of the Response to Carbon Dioxide Challenge in Panic Disorder

Marco Battaglia, MD; Silvana Bertella, MD; Anna Ogliari, MD; Laura Bellodi, MD; Enrico Smeraldi, MD

Background: Panic attacks can be induced in persons Results: According to patients’ self-ratings of subjec- with panic disorder by inhalation of carbon dioxide. Hy- tive anxiety, inhalation of the carbon dioxide/oxygen percapnia also elicits a reflex hyperventilation, which is mixture provoked a significant and intense response controlled in part by cholinergic mechanisms. This study after treatment with pirenzepine and placebo. After investigated whether the exaggerated response to car- biperiden treatment, however, hypercapnia elicited a bon dioxide in panic disorder (PD) can be modulated by response profile similar to that elicited by air, whereby antagonists of muscarinic cholinergic receptors. subjective anxiety remained similar to preinhalation levels. Methods: Twelve patients with PD received biperiden hydrochloride (a muscarinic antagonist that crosses the Conclusions: Consistent with the hypothesis of the study, blood-brain barrier), pirenzepine hydrochloride (a mus- a centrally active muscarinic antagonist can block the re- carinic antagonist that does not cross the blood-brain bar- sponse to carbon dioxide commonly observed in sub- rier), or placebo 2 hours before a 35% carbon dioxide– jects with PD. 65% oxygen respiratory challenge (vs air as a placebo) on 3 separate days, in a double-blind, random crossover design. Arch Gen Psychiatry. 2001;58:114-119

N SUBJECTS who are susceptible to to CO2 with the CO2 hypersensitivity char- panic disorder (PD), the admin- acteristic of patients with PD. istration of carbon dioxide (CO2) In mammals and humans, the physi- provokes a sudden increase in ologic reflex response to heightened con- ventilation followed quickly by a centrations of inhaled CO2 (hypercapnia) surge in anxiety1-6 similar to what occurs is an increase in ventilatory rate and I 10 in spontaneous attacks. Such effects are arousal. This effect of CO2 is indepen- usually not elicited in normal control dent of its effects on hydrogen ion concen- subjects2 or patients with psychiatric dis- trations in the extracerebral fluids.10,11 Cen- orders other than PD,3 whereas the ad- tral chemosensitivity has been localized to ministration of sodium lactate provokes flashbacks in subjects with posttrau- See also page 123 matic stress disorder.7 The evidence of hypersensitivity to the ventral medulla, an area rich in mus- 12 From the Developmental CO2 led to the hypothesis that PD is a dis- carinic receptors. Direct stimulation of this 4 Psychopathology Unit, turbance in a suffocation alarm system in area by CO2 elicits hyperventilation, which Department of Psychology, which a physiological misinterpretation of can be dramatically decreased by the topi- University Vita-Salute San suffocative stimuli produces respiratory cal application of muscarinic antago- Raffaele (Drs Battaglia and distress, hyperventilation, and panic. Al- nists.12,13 Other physiological studies also Ogliari), and the Department though the involvement of central che- suggest that ventilatory neurons are of Neuropsychiatric Sciences, moceptors, and of various neurotransmit- activated through muscarinic mecha- Istituto Scientifico San Raffaele ters including norepinephrine, serotonin, nisms.11,14 Functional magnetic resonance Hospital (Drs Battaglia, ␥ and -aminobutyric acid, have been in- imaging studies of men breathing a 5% CO2 Ogliari, Bellodi, and Smeraldi), 5,8,9 Milan, and the Department voked in PD, the central mechanisms gas mixture showed significant metabolic of Child Psychiatry, Istituto underlying laboratory-induced panic re- and cerebral flow changes localized pri- 15 Scientifico Eugenio Medea, main undescribed. Only limited efforts marily to the ventral medulla. Bosisio Parini, Lecco have been made to link the functional bases One study16 of normal volunteers in- (Dr Bertella), Italy. of the normal reflex ventilatory response vestigated the roles of peripheral vs cen-

(REPRINTED) ARCH GEN PSYCHIATRY/ VOL 58, FEB 2001 WWW.ARCHGENPSYCHIATRY.COM 114

©2001 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/27/2021 SUBJECTS AND METHODS xanthine-containing beverages for at least 8 hours, from smoking for at least 6 hours, and from eating for at least 2 hours before the tests. After complete description, an in- SUBJECTS formed consent was obtained. The procedure of the CO2 mixture challenge has been approved by our hospital’s Ethi- Subjects were recruited consecutively from outpatients of cal Committee. More than 600 subjects in Europe have un- an anxiety treatment facility at San Raffaele Hospital, Mi- dergone the challenge without adverse events. lan, Italy. They had to be aged 18 through 35 years and have a clinical diagnosis of PD based on the DSM III-R,17 which PROCEDURES was confirmed independently by the Mental Health Diag- nostic Interview Schedule Revised.18-20 All subjects received orally 100 mg of pirenzepine hydro- Exclusion criteria were cardiocirculatory and respi- chloride, 4 mg of biperiden hydrochloride, and placebo at ratory disorders, personal or familial history of aneurysm, 48-hour intervals before undergoing a CO2 mixture/air chal- hypertension (systolic blood pressure, Ͼ160 mm Hg; di- lenge on 3 different days, in a double-blind, random cross- astolic blood pressure, Ͼ100 mm Hg), pregnancy, epi- over design. lepsy, intolerance of or hypersensitivity to biperiden or piren- Pirenzepine is a muscarinic antagonist used to treat zepine, glaucoma or ocular hypertension, gastrointestinal gastric ulcer. The low lipid solubility largely prevents pas- tract stenosis, megacolon or dysfunctions in gut motility, sage of the blood-brain barrier.23 After oral administra- urinary retention, prostatic hypertrophia, and history of al- tion, it reaches peak plasma concentrations within 2 to 2.5 cohol, benzodiazepine, or other drug dependence. hours. The mean elimination half-life is 11 hours. With these criteria, 15 subjects were recruited during Biperiden is a muscarinic antagonist used as an anti- 13 months. After the first or second 35% CO2–65% oxygen parkinsonian agent. It easily penetrates the blood-brain bar- (O2) (hereafter referred to as CO2 mixture) challenge, 3 sub- rier and distributes in high concentrations to the brain- jects found the test excessively anxiogenic and dropped out. stem.24 After oral administration, it reaches peak plasma Therefore, this study is based on the 12 subjects who com- concentrations within 2 hours, and declines biphasically, pleted the protocol. Although we used the Italian version of with half times of 1.5 and 24 hours.24,25 the Mental Health Diagnostic Interview Schedule, Re- Two hours after administration of placebo, piren- 19 19,20 vised interview, for which reliability data are available, zepine, or biperiden, subjects underwent the CO2 mixture/ all 12 subjects also satisfied the DSM-IV diagnosis of PD,21 air respiratory challenge.2 according to reviews of interviews and clinical information. Once inhaled, CO2 quickly permeates the blood- All had already been treated for PD with short half-life ben- brain barrier and stimulates the central nervous system.5 zodiazepines at some time in their life. None had taken tri- Responders experience an intense increase in anxiety, usu- cyclic antidepressants, selective serotonin reuptake inhibi- ally described as closely resembling a spontaneous panic tors, or monoamine oxidase inhibitors. At the time of attack1,2,5 and lasting from a few seconds to less than 1 recruitment, 5 subjects were taking benzodiazepines (alpra- minute. A receiver operating characteristic analysis showed zolam, chlordemethildiazepam, etizolam, bromazepam) at that this challenge discriminates between patients with PD a mean daily dose of 0.5 mg alprazolam-equivalent. For these and healthy controls on the basis of subjective anxiety af- subjects, tapering was initiated, with monitoring of pos- ter the test, with a positive predictive power of 91% and a sible symptoms of withdrawal each second day. negative predictive power of 75%.26 In the week before the experiment, the severity of PD The following 2 gas mixtures were used: compressed was assessed using the Panic-Associated Symptoms Scale, air as placebo, and a mixture of 35% CO2 and 65% O2 mix- with scores ranging from 0 to 28.22 ture. Both gases were inhaled through the same self- At initiation of the protocol, subjects had to have been administration mask connected to a respirometer to mea- free of psychotropic drugs for at least 2 weeks, and receive sure vital capacity and gas volume delivered at each inhalation. no medications for at least 1 week. Subjects were asked to refrain from consuming alcohol for at least 36 hours and Continued on next page

tral muscarinic receptors in mediating the physiologic subjects, they can be important in those individuals 16 response to CO2 inhalation. Subjects were treated with who have greater chemosensitivity to hypercapnia. intravenous placebo, pirenzepine hydrochloride (a Central cholinergic receptors can, therefore, contribute muscarinic antagonist that does not cross the blood- to the central CO2-sensing mechanisms in humans, but brain barrier), and biperiden hydrochloride (a musca- the degree of their involvement seems to vary among rinic antagonist that crosses the blood-brain barrier) individuals. before undergoing progressive hyperoxic hypercapnia. This study was performed to investigate whether the There were no significant differences in the delta (⌬) response to hypercapnic stimulation in subjects with PD ⌬ minute ventilation to the end-tidal CO2 pressure ratio could be modified by the administration of muscarinic among the 3 treatments, but the difference in the hyper- antagonists. capnic ventilatory response between biperiden and placebo had a significant negative correlation with the hy- RESULTS percapnic response after placebo.16 Although central muscarinic receptors play no prominent or consistent There were 6 men and 6 women in the study group. Mean role in the physiologic hypercapnic reflex in normal age was 27.35±5.1 years, and mean age at onset of PD

(REPRINTED) ARCH GEN PSYCHIATRY/ VOL 58, FEB 2001 WWW.ARCHGENPSYCHIATRY.COM 115

©2001 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/27/2021 Subjects were informed before the challenge that negative, then ⌬% VASA=(⌬ VASAϫ100)/(VASA before they would inhale 2 different harmless gas mixtures con- inhalation). taining different percentages of O2 and CO2, that the By taking into account the pretest baseline value of procedure might elicit some sensations of discomfort anxiety, this transformation allows us to calculate sym- ranging from a few physical symptoms to a clear sensa- metrically for positive and negative ⌬ values the maxi- tion of anxiety, and that the aim of the study was to mum possible increment or decrement of anxiety after in- investigate whether the drug they had received could halation. influence their response to the challenge.The possibility Similarly, the rating at the PSL-III-R was transformed that a full panic attack might take place was not men- as follows: (1) if postinhalation PSL-III-R values minus pre- tioned, however, to avoid possible negative cognitive inhalation values (⌬ PSL-III-R score) was positive, then ⌬% biases related to expectation.2,3,26,27 PSL-III-R=(⌬ PSL-III-Rϫ100)/(52−PSL-III-R before in- After vital capacity was measured, subjects inhaled 1 halation), where ⌬% PSL-III-R is the percentage of maxi- vital capacity of CO2 mixture, or compressed air, in a ran- mum increment or decrement possible on the PSL-III-R; domly assigned order, at an interval of 30 minutes be- and (2) if ⌬ PSL-III-R was negative, then ⌬% PSL-III-R=(⌬ tween both inhalations. At the end of each inhalation, sub- PSL-III-Rϫ100)/(PSL-III-R before inhalation). jects were asked to hold their breath for 4 seconds. According Receiver operating characteristic analyses showed that to this standardized procedure,2,3,26,27 the test is consid- measuring the challenge response with transformed ⌬% ered valid if the subject inhales at least 80% of the vital ca- VASA and ⌬% PSL-III-R values yields better information pacity. content than measuring with untransformed values.26 Criteria for qualitative assessment of response to the MEASURES challenge reflected standardized procedures derived by receiver operating characteristic analyses26,30: reaction to Immediately before and after each inhalation of com- any gas mixture inhalation was considered an induced pressed air or CO2 mixture, subjects were asked to score attack when it included a sensation of fear or panic with a themselves on a Visual Analog Scale for Anxiety (VASA). ⌬% VASA of at least 26 and an increment of at least 2 Like the visual analog scales commonly used in assess- points on the PSL-III-R for at least 4 symptoms of a panic ments of behavioral psychotherapy,28 the VASA depicts on attack. a 10-cm line the degree of global subjective anxiety, following a continuum from 0 (no anxiety present at all) to DATA ANALYSES 100 (the worst anxiety ever imaginable), and has been used consistently and reliably in several studies of panic provo- Given the small sample size, ⌬% VASA and ⌬% PSL-III-R cation.3,26,29-32 values obtained after challenge administration were ana- Likewise, subjects were asked to score themselves on lyzed in 2 separate Friedman analyses of variance. the Panic Symptom List III-R (PSL-III-R29), a self-rating ques- Contrasts between ⌬% VASA and ⌬% PSL-III-R after tionnaire assessing on a 5-point scale (0 indicates absent; CO2 mixture plus biperiden vs CO2 mixture plus placebo, 4, very intense) each of the 13 DSM-III-R and DSM-IV panic and after CO2 mixture plus biperiden vs CO2 mixture plus symptoms (total score range, 0-52). pirenzepine were made by means of the Wilcoxon matched- Evaluation of responses was made according to the fol- pairs test, with nominal ␣ set at .025. lowing standard2,26,30,32 procedures: global anxiety reactiv- Responses to different gas mixtures across different ity was evaluated as the percentage of maximum incre- treatments measured categorically (ie, presence or ab- ment or decrement possible on the VASA scale (⌬% VASA; sence of provoked attack) were compared by means of the range of values, −100 to 100) calculated as follows: (1) if Cochran Q test. postinhalation VASA values minus preinhalation values (⌬ Analyses were performed using the Statistica pack- VASA score) was positive, then ⌬% VASA=(⌬ VASAϫ100)/ age,33 with all test outcomes 2-tailed. Unless otherwise in- (100−VASA before inhalation); and (2) if ⌬ VASA was dicated, data are given as mean±SD.

was 19.9±6.1 years. Mean Panic-Associated Symptom and ⌬% PSL-III-R values obtained after the challenge in Scale score was 9.3±4.3 in the week before the experi- the 3 pharmacological treatments (Figure 1 and ment (mild to moderate severity of symptoms). Accord- Figure 2). There was no anxious reaction to com- ing to clinical records, one man had qualified for alco- pressed air inhalation, regardless of treatment. hol abuse 4 years before the experiment, but abuse and These responses did not differ between those pa- dependence were definitely ruled out at the time of the tients who had (5 patients [41.7%]) vs those who had not study. None of the subjects who were still taking ben- (7 patients [58.3%]) received psychotropic medication zodiazepines at the time of recruitment had experi- (ie, benzodiazepine therapy) recently (Mann-Whitney enced symptoms of withdrawal during the tapering or test for ⌬% VASA [patients with vs those without recent the drug-free periods before the experiment. medication]: after CO2 mixture plus placebo, U=14.5 The 3 subjects who dropped out did not differ from [P=.63]; after CO2 mixture plus biperiden, U=16.0 ⌬ those who completed the protocol for sociodemo- [P=.81]. Mann-Whitney test for % PSL-III-R: after CO2 graphic features, age at onset, duration and current se- mixture plus placebo, U=11.0 [P=.29]; after CO2 mix- verity of PD, or medication history. ture plus biperiden, U=17.0 [P=.93]). Biperiden, but not pirenzepine or placebo, modu- When blind codes were opened at the end of the ⌬ lated the response to CO2 mixture measured as % VASA experiment, the random orders of administration of gas

(REPRINTED) ARCH GEN PSYCHIATRY/ VOL 58, FEB 2001 WWW.ARCHGENPSYCHIATRY.COM 116

©2001 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/27/2021 mixtures (compressed air and then CO2 mixture, or CO2 120 mixture and then compressed air) across 3 different days ± SD ± could be arranged into 5 different sequences, whereas the SE 80 Mean orders of administration of treatments had 6 sequences. The orders of administration of gas mixtures (on 5 levels) and treatments (on 6 levels) were used for coding 40 group memberships as independent variables in a set of % VASA Kruskall-Wallis analyses of variance that analyzed ⌬% ∆ 0 ⌬ VASA and % PSL-III-R after CO2 mixture inhalation in each of the 3 treatments. Differences were nonsignifi- –40 cant when the responses to CO2 mixture were grouped

according to the order of administration of gas mixtures –80 Air + Air + Air + CO + CO + CO + (values closest to significance observed for ⌬% VASA af- 2 2 2 Placebo Pirenzepine Biperiden Placebo Pirenzepine Biperiden ter CO2 mixture and pirenzepine, H4,12=4.03 [P=.40]; for –11.2 ± 18.7 –1.2 ± 27.6 –18.3 ± 46.8 74.7 ± 20.5 75.1 ± 26.2 0.7 ± 45.2 ⌬ % PSL-III-R, H4,12=3.47 [P=.48]). Similarly, differ- Figure 1. Mean (± SD, ± SE) percentage of maximum increment or ences were nonsignificant when the responses to CO2 mix- decrement possible on the Visual Analog Scale for Anxiety (⌬% VASA) after ture were grouped according to the order of administra- placebo, pirenzepine hydrochloride, and biperiden hydrochloride and 35% tion of treatments (values closest to significance observed carbon dioxide–65% oxygen (CO2 mixture) or compressed air (air) ⌬ inhalation. Results of Friedman analysis of variance of ⌬% VASA values for % VASA after CO2 mixture and placebo, H5,12=8.8 ␹2 ⌬ obtained after challenge administration were as follows: 5=38.85 (n=12; [P=.12]; for % PSL-III-R, H5,12=8.7 [P=.12]), in har- PϽ.001); ⌬% VASA after CO2 mixture and biperiden vs ⌬% VASA after CO2 mony with findings of previous studies.3,26 mixture and pirenzepine (Wilcoxon matched-pairs test), z=3.05 ( PՅ.002); ⌬ ⌬ Qualitative responses to the experiment with CO and % VASA after CO2 mixture and biperiden vs % VASA after CO2 2 mixture and placebo (Wilcoxon matched-pairs test), z=2.98 ( PՅ.003). mixture yielded 9 patients (75%) and 7 patients (58%) Nominal ␣ was set at .025. who developed provoked panic attacks after placebo and pirenzepine, respectively, and no provoked panic 60 attacks after biperiden, whereas there were no panic ± SD attacks with compressed air, regardless of the pharma- 40 ± SE Ͻ Mean cologic treatment (Cochran Q test, Q5=39.7 [P .001]). 20

COMMENT 0 –20 The data reveal that the exaggerated response to CO 2 % PSL-III-R observed in subjects with PD can be modulated through ∆ –40 blockade of central muscarinic receptors by biperiden. –60

Consistently, the effect of intravenous biperiden in –80 blocking the ventilatory reponse to CO2 in normal subjects correlates with the degree of hypercapnic ventila- –100 Air + Air + Air + CO2 + CO2 + CO2 + tory response at baseline.16 Placebo Pirenzepine Biperiden Placebo Pirenzepine Biperiden It has been hypothesized4,26 that individual sensi- –20.4 ± 24.0 –30.6 ± 45.8 –33.5 ± 32.8 30.2 ± 17.3 27.1 ± 19.9 –5.1 ± 31.5 tivities to increasing concentrations of inhaled CO2 may Figure 2. Mean (± SD, ± SE) percentage of maximum increment or reflect a continuously distributed, possibly evolution- decrement possible on the Panic Symptom List III-R (⌬% PSL-III-R) after arily derived, developmental trait. The magnitude of the placebo, pirenzepine hydrochloride, and biperiden hydrochloride treatment and 35% carbon dioxide–65% oxygen (CO2 mixture) or compressed air (air) response to the CO2 challenge among individuals may inhalation. Results of Friedman analysis of variance of ⌬% PSL-III-R values ␹2 parallel such a distribution, with patients with PD at obtained after challenge administration were as follows: 5=36.14 (n=12; Ͻ ⌬ ⌬ one extreme.4,26 It is suggested that muscarinic recep- P .001); % PSL-III-R after CO2 mixture and biperiden vs % PSL-III-R after CO2 mixture and pirenzepine (Wilcoxon matched-pairs test), z=2.43 tors can account at least partially for such individual ( PՅ.02); and ⌬% PSL-III-R after CO2 mixture and biperiden vs ⌬% variability. PSL-III-R after CO2 mixture and placebo (Wilcoxon matched-pairs test), It appears unlikely, however, that dysfunction of z=2.82 ( PՅ.005). Nominal ␣ was set at .025. any single neurotransmitter system, or stimulation of the medullary chemoceptor(s) by CO2 alone, can frontal cortex, other salient features of PD, such as anti- account for the complex phenomenology of panic cipatory anxiety and phobic avoidance, can be primed attacks and PD. A neuroanatomical model connecting and promoted4,5 after the onset of 1 or more acute at- the acute attack, anticipatory anxiety, and phobic avoid- tack(s) generated in the brainstem. ance components of PD to the brainstem, limbic sys- Some pathologic conditions also allow some links tem, and prefrontal cortex, respectively, has been to be drawn between extreme variations in sensitivity to 5 offered. According to this model, stimulation of the CO2, muscarinic receptor function, and anxiety. In con- medullary chemoceptor(s) by CO2 can provoke a dose- genital central hypoventilation syndrome, which is of- dependent increase in the firing rate of the locus ten associated with a hypoplastic or absent arcuate nucleus ceruleus,5,34 an area with a pivotal role in anxiety,5,8 per- at the ventral medullary surface,36 children cannot per- haps through the projections of the medullary nucleus ceive ambient CO2 variations, and have significantly less reticularis gigantocellularis.5,35 Through connections of anxiety symptoms than age-matched controls.37 Insen- the locus ceruleus to the hippocampus and to the pre- sitivity to the suffocation stimulus that occurs when chil-

(REPRINTED) ARCH GEN PSYCHIATRY/ VOL 58, FEB 2001 WWW.ARCHGENPSYCHIATRY.COM 117

©2001 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/27/2021 dren rebreathe their own exalations in the prone posi- CO2 does not necessarily suggest an intrinsic role of cho- tion during sleep makes liable infants subject to sudden linergic mechanisms in the pathophysiology of PD and infant death syndrome. A significant decrease in bind- calls for further investigations in larger samples of patient ing of the muscarinic antagonist tritiated quinuclidinyl and nonpatient subjects. benzilate to muscarinic receptors in the arcuate nucleus has been demonstrated in these infants.38 This led re- Accepted for publication June 20, 2000. searchers to hypothesize that sudden infant death syn- An interim evaluation of part of these data was pre- drome is intrinsically associated with a dysfunction of sented at a Workshop on the Panic and Respiration Con- muscarinic receptors in the arcuate nucleus, the same nection, Milan, Italy, February 10, 1998. muscarinic receptors that might constitute all or part of The statistical consultation of Andrea Fossati, MD, and the central chemoceptor mediating responses to hyper- Clelia Di Serio, PhD, is gratefully acknowledged. capnia.38 Corresponding author and reprints: Marco Battaglia, Some limitations and caveats need to be taken into MD, Department of Neuropsychiatric Sciences, Istituto account in this study. First, this is a small patient sample. Scientifico San Raffaele Hospital, 29 via Prinetti, 20127 Second, measures of subjective increase in anxiety39 have Milan, Italy (e-mail: [email protected]). been the only index of response adopted here; assess- ments of ventilation would have been desirable to better REFERENCES delineate the role of chemosensitivity in CO2 vulnerabil- 6 ity. Third, the inclusion of objective measures of pe- 1. Price LH, Goddard AW, Barr LC, Goodman WK. Pharmacological challenges in ripheral anticholinergic effects would have been valu- anxiety disorders. In: Bloom FE, Kupfer DJ, eds. Psychopharmacology: The Fourth able; the central effect of biperiden as the only explanation Generation of Progress. New York, NY: Raven Press; 1995:1311-1323. for the different response from pirenzepine would have 2. Griez E, Lousberg H, van den Hout MA, van der Molen GM. CO2 vulnerability in panic disorder. Psychiatry Res. 1987;31:193-199. been more reliably interpreted in the presence of simi- 3. Perna G, Bertani A, Arancio C, Ronchi P, Bellodi L. Laboratory response of pa-

lar peripheral anticholinergic effects of biperiden and tients with panic and obsessive-compulsive disorders to 35% CO2 challenge. Am pirenzepine. Fourth, beyond double-blindedness and ran- J Psychiatry. 1995;152:85-89. domization, we have not controlled for the possible ef- 4. Klein DF. False suffocation alarms, spontaneous panic, and related conditions. fect of patients’ expectations of the challenge. Gener- Arch Gen Psychiatry. 1993;50:306-317. 5. Gorman JM, Liebowitz MR, Fyer AJ, Stein J. A neuroanatomical hypothesis for ally, however, it has been shown that patients’ expectancy panic disorder. Am J Psychiatry. 1989;146:148-161. 32 has little influence on the challenge outcome. Fifth, al- 6. Papp LA, Martinez JM, Klein DF, Coplan JD, Norman RG, Cole R, de Jesus MJ, though none of the subjects had spontaneously re- Ross D, Goetz R, Gorman JM. Respiratory psychophysiology of panic disorder: ported any effects that could be linked specifically to cen- three respiratory challenges in 98 subjects. Am J Psychiatry. 1997;154:1557- 1565. tral cholinergic blockade (eg, drowsiness) in the 2 hours 7. Jensen CF, Keller TW, Peskind ER, McFall ME, Veith RC, Martin D, Wilkinson after administration of biperiden, we have not system- CW, Raskind MA. Behavioral and neuroendocrine responses to sodium lactate atically controlled for possible nonspecific effects that infusion in subjects with posttraumatic stress disorder. Am J Psychiatry. 1997; 154:266-268. might have contributed to blunting the response to CO2 with this drug. 8. Goddard AW, Charney DS. Toward an integrated neurobiology of panic disorder. J Clin Psychiatry. 1997;58(suppl):4-11. General caveats include the only partial overlap be- 9. Heninger GR. Catecholamines and pathogenesis in panic disorder. Arch Gen Psy- 30 tween the diagnosis of PD and CO2 -induced panic, sug- chiatry. 1998;55:522-523. gesting at least partially different determinants for these 10. Millhorn DE, Eldridge FL. Role of ventrolateral medulla in the regulation of res- 2 phenotypes.40 piratory and cardiovascular systems. J Appl Physiol. 1986;61:1249-1263. 11. Burton MD, Johnson D, Kazemi H. CSF acidosis augments ventilation through Moreover, the 35% CO2–65% O2 mixture is consid- cholinergic mechanisims. J Appl Physiol. 1989;66:2565-2572. ered a robust stimulus, but it may be less specific than 12. Nattie EE, Wood J, Mega A, Goritski W. Rostral ventrolateral medulla musca- lower concentrations of CO2. Stimulation with different rinic receptors involvement in central ventilatory chemosensitivity. J Appl Physiol. gas concentrations may provide different results to those 1989;66:1462-1470. reported herein, and the short-term reliability of the test 13. Nattie EE, Aihua L. Ventral medulla muscarinic receptor subtypes involved in car- diorespiratory control. J Appl Physiol. 1990;69:33-41. is unknown. 14. Burton MD, Nouri M, Kazemi H. Acetylcholine and central respiratory control: Attenuation of the response to CO2 mixture inha- perturbations of acetylcholine synthesis in the isolated brainstem of the neona- lation in patients with PD can be achieved with a variety tal rat. Brain Res. 1995;670:39-47. of agents. These include tricyclic antidepressants and se- 15. Gozal D, Hathout G, Kirlew K, Tang H, Woo MS, Zhang J, Lufkin RB, Harper RM. lective serotonin reuptake inhibitors,41 usually after 7 to Localization of putative neural respiratory regions in human by functional MRI. 29 J Appl Physiol. 1994;76:2076-2083. 30 days of treatment, and benzodiazepines, but the data 16. Akiyama Y, Nishimura M, Kobayashi S, Yamamoto M, Miyamoto K, Kawakami presented herein show very substantial reduction in re- Y. Effects of M1 selective antimuscarinics in respiratory chemosensitivity in sponsiveness to CO2 mixture inhalation after a single dose humans. Respir Physiol. 1996;103:127-135. of biperiden, a drug which is not considered therapeu- 17. American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, Revised Third Edition. Washington, DC: American Psychiatric Asso- tic for PD. ciation; 1987. Although it must be remembered that the global 18. Robins L, Helzer JE, Cottler L, Goldring E. NIMH Diagnostic Interview Schedule: anxiety response to laboratory provocation of panic is a Version III-R. St Louis, Mo: Washington University; 1989. complex index that is likely to be affected by different 19. Battaglia M, Bertella S, Politi E, Bernardeschi L, Perna G, Gabriele A, Bellodi L. neurochemical pathways, to our knowledge this is the Age at onset of panic disorder: influence of familial liability to the disease and of childhood separation anxiety disorder. Am J Psychiatry. 1995;152:1362- first report to show a specific role for acetylcholine, a neu- 1364. rotransmitter that is not usually associated with PD. Like- 20. Battaglia M, Bertella S, Bajo S, Binaghi F, Bellodi L. Anticipation of age at onset wise, the ability of biperiden to modulate the response to in panic disorder. Am J Psychiatry. 1998;155:590-595.

(REPRINTED) ARCH GEN PSYCHIATRY/ VOL 58, FEB 2001 WWW.ARCHGENPSYCHIATRY.COM 118

©2001 American Medical Association. All rights reserved. Downloaded From: https://jamanetwork.com/ on 09/27/2021 21. American Psychiatric Association. Diagnostic and Statistical Manual of Mental 31. Fyer MR, Uy J, Martinez J, Goetz R, Klein DF, Fyer A, Liebowitz MR, Gorman J.

Disorders, Fourth Edition. Washington, DC: American Psychiatric Association; CO2 challenge of patients with panic disorder. Am J Psychiatry. 1987;144:1080- 1994. 1082. 22. Argyle N, Deltito J, Allerup P, Maier W, Albus M, Nutzinger D, Rasmussen S, 32. Perna G, Battaglia M, Garberi A, Arancio C, Bertani A, Bellodi L. Carbon dioxide/ Ayuso JL, Bech P. The Panic-Associated Symptoms Scale: measuring the se- oxygen challenge test in panic disorder. Psychiatry Res. 1994;52:159-171. verity of panic disorder. Acta Psychiatr Scand. 1991;83:20-26. 33. Statistica for Windows. Tulsa, Okla: Statsoft; 1993. 23. Carmine AA, Brogden RN. Pirenzepine: a review of its pharmacodynamic and phar- 34. Aston-Jones G, Ennis M, Pieribone VA, Nickell WT, Shipley MT. The brain nucleus macokinetic properties and therapeutic efficacy in peptic ulcer disease and other locus coeruleus: restricted afferent control of a broad efferent network. Science. allied diseases. Drugs. 1985;30:85-126. 1986;234:734-737. 24. Yokogawa K, Nakashima E, Ishizaki M, Hasegawa J, Kido H, Ichimura F. Brain 35. Torvik A, Bredal A. The origin of the reticulospinal fibers in the cat: an experi- regional pharmacokinetics of biperiden in rats. Biopharm Drug Dispos. 1992; mental study. Anat Rec. 1957;128:113-137. 13:131-140. 36. Folgering H, Kuyper F, Kille JF. Primary alveolar hypoventilation (Ondine’s curse) 25. Grimaldi RE, Perucca E, Ruberto G, Gelmi C, Trimarchi F, Hollmann M, Crema A. in an infant without external arcuate nucleus. Bull Eur Physiopathol Respir. 1979; Pharmacokinetic and pharmacodynamic studies following the intravenous and 15:659-665. oral administration of the antiparkinsonian drug biperiden to normal subjects. 37. Pine DS, Weese-Mayer DE, Silvestri JM, Davies M, Whitaker AH, Klein DF. Eur J Clin Pharmacol. 1986;29:735-737. Anxiety and central hypoventilation syndrome. Am J Psychiatry. 1994;151:

26. Battaglia M, Perna G. The 35% CO2 challenge in panic disorder: optimization by 864-870. receiver operating characteristic analysis. J Psychiatr Res. 1995;29:111-119. 38. Kinney HC, Filiano JJ, Sleeper LA, Mandell F, Valdes-Dapena M, Frost-White W. 27. Harrington PJ, Schmidt NB, Telch MJ. Prospective evaluation of panic potentia- Decreased muscarinic receptor binding in the arcuate nucleus in sudden infant

tion following 35% CO2 challenge in nonclinical subjects. Am J Psychiatry. 1996; death syndrome. Science. 1995;269:1446-1450.

153:823-825. 39. Coryell W, Arndt S. The 35% CO2 inhalation procedure: test-retest reliability. Biol 28. Wolpe J. The Practice of Behavior Therapy. Elmsford, NY: Pergamon Press Inc; Psychiatry. 1999;45:923-927. 1973. 40. Roy-Byrne PP, Cowley DS. Search for pathophysiology of panic disorder. Lan- 29. Pols H, Zandbergen J, de Loof C, Griez E. Attenuation of carbon dioxide– cet. 1998;352:1646-1647. induced panic after clonazepam treatment. Acta Psychiatr Scand. 1991;84:585- 41. Bertani A, Perna G, Caldirola D, Arancio C, Bellodi L. Pharmacologic effect of imip- 586. ramine, paroxetine, and sertraline on 35% carbon dioxide hypersensitivity in panic

30. Bellodi L, Perna G, Caldirola D, Arancio C, Bertani A, Di Bella D. CO2-induced panic patients: a double-blind, random, placebo-controlled study. J Clin Psychophar- attacks: a twin study. Am J Psychiatry. 1998;155:1184-1188. macology. 1997;17:97-101.

(REPRINTED) ARCH GEN PSYCHIATRY/ VOL 58, FEB 2001 WWW.ARCHGENPSYCHIATRY.COM 119