Psychosomatic Medicine Tips for Better Browsing
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Wilhelm, F. H.
Right arrow Articles by Roth, W. T.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Wilhelm, F. H.
Right arrow Articles by Roth, W. T.
Related Collections
Right arrow Anxiety
Right arrow Pulmonary
Psychosomatic Medicine 63:638-649 (2001)
© 2001 American Psychosomatic Society

ORIGINAL ARTICLES

Slow Recovery From Voluntary Hyperventilation in Panic Disorder

Frank H. Wilhelm, PhD, Alexander L. Gerlach, Dr rer nat and Walton T. Roth, MD

From the Department of Psychiatry and Behavioral Sciences, Stanford University, Stanford and the Department of Veterans Affairs Health Care System, Palo Alto, California. Dr. Gerlach is currently at the Westfälische Wilhelms-Universität Münster, Germany.

Address reprint requests to: Frank Wilhelm, Stanford University/VAPAHCS (116F-PAD), 3801 Miranda Ave., Palo Alto, CA 94304. Email: fwilhelm{at}stanford.edu


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 ACKNOWLEDGMENTS
 REFERENCES
 
OBJECTIVE: Because hyperventilation has figured prominently in theories of panic disorder (PD) but not of social phobia (SP), we compared predictions regarding diagnosis-specific differences in psychological and physiological measures before, during, and after voluntary hyperventilation.

METHOD: Physiological responses were recorded in 14 patients with PD, 24 patients with SP, and 24 controls during six cycles of 1-minute of fast breathing alternating with 1 minute of recovery, followed by 3 minutes of fast breathing and 10 minutes of recovery. Speed of fast breathing was paced by a tone modulated at 18 cycles/minute, and depth by feedback aimed at achieving an end-tidal pCO2 of 20 mm Hg. These values were reached equally by all groups.

RESULTS: During fast breathing, PD and SP patients reported more anxiety than controls, and their feelings of dyspnea and suffocation increased more from baseline. Skin conductance declined more slowly in PD over the six 1-minute fast breathing periods. At the end of the final 10-minute recovery, PD patients reported more awareness of breathing, dyspnea, and fear of being short of breath, and their pCO2s, heart rates, and skin conductance levels had returned less toward normal levels than in other groups. Their lower pCO2s were associated with a higher frequency of sigh breaths.

CONCLUSIONS: PD and SP patients report more distress than controls to equal amounts of hypocapnia, but PD differ from SP patients and controls in having slower symptomatic and physiological recovery. This finding was not specifically predicted by hyperventilation, cognitive-behavioral, or suffocation alarm theories of PD.

Key Words: Panic disorder • social phobia • hypocapnia • respiration • respiratory sinus arrhythmia • autonomic nervous system

Abbreviations: HV = hyperventilation; PD = panic disorder; SP = social phobia; CO2 = carbon dioxide; pCO2 = partial pressure of CO2; RSA = respiratory sinus arrhythmia; SC = skin conductance; STAI = Spielberger State-Trait Anxiety Inventory; DSM-IV = Diagnostic and Statistical Manual of Mental Disorders, 4th Edition; FB = fast breathing; ANOVA = analysis of variance; ANCOVA = analysis of covariance; SD = standard deviation


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
METHODS
 RESULTS
 DISCUSSION
 ACKNOWLEDGMENTS
 REFERENCES
 
Panic disorder (PD) is characterized by recurrent unexpected panic attacks about which there is persistent concern - typically about having additional attacks and about somatic or psychological implications of attacks, which often leads to significant behavior changes (1). Thus in its current conceptualization, PD is a syndrome that at its core includes both disordered physiology and persistent apprehension about symptoms (2). One specific physiological dysregulation that has received much attention in PD research is hyperventilation (HV). HV is defined as breathing in excess of metabolic needs and acutely leads to a reduction in the level of arterial pCO2 (hypocapnia). This respiratory abnormality has been connected with PD in several ways.

First, long before the diagnostic category of PD became accepted, a hyperventilation syndrome (3) had been described, and a symptomatic overlap in the two entities is apparent (4, 5). In its basic form, this perspective postulates that HV itself produces anxiety and related symptoms (6); PD patients experience panic attacks because they hyperventilate acutely. Recent results, however, have questioned the importance of HV for panic: voluntary HV is a relatively weak panic provocation (7–9), and spontaneous panic attacks often are not accompanied by HV (10–12), at least as detected by transcutaneous measurement. Yet empirically, in laboratory studies, lower end-tidal pCO2 has repeatedly, although not invariably (13, 14), emerged as a difference between PD patients and comparison groups, either during baseline periods (9, 15–18) or during recovery from voluntary HV (7, 19) (see below).

Second, the cognitive-behavioral theory of PD emphasizes hypochondriacal ideas (catastrophic cognitions) that lead PD patients to interpret sensations originating from benign physiological changes, such as slight accelerations in heart rate or mild HV, as an indication of imminent harm (20). The anxiety associated with these cognitive misinterpretations—and to some extent, phobic emotional conditioning from previous panic attacks—leads to more intense sensations, which in a positive-feedback loop causes an ascending spiral of anxiety or panic (21). As support for the theory it has been shown that PD patients exhibit elevated levels of anxiety sensitivity, a disposition to react fearfully to anxiety sensations (22). Within this theory, HV has an important place because it can result in a variety of sensations that PD patients can misinterpret catastrophically.

Third, and most recently, the suffocation false alarm theory (23, 24) assigns HV a secondary role in PD. It postulates that in patients with PD, a hypothetical biological mechanism is faulty and overreacts to stimuli indicating a suffocation threat (typically signified by an increase in arterial pCO2) and consequently triggers feelings of dyspnea (shortness of breath) and panic. As a preventive measure, PD patients hyperventilate chronically to produce a buffer of hypocapnia, which makes it less likely that pCO2 levels will edge their way up to where they may trigger the suffocation alarm. According to this theory, a rise in arterial pCO2, not a fall, is a more likely initial trigger of panic. Evidence for this theory comes from experiments where PD patients react to increases in inspired CO2 with panic attacks and augmented respiratory adjustments (25).

To shed more light on the empirical support for the three theories, we studied patterns of physiological and psychological response to HV in two anxiety disorders, PD and social phobia (SP), and compared responses in these groups with psychiatrically healthy controls. SP served as a clinical control group to test for the specificity of reactions to a diagnosis of PD, not to state anxiety per se. The experiment consisted of seven episodes of increased breathing at tightly controlled respiratory rates and to specified pCO2 levels, followed by recovery periods. To prevent drop-outs, breathing rates, pCO2 reduction goals, and length of HV were set to levels that should have been attainable by most, if not all of the participants.

The three perspectives on the role of HV in PD lead to specific predictions regarding reactions to voluntary HV in the three groups: From the standpoint of the HV syndrome perspective, voluntary HV would be expected to produce equivalent increases in anxiety and related symptoms in all three groups, irrespective of diagnosis. In contrast, cognitive-behavioral theory would predict that voluntary HV would increase anxiety and symptoms in the PD group specifically. In sharp contrast, the most straightforward prediction of the suffocation false alarm theory is that voluntary HV decreases panic anxiety in PD patients, because with lower pCO2 levels, false alarms are less likely to be triggered.

None of the perspectives above specifically addresses what happens when voluntary HV is suddenly stopped, although ideas that panic disorder represents a kind of brainstem autonomic dysregulation, might imply peculiarities of recovery as well (26). Two sets of empirical observations indicate an attenuated course of physiological recovery in PD patients. First, slowed recovery of pCO2 levels after voluntary HV compared with healthy controls has been observed in HV syndrome patients (27–30) and in PD patients panicking during HV (7, 19). Negative studies (31) had shorter HV (<3 minutes) or recovery periods (<5 minutes) than positive ones. Second, PD patients have demonstrated slowed recovery of skin conductance response levels after HV (32, 33), an indication of prolonged sympathetic response to the respiratory challenge. Thus, slow recovery in PD is apparently not confined to the respiratory system.

To test the predictions of the different perspectives on HV in PD, we assessed heart rate, skin conductance level, and RSA, as the primary physiological measures of autonomic dysregulation in anxiety (34, 35); end-tidal pCO2, respiratory rate, tidal volume, and their product, minute volume as indices of the quality and quantity of respiratory changes; and self-reported anxiety, a set of symptoms salient for panic and HV, and a set of respiratory sensations related to interoception and the fear of those interceptions as psychological measures. In addition, we measured tidal volume instability because elevations in this variable have been observed repeatedly in PD patients during baseline and anxiety provocations (36–39).


   METHODS
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
RESULTS
 DISCUSSION
 ACKNOWLEDGMENTS
 REFERENCES
 
Subjects
Fourteen patients with PD and 24 patients with SP (DSM-IV) (1) were recruited by advertisements in local newspapers offering evaluation and treatment. Thirteen of the SP patients had been recruited specifically for complaints about blushing in social situations. In addition, 24 controls were selected to age- and sex-match the PD and SP patients. The advertisement for controls asked for people comfortable in social situations and without current emotional problems. Written informed consent was obtained after the procedures had been fully explained. Subjects were then interviewed using the Anxiety Disorder Interview Schedule for DSM-IV (40). None of the PD patients but 6 (25%) of the SP patients had additional diagnoses of major depression or generalized anxiety disorder. Controls were selected to be without lifetime psychiatric diagnoses. All subjects denied taking psychoactive or cardiovascularly active medication in the 2 weeks before testing. As can be seen in Table 1, groups were successfully selected not to differ in the proportion of women or in body mass index, which is known to influence respiratory function (41).


View this table:
[in this window]
[in a new window]
  		Table 1. Means (SD) of Sample Characteristics
 
In the week before testing, subjects filled out several questionnaires that had been mailed to them. These included the Beck Depression Inventory (42), the State-Trait Anxiety Inventory (STAI) Trait Form (43), the Anxiety Sensitivity Index (44), the Body Sensations Questionnaire (45), the Agoraphobic Cognitions Questionnaire (45), the Mobility Inventory (46), and the Social Avoidance and Distress Scale (47). Table 1 shows that on many of the questionnaires both anxious groups had higher scores than controls, but did not differ from each other. The two groups were matched on depression and anxiety sensitivity. Blushing and nonblushing SP patients had very similar scores on the questionnaires, including Anxiety Sensitivity Index (blushers = 26.6 ± 13.7; nonblushers = 28.9 ± 12.4.) As expected, SP patients were more socially avoidant and distressed than PD patients. Mobility Inventory scores in PD patients were below the agoraphobic range (46) and not different from those of the SP patients.

Procedures
Testing took place in a sound-attenuated chamber containing two small rooms. The subject sat in a comfortable chair in one room and the experimenter in the other. The experimenter could communicate with subjects by intercom and observe them through a one-way mirror. SP patients and some of the controls, but not PD patients, had been given other tests in this setting on a previous occasion. The following seven tests were given in fixed order: calibration of the respiration belts, a 10-minute baseline, repeated breath holding, slow and shallow breathing, relaxation, comfortable breathing at three speeds (for controlled assessment of respiratory sinus arrhythmia), and finally the fast- and deep-breathing test, which is the topic of this report. It was given last out of concern that it might induce so much anxiety that subjects would decline to be tested further.

The fast- and deep-breathing test lasted about 30 minutes. The sequence of events was a) 3 minutes of normal breathing, b) 1 minute of fast breathing alternating with 1 minute of recovery repeated six times, c) 3 minutes of fast breathing, and d) 10 minutes of recovery. During the fast-breathing periods, breathing speed was paced by a tone of varying pitch modulated sinusoidally at 18 cycles/minute. Subjects were instructed to breathe in as the pitch rose and out as it fell. Correct breathing depth was signaled by the illumination of a green LED in a 10-cm vertical column of 10 colored LEDs. The experimenter, watching a display of the subject’s end-tidal pCO2, manually controlled which LED was illuminated with the goal of inducing each subject to quickly reach and maintain a pCO2 of 20 mmHg.

Physiological Measures

  1. From an electrocardiogram lead, RR intervals were measured and instantaneous heart rate was calculated.
  2. Two channels of respiration (12-Hz sampling rate) were measured with inductive plethysmography using Respibands (Respitrace Corporation, Ardsley, NY) placed around the chest and abdomen. Calibration against spirometry was accomplished by the least-squares method (48). Inductive plethysmography is the least intrusive method for measuring respiratory volumes; measurement methods requiring a facemask or mouthpiece are known to alter natural breathing (49). Using two belts helps insure the accuracy of the volume estimations despite variation in proportions of abdominal and thoracic respiration with different depths and speeds of respiration. Eighty-nine percent of the tidal volumes measured by this method have been shown to be within ±10% of simultaneous spirometric measurements, and 100% within ±20%. Mean tidal volumes are not systematically biased (50). Several respiratory parameters were calculated breath-by-breath using customized programs: instantaneous respiratory rate, tidal volume, minute volume, mean inspiratory flow rate, and duty cycle (inspiratory time/total time). For the assessment of breath-to-breath within-subject variability (instability) in respiratory rate and tidal volume, values for each breath were converted into equidistant time series using cubic spline interpolation and resampling at 4 Hz. Resulting signals were subsequently analyzed using complex demodulation, a nonlinear time-domain method of time series analysis suitable for quantifying nonstationary oscillations in defined frequency ranges (51). Spectral bands were set to 0.017 to 0.15 Hz, corresponding to period lengths of 60 to 6.6 seconds. Most breath-by-breath variability in respiratory parameters is contained within this band. Complex demodulation is effective in characterizing nonstationary oscillations. If oscillations drift in frequency and amplitude or occur in bursts, the power spectrum or autocorrelation function can be insensitive or misleading.
  3. RSA as a measure of cardiac vagal control was assessed by means of cross-spectral analysis as the magnitude of the transfer function relating instantaneous RR interval (resampled to 4 Hz) to lung volume oscillations at the peak respiratory frequency (52, 53). The peak respiratory frequency was automatically detected as the greatest local maximum in the 0.15 to 0.50 Hz lung volume power spectral density. Spectral coherence at this frequency was required to be at least 0.5. Epochs with peak respiratory frequency below the 0.15 to 0.50 Hz band were excluded. For computation of the spectra, the RR interval and lung volume time series were first linearly detrended, and power density functions were computed for each period using the fast Fourier transform and the Welch algorithm (54). RSA measured in this way is adjusted for tidal volume, providing a more accurate index of cardiac vagal control than simple RR interval spectral power (55).
  4. Expiratory pCO2 was measured continuously by a calibrated infrared capnograph (Datex B, Puritan-Bennett Corporation, San Ramon, CA) into which air was drawn with a flow rate of 150 ml/min through a 1.2-mm diameter plastic tube ending in a dual nostril prong. Subjects were instructed to breathe only through their nose. End-tidal pCO2 was determined as the level at which pCO2 stopped rising at the end of expiration (final maximum). Expirations with a low percentage of alveolar air can be recognized by the pCO2 waveform not reaching a plateau (56), and these were excluded.
  5. Skin conductance level was recorded from the palmar surface of the middle phalanges of digits 3 and 4 of the left hand. Disposable Ag/AgCl electrodes with a contact surface area of 2.0 cm2 and an isotonic electrode paste were used. A constant 0.5 V was applied across the electrodes.

Self-Report Measures
We administered three questionnaires at two points in time—once at the end of an initial 10-minute baseline that took place before any of the respiratory challenge paradigms (about 50 minutes before the beginning of fast breathing) and once at the end of the final fast breathing recovery. On the second occasion, subjects were asked to fill out one set of questionnaires to describe how they felt at that point of time (end of recovery) and then to fill out a second set to describe retrospectively how they felt during the fast breathing. To test our theoretical predictions, eight items were selected a priori from these questionnaires for the present analysis. Detailed results from the questionnaires across several breathing provocations will be reported elsewhere.

The first was a Mood Questionnaire with 15 adjectives such as nervous, anxious, sad, angry, to be answered on a scale from 1 (not at all) to 10 (extremely). Only one item, anxious, was selected for the current analysis. In addition, anxiety on a 1 (not at all) to 10 (extremely) scale was assessed after each of the 1-minute fast breathing periods by prompting subjects to give a verbal response. The second questionnaire was a Symptom Questionnaire with 15 symptoms derived from the list of 13 in DSM-IV to be rated from 0 (not at all) to 4 (extremely). Feeling dizzy or unsteady was separated from feeling faint, and fear of losing control was separated from fear of going crazy. Three items were selected for the present analysis: shortness of breath (associated with the suffocation false-alarm theory), feeling dizzy/unsteady (associated with HV), and tingling/numbness (most specific to HV). The third was a Breathing Questionnaire with 25 questions to be answered on a scale from 0 (not at all) to 4 (extremely). The questions were based on empirically derived dyspnea dimensions (57–59) such as air hunger, resistance to inspiration and expiration, feelings of suffocation, and sensations in the chest. On all items, more positive scores indicated more disturbed breathing. Four items were selected for the present analysis: "I was aware of my breathing," "It scared me to feel short of breath," "My breathing varied from breath to breath" (all three relevant to cognitive-behavioral theory), and "I felt I was suffocating" (suffocation false-alarm theory).

Following the definition for panic attacks in DSM-IV (1), on the second occasion, subjects were also asked to answer for the fast-breathing period and for the final recovery period, "Did you have an attack during the . . . period when you suddenly felt more frightened, anxious, or extremely uncomfortable? A panic attack was defined as an affirmative answer to this question together with four symptoms (with an intensity of 1 or more on a scale from 0–4) on the Symptom Questionnaire for that period.

Statistical Analysis
One set of analyses assessed anxiety self-report and physiological reactions to the six 1-minute fast-breathing and the six 1-minute recovery periods. Another set looked at self-report and physiological measures for the three groups at the baseline, the 3-minute fast-breathing period, and the final recovery period. Within these periods, representative segments were selected for physiological measures: the mean of minutes 8 and 9 of the initial 10-minute baseline (minute 10 was not used because data were incomplete for several subjects), the mean of the last 2 minutes of the last fast-breathing period (where pCO2 was lowest), and the mean of minutes 8 and 9 of the 10-minute recovery. The sets of 8 self report and 8 physiological variables were entered into two 8 x 3 x 3 MANOVAs with the repeated measures factors, Variable and Task, and the factor Group. When appropriate, self-report measures were analyzed further with nonparametric tests such as the Kruskal-Wallis ANOVA by ranks and the Mann-Whitney U test. Physiological measures were analyzed by parametric tests: one-way analysis of variance (ANOVA) followed when significant by pairwise comparisons using the Tukey Honest Significant Difference test, or analysis of covariance (ANCOVA). P levels were corrected for nonsphericity using the Greenhouse-Geisser {varepsilon} where necessary.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
DISCUSSION
 ACKNOWLEDGMENTS
 REFERENCES
 
Self Report Measures
Figure 1 shows the self-reported anxiety at the baseline, the 7 fast-breathing periods, and the final recovery period. ANOVA of the six 1-minute fast-breathing periods confirmed that PD and SP patients were more anxious than controls at these times but did not differ from each other (group effect: F(2,58) = 8.15, p < .001, Tukey post-hoc tests, p values <.05).



View larger version (25K):
[in this window]
[in a new window]
  		Fig. 1. Means and standard errors of self-reported anxiety and minute-by-minute mean end-tidal pCO2 for the three groups over the course of the experiment. BASE is the baseline. 6 x FB/REC refers to the six 1-minute fast-breathing periods and 1-minute recovery periods, the following FB refers to the 3-minute fast-breathing period, and REC refers to the 10-minute final recovery period. Anxiety ratings were not evaluated during the initial six recovery periods. a marks the first time point where the panic disorder patients differ from the other groups (ANCOVA with last 2 minutes of FB as covariate: F(2,53) = 4.01, p < .02).

 
Table 2 presents information about self-report items for the three groups at the baseline, the 3-minute fast-breathing period, and the final recovery period. The MANOVA showed a significant effect for group (Rao’s R (16,94) = 3.95, p < .0001), Task (Rao’s R (16,39) = 3.95, p < .0001), and group x Task (Rao’s R (32,78) = 3.18, p < .0001). In all three periods, PD and SP patients reported equally higher anxiety than controls. During fast breathing, both patient groups had equally higher scores than controls on all other self-report items, except awareness of breathing and tingling/numbness, but during recovery, PD patients had higher scores than SP patients on most respiratory items.


View this table:
[in this window]
[in a new window]
  		Table 2. Means (SD) of Self-Report Measures
 
Table 3 presents the statistical significance of self-reported changes of the same items between the 3-minute fast breathing period and baseline (activation) and between that fast breathing period and recovery (deactivation). (We use activation and deactivation only to mean temporal change of individual measures between experimental conditions.) The fast breathing produced no more activation of anxiety in patients than controls, but more activation of shortness of breath and "I felt I was suffocating" in both patient groups compared with controls. During recovery, PD patients failed to deactivate as much as SP patients or controls on "I was aware of my breathing" and "It scared me to feel short of breath." Note that floor effects could reduce the significance of deactivation in controls.


View this table:
[in this window]
[in a new window]
  		Table 3. Statistical Significance of Self-Report Change Measures
 
Reports of Panic Attacks
Panic attacks were not assessed during baseline. During fast breathing, self-report criteria for panic attacks were met by 54% of SP patients, 36% of PD patients, and 4% of controls ({chi}2 (2) = 13.77, p < .001), and during recovery by 8% of SP patients, 43% of PD patients, and 0% of controls ({chi}2 (2) = 16.18, p < .001). During fast breathing, percentages were not different between SP and PD patients ({chi}2(1) = 1.21, p > .2), but the percentage was higher in PD than SP patients during recovery ({chi}2 (1) = 7.12, p < .01). Panic rates were almost identical for blushers and nonblushers among the SP patients during both fast breathing and recovery.

Physiological Measures
Figure 1 shows the time course of end-tidal pCO2. The HV instructions succeeded in bringing all groups to the same levels, which were approximately 27 mmHg during the first 1-minute fast breathing period and 20 mmHg in the last minute of the 3-minute fast breathing period. ANOVAs of the six 1-minute fast breathing and the six 1-minute recovery periods confirmed that pCO2 did not differ between groups for either fast breathing or recovery. None of the physiological variables except heart rate and skin conductance level showed main Group effects or Group x Repetition interactions for these periods. Heart rate level showed Group effects during both breathing (F(2,57) = 3.53, p < .04) and recovery (F(2,57) = 3.54, p < .04) because of higher levels in SP patients than controls (Tukey post-hoc tests: p < .05). Skin conductance level showed a Group x Repetition interaction during fast breathing (F(10,295) = 2.43, {varepsilon} = 0.409, p < .05). Pairwise linear contrasts showed that PD patients declined in skin conductance less than SP patients or controls (p < .05), who were not different.

Figure 2 gives information about physiological measures (except for pCO2, which is shown in more detail in Figure 1) for the three groups at the baseline, the 3-minute fast-breathing period, and the final recovery period. The MANOVA, including all physiological measures, showed a significant effect for Group (Rao’s R(16,92) = 1.80, p < .04), Task (Rao’s R(16,38) = 103.73, p < .0001), and Group x Task (Rao’s R(32,76) = 1.76, p < .02). One-way ANOVAs of the measures for each period were significant only for end-tidal pCO2 (F(2,53) = 7.21, p < .005) and tidal volume instability (F(2,59) = 4.94, p < .01). Post-hoc pairwise comparisons (p < .05) indicated that PD patients had lower end-tidal pCO2 and more tidal volume instability than SP patients and controls during recovery. ANOVAs for skin conductance level showed no statistically significant group differences for any period (p > .05). For exploratory purposes, duty cycle, inspiratory flow rate, and respiratory rate instability were examined also. None of these variables showed significant Group effects, except for respiratory rate instability, which during baseline was higher in PD patients (2.41 ± 1.09) than controls (1.59 ± 0.59; F(2,59) = 4.78, p < .01, post-hoc tests p < .05). SP patients were intermediate (1.97 ± 0.78).



View larger version (27K):
[in this window]
[in a new window]
  		Fig. 2. Means and standard errors of selected physiological measures for the three groups at the baseline (BASE), the 3-minute fast-breathing period (FB), and the final recovery period (REC).

 
Table 4 presents the statistical significance of an ANCOVA of the 3-minute fast-breathing period covaried with baseline levels (activation) and an ANCOVA of final recovery levels covaried with the 3-minute fast breathing levels (deactivation). ANCOVA can reduce some of the between-subject physiological variance irrelevant to the effects we want to examine (60). End-tidal pCO2, heart rate, and SC level showed less deactivation in PD patients than in SP patients or controls, and tidal volume instability showed less deactivation in PD patients than in controls. The exploratory variables (duty cycle, inspiratory flow rate, respiratory rate instability) showed no significant effects in these analyses. The pattern of results (not found in Table 4) is the same when deactivation is conceptualized as change from baseline, using baseline values as covariates for recovery levels, with the exception of SC level, which shows equal deactivation between groups (p > .1).


View this table:
[in this window]
[in a new window]
  		Table 4. Statistical Significance of Physiological Change Measures
 
Table 5 presents intercorrelations between deactivation change scores (recovery minus fast breathing) for five major variables—pCO2, tidal volume instability, heart rate, skin conductance level, and anxiety. Including all subjects, correlations were all significant except for skin conductance level and pCO2, and skin conductance level and anxiety. Within PD patients, correlations were generally similar, but with the lower number of subjects were only significant for pCO2 and tidal volume instability.


View this table:
[in this window]
[in a new window]
  		Table 5. Intercorrelations Between Deactivation Change Scores (Recovery to Fast Breathing) of Five Major Variables Across All Subjects (Below the Diagonal) and Within PD Patients (Above the Diagonal)
 
Panickers Vs Nonpanickers Among PD Patients
The small number of PD patients who panicked during HV did not permit statistical comparisons with nonpanickers. However, for panic attacks during final recovery, we compared the six patients with PD who panicked with the eight who did not panick, although these numbers were also small. As expected from the definition of panic attacks, self-reported anxiety was significantly higher in panickers with PD than the nonpanickers during recovery (7.7 ± 2.4 vs. 3.1 ± 2.7, t(12) = 3.13, p < .01), but not during baseline (p > .5) or HV (p > .07). None of the physiological measures during recovery was significantly different between these PD subgroups. The measure closest to significance was pCO2 (28.9 ± 2.6 vs. 32.9 ± 5.0, t(11) = 1.70, p = .11).

Respiratory Determinants of pCO2 During Recovery
Because a recent analysis of another sample of PD patients had shown that hypocapnia was associated with and probably caused by a higher frequency of sighs (defined as tidal volumes two times or more the subject mean) (61, 62), we assessed the number of sighs in the last 5 minutes of the recovery period. (We used a 5-minute period at the end of recovery rather than just minutes 8 and 9 because a reliable assessment of sigh frequency, which is below 0.2/min in some subjects, requires longer assessment periods.) Sigh frequency was significantly higher in PD patients (0.57 ± 0.35) than SP patients (0.30 ±0.33) or controls (0.30 ± 0.31) (Kruskal-Wallis H = 7.03, p < .03, post-hoc tests p < .05). Because higher sigh frequency implies more respiratory instability, these two variables correlated significantly across groups (r(62) = 0.40 p < .002). Within PD patients, however, this correlation was low (r(14) = 0.06, p > .8). We next calculated correlations between pCO2 during recovery and various of its possible determinants: minute volume, tidal volume, respiratory rate, and sigh frequency. For all groups combined, only sigh frequency (r(57) = -0.33, p < .02) and tidal volume (r(57) = -0.29, p < .03) were correlated with pCO2. In a stepwise multiple regression analysis with pCO2 as the dependent variable, sigh frequency was the first variable to be selected with an F-to-enter of 6.70 (p < .02) and tidal volume was the second, with an F-to-enter of 5.03 (p < .03). Within the PD patients, tidal volume (r(13) = -0.57, p < .04) was correlated significantly with pCO2 but sigh frequency was not (r(13) = -0.32, p < .3).


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
ACKNOWLEDGMENTS
 REFERENCES
 
Our use of paced breathing with pCO2 feedback was successful in that respiratory rate was increased and pCO2 was lowered to equal levels in all three diagnostic groups. Thus, between-group differences in self-report or in other physiological measures during the HV itself cannot be attributed to differences in pCO2 levels. We avoided dropouts by not requiring too extreme or prolonged respiratory efforts. Surprisingly, although results indicate clear diagnosis-specific patterns of reactivity to and recovery from voluntary HV, many core predictions of the three theoretical perspectives for understanding HV in PD were not supported by our data.

First, against the HV syndrome perspective, pCO2 failed to correspond to anxiety: In the initial baseline, PD patients were much more anxious than controls but did not have significantly lower pCO2 levels, and during paced breathing they were much more anxious than controls despite identically lowered pCO2 levels. However, within each group, voluntary HV-induced pCO2 reduction was associated with anxiety increase. Furthermore, the PD group had the lowest pCO2 levels during recovery and reported the highest anxiety levels.

Second, against the cognitive-behavioral theory, both patient groups showed equally greater anxiety levels during HV than controls, and the anxiety increase from baseline was not different for any group. In defense of the theory, Anxiety Sensitivity Index scores were also equally higher in PD and SP patients than controls, although this raises doubts about the theory as an explanation of panic anxiety or the validity of the index. Our mean score in SP patients was not elevated by the inclusion of patients principally concerned with blushing because blushers and nonblushers had similar scores, and was not atypically high compared with other published studies (63, 64). Nor was the Anxiety Sensitivity Index score for PD patients atypically low because of a lack of psychiatric comorbidity (65, 66).

Third, our predictions from the suffocation false alarm theory were the opposite of what actually happened: Voluntary HV increased anxiety and feelings of both shortness of breath and suffocation in both PD patients and SP patients. Furthermore, during the recovery period, PD patients reported more dyspneic breathing sensations than controls, although their pCO2 was lower. Hence, both groups of patients were receiving a false alarm during HV, and that alarm was particularly false in PD patients during the later recovery period, although pCO2 is rising. A possible reason for the failure of our initial predictions is that voluntary HV, which evokes sensations that may previously have been associated with anxiety and stress, may through emotional conditioning have produced heightened anticipatory anxiety. Against this possibility is that rather parallel increases in the incidence of panic attacks were reported. The high 54% incidence may be because our definition of panic attacks may be less conservative than that of other investigators (7), although it follows DSM-IV.

Consistent with some previous reports, PD patients were slower to recover from 3 minutes of HV symptomatically and physiologically than SP patients or controls. Because during the baseline and HV period, the two patient groups reported equal anxiety levels, the diagnosis of PD seems to carry with it a specific reaction pattern to this ventilatory challenge. By correlational analyses, the pattern included anxiety, sighing, hypocapnia, and elevated heart rate and skin conductance. Previous failures to observe slower recovery may be explained by shorter HV or recovery periods. For example, in one such study, only 3 minutes of recovery was assessed (67). In others, there was only 90 seconds of HV (31) or only 2 minutes of HV and 2 minutes of recovery (15). In the current study, we failed to observe physiological differences in 1-minute recovery periods.

The slow symptomatic and physiological recovery in PD could have resulted from alterations in neural control at several levels. First, in both humans and subhumans changes in breathing last longer than the stimulus for changed respiratory drive. This is called short-term potentiation or after-discharge (68), and presumably depends on networks located close to basic respiratory control centers. Perhaps after-discharge lasts longer in PD. Second, neural mechanisms controlling recovery from general emotional activation may be responsible for the slow recovery of hypocapnia, which would be consistent with the delayed recovery of skin conductance level and heart rate that we observed after the 3-minute HV period. In addition, skin conductance level had declined less over the six 1-minute fast-breathing periods, similar to many earlier studies of skin conductance habituation in anxiety patients (69). Transfer function RSA, a measure of cardiac vagal control adjusted for the confounding factor respiratory depth, did not show a group difference at any time period, and activation and deactivation were equal between groups. Another confounding factor, respiratory rate, was hardly different between groups at all points, and thus cannot explain lack of RSA group discrimination. (Respiratory rate increase reduces RSA without changes in vagal control of heart rate (70). This also explains the large degree of RSA attenuation in all groups during HV.) From the less skin conductance recovery in PD patients unaccompanied by less RSA recovery, we infer that slowed recovery from emotional activation in PD patients must have been predominantly sympathetically mediated.

Third, alterations at the level of chemoreflex regulation could have played a role. Post-hoc, this would be most in line with the suffocation false-alarm theory, which allows for the possibility of acute adjustments of the CO2 setpoint in PD patients. Group pCO2 levels did not differ significantly before HV, but did afterward, suggesting that repeated voluntary HV producing more downward adjustment in the CO2 sensor in PD patients (Klein, personal communication, 2000). Their setpoint could be particularly susceptible to being lowered by periods of hypocapnia or by anxiety. Normal people react to anxiety or stress with decreased pCO2 (71), and this tendency may be exaggerated in PD patients, because their fear of suffocation is paradoxically stimulated by HV via interoceptive conditioning. Voluntary HV can result in sensations similar to those experienced during CO2 inhalation (9) or breath holding (72). In addition, during recovery from HV, PD patients may be more sensitive to the rising than to the absolute levels of CO2, stimulating their feelings of suffocation and increasing sighing, which serves to avoid this feeling.

Finally, neural mechanisms at the level of cognition could be relevant in two ways. It is possible that our PD patients believed that somatic danger would continue after the events that had provoked it had ceased. Selective attention to remaining physiological activation during the quiet recovery period could have amplified such fears. Furthermore, if PD patients tried to suppress catastrophic cognitions during HV, the frequency of these cognitions and their attendant anxiety may have increased afterward for that reason (for a review, see 73).

In terms of immediate physiological mechanism, the slow recovery of hypocapnia oddly was not accompanied by larger minute volumes. This discrepancy could be because of a decreased production rate of CO2, a shift in metabolic pathways, or a statistical distribution of tidal volumes in which CO2 exchange is enhanced without a significant increase in mean minute volume. Deep breaths lower pCO2 disproportionally (61), an asymmetry that can produce lower pCO2s in individuals who show large tidal volume variability. Of the possible reasons, less CO2 production is unlikely because anxious PD patients should be more, not less, motorically restless or isometrically contracted than the other groups. A shift toward anaerobic glycolysis is not implausible, because after HV, blood lactate and spectroscopically measured brain lactate rise more in PD patients than in controls (74, 75). In support of a different distribution of tidal volumes, we observed elevated tidal volume instability in PD patients during recovery, which was associated with more sigh breaths. Sigh frequency emerged as the best predictor of pCO2 levels during recovery in our combined sample, and was the best predictor of pCO2 during extended quiet sitting in another study with PD patients showing reduced pCO2 levels (61).

The conclusions that can be drawn from this one experiment are limited. Our relatively small number of PD patients prevented us from systematically examining subgroups of panickers and nonpanickers and from trying to correlate symptoms of natural panic attacks with reactions to HV. For the same reason, we could not divide them into subtypes with and without prominent respiratory symptoms (76–78). The respiratory subtype has been reported to be more physiologically reactive to CO2 challenge (79) and to voluntary HV (15). Finally, our design of repeated short fast-breathing periods followed by a longer one leaves us uncertain as to the relative contributions of the shorter periods and the longer one to what happens in the final recovery period. Probably similar recovery delays in PD patients would have been found with 6 to 10 minutes of sustained HV.

Despite these limitations, the present results contribute to the evidence that respiratory measures deviate in specific ways in PD (7, 19, 38, 39), although as usual, we cannot be sure as to what extent respiratory abnormalities are driving the anxiety circuits of the brain or being driven by them. In any case, slow physiological and symptomatic recovery from HV in PD is a finding that theories need to address. Prolonged post-HV hypocapnia and its concomitants may be giving us a glimpse of an impaired inhibitory or an enhanced sensitization mechanism, which intermittently is even more deranged, allowing the disabling explosions of anxiety characteristic of PD.


   ACKNOWLEDGMENTS
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 ACKNOWLEDGMENTS
REFERENCES
 
This work was supported by the Department of Veterans Affairs and National Institutes of Health Grant MH56094. The authors thank Donald F. Klein for his comments and suggestions on an earlier version of this paper.

Received for publication September 27, 1999.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 ACKNOWLEDGMENTS
 REFERENCES
 

  1. American Psychiatric Association. Diagnostic and statistical manual of mental disorders. 4th ed. Washington DC: American Psychiatric Association; 1994.
  2. McNally RJ. Panic disorder: a critical analysis. New York: The Guilford Press; 1994.
  3. Kerr WJ, Dalton JW, Gliebe PA. Some physical phenomena associated with the anxiety states and their relation to hyperventilation. Ann Intern Med 1937; 11: 961–92.
  4. Ley R. Blood, breath, and fears: a hyperventilation theory of panic attacks and agoraphobia. Clin Psychol Rev 1985; 5: 271–85.
  5. Hornsveld H, Garssen B. Hyperventilation syndrome: an elegant but scientifically untenable concept. Neth J Med 1997; 50: 13–20.[Medline]
  6. Lum LC. Hyperventilation syndromes in medicine and psychiatry: a review. J R Soc Med 1987: 80: 229–31.[Medline]
  7. Gorman JM, Fyer MR, Goetz R, Askanazi J, Liebowitz MR, Fyer AJ. Ventilatory physiology of patients with panic disorder. Arch Gen Psychiatry 1988; 45: 31–9.[Abstract]
  8. Gorman JM, Papp LA, Coplan JD, Martinez JM, Lennon S, Goetz RR, Ross D, Klein DF. Anxiogenic effects of CO2 and hyperventilation in patients with panic disorder. Am J Psychiatry 1994; 151: 547–53.[Abstract]
  9. Papp LA, Martinez JM, Klein DF, Coplan JD, Norman RG, Cole R, de Jesus MJ, Ross D, Goetz R, Gorman JM. Respiratory psychophysiology of panic disorder: three respiratory challenges in 98 subjects. Am J Psychiatry 1997; 154: 1557–65.[Abstract/Free Full Text]
  10. Garssen B, Buikhuisen M, van Dyck R. Hyperventilation and panic attacks. Am J Psychiatry 1996; 153: 513–8.[Abstract/Free Full Text]
  11. Hibbert GA, Pilsbury D. Hyperventilation in panic attacks. Ambulant monitoring of transcutaneous carbon dioxide. Br J Psychiatry 1988; 153: 76–80.[Abstract/Free Full Text]
  12. Hibbert G, Pilsbury D. Hyperventilation: is it a cause of panic attacks. Br J Psychiatry 1989; 155: 805–9.[Abstract/Free Full Text]
  13. Holt PD, Andrews G. Hyperventilation and anxiety in panic disorder, social phobia, GAD, and normal controls. Behav Res Ther 1998; 27: 453–60.
  14. Woods SW, Charney DS, Loke J, Goodman WK, Redmond DE, Heninger GR. Carbon dioxide sensitivity in panic anxiety. Ventilatory and anxiogenic response to carbon dioxide in healthy subjects and patients with panic anxiety before and after alprazolam treatment. Arch Gen Psychiatry 1986; 43: 900–9.[Abstract]
  15. Hegel MT, Ferguson RJ. Psychophysiological assessment of respiratory function in panic disorder: evidence for a hyperventilation subtype. Psychosom Med 1997; 59: 224–30.[Abstract/Free Full Text]
  16. Rapee R. Differential response to hyperventilation in panic disorder and generalized anxiety disorder. J Abnorm Psychol 1986; 95: 24–8.[Medline]
  17. Munjack DJ, Brown RA, McDowell DE. Existence of hyperventilation in panic disorder with and without agoraphobia, GAD, and normals: implications for a cognitive theory of panic. J Anxiety Disord 1993; 7: 37–48.
  18. van den Hout MA, Hoekstra R, Arntz A, Christiaanse M, Ranschaert W, Schouten E. Hyperventilation is not diagnostically specific to panic patients. Psychosom Med 1992; 53: 182–91.
  19. Maddock RJ, Carter CS. Hyperventilation-induced panic attacks in panic disorder with agoraphobia. Biol Psychiatry 1991; 29: 843–54.[Medline]
  20. Clark DM. A cognitive approach to panic. Behav Res Ther 1986; 24: 461–70.[Medline]
  21. Margraf J, Ehlers A, Roth WT. Biological models of panic disorder and agoraphobia: A review. Behav Res Ther 1986; 24: 553–67.[Medline]
  22. Taylor S, Koch WJ, McNally RJ. How does anxiety sensitivity vary across the anxiety disorders? J Anxiety Disord 1992; 6: 249–59.
  23. Klein DF. False suffocation alarms, spontaneous panics, and related conditions. An integrative hypothesis. Arch Gen Psychiatry 1992; 50: 306–17.[Abstract]
  24. Klein DF. Testing the suffocation false alarm theory of panic disorder. Anxiety 1994; 1: 1–7.[Medline]
  25. Papp LA. Respiratory neurobiology of panic. In: Asnis GM, Van Praag HM, editors. Panic disorder: clinical, biological and treatment aspects. New York: John Wiley & Sons; 1995. p. 255–75.
  26. Gorman JM, Liebowitz MR, Fyer AJ, Stein J. A neuroanatomical hypothesis for panic disorder. Am J Psychiatry 1989; 146: 148–51.[Abstract/Free Full Text]
  27. Hardonk HJ, Beumer HM. Hyperventilation syndrome. In: Vinken PJ, Bruyn GW, editors. Handbook of clinical neurology, Vol. 38. Amsterdam: North-Holland Publishing Company; 1979. p. 309–60.
  28. Brodtkorb E, Gimse R, Antonaci F, Ellertsen B, Sand T, Sulg I, Sjaastad O. Hyperventilation syndrome: clinical, ventilatory, and personality characteristics as observed in neurological practice. Acta Neurol Scand 1990; 81: 307–13.[Medline]
  29. Stoop A, de Boo T, Lemmens WHF. Hyperventilation syndrome: measurement of objective symptoms and subjective complaints. Respiration 1986; 49: 37–44.[Medline]
  30. Han JN, Stegen K, Simkens K, Cauberghs M, Schepers R, Van den Bergh O, Clement J, Van de Woestijne KP. Unsteadiness of breathing in patients with hyperventilation syndrome and anxiety disorders. Eur Respir J 1997; 10: 167–76.[Abstract]
  31. Rapee RM, Brown TA, Antony MM, Barlow DH. Response to hyperventilation and inhalation of 5.5% carbon dioxide-enriched air across the DSM-III-R anxiety disorders. J Abnorm Psychol 1992; 101: 538–52.[Medline]
  32. Roth WT, Telch MJ, Taylor CB, Sachitano JA, Gallen CC, Kopell M, McClenahan KL, Agras WS, Pfefferbaum A. Autonomic characteristics of agoraphobia with panic attacks. Biol Psychiatry 1986; 21: 1133–54.[Medline]
  33. Lader MH, Wing L. Physiological measures, sedative drugs, and morbid anxiety. London: Oxford University Press; 1966.
  34. Wilhelm FH, Roth WT. Taking the laboratory to the skies: ambulatory assessment of self-report, autonomic, and respiratory responses in flying phobia. Psychophysiology 1998; 35: 596–606.[Medline]
  35. Watkins LL, Grossman P, Krishnan R, Sherwood A. Anxiety and vagal control of heart rate. Psychosom Med 1998; 60: 498–502.[Abstract/Free Full Text]
  36. Abelson JL, Weg JG, Nesse RM, Curtis GC. Persistent respiratory instability in patients with panic disorder. Biol Psychiatry 2001; 49: 588–95.[Medline]
  37. Martinez JM, Papp LA, Coplan JD, Anderson DE, Mueller CM, Klein DF, Gorman JM. Ambulatory monitoring of respiration in anxiety. Anxiety 1996; 2: 296–302.[Medline]
  38. Papp LA, Welkowitz A, Martinez JM, Klein DF, Browne S, Gorman JM. Instructional set does not alter outcome of respiratory challenges in panic disorder. Biol Psychiatry 1995; 38: 826–30.[Medline]
  39. Stein MB, Millar TW, Larsen DK, Kryger MH. Irregular breathing patterns during sleep in patients with panic disorder. Am J Psychiatry 1995; 152: 1168–73.[Abstract/Free Full Text]
  40. DiNardo PA, Brown TA, Barlow DH. Anxiety disorders interview schedule for DSM-VI: lifetime version (ADIS-IV-L). Albany, NY: Graywind Publications Inc; 1994.
  41. Pelosi P, Croci M, Ravagnan I, Tredici S, Pedoto A, Lissoni A, Gattinoni L. The effects of body mass on lung volumes, respiratory mechanics, and gas exchange during general anesthesia. Anesth Analg 1998; 87: 654–60.[Abstract/Free Full Text]
  42. Beck AT, Ward CH, Mendelson M, Mock J, Erbaugh J. An inventory for measuring depression. Arch Gen Psychiatry 1961; 41: 561–71.
  43. Spielberger CD, Gorsuch RL, Luchene RE. State-trait anxiety inventory. Palo Alto: Consulting Psychology Press; 1970.
  44. Reiss S, Peterson RA, Gursky DM, McNally RJ. Anxiety sensitivity, anxiety frequency, and the prediction of fearfulness. Behav Res Ther 1986; 24: 1–8.[Medline]
  45. Chambless DL, Caputo GC, Bright P, Gallaher R. Assessment of fear of fear in agoraphobics: the body sensations questionnaire and the agoraphobic cognitions questionnaire. J Consult Clin Psychol 1984; 52: 1090–7.[Medline]
  46. Chambless DL, Caputo GC, Hasin SE, Gracely EJ, Williams C. The mobility inventory for agoraphobia. Behav Res Ther 1985; 23: 35–44.[Medline]
  47. Turner SM, McCanna M, Beidel DC. Discriminative validity of the social avoidance and distress and fear of negative evaluation scale. Behav Res Ther 1987; 25: 113–5.[Medline]
  48. Morel DR, Forster A, Suter PM. Noninvasive ventilatory monitoring with bellows pneumographs in supine subjects. J Appl Physiol 1983; 55: 598–606.[Abstract/Free Full Text]
  49. Askanazi J, Silverberg PA, Foster RJ, Hyman AI, Milic-Emili J, Kinney JM. Effects of respiratory apparatus on breathing pattern. J Appl Physiol 1980; 48: 577–80.[Abstract/Free Full Text]
  50. Tobin MJ, Guenther SM, Perez W, Mador MJ. Accuracy of the respiratory inductive plethysmograph during loaded breathing. J Appl Physiol 1987; 62: 497–505.[Abstract/Free Full Text]
  51. Bloomfield P. Complex demodulation. In Bloomfield P, editor. Fourier analysis of time series: an introduction. New York: Wiley; 1976. p. 118–50.
  52. Saul JP, Berger RD, Albrecht P, Stein SP, Chen MH, Cohen RJ. Transfer function analysis of the circulation: unique insights into cardiovascular regulation. Am J Physiol 1991; 261: H1231–45.[Abstract/Free Full Text]
  53. Wilhelm FH, Grossman P, Roth WT. Analysis of cardiovascular regulation. Biomed Sci Instrum 1999; 35: 135–40.[Medline]
  54. Welch PD. The use of fast Fourier transform for the estimation of power spectra: a method based on time averaging over short modified periodograms. IEEE Trans Audio Electroacoustics 1967; 15: 70–3.
  55. Brown T, Beightol L, Koh J, Eckberg D. Important influence of respiration on human RR interval power spectra is largely ignored. J Appl Physiol 1993; 75: 2310–7.[Abstract/Free Full Text]
  56. Bass C, Gardner WN. Respiratory and psychiatric abnormalities in chronic symptomatic hyperventilation. BMJ 1985; 290: 1387–90.
  57. Simon PM, Schwartzstein RM, Weiss JW, Fencl V, Teghtsoonian M, Weinberger SE. Distinguishable types of dyspnea in patients with shortness of breath. Am Rev Respir Dis 1990; 142: 1009–14.[Medline]
  58. Elliott MW, Adams L, Cockcroft A, MacRae KD, Murphy K, Guz A. The language of breathlessness. Use of verbal descriptors by patients with cardiopulmonary disease. Am Rev Respir Dis 1991; 144: 826–32.[Medline]
  59. Mahler DA, Harver A, Lentine T, Scott JA, Beck K, Schwartzstein RM. Descriptors of breathlessness in cardiorespiratory diseases. Am J Respir Crit Care Med 1996; 154: 1357–63.[Abstract]
  60. Stemmler G. Implicit measurement models in methods for scoring physiological reactivity. J Psychophysiol 1987; 1: 113–25.
  61. Wilhelm FH, Trabert W, Roth WT. Characteristics of sighing in panic disorder. Biol Psychiatry 2001; 49: 606–14.[Medline]
  62. Wilhelm FH, Trabert W, Roth WT. Physiological instability in panic disorder and generalized anxiety disorder. Biol Psychiatry 2001; 49: 596–605.[Medline]
  63. Asmundson GJG, Stein MB. Selective processing of social threat in patients with generalized social phobia: evaluation using a dot-probe paradigm. J Anxiety Disord 1994; 8: 107–17.
  64. Ball SG, Otto MW, Pollack MH, Uccello R. Differentiating social phobia and panic disorder: a test of core beliefs. Cog Therapy Res 1995; 19: 473–81.
  65. Brown TA, 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: 408–18.[Medline]
  66. Ehlers A, Margraf J. Fear of fear: a new concept in the assessment of anxiety disorders. Verhaltenstherapie 1993; 3: 14–24.
  67. Hornsveld HK, Garssen B Fiedeldij Dop MJC, van Spiegel P, de Haes JCJM. Double-blind placebo-controlled study of the hyperventilation provocation test and the validity of the hyperventilation syndrome. Lancet 1996; 348: 152–8.
  68. Folgering H. The hyperventilation syndrome. In: Altose MD, Kawakami Y, editors. Control of breathing in health and disease. New York: Marcel Dekker; 1999. p. 633–60.
  69. Roth WT, Ehlers A, Taylor CB, Margraf J, Agras WS. Skin conductance habituation in panic disorder patients. Biol Psychiatry 1990; 27: 1231–43.[Medline]
  70. Wilhelm FH, Berkowitz J, Hansen M, Grossman P, Roth WT, Gross JJ. RSA estimation during spontaneous breathing using a paced breathing calibration. Psychophysiology 1998; 35: S88.
  71. Schleifer L, Ley R. End-tidal pCO2 as an index of psychophysiological activity during VDT data-entry work and relaxation. Ergonomics 1994; 245–54.
  72. Roth WT, Wilhelm FH, Trabert W. Voluntary breath holding in panic and generalized anxiety disorders. Psychosom Med 1998; 60: 671–9.[Abstract/Free Full Text]
  73. Purdon C. Thought suppression and psychopathology. Behav Res Ther 1999; 37: 1029–54.[Medline]
  74. Maddock RJ, Mateo-Bermudez J. Elevated serum lactate following hyperventilation during glucose infusion in panic disorder. Biol Psychiatry 1990; 27: 411–8.[Medline]
  75. Dager SR, Strauss WL, Marro KI, Richards TL, Metzger GD, Artru AA. Proton magnetic responance spectroscopy investigation of hyperventilation in subjects with panic disorder and comparison subjects. Am J Psychiatry 1995; 152: 666–72.[Abstract/Free Full Text]
  76. Briggs AC, Stretch DD, Brandon S. Subtyping of panic disorder by symptom profile. Br J Psychiatry 1993; 163: 201–9.[Abstract/Free Full Text]
  77. Aronson TA, Logue CM. Phenomenology of panic attacks: a descriptive study of panic disorder patients’ self-reports. J Clin Psychiatry 1988; 49: 8–13.
  78. Shioiri T, Someya T, Murashita J, Takahashi S. The symptom structure of panic disorder: a trial using factor and cluster analysis. Acta Psychiatr Scand 1996; 93: 80–6.[Medline]
  79. Biber B, Alkin T. Panic disorder subtypes: differential responses to CO2 challenge. Am J Psychiatry 1999; 156: 739–44.[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 Psychosom. Med.Home page
J. Blechert, T. Michael, P. Grossman, M. Lajtman, and F. H. Wilhelm
Autonomic and Respiratory Characteristics of Posttraumatic Stress Disorder and Panic Disorder
Psychosom Med, November 1, 2007; 69(9): 935 - 943.
[Abstract] [Full Text] [PDF]

Home page
 Psychosom. Med.Home page
T. Ritz and B. Dahme
Implementation and Interpretation of Respiratory Sinus Arrhythmia Measures in Psychosomatic Medicine: Practice Against Better Evidence?
Psychosom Med, July 1, 2006; 68(4): 617 - 627.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. PsychiatryHome page
S. D. Friedman, C. M. Mathis, C. Hayes, P. Renshaw, and S. R. Dager
Brain pH Response to Hyperventilation in Panic Disorder: Preliminary Evidence for Altered Acid-Base Regulation
Am J Psychiatry, April 1, 2006; 163(4): 710 - 715.
[Abstract] [Full Text] [PDF]

Home page
 Psychosom. Med.Home page
T. Ritz, F. H. Wilhelm, A. L. Gerlach, A. Kullowatz, and W. T. Roth
End-Tidal pCO2 in Blood Phobics During Viewing of Emotion- and Disease-Related Films
Psychosom Med, July 1, 2005; 67(4): 661 - 668.
[Abstract] [Full Text] [PDF]

Home page
 Chronic Respiratory DiseaseHome page
P M Meek
Measurement of dyspnea in chronic obstructive pulmonary disease: what is the tool telling you?
Chronic Respiratory Disease, January 1, 2004; 1(1): 29 - 37.
[Abstract] [PDF]

Home page
 Behav ModifHome page
F. H. Wilhelm, W. T. Roth, and M. A. Sackner
The LifeShirt: An Advanced System for Ambulatory Measurement of Respiratory and Cardiac Function
Behav Modif, October 1, 2003; 27(5): 671 - 691.
[Abstract] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager