- Email: [email protected]
Doxapram-induced panic attacks and cortisol elevation
Psychiatry Research 133 (2005) 253 – 261 www.elsevier.com/locate/psychres
Doxapram-induced panic attacks and cortisol elevation David A. Gutmana,*, Jeremy Coplanb, Laszlo Pappc, Jose Martinezd, Jack Gormand,e a
Department of Psychiatry, Columbia University School of Medicine, 1051 Riverside Drive, New York, NY 10032, USA b Department of Psychiatry, State University of New York, Downstate, Brooklyn, NY, USA c New York State Psychiatric Institute, New York, NY, USA d Department of Psychiatry, Mount Sinai School of Medicine, New York, NY, USA e Department of Neuroscience, Mount Sinai School of Medicine, New York, NY, USA Received 29 August 2004; accepted 26 October 2004
Abstract Numerous agents with differing biological properties and central nervous system (CNS) effects can induce panic attacks in predisposed individuals. A potential explanation of this finding is that panic disorder patients are more likely to panic than normal control subjects when given a panicogen due to an excessive fear response to somatic arousal. We test this hypothesis by using doxapram, a panicogen with minimal CNS effects, to induce panic in patients and control subjects. Doxapram was given to six subjects with panic disorder with or without agoraphobia and four healthy volunteers. Measures comprised the Acute Panic Inventory, the Borg Exertion scale, the 10-point Anxiety Scale, the 10-point Apprehension Scale, cortisol, prolactin, and MHPG, all obtained at baseline and multiple time points after the doxapram infusion. All panic disorder patients panicked with doxapram, whereas no control subjects had a panic attack. Panic patients had similar levels of breathlessness with doxapram compared with control subjects. Although panic patients had higher levels of anxiety and apprehension, these did not change significantly with doxapram compared with control levels. Doxapram led to similar increases in cortisol and prolactin in both groups, and MHPG was consistently elevated in panic patients, but unaffected by doxapram. These results show that doxapram is a useful panicogen in the study of panic disorder. Since the panic patients and control subjects had similar levels of physiological and psychological arousal, but the panic patients were more likely to have a panic attack, this lends support to the concept of a sensitized fear network in panic disorder patients. D 2004 Elsevier Ireland Ltd. All rights reserved. Keywords: Doxapram; Panic attacks; Ventilatory stimulant; Cortisol
1. Introduction
* Corresponding author. Tel.: +1 212 342 1101; fax: +1 419 730 3208. E-mail address: [email protected] (D.A. Gutman).
A number of agents with varying biological properties and central nervous system (CNS) effects have been shown to produce panic attacks in predisposed individuals. These bpanicogensQ include
0165-1781/$ - see front matter D 2004 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.psychres.2004.10.010
254
D.A. Gutman et al. / Psychiatry Research 133 (2005) 253–261
compounds that produce osmotic stress (sodium lactate and hyperosmotic sodium chloride; stimulate neurotransmitter systems (yohimbine and m-CPP); stimulate brain peptide systems (cholecystokinin); and stimulate breathing (carbon dioxide and doxapram). Although many theories have been suggested to explain the mechanism of action of these panicogens, what they have in common is the capacity to produce significant somatic distress. We have proposed the hypothesis that panic disorder patients are more likely to panic when given a panicogen because they develop a greater fear response when somatic systems are aroused (Gorman et al., 2000). This is consistent with the catastrophic misinterpretation hypothesis of panic disorder that has been articulated by cognitive theorists (Austin and Richards, 2001). Animal models of conditioned fear have demonstrated a coordinated brain circuitry, with the amygdala at its core, that can be linked to the excessive fear in panic disorder patients. A test of this hypothesis is to administer an agent that produces somatic symptoms but does not have an appreciable CNS action at panicogenic doses. Doxapram is one such substance. The Ann Arbor, Michigan group has done interesting and significant work in exploring the properties of doxapram as a panicogen. Doxapram does not directly affect vascular flow, appears not to activate centrally mediated neurohormonal systems and minimally stimulates a noradrenergic response (Folgering et al., 1981; Calverley et al., 1983; Winnie, 1973; Burki, 1984; Abelson et al., 1995, 1996a). It induces panic attacks quickly in 75–80% of subjects and only 12.5–20% of controls (Lee et al., 1993; Abelson et al., 1996a,b). We hypothesize that doxapram will produce equivalent physiological changes in patients with panic disorder and control subjects, but that the former will experience more fear and therefore be more prone to panic. In this study, we will evaluate this hypothesis as well as validate the effectiveness of doxapram as an efficient precipitant of panic attacks in panic patients by measuring the anxiety and physiological responses to doxapram. We propose doxapram as an effective agent for precipitating panic attacks for imaging purposes, based on work by Abelson and Lee (Lee et al., 1993; Abelson et al., 1996a,b). As mentioned previously, doxapram does not directly affect vascular
flow, appears not to activate centrally mediated neurohormonal systems, minimally stimulates a noradrenergic response, is sensitive and specific for inducing panic in panic disorder subjects versus control subjects, and has a rapid and predicable onset (Lee et al., 1993; Abelson et al., 1996a,b). In this study, we examine whether doxapram will produce increases in systems linked to the fear response during panic attacks by measuring MHPG (noradrenergic systems), cortisol, and prolactin (neuroendocrine systems). Furthermore, we will validate the effectiveness of doxapram as a precipitant of panic attacks in patients with panic disorder and not in control subjects by replicating the work of Abelson and Lee (Lee et al., 1993; Abelson et al., 1996a,b).
2. Methods 2.1. Subjects Six subjects with panic disorder with or without agoraphobia (DSM-III-R; 1987) and four healthy volunteers participated in the study. The normal controls consisted of two males and two females, with a mean age of 29.75F10.94. The patients consisted of five males and one female, with a mean age of 37.33F7.74. The age was not significantly different (t= 1.30; df=8; Pb0.23). Subjects signed informed consent after a thorough description of the study. Psychiatric diagnoses were made by a psychiatric interview by a psychiatrist, and the Structured Clinical Interview for DSM-III-R (Spitzer et al., 1990) was performed by a trained rater. A medical history, physical examination, EKG, blood and urine tests including a thyroid function test, pregnancy test, and toxicology screen were performed. All subjects were required to be medically healthy, without a history of hypertension, cerebrovascular or cardiac disease, nonsubstance abusing, and medication-free for 2 weeks prior to the study. Any medication known to affect neuroendocrine function (e.g., phenytoin, phenobarbital, oral contraceptives, thyroid replacement, or other hormones) had to be discontinued at least 3 months prior to the study. A current or past diagnosis of an affective disorder, eating disorder, suicidal or homicidal ideation, schizophrenia, or other psychotic
D.A. Gutman et al. / Psychiatry Research 133 (2005) 253–261
illness excluded subjects from the study. Control subjects did not meet criteria for Axis-I or Axis-II disorders (DSM-III-R; American Psychiatric Association, 1987). Female subjects could not be pregnant or lactating and had a negative urine pregnancy test the day of the study. All women in the study were within the first 10 days of their menstrual cycle. 2.2. Experimental procedures Subjects were requested to abstain from food, alcohol, tobacco, and caffeine for 8 h before biological tests and arrived at the Biological Studies Unit at the New York State Psychiatric Institute after overnight fasting at 8 AM on 2 separate days separated by at least 1 day but within 4 days. Patients were told that on 1 study day they would receive an infusion of salt water and on another day they would receive an infusion of doxapram, which is routinely used in anesthesiology practice to stimulate ventilation. An intravenous angiocatheter line was started for the periodic withdrawal of blood samples for cortisol, prolactin, and MHPG levels (Papp et al., 1993; Coplan et al., 1995, 1998a,b; Mathew et al., 2001). An automated sphygmomanometer and EKG leads were used for continuous cardiovascular monitoring. Respiratory measures were acquired using the Respitrace system and a ventilatory canopy (a clear headchamber for the monitoring of gas exchange as the subject breathes room air), but will not be discussed in this article. At 40 min prior to doxapram administration, collection of vital signs and ratings commenced, and blood sample #1 was drawn (Table 1). Rating instruments were administered by a master’s or doctoral level clinician who was blind to both diagnosis and nature of substance administered at each time point of the four blood draws. These scales included the Acute Panic Inventory (API), a 17-item questionnaire specific to panic attacks that grades symptoms from none to very severe and is a reliable measure of laboratory-induced panic (Dillon et al., 1987); the Borg Exertion Dyspnea Scale, a measure of exertion on a 10-point Likert scale (Borg, 1982); the 10-point Apprehension Scale and the 10-point Anxiety Scale, which measure apprehension and anxiety, respectively, on a 10-point Likert scale. These measures
255
Table 1 Flow sheet for double-blind saline-controlled doxapram infusions counterbalanced for day of infusion Time
Experimental phase
Blood sample #a
40 min 10 min 0 min
BSU arrival Preinfusion baseline Double-blind doxapram infusion, saline-controlled Immediate postinfusion Delayed postinfusion
#1 #2
+90 s +20 min
#3 #4
a Acute Panic Inventory (API), 10-point Anxiety and Apprehension Scale and Borg Exertional Dyspnea Scale were collected at each blood draw time point. Doxapram/saline infusions were performed while subjects were in a sealed canopy for ventilatory monitoring. Doxapram was administered at a dose of 0.5 mg/kg in 10 ml of saline over 15 s and saline over the identical time period. BSU—Biological Studies Unit, New York State Psychiatric Institute.
were also performed 10 min prior to the intravenous infusion of doxapram or saline, at the time of infusion, at 90 s afterwards to correspond with panic attacks, and at 20 min postinfusion. Subjects received either a doxapram (0.5 mg/kg in 10 ml saline over 15 s; Lee et al., 1993) or saline intravenous infusion in a randomized order. At the end of the protocol, the subject was debriefed and discharged from the unit. 2.3. Data analysis Repeated measures analysis of variance (RMANOVA) was performed using group as the independent variable (patient versus control) and two within factors: condition (doxapram day versus saline day) and within-infusion time effects (in minutes: 40, 10, +1.5, +20). Significant RM-ANOVA were followed by planned comparisons using the Least Squares Difference post hoc testing method. This statistical approach was used for psychological scales and blood parameters.
3. Results 3.1. Panic rates Six of six panic disorder patients were rated by the blinded raters to have panicked to the doxapram infusions. Only five had the saline control, none of
256
D.A. Gutman et al. / Psychiatry Research 133 (2005) 253–261
whom panicked to saline. One subject did not complete the study after receiving doxapram for the first infusion. None of four control subjects experienced panic attacks during either the doxapram or saline infusions. This distribution of proportions is significantly different (v 2=8.43; df=1, P=0.008), and despite the small number of subjects in the current study, implies a panicogenic effect for doxapram. 3.2. Psychological scales 3.2.1. Acute Panic Inventory (see Fig. 1) There is a significant time effect [ F(1,21)=12.30; P=0.00007] and condition effect [ F(1,7)=9.96; P =0.016], but no significant group effect [ F(1,7)=3.58; P=0.1] (Fig. 1). A significant group by time effect [ F(3,21)=5.75; P=0.005] and time by condition effect [ F(3,21)=10.58; P=0.0002] were noted, but no significant group by condition effect [ F(1,7)=4.93; P=0.062]. A significant group by time by condition effect was noted [ F(3,21)=5.75; P=0.005]. Doxapram produced robust increases of the API scores within 90 s following inception of the doxapram infusion, with API scores at that time point clearly higher than all other time points that were measured. Thus, panic disorder patients were more
sensitive than control subjects to the panicogenic effects of doxapram as measured by the API. 3.2.2. Borg Exertional Dyspnea Scale There is a significant time effect [ F(3,21)=4.15; P=0.018] and a time by condition effect [ F(3,21)=4.15; P=0.018], but the analysis fails to reveal a corresponding group by time by condition effect [ F(3,21)=0.88; Pb0.46]. This suggests that the increases in Borg scores following doxapram were not distinguishable between patients and controls, and that both groups experienced similar levels of self-reported dyspnea. 3.2.3. Ten-point Anxiety Scale A group effect was noted for anxiety [ F(1,7)=6.65; P=0.036] with patients exhibiting higher overall levels of anxiety. A significant time by condition effect [ F(3,21)=4.64; P=0.01] reflects doxapram’s overall effect of increasing anxiety scores, but there was no group by time by condition effect [ F(3,21)=1.02; Pb0.4]. 3.2.4. Ten-point Apprehension Scale An overall group effect was noted for apprehension scores, with patients scoring higher than controls [ F(1,7)=4.7; P=0.017). Doxapram increased appre-
Fig. 1. For the Acute Panic Inventory, a group by time by condition was noted. Using least square differences (LSD) for planned comparisons, posttesting revealed that the +1.5 min postdoxapram infusion Acute Panic Inventory scores were significantly increased ( P b0.001; see symbol @ in figure) in comparison to all other time points. In addition, a significant time effect, condition effect, group by time effect, and time by condition effect was evident, but these effects were all best explained by the group by time by condition effect (statistics available in text).
D.A. Gutman et al. / Psychiatry Research 133 (2005) 253–261
hension in comparison to saline [ F(3,21)=5.38; Pb0.007], and a trend for a group by time by condition was noted [ F(3,21)=2.8; P=0.065), suggesting, in parallel to the API scores, an enhanced effect of doxapram on apprehension in patients. 3.3. Physiological results 3.3.1. Cortisol results (see Fig. 2) There was a significant time effect [ F(3,21)=8.86; P=0.0005] and a significant condition by time effect [ F(3,21)=8.47; P=0.0007] but an absence of a group by time by condition effect [ F(3,21)=1.19; P=0.33). Planned comparisons using post hoc testing using Least Squares Difference (LSD) indicated that baseline ( 40 min) cortisol in all groups prior to saline infusions was elevated relative to levels at +1.5 and +20 min (see Fig. 2), indicating a progressive decrease of cortisol levels in both groups over time during the saline day. On the doxapram infusion day, both the baseline ( 40 min) and delayed postdoxapram (+20 min) cortisol levels were elevated in comparison to the 10 and +1.5 min time points. In addition, only the
257
delayed (+20 min) postdoxapram cortisol level was strongly significant in comparison to the corresponding time point of the saline infusions ( P=0.0006). 3.3.2. Prolactin results (see Fig. 3) There was a significant condition by time effect [ F(3,18)=5.22; P=0.009) but no other significant statistical effects. Post hoc testing using specified comparisons and least squares differences (see Fig. 3) indicated that all subjects exhibited significant prolactin elevations +20 min after doxapram in comparison to all other time points ( Pb0.05; see Fig. 3). At +20 min, prolactin levels were significantly elevated with doxapram compared with saline (LSD difference is P=0.0002 level). 3.3.3. MHPG results (see Fig. 4) Since there were multiple missing data points for MHPG during the saline infusions, only the doxapram infusion component of the study was analyzed. An overall group by time interaction was evident for the doxapram component (see Fig. 4 for statistics). Using Least Squares Difference for planned comparisons,
Fig. 2. For plasma cortisol, there was a significant time effect and a significant condition by time effect (see figure for statistics) but an absence of a group by time by condition effect [ F(3,21)=1.19; P=0.33). Planned comparisons using post hoc testing using Least Squares Difference indicated that baseline ( 40 min) cortisol in both groups prior to saline infusions was elevated relative to the levels at +1.5 and +20 min (see symbol * left panel). Post hoc testing indicated that for the doxapram infusion day (right panel), both the baseline ( 40 min) and delayed postdoxapram (+20 min) levels were elevated in comparison to the 10 and +1.5 min time points (indicated by symbol #). In addition, the delayed (+20 min) postdoxapram saline cortisol level was differed strongly from the level at the corresponding point of the saline infusions ( P=0.0006).
258
D.A. Gutman et al. / Psychiatry Research 133 (2005) 253–261
Fig. 3. For the plasma prolactin analysis, there was a significant condition by time effect [ F(3,18)=5.22; P=0.009)] but no other significant statistical effects. Post hoc testing using planned comparisons and Least Squares Differences (LSD) indicated that all subjects exhibited prolactin elevations +20 min after doxapram in comparison to all other time points ( Pb0.05; symbol ##); for the corresponding time point of the saline infusion, the LSD difference is at the P=0.0002 significance level.
patients with panic disorder showed an overall increase in plasma MHPG ( Pb0.001), but this effect was most marked at baseline ( Pb0.000001), when patients demonstrated relatively higher MHPG levels
in comparison to controls. Although differences narrowed between the groups, matched time points continued to indicate significant elevations during the infusion in the patients versus controls (see Fig. 4,
Fig. 4. An overall group by time interaction was evident for the doxapram component (see statistics). Using least square differences for planned comparisons, patients with panic disorder showed an overall increase in plasma MHPG (see symbol *; Pb0.001), but this effect was most marked at baseline (see symbol **; Pb0.000001), where patients demonstrated relatively higher MHPG levels in comparison to controls, who demonstrated relatively lower baseline MHPG levels.
D.A. Gutman et al. / Psychiatry Research 133 (2005) 253–261
legend). There was no discernible effect of doxapram on MHPG in either the patients or controls.
4. Discussion We demonstrated that doxapram is an effective panicogen with a 100% panic rate in panic disorder patients and a 0% rate in normal controls, thereby replicating the findings of Abelson et al. (1996a,b) and Lee et al. (1993). Our sample is small and requires further validation, although additional unpublished data (Kent et al., in press) from our group demonstrate similar rates in another group of subjects, with 6/6 (100%) subjects with panic disorder but only 1/6 (16.67)% controls having a panic attack after IV doxapram. Some considerations are necessary when reviewing these data. Although we have subject reports of breathlessness, we do not have respiratory data, which would clarify whether all subjects had similar physiological responses. Furthermore, our data demonstrate that doxapram induces a cortisol surge that is not related to diagnosis or the occurrence of panic attacks. This is in contrast to the Abelson et al. (1996a) finding of no change in cortisol level with doxapram infusion. MHPG was elevated in panic patients compared with control subjects, which has been previously demonstrated (Coplan et al., 1997), but did not increase with doxapram, as did the cortisol and prolactin levels. Our subject sample is small, and further studies with larger samples may clarify these findings as well as provide necessary replication. Other limitations to this study include the following: (1) we only employed one dose of doxapram and it is possible that less difference between patients and controls would be observed with higher doses. Doseresponse studies would be necessary to clarify this issue. (2) We did not include groups with diagnoses other than panic disorder, and therefore the diagnostic specificity of doxapram-induced panic cannot be inferred. (3) We did not obtain an assessment of the similarity of doxapram-induced and naturally occurring panic attacks. These shortcomings could be easily addressed in future studies. Although some have previously divided panicogens into respiratory and HPA-activating agents (Coplan and Klein, 1996), in our results, doxapram
259
appears to be like cholecystokinin, with both actions represented (Koszycki et al., 1998; Bradwejn and Koszycki, 2001). In fact, we found that panic patients experienced similar levels of physiological arousal— increases in cortisol, prolactin, and dyspnea—as control subjects. The panic patients had panic attacks and a greater increase on the API, whereas the control subjects did not panic. In other words, it appears that panic patients may be more fearful of the stimulation aroused by doxapram. This finding provides support for a model of panic disorder proposed by Gorman et al. (2000) based on a bfear networkQ in the brain that is centered in the amygdala and involves its interaction with the hippocampus and the medial prefrontal cortex. This bfear networkQ is derived from animal work on classical conditioning (LeDoux et al., 1988; Davis, 1992; LeDoux, 1996). The model notes that there are many stimulants with various differing mechanisms, but all can induce panic attacks in patients with panic disorder (Charney et al., 1984; Liebowitz et al., 1985; Pyke and Greenberg, 1986; Gorman et al., 1988; Targum and Marshall, 1989; Klein et al., 1991; Jensen et al., 1991; Bradwejn and Koszycki, 1994; Veltman et al., 1996). What they may have in common is physiological arousal of the subject, which in vulnerable patients leads to misinterpretation of the sensations as a fear cue. The common final pathway is stimulation of the amygdala, which leads to a panic attack. Indeed, doxapram causes an increase in c-fos expression in the central nucleus of the amygdala in rat models of contextual fear conditioning (Sullivan et al., 2003). Future directions include replication of these findings with a larger sample size and brainimaging studies to examine activity in areas predicted by the fear model of panic disorder. Acknowledgments The research reported was supported in part by NIMH Grants P50MH58911 and RO1MH60951. References Abelson, J.L., Curtis, G.C., Nesse, R., Fantone, R., Pyke, R.E., Bammert-Adams, J., 1995. The effects of central cholecystokinin receptor blockade on hypothalamic–pituitary–adrenal and
260
D.A. Gutman et al. / Psychiatry Research 133 (2005) 253–261
symptomatic responses to overnight withdrawal from alprazolam. Biological Psychiatry 37, 56 – 59. Abelson, J.L., Nesse, R.M., Weg, J.G., Curtis, G.C., 1996a. Respiratory psychophysiology and anxiety: cognitive intervention in the doxapram model of panic. Psychosomatic Medicine 58, 302 – 313. Abelson, J.L., Weg, J.G., Nesse, R.M., Curtis, G.C., 1996b. Neuroendocrine responses to laboratory panic: cognitive intervention in the doxapram model. Psychoneuroendocrinology 21, 375 – 390. American Psychiatric Association, 1987. The Diagnostic and Statistical Manual of Mental Disorders (DSM-III R). Author, Washington, DC. Austin, D.W., Richards, J.C., 2001. The catastrophic misinterpretation model of panic disorder. Behaviour Research and Therapy 39, 1277 – 1291. Borg, G.A., 1982. Psychophysical bases of perceived exertion. Medicine and Science in Sports and Exercise 14, 377 – 381. Bradwejn, J., Koszycki, D., 1994. The cholecystokinin hypothesis of anxiety and panic disorder. Annals of the New York Academy of Sciences 713, 273 – 282. Bradwejn, J., Koszycki, D., 2001. Cholecystokinin and panic disorder: past and future clinical research strategies. Scandinavian Journal of Clinical & Laboratory Investigation Supplement 234, 19 – 27. Burki, N.K., 1984. Ventilatory effects of doxapram in conscious human subjects. Chest 85, 600 – 604. Calverley, P.M., Robson, R.H., Wraith, P.K., Prescott, L.F., Flenley, D.C., 1983. The ventilatory effects of doxapram in normal man. Clinical Science 65, 65 – 69. Charney, D.S., Heninger, G.R., Breier, A., 1984. Noradrenergic function in panic anxiety. Effects of yohimbine in healthy subjects and patients with agoraphobia and panic disorder. Archives of General Psychiatry 41, 751 – 763. Coplan, J., Klein, D., 1996. Pharmacological probes in panic disorder. In: Westenberg, H., Boer, J., Murphy, D. (Eds.), Advances in the Neurobiology of Anxiety Disorders. Wiley, Chichester, pp. 173 – 196. Coplan, J.D., Pine, D., Papp, L., Martinez, J., Cooper, T., Rosenblum, L.A., Gorman, J.M., 1995. Uncoupling of the noradrenergic–hypothalamic–pituitary–adrenal axis in panic disorder patients. Neuropsychopharmacology 13, 65 – 73. Coplan, J.D., Papp, L.A., Pine, D., Martinez, J., Cooper, T., Rosenblum, L.A., Klein, D.F., Gorman, J.M., 1997. Clinical improvement with fluoxetine therapy and noradrenergic function in patients with panic disorder. Archives of General Psychiatry 54, 643 – 648. Coplan, J.D., Trost, R.C., Owens, M.J., Cooper, T.B., Gorman, J.M., Nemeroff, C.B., Rosenblum, L.A., 1998a. Cerebrospinal fluid concentrations of somatostatin and biogenic amines in grown primates reared by mothers exposed to manipulated foraging conditions. Archives of General Psychiatry 55, 473 – 477. Coplan, J.D., Goetz, R., Klein, D.F., Papp, L.A., Fyer, A.J., Liebowitz, M.R., Davies, S.O., Gorman, J.M., 1998b. Plasma cortisol concentrations preceding lactate-induced panic. Psychological, biochemical, and physiological correlates. Archives of General Psychiatry 55, 130 – 136.
Davis, M., 1992. The role of the amygdala in fear and anxiety. Annual Review of Neuroscience 15, 353 – 375. Dillon, D.J., Gorman, J.M., Liebowitz, M.R., Fyer, A.J., Klein, D.F., 1987. Measurement of lactate-induced panic and anxiety. Psychiatry Research 20, 97 – 105. Folgering, H., Vis, A., Ponte, J., 1981. Ventilatory and circulatory effects of doxapram, mediated by carotid body chemoreceptors. Bulletin Europeen de Physiopathologie Respiratoire 17, 237 – 241. Gorman, J.M., Fyer, M.R., Goetz, R., Askanazi, J., Liebowitz, M.R., Fyer, A.J., Kinney, J., Klein, D.F., 1988. Ventilatory physiology of patients with panic disorder. Archives of General Psychiatry 45, 31 – 39. Gorman, J.M., Kent, J.M., Sullivan, G.M., Coplan, J.D., 2000. Neuroanatomical hypothesis of panic disorder, revised. American Journal of Psychiatry 157, 493 – 505. Jensen, C.F., Peskind, E.R., Veith, R.C., Hughes, J., Cowley, D.S., Roy-Byrne, P., Raskind, M.A., 1991. Hypertonic saline infusion induces panic in patients with panic disorder. Biological Psychiatry 30, 628 – 630. Kent, J.M., Coplan J.D., Mawlawi O., Martinez J.M., Browne S.T., Slifstein M., Martinez D., Abi-Dargham A., Laruelle M., Gorman J.M., in press. Reduced orbitofrontal CBF predicts panic response to a respiratory stimulant in panic disorder. American Journal of Psychiatry. Klein, E., Zohar, J., Geraci, M.F., Murphy, D.L., Uhde, T.W., 1991. Anxiogenic effects of m-CPP in patients with panic disorder: comparison to caffeine’s anxiogenic effects. Biological Psychiatry 30, 973 – 984. Koszycki, D., Zacharko, R.M., Le Melledo, J.M., Bradwejn, J., 1998. Behavioral, cardiovascular, and neuroendocrine profiles following CCK-4 challenge in healthy volunteers: a comparison of panickers and nonpanickers. Depression and Anxiety 8, 1 – 7. LeDoux, J., 1996. The Emotional Brain: The Mysterious Underpinnings of Emotional Life. Simon and Schuster, New York. LeDoux, J.E., Iwata, J., Cicchetti, P., Reis, D.J., 1988. Different projections of the central amygdaloid nucleus mediate autonomic and behavioral correlates of conditioned fear. Journal of Neuroscience 8, 2517 – 2529. Lee, Y.J., Curtis, G.C., Weg, J.G., Abelson, J.L., Modell, J.G., Campbell, K.M., 1993. Panic attacks induced by doxapram. Biological Psychiatry 33, 295 – 297. Liebowitz, M.R., Gorman, J.M., Fyer, A.J., Levitt, M., Dillon, D., Levy, G., Appleby, I.L., Anderson, S., Palij, M., Davies, S.O., Klein, D.F., 1985. Lactate provocation of panic attacks: II. Biochemical and physiological findings. Archives of General Psychiatry 42, 709 – 719. Mathew, S.J., Coplan, J.D., Perko, K.A., Goetz, R.R., de la Neuz, M., Hollander, E., Liebowitz, M.R., Fallon, B.A., 2001. Neuroendocrine predictors of response to intravenous clomipramine therapy for refractory obsessive–compulsive disorder. Depression and Anxiety 14, 199 – 208. Papp, L.A., Klein, D.F., Martinez, J., Schneier, F., Cole, R., Liebowitz, M.R., Hollander, E., Fyer, A.J., Jordan, F., Gorman, J.M., 1993. Diagnostic and substance specificity of carbondioxide-induced panic. American Journal of Psychiatry 150, 250 – 257.
D.A. Gutman et al. / Psychiatry Research 133 (2005) 253–261 Pyke, R.E., Greenberg, H.S., 1986. Norepinephrine challenges in panic patients. Journal of Clinical Psychopharmacology 6, 279 – 285. Spitzer, R., Williams, J., Gibbon, M., First, M., 1990. Structured Clinical Interview for the DSM III-R. American Psychiatric Association. Sullivan, G.M., Apergis, J., Gorman, J.M., LeDoux, J.E., 2003. Rodent doxapram model of panic: behavioral effects and c-fos immunoreactivity in the amygdala. Biological Psychiatry 53, 863 – 870.
261
Targum, S.D., Marshall, L.E., 1989. Fenfluramine provocation of anxiety in patients with panic disorder. Psychiatry Research 28, 295 – 306. Veltman, D.J., van Zijderveld, G.A., van Dyck, R., 1996. Epinephrine infusions in panic disorder: a double-blind placebo-controlled study. Journal of Affective Disorders 39, 133 – 140. Winnie, A.P., 1973. Chemical respirogenesis: a comparative study. Acta Anaesthesiologica Scandinavica. Supplement 51, 1 – 32.
Doxapram-induced panic attacks and cortisol elevation David A. Gutmana,*, Jeremy Coplanb, Laszlo Pappc, Jose Martinezd, Jack Gormand,e a
Department of Psychiatry, Columbia University School of Medicine, 1051 Riverside Drive, New York, NY 10032, USA b Department of Psychiatry, State University of New York, Downstate, Brooklyn, NY, USA c New York State Psychiatric Institute, New York, NY, USA d Department of Psychiatry, Mount Sinai School of Medicine, New York, NY, USA e Department of Neuroscience, Mount Sinai School of Medicine, New York, NY, USA Received 29 August 2004; accepted 26 October 2004
Abstract Numerous agents with differing biological properties and central nervous system (CNS) effects can induce panic attacks in predisposed individuals. A potential explanation of this finding is that panic disorder patients are more likely to panic than normal control subjects when given a panicogen due to an excessive fear response to somatic arousal. We test this hypothesis by using doxapram, a panicogen with minimal CNS effects, to induce panic in patients and control subjects. Doxapram was given to six subjects with panic disorder with or without agoraphobia and four healthy volunteers. Measures comprised the Acute Panic Inventory, the Borg Exertion scale, the 10-point Anxiety Scale, the 10-point Apprehension Scale, cortisol, prolactin, and MHPG, all obtained at baseline and multiple time points after the doxapram infusion. All panic disorder patients panicked with doxapram, whereas no control subjects had a panic attack. Panic patients had similar levels of breathlessness with doxapram compared with control subjects. Although panic patients had higher levels of anxiety and apprehension, these did not change significantly with doxapram compared with control levels. Doxapram led to similar increases in cortisol and prolactin in both groups, and MHPG was consistently elevated in panic patients, but unaffected by doxapram. These results show that doxapram is a useful panicogen in the study of panic disorder. Since the panic patients and control subjects had similar levels of physiological and psychological arousal, but the panic patients were more likely to have a panic attack, this lends support to the concept of a sensitized fear network in panic disorder patients. D 2004 Elsevier Ireland Ltd. All rights reserved. Keywords: Doxapram; Panic attacks; Ventilatory stimulant; Cortisol
1. Introduction
* Corresponding author. Tel.: +1 212 342 1101; fax: +1 419 730 3208. E-mail address: [email protected] (D.A. Gutman).
A number of agents with varying biological properties and central nervous system (CNS) effects have been shown to produce panic attacks in predisposed individuals. These bpanicogensQ include
0165-1781/$ - see front matter D 2004 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.psychres.2004.10.010
254
D.A. Gutman et al. / Psychiatry Research 133 (2005) 253–261
compounds that produce osmotic stress (sodium lactate and hyperosmotic sodium chloride; stimulate neurotransmitter systems (yohimbine and m-CPP); stimulate brain peptide systems (cholecystokinin); and stimulate breathing (carbon dioxide and doxapram). Although many theories have been suggested to explain the mechanism of action of these panicogens, what they have in common is the capacity to produce significant somatic distress. We have proposed the hypothesis that panic disorder patients are more likely to panic when given a panicogen because they develop a greater fear response when somatic systems are aroused (Gorman et al., 2000). This is consistent with the catastrophic misinterpretation hypothesis of panic disorder that has been articulated by cognitive theorists (Austin and Richards, 2001). Animal models of conditioned fear have demonstrated a coordinated brain circuitry, with the amygdala at its core, that can be linked to the excessive fear in panic disorder patients. A test of this hypothesis is to administer an agent that produces somatic symptoms but does not have an appreciable CNS action at panicogenic doses. Doxapram is one such substance. The Ann Arbor, Michigan group has done interesting and significant work in exploring the properties of doxapram as a panicogen. Doxapram does not directly affect vascular flow, appears not to activate centrally mediated neurohormonal systems and minimally stimulates a noradrenergic response (Folgering et al., 1981; Calverley et al., 1983; Winnie, 1973; Burki, 1984; Abelson et al., 1995, 1996a). It induces panic attacks quickly in 75–80% of subjects and only 12.5–20% of controls (Lee et al., 1993; Abelson et al., 1996a,b). We hypothesize that doxapram will produce equivalent physiological changes in patients with panic disorder and control subjects, but that the former will experience more fear and therefore be more prone to panic. In this study, we will evaluate this hypothesis as well as validate the effectiveness of doxapram as an efficient precipitant of panic attacks in panic patients by measuring the anxiety and physiological responses to doxapram. We propose doxapram as an effective agent for precipitating panic attacks for imaging purposes, based on work by Abelson and Lee (Lee et al., 1993; Abelson et al., 1996a,b). As mentioned previously, doxapram does not directly affect vascular
flow, appears not to activate centrally mediated neurohormonal systems, minimally stimulates a noradrenergic response, is sensitive and specific for inducing panic in panic disorder subjects versus control subjects, and has a rapid and predicable onset (Lee et al., 1993; Abelson et al., 1996a,b). In this study, we examine whether doxapram will produce increases in systems linked to the fear response during panic attacks by measuring MHPG (noradrenergic systems), cortisol, and prolactin (neuroendocrine systems). Furthermore, we will validate the effectiveness of doxapram as a precipitant of panic attacks in patients with panic disorder and not in control subjects by replicating the work of Abelson and Lee (Lee et al., 1993; Abelson et al., 1996a,b).
2. Methods 2.1. Subjects Six subjects with panic disorder with or without agoraphobia (DSM-III-R; 1987) and four healthy volunteers participated in the study. The normal controls consisted of two males and two females, with a mean age of 29.75F10.94. The patients consisted of five males and one female, with a mean age of 37.33F7.74. The age was not significantly different (t= 1.30; df=8; Pb0.23). Subjects signed informed consent after a thorough description of the study. Psychiatric diagnoses were made by a psychiatric interview by a psychiatrist, and the Structured Clinical Interview for DSM-III-R (Spitzer et al., 1990) was performed by a trained rater. A medical history, physical examination, EKG, blood and urine tests including a thyroid function test, pregnancy test, and toxicology screen were performed. All subjects were required to be medically healthy, without a history of hypertension, cerebrovascular or cardiac disease, nonsubstance abusing, and medication-free for 2 weeks prior to the study. Any medication known to affect neuroendocrine function (e.g., phenytoin, phenobarbital, oral contraceptives, thyroid replacement, or other hormones) had to be discontinued at least 3 months prior to the study. A current or past diagnosis of an affective disorder, eating disorder, suicidal or homicidal ideation, schizophrenia, or other psychotic
D.A. Gutman et al. / Psychiatry Research 133 (2005) 253–261
illness excluded subjects from the study. Control subjects did not meet criteria for Axis-I or Axis-II disorders (DSM-III-R; American Psychiatric Association, 1987). Female subjects could not be pregnant or lactating and had a negative urine pregnancy test the day of the study. All women in the study were within the first 10 days of their menstrual cycle. 2.2. Experimental procedures Subjects were requested to abstain from food, alcohol, tobacco, and caffeine for 8 h before biological tests and arrived at the Biological Studies Unit at the New York State Psychiatric Institute after overnight fasting at 8 AM on 2 separate days separated by at least 1 day but within 4 days. Patients were told that on 1 study day they would receive an infusion of salt water and on another day they would receive an infusion of doxapram, which is routinely used in anesthesiology practice to stimulate ventilation. An intravenous angiocatheter line was started for the periodic withdrawal of blood samples for cortisol, prolactin, and MHPG levels (Papp et al., 1993; Coplan et al., 1995, 1998a,b; Mathew et al., 2001). An automated sphygmomanometer and EKG leads were used for continuous cardiovascular monitoring. Respiratory measures were acquired using the Respitrace system and a ventilatory canopy (a clear headchamber for the monitoring of gas exchange as the subject breathes room air), but will not be discussed in this article. At 40 min prior to doxapram administration, collection of vital signs and ratings commenced, and blood sample #1 was drawn (Table 1). Rating instruments were administered by a master’s or doctoral level clinician who was blind to both diagnosis and nature of substance administered at each time point of the four blood draws. These scales included the Acute Panic Inventory (API), a 17-item questionnaire specific to panic attacks that grades symptoms from none to very severe and is a reliable measure of laboratory-induced panic (Dillon et al., 1987); the Borg Exertion Dyspnea Scale, a measure of exertion on a 10-point Likert scale (Borg, 1982); the 10-point Apprehension Scale and the 10-point Anxiety Scale, which measure apprehension and anxiety, respectively, on a 10-point Likert scale. These measures
255
Table 1 Flow sheet for double-blind saline-controlled doxapram infusions counterbalanced for day of infusion Time
Experimental phase
Blood sample #a
40 min 10 min 0 min
BSU arrival Preinfusion baseline Double-blind doxapram infusion, saline-controlled Immediate postinfusion Delayed postinfusion
#1 #2
+90 s +20 min
#3 #4
a Acute Panic Inventory (API), 10-point Anxiety and Apprehension Scale and Borg Exertional Dyspnea Scale were collected at each blood draw time point. Doxapram/saline infusions were performed while subjects were in a sealed canopy for ventilatory monitoring. Doxapram was administered at a dose of 0.5 mg/kg in 10 ml of saline over 15 s and saline over the identical time period. BSU—Biological Studies Unit, New York State Psychiatric Institute.
were also performed 10 min prior to the intravenous infusion of doxapram or saline, at the time of infusion, at 90 s afterwards to correspond with panic attacks, and at 20 min postinfusion. Subjects received either a doxapram (0.5 mg/kg in 10 ml saline over 15 s; Lee et al., 1993) or saline intravenous infusion in a randomized order. At the end of the protocol, the subject was debriefed and discharged from the unit. 2.3. Data analysis Repeated measures analysis of variance (RMANOVA) was performed using group as the independent variable (patient versus control) and two within factors: condition (doxapram day versus saline day) and within-infusion time effects (in minutes: 40, 10, +1.5, +20). Significant RM-ANOVA were followed by planned comparisons using the Least Squares Difference post hoc testing method. This statistical approach was used for psychological scales and blood parameters.
3. Results 3.1. Panic rates Six of six panic disorder patients were rated by the blinded raters to have panicked to the doxapram infusions. Only five had the saline control, none of
256
D.A. Gutman et al. / Psychiatry Research 133 (2005) 253–261
whom panicked to saline. One subject did not complete the study after receiving doxapram for the first infusion. None of four control subjects experienced panic attacks during either the doxapram or saline infusions. This distribution of proportions is significantly different (v 2=8.43; df=1, P=0.008), and despite the small number of subjects in the current study, implies a panicogenic effect for doxapram. 3.2. Psychological scales 3.2.1. Acute Panic Inventory (see Fig. 1) There is a significant time effect [ F(1,21)=12.30; P=0.00007] and condition effect [ F(1,7)=9.96; P =0.016], but no significant group effect [ F(1,7)=3.58; P=0.1] (Fig. 1). A significant group by time effect [ F(3,21)=5.75; P=0.005] and time by condition effect [ F(3,21)=10.58; P=0.0002] were noted, but no significant group by condition effect [ F(1,7)=4.93; P=0.062]. A significant group by time by condition effect was noted [ F(3,21)=5.75; P=0.005]. Doxapram produced robust increases of the API scores within 90 s following inception of the doxapram infusion, with API scores at that time point clearly higher than all other time points that were measured. Thus, panic disorder patients were more
sensitive than control subjects to the panicogenic effects of doxapram as measured by the API. 3.2.2. Borg Exertional Dyspnea Scale There is a significant time effect [ F(3,21)=4.15; P=0.018] and a time by condition effect [ F(3,21)=4.15; P=0.018], but the analysis fails to reveal a corresponding group by time by condition effect [ F(3,21)=0.88; Pb0.46]. This suggests that the increases in Borg scores following doxapram were not distinguishable between patients and controls, and that both groups experienced similar levels of self-reported dyspnea. 3.2.3. Ten-point Anxiety Scale A group effect was noted for anxiety [ F(1,7)=6.65; P=0.036] with patients exhibiting higher overall levels of anxiety. A significant time by condition effect [ F(3,21)=4.64; P=0.01] reflects doxapram’s overall effect of increasing anxiety scores, but there was no group by time by condition effect [ F(3,21)=1.02; Pb0.4]. 3.2.4. Ten-point Apprehension Scale An overall group effect was noted for apprehension scores, with patients scoring higher than controls [ F(1,7)=4.7; P=0.017). Doxapram increased appre-
Fig. 1. For the Acute Panic Inventory, a group by time by condition was noted. Using least square differences (LSD) for planned comparisons, posttesting revealed that the +1.5 min postdoxapram infusion Acute Panic Inventory scores were significantly increased ( P b0.001; see symbol @ in figure) in comparison to all other time points. In addition, a significant time effect, condition effect, group by time effect, and time by condition effect was evident, but these effects were all best explained by the group by time by condition effect (statistics available in text).
D.A. Gutman et al. / Psychiatry Research 133 (2005) 253–261
hension in comparison to saline [ F(3,21)=5.38; Pb0.007], and a trend for a group by time by condition was noted [ F(3,21)=2.8; P=0.065), suggesting, in parallel to the API scores, an enhanced effect of doxapram on apprehension in patients. 3.3. Physiological results 3.3.1. Cortisol results (see Fig. 2) There was a significant time effect [ F(3,21)=8.86; P=0.0005] and a significant condition by time effect [ F(3,21)=8.47; P=0.0007] but an absence of a group by time by condition effect [ F(3,21)=1.19; P=0.33). Planned comparisons using post hoc testing using Least Squares Difference (LSD) indicated that baseline ( 40 min) cortisol in all groups prior to saline infusions was elevated relative to levels at +1.5 and +20 min (see Fig. 2), indicating a progressive decrease of cortisol levels in both groups over time during the saline day. On the doxapram infusion day, both the baseline ( 40 min) and delayed postdoxapram (+20 min) cortisol levels were elevated in comparison to the 10 and +1.5 min time points. In addition, only the
257
delayed (+20 min) postdoxapram cortisol level was strongly significant in comparison to the corresponding time point of the saline infusions ( P=0.0006). 3.3.2. Prolactin results (see Fig. 3) There was a significant condition by time effect [ F(3,18)=5.22; P=0.009) but no other significant statistical effects. Post hoc testing using specified comparisons and least squares differences (see Fig. 3) indicated that all subjects exhibited significant prolactin elevations +20 min after doxapram in comparison to all other time points ( Pb0.05; see Fig. 3). At +20 min, prolactin levels were significantly elevated with doxapram compared with saline (LSD difference is P=0.0002 level). 3.3.3. MHPG results (see Fig. 4) Since there were multiple missing data points for MHPG during the saline infusions, only the doxapram infusion component of the study was analyzed. An overall group by time interaction was evident for the doxapram component (see Fig. 4 for statistics). Using Least Squares Difference for planned comparisons,
Fig. 2. For plasma cortisol, there was a significant time effect and a significant condition by time effect (see figure for statistics) but an absence of a group by time by condition effect [ F(3,21)=1.19; P=0.33). Planned comparisons using post hoc testing using Least Squares Difference indicated that baseline ( 40 min) cortisol in both groups prior to saline infusions was elevated relative to the levels at +1.5 and +20 min (see symbol * left panel). Post hoc testing indicated that for the doxapram infusion day (right panel), both the baseline ( 40 min) and delayed postdoxapram (+20 min) levels were elevated in comparison to the 10 and +1.5 min time points (indicated by symbol #). In addition, the delayed (+20 min) postdoxapram saline cortisol level was differed strongly from the level at the corresponding point of the saline infusions ( P=0.0006).
258
D.A. Gutman et al. / Psychiatry Research 133 (2005) 253–261
Fig. 3. For the plasma prolactin analysis, there was a significant condition by time effect [ F(3,18)=5.22; P=0.009)] but no other significant statistical effects. Post hoc testing using planned comparisons and Least Squares Differences (LSD) indicated that all subjects exhibited prolactin elevations +20 min after doxapram in comparison to all other time points ( Pb0.05; symbol ##); for the corresponding time point of the saline infusion, the LSD difference is at the P=0.0002 significance level.
patients with panic disorder showed an overall increase in plasma MHPG ( Pb0.001), but this effect was most marked at baseline ( Pb0.000001), when patients demonstrated relatively higher MHPG levels
in comparison to controls. Although differences narrowed between the groups, matched time points continued to indicate significant elevations during the infusion in the patients versus controls (see Fig. 4,
Fig. 4. An overall group by time interaction was evident for the doxapram component (see statistics). Using least square differences for planned comparisons, patients with panic disorder showed an overall increase in plasma MHPG (see symbol *; Pb0.001), but this effect was most marked at baseline (see symbol **; Pb0.000001), where patients demonstrated relatively higher MHPG levels in comparison to controls, who demonstrated relatively lower baseline MHPG levels.
D.A. Gutman et al. / Psychiatry Research 133 (2005) 253–261
legend). There was no discernible effect of doxapram on MHPG in either the patients or controls.
4. Discussion We demonstrated that doxapram is an effective panicogen with a 100% panic rate in panic disorder patients and a 0% rate in normal controls, thereby replicating the findings of Abelson et al. (1996a,b) and Lee et al. (1993). Our sample is small and requires further validation, although additional unpublished data (Kent et al., in press) from our group demonstrate similar rates in another group of subjects, with 6/6 (100%) subjects with panic disorder but only 1/6 (16.67)% controls having a panic attack after IV doxapram. Some considerations are necessary when reviewing these data. Although we have subject reports of breathlessness, we do not have respiratory data, which would clarify whether all subjects had similar physiological responses. Furthermore, our data demonstrate that doxapram induces a cortisol surge that is not related to diagnosis or the occurrence of panic attacks. This is in contrast to the Abelson et al. (1996a) finding of no change in cortisol level with doxapram infusion. MHPG was elevated in panic patients compared with control subjects, which has been previously demonstrated (Coplan et al., 1997), but did not increase with doxapram, as did the cortisol and prolactin levels. Our subject sample is small, and further studies with larger samples may clarify these findings as well as provide necessary replication. Other limitations to this study include the following: (1) we only employed one dose of doxapram and it is possible that less difference between patients and controls would be observed with higher doses. Doseresponse studies would be necessary to clarify this issue. (2) We did not include groups with diagnoses other than panic disorder, and therefore the diagnostic specificity of doxapram-induced panic cannot be inferred. (3) We did not obtain an assessment of the similarity of doxapram-induced and naturally occurring panic attacks. These shortcomings could be easily addressed in future studies. Although some have previously divided panicogens into respiratory and HPA-activating agents (Coplan and Klein, 1996), in our results, doxapram
259
appears to be like cholecystokinin, with both actions represented (Koszycki et al., 1998; Bradwejn and Koszycki, 2001). In fact, we found that panic patients experienced similar levels of physiological arousal— increases in cortisol, prolactin, and dyspnea—as control subjects. The panic patients had panic attacks and a greater increase on the API, whereas the control subjects did not panic. In other words, it appears that panic patients may be more fearful of the stimulation aroused by doxapram. This finding provides support for a model of panic disorder proposed by Gorman et al. (2000) based on a bfear networkQ in the brain that is centered in the amygdala and involves its interaction with the hippocampus and the medial prefrontal cortex. This bfear networkQ is derived from animal work on classical conditioning (LeDoux et al., 1988; Davis, 1992; LeDoux, 1996). The model notes that there are many stimulants with various differing mechanisms, but all can induce panic attacks in patients with panic disorder (Charney et al., 1984; Liebowitz et al., 1985; Pyke and Greenberg, 1986; Gorman et al., 1988; Targum and Marshall, 1989; Klein et al., 1991; Jensen et al., 1991; Bradwejn and Koszycki, 1994; Veltman et al., 1996). What they may have in common is physiological arousal of the subject, which in vulnerable patients leads to misinterpretation of the sensations as a fear cue. The common final pathway is stimulation of the amygdala, which leads to a panic attack. Indeed, doxapram causes an increase in c-fos expression in the central nucleus of the amygdala in rat models of contextual fear conditioning (Sullivan et al., 2003). Future directions include replication of these findings with a larger sample size and brainimaging studies to examine activity in areas predicted by the fear model of panic disorder. Acknowledgments The research reported was supported in part by NIMH Grants P50MH58911 and RO1MH60951. References Abelson, J.L., Curtis, G.C., Nesse, R., Fantone, R., Pyke, R.E., Bammert-Adams, J., 1995. The effects of central cholecystokinin receptor blockade on hypothalamic–pituitary–adrenal and
260
D.A. Gutman et al. / Psychiatry Research 133 (2005) 253–261
symptomatic responses to overnight withdrawal from alprazolam. Biological Psychiatry 37, 56 – 59. Abelson, J.L., Nesse, R.M., Weg, J.G., Curtis, G.C., 1996a. Respiratory psychophysiology and anxiety: cognitive intervention in the doxapram model of panic. Psychosomatic Medicine 58, 302 – 313. Abelson, J.L., Weg, J.G., Nesse, R.M., Curtis, G.C., 1996b. Neuroendocrine responses to laboratory panic: cognitive intervention in the doxapram model. Psychoneuroendocrinology 21, 375 – 390. American Psychiatric Association, 1987. The Diagnostic and Statistical Manual of Mental Disorders (DSM-III R). Author, Washington, DC. Austin, D.W., Richards, J.C., 2001. The catastrophic misinterpretation model of panic disorder. Behaviour Research and Therapy 39, 1277 – 1291. Borg, G.A., 1982. Psychophysical bases of perceived exertion. Medicine and Science in Sports and Exercise 14, 377 – 381. Bradwejn, J., Koszycki, D., 1994. The cholecystokinin hypothesis of anxiety and panic disorder. Annals of the New York Academy of Sciences 713, 273 – 282. Bradwejn, J., Koszycki, D., 2001. Cholecystokinin and panic disorder: past and future clinical research strategies. Scandinavian Journal of Clinical & Laboratory Investigation Supplement 234, 19 – 27. Burki, N.K., 1984. Ventilatory effects of doxapram in conscious human subjects. Chest 85, 600 – 604. Calverley, P.M., Robson, R.H., Wraith, P.K., Prescott, L.F., Flenley, D.C., 1983. The ventilatory effects of doxapram in normal man. Clinical Science 65, 65 – 69. Charney, D.S., Heninger, G.R., Breier, A., 1984. Noradrenergic function in panic anxiety. Effects of yohimbine in healthy subjects and patients with agoraphobia and panic disorder. Archives of General Psychiatry 41, 751 – 763. Coplan, J., Klein, D., 1996. Pharmacological probes in panic disorder. In: Westenberg, H., Boer, J., Murphy, D. (Eds.), Advances in the Neurobiology of Anxiety Disorders. Wiley, Chichester, pp. 173 – 196. Coplan, J.D., Pine, D., Papp, L., Martinez, J., Cooper, T., Rosenblum, L.A., Gorman, J.M., 1995. Uncoupling of the noradrenergic–hypothalamic–pituitary–adrenal axis in panic disorder patients. Neuropsychopharmacology 13, 65 – 73. Coplan, J.D., Papp, L.A., Pine, D., Martinez, J., Cooper, T., Rosenblum, L.A., Klein, D.F., Gorman, J.M., 1997. Clinical improvement with fluoxetine therapy and noradrenergic function in patients with panic disorder. Archives of General Psychiatry 54, 643 – 648. Coplan, J.D., Trost, R.C., Owens, M.J., Cooper, T.B., Gorman, J.M., Nemeroff, C.B., Rosenblum, L.A., 1998a. Cerebrospinal fluid concentrations of somatostatin and biogenic amines in grown primates reared by mothers exposed to manipulated foraging conditions. Archives of General Psychiatry 55, 473 – 477. Coplan, J.D., Goetz, R., Klein, D.F., Papp, L.A., Fyer, A.J., Liebowitz, M.R., Davies, S.O., Gorman, J.M., 1998b. Plasma cortisol concentrations preceding lactate-induced panic. Psychological, biochemical, and physiological correlates. Archives of General Psychiatry 55, 130 – 136.
Davis, M., 1992. The role of the amygdala in fear and anxiety. Annual Review of Neuroscience 15, 353 – 375. Dillon, D.J., Gorman, J.M., Liebowitz, M.R., Fyer, A.J., Klein, D.F., 1987. Measurement of lactate-induced panic and anxiety. Psychiatry Research 20, 97 – 105. Folgering, H., Vis, A., Ponte, J., 1981. Ventilatory and circulatory effects of doxapram, mediated by carotid body chemoreceptors. Bulletin Europeen de Physiopathologie Respiratoire 17, 237 – 241. Gorman, J.M., Fyer, M.R., Goetz, R., Askanazi, J., Liebowitz, M.R., Fyer, A.J., Kinney, J., Klein, D.F., 1988. Ventilatory physiology of patients with panic disorder. Archives of General Psychiatry 45, 31 – 39. Gorman, J.M., Kent, J.M., Sullivan, G.M., Coplan, J.D., 2000. Neuroanatomical hypothesis of panic disorder, revised. American Journal of Psychiatry 157, 493 – 505. Jensen, C.F., Peskind, E.R., Veith, R.C., Hughes, J., Cowley, D.S., Roy-Byrne, P., Raskind, M.A., 1991. Hypertonic saline infusion induces panic in patients with panic disorder. Biological Psychiatry 30, 628 – 630. Kent, J.M., Coplan J.D., Mawlawi O., Martinez J.M., Browne S.T., Slifstein M., Martinez D., Abi-Dargham A., Laruelle M., Gorman J.M., in press. Reduced orbitofrontal CBF predicts panic response to a respiratory stimulant in panic disorder. American Journal of Psychiatry. Klein, E., Zohar, J., Geraci, M.F., Murphy, D.L., Uhde, T.W., 1991. Anxiogenic effects of m-CPP in patients with panic disorder: comparison to caffeine’s anxiogenic effects. Biological Psychiatry 30, 973 – 984. Koszycki, D., Zacharko, R.M., Le Melledo, J.M., Bradwejn, J., 1998. Behavioral, cardiovascular, and neuroendocrine profiles following CCK-4 challenge in healthy volunteers: a comparison of panickers and nonpanickers. Depression and Anxiety 8, 1 – 7. LeDoux, J., 1996. The Emotional Brain: The Mysterious Underpinnings of Emotional Life. Simon and Schuster, New York. LeDoux, J.E., Iwata, J., Cicchetti, P., Reis, D.J., 1988. Different projections of the central amygdaloid nucleus mediate autonomic and behavioral correlates of conditioned fear. Journal of Neuroscience 8, 2517 – 2529. Lee, Y.J., Curtis, G.C., Weg, J.G., Abelson, J.L., Modell, J.G., Campbell, K.M., 1993. Panic attacks induced by doxapram. Biological Psychiatry 33, 295 – 297. Liebowitz, M.R., Gorman, J.M., Fyer, A.J., Levitt, M., Dillon, D., Levy, G., Appleby, I.L., Anderson, S., Palij, M., Davies, S.O., Klein, D.F., 1985. Lactate provocation of panic attacks: II. Biochemical and physiological findings. Archives of General Psychiatry 42, 709 – 719. Mathew, S.J., Coplan, J.D., Perko, K.A., Goetz, R.R., de la Neuz, M., Hollander, E., Liebowitz, M.R., Fallon, B.A., 2001. Neuroendocrine predictors of response to intravenous clomipramine therapy for refractory obsessive–compulsive disorder. Depression and Anxiety 14, 199 – 208. Papp, L.A., Klein, D.F., Martinez, J., Schneier, F., Cole, R., Liebowitz, M.R., Hollander, E., Fyer, A.J., Jordan, F., Gorman, J.M., 1993. Diagnostic and substance specificity of carbondioxide-induced panic. American Journal of Psychiatry 150, 250 – 257.
D.A. Gutman et al. / Psychiatry Research 133 (2005) 253–261 Pyke, R.E., Greenberg, H.S., 1986. Norepinephrine challenges in panic patients. Journal of Clinical Psychopharmacology 6, 279 – 285. Spitzer, R., Williams, J., Gibbon, M., First, M., 1990. Structured Clinical Interview for the DSM III-R. American Psychiatric Association. Sullivan, G.M., Apergis, J., Gorman, J.M., LeDoux, J.E., 2003. Rodent doxapram model of panic: behavioral effects and c-fos immunoreactivity in the amygdala. Biological Psychiatry 53, 863 – 870.
261
Targum, S.D., Marshall, L.E., 1989. Fenfluramine provocation of anxiety in patients with panic disorder. Psychiatry Research 28, 295 – 306. Veltman, D.J., van Zijderveld, G.A., van Dyck, R., 1996. Epinephrine infusions in panic disorder: a double-blind placebo-controlled study. Journal of Affective Disorders 39, 133 – 140. Winnie, A.P., 1973. Chemical respirogenesis: a comparative study. Acta Anaesthesiologica Scandinavica. Supplement 51, 1 – 32.