La patofiología del trastorno de pánico.

La patofiología del trastorno de pánico.

(The pathophysiology of panic disorder)

K. Schruers; E. Griez.

Maastricht University
Institute Brain and Behaviour
Department of Psychiatry and Neuropsychology
PO Box 88
Maastricht AB 6200
The Netherlands

PALABRAS CLAVE: trastorno de angustia, trastorno de pánico, Pruebas de provocación, Lactato, Dióxido de carbono, Asfixia.

(KEYWORDS: Panic disorder, Challenge tests, Lactate, Carbon dioxide, Suffocation. )

Resumen

Los ataques de pánico espontáneos son impredecibles, y por lo tanto difíciles de estudiar. Para poder estudiar directamente la enfermedad se han desarrollado diferentes modelos de pánico utilizando diversos estímulos de provocación para conseguir provocar las crisis en condiciones de laboratorio. La infusión de lactato fue el primer modelo utilizado. Provoca una crisis en el 60-80% de los pacientes pero sólo en el 0-20% de los controles sanos. El hecho de que el pánico desencadenado por lactato se asocia a hiperventilación y que la hipocapnia es mayor en pacientes, impulsó la investigación de la hiperventilación en el trastorno de Angustia (TA). En la prueba de provocación con dióxido de carbono los pacientes responden con más ansiedad que los controles, los pacientes con ansiedad generalizada, los obsesivos, y los que presentan fobias simples. Estos hallazgos han renovado el interés en una hipótesis metabólica que explique la vulnerabilidad de los pacientes con TA al lactato y al CO2. La hipótesis de la amenaza vital formulada por Klein en 1993 explica que tras un exceso de lactato o CO2 se puede desarrollar una alarma de ahogo innata. Esta alarma, desarrollada filogenéticamente provoca un estrés agudo, sensación de ahogo y una conducta de huída de lugares cerrados. Los ataques de pánico pueden constituir falsas alarmas en personas predispustas que poseen un "sistema de detección de ahogo" hipersensible. Aunque los mecanismos finales del pánico distan de conocerse, las pruebas de provocación pueden constituir herramientas útiles a la hora de validar diagnósticos clínicos.

Abstract

Spontaneous panic attacks are unpredictable and therefore hard to study. To be able to study the phenomena of this illness more directly, models of panic were developed, using different types of challenges to provoke panic under laboratory conditions. Lactate was the first model. It induces panic attacks in about 60-80% of panic disorder (PD) patient compared to 0-20% of healthy controls. The finding that hyperventilation is associated with lactate triggered panic, and that hypocapnia is deeper in patients than controls fostered lactate investigators to explore the role of hyperventilation in the genesis of panic. In the carbon dioxide challenge PD subjects reported significantly more anxiety than controls, generalized anxiety, obsessive patients, and those with simple phobia. These findings raise a regain of interest in a metabolic hypothesis to account for the vulnerability of PD subjects to both lactate and CO2. An appealing hypothesis has been formulated by Klein (1993). The vital threat represented by an excess of lactate or CO2 may trigger an inborn suffocation alarm. This phylogenetically evolved alarm releases acute distress, and breathlessness, urging air hunger and flight to open surroundings. Panic attacks may be false alarms in sensitive individuals who have a disregulated hypersensitive suffocation detector. Even though the underlying mechanisms of panic are far from fully unraveled, laboratory challenges may become useful tools to validate clinical diagnosis.



Part 1. The concept of panic disorder

Clinical picture

Panic disorder (PD) is characterized by the spontaneous, unexpected occurrence of panic attacks. Panic attacks are relatively short-lived periods of intense fear or discomfort, accompanied by a number of somatic or psychological symptoms. According to DSM IV, at least four symptoms are required, that have to develop abruptly and reach a peak within 10 minutes. These symptoms are:

Palpitations, pounding heart or accelerated heart rate
Sweating
Trembling on shaking
Sensations of shortness of breath or smothering
Feeling of choking
Chest pain or discomfort
Nausea or abdominal distress
Feeling dizzy, unsteady, lightheaded or faint
Derealization (feelings of unreality) or depersonalization (feeling detached from oneself)
Fear of losing control or going crazy
Fear of dying
Paresthesias (numbness or tingling sensations)
Chills or hot flushes

The symptoms of panic may be present in various other situations: use or withdrawal of a drug or a medical condition, such as hyperthyroidism or pheochromocytoma. For this reason, DSM IV requires the exclusion of substance use or a general medical condition.

Three types of panic attacks are distinguished, based on the presence or absence of situational triggers. Firstly, unexpected or “spontaneous” panic attacks are not associated with a situational trigger. Secondly situationally bound panic attacks occur almost invariably on exposure to, or in anticipation of, a situational cue.

Thirdly, the situationally predisposed panic attacks are more likely to occur on exposure to the situational cue or trigger, but are not invariably associated with the cue and do not necessarily occur immediately upon exposure (Faravelli and Paionni 2000).

The unexpected panic attacks are considered to be the hallmark of panic disorder. DSM IV requires at least two unexpected attacks for the diagnosis of PD. After the first attack, most patients develop the fear that another attack may occur. Therefore, in the periodes in between the panic attacks, the level of diffuse, anticipatory anxiety increases. DSM IV takes this anticipatory anxiety into account by requiring for the diagnosis of PD at least one of the following:

persistent concern about having additional attacks
worry about the implications of the attack or its consequences
a significant change in behavior related to the attacks.

Panic disorder is often accompanied by agoraphobia, the fear of being in places or situations from which escape might be difficult or embarrassing or in which help may not be available in the event of a panic attack. Although some controversy exists in the literature, most evidence suggests that agoraphobia develops after panic attacks begin, and are due to the above mentioned anticipatory anxiety (McNally 1994).

Syndromal validity

The use of explicit diagnostic criteria and of structured interviews have greatly improved diagnostic reliability of PD, basis for its syndromal validity. Interrater reliability can be expressed as kappa, a coefficient that expresses the level of agreement corrected for chance. The kappa coefficient ranges from negative values through 0 to 1. With negative values, the observer agreement is less than might be expected from chance agreement.

With a value of 0, the agreement is entirely explained by chance. With positive values, the agreement is higher than chance. By conventional standards, values below . 50 indicate poor reliability, those between . 50 and . 70 indicate fair reliability and those above . 70 indicate good reliability (McNally 1994).

It has been shown that PD can be reliably diagnosed in anxiety disorder clinics: kappa’s range from . 75 to . 90 for panic disorder or panic disorder with agoraphobia, at least in settings where the rate of diagnosis is high and expert clinicians conduct structured interviews (McNally 1994). Therefore in the setting where the research for the present thesis was conducted, a tertiary referral clinic, diagnostic reliability and hence syndromal validity can be considered as good.

Epidemiology

Twelve month prevalence of PD is around 1%. Lifetime prevalence rates range from 1, 4% to 3, 8%. Across epidemiological surveys, prevalence is about twice as high in women, compared to men (Faravelli and Paionni 2000; Overbeek et al 2000).

PD most commonly develops around the age of 25, but the age of onset varies considerably. Hazard rates for women range from 25 to 35 years, for men between 30 and 45 (Wittchen and Essau 1993).

Panic disorder has a high degree of comorbidity with other anxiety disorders like social phobia (20-75%), generalized anxiety disorder (20%), obsessive compulsive disorder (14%) and post-traumatic stress disorder (6%) (Goisman et al 1994; Overbeek et al 2000; Pelissolo and Lepine 1998). There is also a considerable comorbidity with major depression, with reported rates ranging between 30 and 60% (Weissman et al 1997). Substance abuse is common in PD, with a prevalence of 36% according to the Epidemiological Catchment Area Study (Regier et al 1990).

Quality of life and economic burden
Studies in community samples show that 35% of the people with panic disorder felt they were in fair of poor physical health and 38% felt they were in poor emotional health, whereas people with no panic disorder had significantly lower levels of negative perceptions of their physical and mental health (24% and 16%). Around 27% of the persons with panic disorder were receiving welfare or some form of disability compensation, a much higher rate than among people without panic disorder (Markowitz et al 1989; Mendlowicz and Stein 2000; Weissman 1991). Studies in clinical samples also show a considerably decreased quality of life in patients with panic disorder, compared to normal subjects (Mendlowicz and Stein 2000).

There is little direct information on the economic costs of panic disorder to society. However, because of the association with increased utilization of medical care, the costs are considerable. One study examining both the direct costs, including all the medical care, and the indirect costs due to lost productivity showed a total cost of $ 94801 per year, before treatment. After treatment those costs were reduced to $ 60139, (Salvador Carulla et al 1995; Zambler and Katon 1998).

Conclusion

In summary, it may be concluded that PD, being a relatively recent concept, is a clinically valid syndrome. It refers to a quite common condition that significantly affects people's quality of life and may represent an important economical burden at societal level.

Part 2 : The Experimental pathophysiology of panic

1. The importance of challenge tests

Spontaneous panic attacks are unpredictable and therefore hard to study. The information we obtain from panic disorder patients about their panic attacks is subjective and entirely retrospective.

To be able to study the phenomena of this illness more directly, models of panic were developed, using different types of challenges to provoke panic under laboratory conditions.

Such a model should respond to certain criteria (Uhde and Tancer 1990 based on Guttmacher et al 1983 and Gorman et al 1987):

1. The symptoms of an induced panic attack must closely mimic the symptoms of a spontaneous attack.
2. Reaction to the test should be specific for panic disorder patients.
3. Clinically effective medication for panic disorder should also lessen the reaction to the challenge and clinically ineffective agents should fail to do so.
4. The results of the challenge should be replicable.

Up to now there is no such thing as a definite diagnostic test for panic disorder: The currently available laboratory models respond in a different degree to these criteria of an ideal model, as will be discussed further.

Further, challenge tests are important in unraveling the role of different neuronal systems in the mechanism of panic attacks.

2. History

In 1871 the American military surgeon Da Costa described the syndrome of “Irritable heart”. Da Costa’s syndrome was characterized by spells of dizziness, breathlessness and palpitations, chest discomfort and anxiety (Da Costa 1871). One year later, Westphal published his famous account of a series of patients whose chief complaints existed of anxiety attacks, inability to walk in certain streets and to cross open squares and a tendency to the generalization of behavioral incapacitation (Westphal 1872). Because of the inability to walk across certain public places, Westphal called the condition agoraphobia. In fact, both Da Costa and Westphal were providing early detailed clinical descriptions of cases of panic disorder and agoraphobia.

At about the same time, a German physician called Cordes, reported on 27 cases of “Platzangst” (the German term for agoraphobia). Suffering himself with the disorder he had observed that his condition worsened with bright light and in air polluted with carbon dioxide. “Alcohol”, he noted, “always helps” (Cordes 1872). Half a century later, Drury showed that patients with irritable heart indeed develop intolerable hyperpnea at a lower level of inspired carbon dioxide than healthy controls (Drury 1920). However, this very first experimental approach of panic remained largely unnoticed.

In a comprehensive overview dated 1951, Cohen and White reported that a number of abnormal physiologic responses could be found in subjects with the disorder (Cohen and White 1951). In particular they noticed metabolic anomalies, with abnormal oxygen consumption and excessive lactate production during muscular exercise. In line with Drury, they also found these patients to be overreactive to carbon dioxide, with “intolerable hyperpnea, increased sighing, and symptoms identical with anxiety attacks” when breathing a mixture containing an excess of carbon dioxide (Cohen and White 1951). The alleged excessive lactate production suggested that lactate administration may bring about the symptoms of anxiety neurosis. This inspired Pitts and McClure’s early experiments with lactate infusion (Pitts and McClure 1967). Curiously, although carbon dioxide hypersensitivity was mentioned in the same report, it took another thirty years to confirm that carbon dioxide administration experimentally induces attacks in panic disorder patients. Since then, both lactate infusion and carbon dioxide inhalation have emerged as experimental models of high heuristic value in anxiety research. Both models will be reviewed at some length in the present paper.

3. The lactate model

Assuming that the lactate ion itself may produce anxiety attacks in vulnerable persons, Pitts and McClure conducted a double blind placebo controlled study in 14 patients with anxiety neurosis and 10 healthy controls. The experiment was carefully designed, the effects of the infusion being assessed by a “blind” observer, not aware of the purpose of the study.

On basis of a standard checklist and subjects self report, symptoms of an anxiety attack were found in 13 out of the 14 patients after lactate infusion, but only in 2 out of the 10 controls. Glucose in saline produced no attacks at all. Adding calcium to the lactate significantly reduced the response to the procedure. All patients rated the lactate induced state as a very similar if not identical to their real life attacks.

Pitts and McClure interpreted their findings as being the first instance of experimentally produced anxiety symptoms in susceptible individuals.

They speculated that anxiety attacks may be brought about by a lactate specific mechanism, probably chelation of ionized calcium by lactate at the surface of excitable membranes.
Whatsoever, Pitts and McClure’s findings fostered an impetus of interest in the underlying pathophysiology of anxiety neurosis. In the early seventies, there were at least three controlled studies which replicated the original finding of lactate precipitated anxiety attacks (Fink et al 1969; Kelly et al 1971; Bonn et al 1971). For lactate, all studies yielded a response rate of 80-100% in patients, comparing to 0-25% in normals. However, beside these repeated confirmations of lactate’s anxiogenic effect in anxiety neurosis, no progress was made in the understanding of the underlying mechanisms. The early speculations on ionized calcium were challenged by the finding that EDTA (ethylene diamine tetra acetate), a potent chelator of free calcium, failed to reproduce the effect of lactate in patients (Pitts and Allen 1979). Shifts in acid base equilibrium were invoked but it was immediately pointed out that lactate infusion does not mimic the physiological state of naturally occurring hyperlactacidemia. A rise in endogenous lactate induces a metabolic acidosis, exogenous administration of lactate results in a systemic alkalosis.

In the eighties, Liebowitz, Gorman, Fyer, Klein and coworkers (1984) initiated a large scale investigation on the effects of lactate provocation in patients with panic disorder (PD).

Relying on the predefined diagnostic criteria of the DSM III they established that lactate induces panic attacks (PA) in about 60-80% of PD patient compared to 0-20% of healthy controls. Subsequent studies strongly suggested lactate to have diagnostic specificity amongst DSM III anxiety disorders. Lactate does neither affect patients with obsessive compulsive disorder (OCD, Gorman et al 1987), nor subjects with social phobia (Liebowitz et al 1985a). Results are less clear cut in generalized anxiety (GAD): in one study (Lapierre et al 1984) only 2 of 16 GAD patients panicked but the response rate in PD’s in that sample was very low too, as only 6 out of 23 had a PA. In another report, 12 out of 22 PD subjects and 5 out of 9 GAD’s proved lactate sensitive (Cowley et al 1988).

Lactate vulnerability is sensitive to clinically successful treatment: when panic free under effective antidepressant medication, 13 patients became insensitive to lactate induced PA. After drug discontinuation, 3 of them had a PA under lactate (Fyer et al 1985). These results show that effective pharmacological treatment protects against the effect of lactate infusion. However, lactate vulnerability may exist in clinically well, but panic prone individuals. To this extent, lactate vulnerability may be a trait marker.

Contrasting with the growing body of data on the validity of the lactate infusion as a laboratory model for panic, no convincing progress has been made in answering the question why patients panic from lactate. Numerous biochemical and physiological measures during lactate challenges yielded only meager results (Liebowitz et al 1985b).

Neither hypocalcemia, nor metabolic alkalosis appear to play a crucial role in lactate precipitated PA’s. In sharp contrast with anticipatory and generalized anxiety, lactate elicited PA’s are not attended by a large increase in plasma cortisol, suggesting a lack of activitation of the HPA axis in panic (Hollander et al 1989).

Circulating catecholamines show an inconstant and modest increase (Liebowitz et al 1986). Cardiovascular modifications seen during lactate infusion in panickers may be compatible with sympathetic activation of central origin (Liebowitz et al 1985b; Liebowitz et al 1986). However, propranolol fails to block lactate induced PA’s, and isoproterenol shows inconstant effects (Gorman et al 1984; Nesse et al 1984; Rainey et al 1984). Clonidine, an alpha-2 agonist, partially prevented lactate induced PA’s (Coplan et al 1992a). The most striking effect of a lactate infusion is that it causes subjects to hyperventilate, patients more than controls, and those who panic more than those who do not. Since lactate infusion induces a systemic metabolic alkolosis, with increased pH and bicarbonate in all subjects, hyperventilation and hypocapnia is unexpected and appears paradoxical (Liebowitz et al 1986).

4. The role of hyperventilation

The finding that hyperventilation is associated with lactate triggered panic, and that hypocapnia is deeper in patients than controls fostered lactate investigators to explore the role of hyperventilation in the genesis of panic. During the post world war II period, a body of clinical literature linking hyperventilation and anxiety had developed, mainly in Europe. Hyperventilation, due to “bad breathing habits” was regarded as a key mechanism in some syndromes of anxiety. Individuals with the so-called hyperventilation syndrome were believed to have spells of self induced hypocapnia, generating physical symptoms which they could not explain, causing in turn further arousal and anxiety (Lum 1975).

To explore the role of hyperventilation in panic, Gorman and associates submitted a group of PD-patients to voluntary hyperventilation. Breathing a slightly hypercapnic mixture (5% CO2) was used as a control condition.

Unexpectedly, not the forced hyperventilation, but the carbon dioxide condition proved to be panicogenic (Gorman et al 1984).
At the same time, our group had become interested in an old fashioned procedure, consisting of one single breath administration of a 35% CO2 mixture oxygen, which was claimed to be anxiolytic (Wolpe 1973). Under controlled conditions, the procedure proved devoid of any anxiolytic effects (Griez and van den Hout 1982; van den Hout and Griez 1982). On the contrary, it appeared to mimic symptoms of panic (Van den Hout and Griez 1984).

The role of hyperventilation was further explored in normals and in patients with PD, carefully controlling for cognitive factors and suggestion. All experiments showed that acute hypocapnia induced by voluntary hyperventilation fails to induce PA’s (Griez et al 1988; van den Hout et al 1988; Zandbergen et al 1990).
Even though patients do hyperventilate as well in real life as in laboratory induced panic, hypocapnia is neither sufficient nor necessary to cause panic (Griez et al 1987a).

5. Carbon dioxide as a model for panic

Essentially, two main procedures have been developed in using carbon dioxide to explore the mechanisms of PD. Continuous breathing of a 5% CO2 mixture for 10-20 minutes is known as the 5% CO2 challenge. Subjects usually breath through a mask for several minutes to the point of panic. Some studies have used a canopy. In analogy with the lactate infusion, CO2 administration ceases either on request of the subjects or upon a PA. The so-called 35% CO2 challenges rely on one single vital capacity inhalation of a mixture containing 35% CO2 and 65% O2.

Subjects use a self administration mask, effects are immediate symptoms of a PA appearing within a matter of seconds upon inhalation and waning after 30 to 60 seconds. The 5% CO2 challenge has been used after Gorman and coworkers’ unexpected finding while trying to control for hypocapnia. The 35% CO2 method was studied when, contrary to old clinical claims (Ley 1985), it appeared to be anxiogenic rather than anxiolytic. Although they each offer some advantages and disadvantages, using the same pharmacologic agent, both may be supposed to explore the very same underlying mechanisms of panic.

Following the results of Gorman et al (1984), the method with continuous breathing of low concentrations of CO2 became widely used. Overall, the effects are in line with those of lactate studies. There have been a number of experiments to assess the specificity and sensitivity of the procedure, but results are difficult to interpret due to methodological differences across the studies (Sanderson and Wetzler 1990). In their review of the literature, Sanderson and Wetzler (1990) point out a number of problems. There are different methods of administration (e. g. mask vs. canopy) the CO2 concentration may vary (some authors use 7% CO2), the way the subjects are instructed is not standardized. An important methodological difficulty is the operational definition of a PA. Even though most studies do refer to DSM III(R) criteria, the way investigators assess the presence of a PA differs widely from one study to another.

An overall evaluation of the single breath 35% CO2 procedure is easier, because the majority of the studies using this approach has been done by two groups working quite closely together. As a consequence there are few methodological differences between the studies. In particular, concentration of CO2, dosage, subjects instruction and assessment of induced symptoms are similar throughout most of the studies. In most of the 35% CO2 studies, the results are not merely expressed as PA vs. no PA. The effect is also assessed as subjective anxiety, expressed on a visual analogue scale immediately upon inhalation. This circumvents the problem of setting arbitrary criteria for the recognition of PA.

After it had become clear that a 35% CO2 inhalation may be panicomimetic (Van den Hout and Griez 1984), this challenge was administered to a group of patients meeting DSM III criteria for PD. Compared to healthy controls, PD subjects reported significantly more anxiety. In fact, they rated the inhalation experience as being strikingly similar to a real life PA.

Normals reported some neurovegetative symptomatology resembling that of a PA, but failed to experience anxiety (Griez et al 1987b). The vulnerability of PD patients to 35% CO2 has been subsequently documented in other studies (Fyer et al 1987; Gorman et al
1990; Perna et al 1994a).

Several studies have addressed the question whether 35% CO2 susceptibility is specific for individuals suffering with PA’s. Perna and coworkers have shown that subjects with sporadic PA’s tend to respond as those meeting the DSM III R defined threshold for PD, in contrast to normals who never had PA’s (Perna et al 1995c).

Griez et al demonstrated that 35% CO2 elicited anxiety acts as a marker of PD amongst a mixed group of subjects with various anxiety disorders, all of them displaying high ratings on anticipatory anxiety (Griez et al 1990b). Therefore, CO2 susceptibility is not a matter of baseline anxiety in highly aroused patients.

In the first of a series of subsequent experiments, a group of PD-patients was compared with OCD patients and healthy controls (Griez et al 1990a). CO2 triggered anxiety was significantly higher in PD than in OCD and in normals, with no difference between the latter groups. These results have been replicated by Perna et al (1995b). In another experiment, 28 subjects with simple phobia and 30 healthy controls were challenged with 35% CO2. Phobics were separated into situational phobias and animal phobias. Those with animal phobias had the same response as controls. Patients with situational phobias, however, reacted differently, and had CO2-induced anxiety which tended to resemble that of PD’s (Verburg et al 1994). These results are in keeping with epidemiological data linking situational phobias and agoraphobia. Worth noting, the above distinction between animal and situational phobias has been acknowledged in the DSM IV.

PD patients have been compared with GAD patients (Verburg et al 1995). While PD patients reacted with a marked increase in anxiety, there was no significant reaction in GAD subjects. Very stringent selection criteria were applied to select pure GAD patients, excluding any lifetime history of PA’s.

Thus, the sample size was small with only 9 GAD patients, and replication in a larger group of patients would be welcome. However, these preliminary results lend experimental validity to the distinction between PD and GAD.

The case of social phobia is less clear cut. Caldirola et al (1997) published results suggesting that patients with a social phobia are sensitive to a 35% CO2 challenge. Data from Gorman and coworkers (1990) seem to support the idea of some CO2 vulnerability in social phobics. However, in the latter study a double, instead of a single inhalation was administered. Our group has unpublished data showing social phobics, with no history of PA’s to be CO2 insensitive (Verburg et al, unpublished). It is noteworthy that social phobics are not sensitive to lactate infusion.

The 35% CO2 challenge has been administered to patients with mood disorders as well. Perna et al (1995a) showed patients with major depression to be not sensitive to the test.

Verburg et al (1998c) compared PD patients with and without comorbid depression. In this study, the presence of a comorbid depression was associated to a higher vulnerability to the challenge.

A number of investigations have been conducted to address the sensitivity of the 35% CO2 test to a clinically successful treatment. Upon retesting after six weeks, an effective treatment with clonazepam attenuated the CO2 triggered reaction (Pols et al 1991).

The same applies to a six weeks treatment with fluvoxamine (Pols et al 1996). Using toloxatone, a MAO inhibitor, a blunting effect on the 35% CO2 challenge could be evidenced after only one week of treatment (Perna et al 1994c). These findings strongly support the validity of the 35% CO2 model (Bertani et al 1997; Perna et al 1997).

Under the assumption that the development of PA relies on a constitutional predisposition, at least some of the relatives of PD patients, though free of manifest PA’s may be expected to have inherited a “panic proneness” trait. In a study by Perna and coworkers (Perna et al 1995d), 23 first degree relatives of PD patients underwent a 35% CO2 challenge. Twenty two of them (96%) had a CO2 triggered PA as compared to 51% of PD patients and 2% of healthy controls. The same group performed a CO2 challenge in a population of twins, showing a high concordance ratio, as far as CO2 proneness is concerned, in monozygotic twins compared to dizygotics (Perna et al 1993).

Carbon dioxide is a basic factor in the control of breathing. The partial pressure of CO2 is continuously monitored by peripheral and central chemoreceptors, and the depth and frequency of respiratory movements are adapted accordingly. Since disrupted breathing is a central feature of panic it is tempting to link the high CO2 vulnerability of PD subjects to a disturbed chemosensitivity. Our group has data which may be consistent with a high CO2 chemosensitivity in PD (Lousberg et al 1988; Pols et al 1993).

However, other studies yielded negative results (Woods et al 1986). In another study by our group, we could not replicate earlier findings on elevated chemosensitivity in patients with PD (Zandbergen et al 1991). Finally, the Columbia group published a study with positive findings in male subjects (Papp et al 1989).

In sum, the issue of an over-sensitivity of chemoreceptive areas in PD subjects is still open to question.

Within the broader context of breathing control, there is a growing evidence from clinical epidemiology that some relationship may exist between PD and chronic obstructive pulmonary diseases (COPD). There is an increased prevalence of PD amongst patients with COPD (Karajgi et al 1990; Shavit et al 1992). Moreover, patients with PD have an increased childhood prevalence of respiratory diseases, in particular asthma and bronchitis (Zandbergen et al 1991; Perna et al 1994d; Verburg et al 1995).

6. Other provocative agents

Various other agents have been used to trigger PA experimentally, though few have been as much investigated as lactate and CO2. The beta agonist isoprotenerol has been reported to produce laboratory PA’s resembling those induced by lactate (Rainey et al 1984) but this evidence is conflicting (Nesse et al 1984). Caffeine stimulates anxiety and caffeine intoxication at times may cause attacks similar to PA’s. PD patients have been shown to be more sensitive to caffeine than controls, but it seems that caffeine is rather an analogue of generalized anxiety than PD (Charney et al 1985).

Patients with PD are very sensitive to yohimbine, an alpha 2 antagonist. This substance induces bursts of anxiety with symptoms resembling true PA’s (Charney et al 1987). Yohimbine apparently triggers an hypersensitivity of adrenergic pathways in patients with a PD. However it is questionable whether yohimbine induced PA’s are as specific as those provoked by lactate or CO2.

Several serotonergic compounds may be anxiogenic (Bourin et al 1995). In particular, m-chlorophenylpiperazine (m-CPP) has been tested. Results are not clear cut: amongst other factors, the effect depends on the route of administration. Moreover, m-CPP may be anxiogenic in OCD as well. Fenfluramine was told to stimulate anxiety in panic (Targum and Marshall 1989). Again it is not clear whether this specifically applies to subjects with PD.

The neuropeptide cholecystokinin (CCK) has been shown to trigger PA’s in PD, which are similar to real life PA’s (Bradwejn et al 1991) CCK reliably reproduces the symptoms of a PA, and PD subjects are more sensitive to the challenge than subjects without real life PA’s (Bradwejn et al 1992). The effect of CCK is sensitive to antidepressants which are effective in the treatment of PD. Therefore, CCK meets a number of criteria for a valid laboratory model of PD, even though to date, its diagnostic specificity is not as well established as for CO2 (Bourin et al 1995). The CCK and 35% CO2 challenge have been compared, and found to be similar in their behavioral effects: both substances elicit, as well in healthy volunteers as in patients, the same type of panic-like symptoms, with a prominent respiratory symptomatology (Bradwejn and Koszycki 1991). Beside lactate and CO2, CCK is a most interesting approach to the experimental modeling of PD. It is striking to note that both CO2 and CCK yield marked stimulating effects on respiration (Bradwejn et al 1998)

7. The metabolic hypothesis

When it became clear that beside lactate, CO2 could trigger PA’s, it was hypothesized that not lactate, but CO2 was the key factor accounting for both panicogenic procedures. Although anionic compounds such as lactate do not pass the blood brain barrier, it was observed that some limited fenestrated areas do exist, through which some amount of lactate may penetrate into the central nervous system (CNS). Carr and Sheehan (1984) postulated that lactate passing via the above areas, may be oxidized intracellularly into water and CO2, resulting in a final intra- neuronal hypercapnia, contrasting with the peripheral metabolic alkalosis (Bradwejn et al 1998).

Bibliografía

Battaglia M, Perna G (1995): ROC Analyses of 35% CO2 test for panic disorder. Psychiatry research 29:111-119.
Bertani A, Perna G, Phd, Arancio C, Caldirola D, Bellodi L (1997): Pharmacologic effect of imipramine, paroxetine, and sertraline on 35% carbon dioxide hypersensitivity in panic patients: a double-blind, random, placebo-controlled study. Journal of clinical psychopharmacology 17:97-101.
Bonn JA, Harrison J, Rees WL (1971): Lactate-induced anxiety: therapeutic application. The British journal of psychiatry; the journal of mental science 119:468-70.
Bourin M, Malinge M, Guitton B (1995): Provocative agents in panic disorder. Therapie 50:301-6.
Bradwejn J, Koszycki D (1991): Comparison of the panicogenic effect of cholecystokinin 30-33 and carbon dioxide in panic disorder. Progress in neuro psychopharmacology and biological psychiatry 15:237-9.
Bradwejn J, Koszycki D, Payeur R, Bourin M, Borthwick H (1992): Replication of action of cholecystokinin tetrapeptide in panic disorder: clinical and behavioral findings. The American journal of psychiatry 149:962-4.
Bradwejn J, Koszycki D, Shriqui C (1991): Enhanced sensitivity to cholecystokinin tetrapeptide in panic disorder. Clinical and behavioral findings. archives of general psychiatry 48:603-10.
Bradwejn J, LeGrand JM, Koszycki D, Bates JH, Bourin M (1998): Effects of cholecystokinin tetrapeptide on respiratory function in healthy volunteers. The American journal of psychiatry 155:280-2.
Caldirola D, Perna G, Arancio C, Bertani A, Bellodi L (1997): The 35% CO2 challenge test in patients with social phobia. Psychiatry research 71:41-8.
Carr DB, Sheehan DV (1984): Panic anxiety: a new biological model. The journal of clinical psychiatry 45:323-30.
Charney DS, Honinger GR, Jatlow PI (1985): Increased anxiogenic effect of caffeine in PD. Arch. gen. Psychiatry 42:233-243.
Charney DS, Woods SW, Goodman WK, Heninger GR (1987): Neurobiological mechanisms of panic anxiety: biochemical and behavioral correlates of yohimbine-induced panic attacks. The American journal of psychiatry 144:1030-6.
Cohen ME, White PD (1951): Life situations, emotions and neurocirculatory asthenia. Proceeding of the Association for Research in Nervous and mental diseases 13:335-357I.
Coplan JD, Liebowitz MR, Gorman JM, et al (1992a): Noradrenergic function in panic disorder. Effects of intravenous clonidine pretreatment on lactate induced panic. biological psychiatry 31:135-46.
Coplan JD, Sharma T, Rosenblum LA, et al (1992b): Effects of sodium lactate infusion on cisternal lactate and carbon dioxide levels in nonhuman primates. The American journal of psychiatry 149:1369-73.
Cordes E (1872): Agoraphobia, a symptom of paratic exhaustion. Arch Psychiatrie 3:521-574.
Cowley DS, Stephen R, Dager R, McClennan J, Roy-Byrne PP, Dunner DL (1988): Response to lactate infusion in generalized anxiety disorders. biological psychiatry 24:409-414.
Da Costa JM (1871): An irritable heart, a clinical form of functional cardiac disorder and its consequences. Am. J. M. Sc. 61:17.
Dager SR, Marro KI, Richards TL, Metzger GD (1994): Preliminary application of magnetic resonance spectroscopy to investigate lactate-induced panic. The American journal of psychiatry 151:57-63.
Dager SR, Strauss WL, Marro KI, Richards TL, Metzger GD, Artru AA (1995): Proton magnetic resonance spectroscopy investigation of hyperventilation in subjects with panic disorder and comparison subjects. The American journal of psychiatry 152:666-72.
Drury AN (1920): The percentage of carbon dioxide in the alveolar air, and the tolerance to accumulating carbon dioxide in cases of so-called "irritable heart". Heart 7 105:1918-1920.
Faravelli C, Paionni N (2000): Panic disorder: clinical course, morbidity and comorbidity. In Griez E, Faravelli C, Nutt D, Zohar J (eds), The European Certificate Textbook of Anxiety Disorders: In press, John Wiley & Sons.
Fink M, Taylor MA, Volavka J (1969): Anxiety precipitated by lactate. The New England journal of medicine 281:1429.
Fyer AJ, Liebowitz MR, Gorman JM, Davies SO, Klein DF (1985): Lactate vulnerability of remitted panic patients. Psychiatry research 14:143-8.
Fyer MR, Uy J, Martinez J, et al (1987): CO2 challenge of patients with panic disorder. The American journal of psychiatry 144:1080-2.
Goisman RM, Warshaw MG, Peterson LG, et al (1994): Panic, agoraphobia, and panic disorder with agoraphobia. Data from a multicenter anxiety disorders study. The Journal of nervous and mental disease 182:72-9.
Gorman JM, Askanazi J, Liebowitz MR, et al (1984): Response to hyperventilation in a group of patients with panic disorder. The American journal of psychiatry 141:857-61.
Gorman JM, Fyer MR, Liebowitz MR, Klein DF (1987): Pharmacologic provocation of panic attacks. In Meltzer HY (ed), Psychopharmacology: the third generation of progress. New York: Raven Press, pp 985-993.

Gorman JM, Papp LA, Martinez J, et al (1990): High-dose carbon dioxide challenge test in anxiety disorder patients. biological psychiatry 28:743-57.
Griez E, de Loof C, Pols H, Zandbergen J, Lousberg H (1990a): Specific sensitivity of patients with panic attacks to carbon dioxide inhalation. Psychiatry research 31:193-9.
Griez E, Pols HJ, van den Hout MA (1987a): Acid-base balance in real life panic. Journal of affective disorders 12:263-6.
Griez E, van den Hout MA (1982): Effects of carbon dioxide-oxygen inhalations on subjective anxiety and some neurovegetative parameters. Journal of behavior therapy and experimental psychiatry 13:27-32.
Griez E, Zandbergen J, Lousberg H, van den Hout M (1988): Effects of low pulmonary CO2 on panic anxiety. Comprehensive psychiatry 29:490-7.
Griez E, Zandbergen J, Pols H, de Loof C (1990b): Response to 35% CO2 as a marker of panic in severe anxiety. The American journal of psychiatry 147:796-7.
Griez EJ, Lousberg H, van den Hout MA, van der Molen GM (1987b): CO2 vulnerability in panic disorder. Psychiatry research 20:87-95.
Guttmacher LB, Murphy DL, Insel TR (1983): Pharmacologic models of anxiety. Comprehensive psychiatry 24:312-26.
Hollander E, Liebowitz MR, Gorman JM, Cohen B, Fyer A, Klein DF (1989): Cortisol and sodium lactate-induced panic. archives of general psychiatry 46:135-40.
Hsiao JK, Bartko JJ, Potter WZ (1989): Diagnosing diagnoses. Receiver operating characteristic methods and psychiatry. archives of general psychiatry 46:664-7.
Karajgi B, Rifkin A, Doddi S, Kolli R (1990): The prevalence of anxiety disorders in patients with chronic obstructive pulmonary disease. The American journal of psychiatry 147:200-1.
Kelly D, Mitchell Heggs N, Sherman D (1971): Anxiety and the effects of sodium lactate assessed clinically and physiologically. The British journal of psychiatry; the journal of mental science 119:129-41.
Klein DF (1993): False suffocation alarms, spontaneous panics, and related conditions. An integrative hypothesis. archives of general psychiatry 50:306-17.
Lapierre YD, Knott VJ, Gray R (1984): Psychophysiological correlates of sodium lactate. psychopharmacology bulletin 20:50-7.
Ley R (1985): Agoraphobia, the panic attack and the hyperventilation syndrome. Behaviour research and therapy 23:79-81.
Liebowitz MR, Coplan JD, Martinez J, et al (1995): Effects of intravenous diazepam pretreatment on lactate-induced panic. Psychiatry research 58:127-38.
Liebowitz MR, Fyer AJ, Gorman JM, et al (1984): Lactate provocation of panic attacks. I. Clinical and behavioral findings. archives of general psychiatry 41:764-70.
Liebowitz MR, Fyer AJ, Gorman JM, et al (1985a): Specificity of lactate infusions in social phobia versus panic disorders. The American journal of psychiatry 142:947-50.
Liebowitz MR, Gorman JM, Fyer A, Dillon D, Levitt M, Klein DF (1986): Possible mechanisms for lactate's induction of panic. The American journal of psychiatry 143:495-502.
Liebowitz MR, Gorman JM, Fyer AJ, et al (1985b): Lactate provocation of panic attacks. II. Biochemical and physiological findings. archives of general psychiatry 42:709-19.
Lousberg H, Griez E, van den Hout MA (1988): Carbon dioxide chemosensitivity in panic disorder. Acta psychiatrica Scandinavica 77:214-8.
Lum LC (1975): Hyperventilation: the tip and the iceberg. Journal of psychosomatic research 19:375-83.
Markowitz JS, Weissman MM, Ouellette R, Lish JD, Klerman GL (1989): Quality of life in panic disorder. archives of general psychiatry 46:984-92.
McNally R (1994): Panic disorder. A critical analysis. New York, USA: The Guifford Press.
Mendlowicz MV, Stein MB (2000): Quality of life in individuals with anxiety disorders. The American journal of psychiatry 157:669-82
Nesse RM, Cameron OG, Curtis GC, McCann DS, Huber Smith MJ (1984): Adrenergic function in patients with panic anxiety. archives of general psychiatry 41:771-6.
Overbeek T, Vermetten E, Griez E (2000): Epidemiology of anxiety disorders. In Griez E, Faravelli C, Nutt D, Zohar J (eds), The European Certificate Textbook of Anxiety Disorders: In press, John Wiley & Sons.
Papp LA, Goetz R, Cole R, et al (1989): Hypersensitivity to carbon dioxide in panic disorder. The American journal of psychiatry 146:779-81.
Pelissolo A, Lepine JP (1998): Epidemiology of depression and anxiety disorders. In Montgomery SA, den Boer JA (eds), SSRI's in depression and anxiety: John Wiley and Sons ltd, pp 1-21.

Perna G, Barbini B, Cocchi S, Bertani A, Gasperini M (1995a): 35% CO2 challenge in panic and mood disorders. Journal of affective disorders 33:189-94.
Perna G, Battaglia M, Garberi A, Arancio C, Bertani A, Bellodi L (1994a): Carbon dioxide/oxygen challenge test in panic disorder. Psychiatry research 52:159-71.
Perna G, Bertani A, Arancio C, Ronchi P, Bellodi L (1995b): Laboratory response of patients with panic and obsessive-compulsive disorders to 35% CO2 challenges. The American journal of psychiatry 152:85-9.
Perna G, Bertani A, Diaferia G, Arancio C, Bellodi L (1994b): Prevalence of respiratory diseases in patients with panic and obsessive compulsive disorders. Anxiety 1:100-1.
Perna G, Bertani A, Gabriele A, Politi E, Bellodi L (1997): Modification of 35% carbon dioxide hypersensitivity across one week of treatment with clomipramine and fluvoxamine: a double-blind, randomized, placebo-controlled study. Journal of clinical psychopharmacology 17:173-8.
Perna G, Cocchi S, Bertani A, Arancio C, Bellodi L (1994c): Pharmacologic effect of toloxatone on reactivity to the 35% carbon dioxide challenge: a single-blind, random, placebo-controlled study. Journal of clinical psychopharmacology 14:414-8.
Perna G, Cocchi S, Bertani A, Arancio C, Bellodi L (1995d): 35% CO2 sensitivity in healthy first degree relatives of patients with panic disorder. The American journal of psychiatry 152:623-5.
Perna G, Cocchi S, Nobile M, Griez E (1993): A new biological marker for genetic studies in panic disorder: preliminary data on the 35% CO2 challenge test. Psych. Genetics 3 (suppl. 3) 184.
Perna G, Gabriele A, Caldirola D, Bellodi L (1995c): Hypersensitivity to inhalation of carbon dioxide and panic attacks. Psychiatry research 57:267-73.
Pitts FN, Allen RE (1979): Biochemical induction of anxiety. In Fann WE (ed), Phenomenology and Treatment of Anxiety. New York: Spectrum Publications.
Pitts FN, Jr. , McClure JN, Jr. (1967): Lactate metabolism in anxiety neurosis. The New England journal of medicine 277:1329-36.
Pols H, Lousberg H, Zandbergen J, Griez E (1993): Panic disorder patients show decrease in ventilatory response to CO2 after clomipramine treatment. Psychiatry research 47:295-6.
Pols H, Zandbergen J, de Loof C, Griez E (1991): Attenuation of carbon dioxide-induced panic after clonazepam treatment. Acta psychiatrica Scandinavica 84:585-6.
Pols HJ, Hauzer RC, Meijer JA, Verburg K, Griez EJ (1996): Fluvoxamine attenuates panic induced by 35% CO2 challenge. The journal of clinical psychiatry 57:539-42.
Rainey JM, Jr. , Pohl RB, Williams M, Knitter E, Freedman RR, Ettedgui E (1984): A comparison of lactate and isoproterenol anxiety states. Psychopathology 17 Suppl 1:74-82.
Regier DA, Farmer ME, Rae DS, et al (1990): Comorbidity of mental disorders with alcohol and other drug abuse. Results from the Epidemiologic Catchment Area (ECA) Study. Jama 264:2511-8.
Salvador Carulla L, Segui J, Fernandez Cano P, Canet J (1995): Costs and offset effect in panic disorders. The British journal of psychiatry. Supplement:23-8.
Sanderson WC, Wetzler S (1990): Five percent carbon dioxide challenge: valid analogue and marker of panic disorder? biological psychiatry 27:689-701.
Shavit RG, Gentil V, Mandetta R (1992): The association of panic-agoraphobia and asthma: contributing factors and clinical implication. General hospital Psychiatry 14:420-423.
Targum SD, Marshall LE (1989): Fenfluramine provocation of anxiety in patients with panic disorder. Psychiatry research 28:295-306.
Uhde TW, Tancer ME (1990): Chemical models of panic: a review and critique. In Tyrer P (ed), psychopharmacology of anxiety. London, Oxford: Oxford University Press.
van den Hout MA, Boek C, van der Molen GM, Jansen A, Griez E (1988): Rebreathing to cope with hyperventilation: experimental tests of the paper bag method. Journal of behavioral medicine 11:303-10.
van den Hout MA, Griez E (1982): Cardiovascular and subjective responses to inhalation of carbon dioxide. A controlled test with anxious patients. Psychotherapy and psychosomatics 37:75-82.
Van den Hout MA, Griez E (1984): Panic symptoms after inhalation of carbon dioxide. The British journal of psychiatry; the journal of mental science 144:503-7.
Verburg C, Griez E, Meijer J (1994): A 35% carbon dioxide challenge in simple phobias. Acta psychiatrica Scandinavica 90:420-3.
Verburg K, Griez E, Meijer J, Pols H (1995): Respiratory disorders as a possible predisposing factor for panic disorder. Journal of affective disorders 33:129-34.
Verburg K, Klaasen T, Pols H, Griez E (1998c): Comorbid depressive disorder increases vulnerability to the 35% carbon dioxide (CO2) challenge in panic disorder patients. Journal of affective disorders 49: 195-201
Verburg K, Perna G, Bellodi L, Griez E (1998b): The 35% panic provocation challenge as a diagnostic test for panic disorder. In Bellodi L, Perna G (ed), The panic respiration connection. Milan: Medical media Srl.

Verburg K, Pols H, de Leeuw M, Griez E (1998a): Reliability of the 35% carbon dioxide panic provocation challenge. Psychiatry research 78:207-14.
Weissman MM (1991): Panic disorder: impact on quality of life. The journal of clinical psychiatry 52 Suppl:6-8; discussion 9.
Weissman MM, Bland RC, Canino GJ, et al (1997): The cross-national epidemiology of panic disorder. archives of general psychiatry 54:305-9.
Westphal C (1872): Agoraphobia: a neuropathic phenomenon. Arch. Psychiatrie Nervenkrankheiten 3:138-161.
Wittchen HU, Essau CA (1993): Epidemiology of panic disorder: progress and unresolved issues. Journal of psychiatric research 27 Suppl 1:47-68.
Wolpe (1973): The practice of behavior therapy (2nd edition): Oxford Pergamon.
Woods SW, Charney DS, Loke J, Goodman WK, Redmond DE, Jr. , Heninger GR (1986): 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. archives of general psychiatry 43:900-9.
Zambler T, Katon W (1998): Panic disorder in the general medical setting. Journal of Psychosomatic Research 44:25-42.
Zandbergen J, Lousberg HH, Pols H, de Loof C, Griez EJ (1990): Hypercarbia versus hypocarbia in panic disorder. Journal of affective disorders 18:75-81.
Zandbergen J, Pols H, de Loof C, Griez EJ (1991): Ventilatory response to CO2 in panic disorder. Psychiatry research 39:13-9