Please, support this site by visiting our sponsors!
Ad Placement By NarrowCast Media... Get Noticed!

Etiology of panic disorder: biological aspects

The essential assumption of the biomedical model is that mental illness is basically a biological disease. In other words, the etiology of the mental disorder can be explained by physical causes, such as infections, genetics, neuroanatomic pathology or malfunctioning biochemistry. Evidence has accumulated, demonstrating that PD may be viewed as a biological disease.

Panic symptoms are physical in nature

The onset of symptoms is abrupt, reaching peak within minutes. It rapidly subsides, leaving only residual anxiety. Typically, the PA come out of nowhere, even though some patients experience situationally bound and situationally predisposed PA. Often, patients are unable to identify anything that could possibly trigger a PA. They experience PA watching cartoons, playing with their children, resting etc., simply put, in situations that do not present any stress or threat to them.

The physical nature of the symptoms of panic attack also provides some reasons for the claim that PA/ PD involves biological modifications. The PA symptoms correspond to a great degree to symptoms of acute activation of the sympathetic branch of the autonomous nervous system, typical for the fight or flight reaction. Indeed, several authors argue that panic attack is, in fact, a fight or flight reaction of the body in absence of a real danger (Rosenhan & Seligman, 1989). When confronted with a real or perceived threat, the automatic "fight or flight" response may be triggered to prepare the body for immediate action. This response is accompanied by peripheral secretion of catecholamines, especially epinephrine and norepinephrine, and glucocorticoids (Carlson, 1992). These hormones increase the availability of the body's energy by glycogenolysis in liver and skeletal muscles thus raising the blood glucose and lactate, lipolysis in adipose tissue, mobilization of free fatty acids, and by increasing temperature. Both epinephrine and norepinephrine also dilate coronary blood vessels. As a consequence, the rate and strength of the heartbeat increases to supply more oxygen to the tissues. While norepinephrine produces vasoconstriction in skin, mucosa, skeletal muscles and most other organs, epinephrine dilates veins in skeletal muscles. These effects result in hypertension and consequently in reflex bradycardia. Other symptoms of a sympatho-adrenergic stimulation involve modifications of breathing, increased temperature, localized sweating, decreased motility and tone of stomach and intestine, constrictions of sphincters in stomach and intestine as well as piloerection. Breathing increases in rate and depth to exchange more oxygen to prepare for exertion. Breathlessness, dizziness, and pain or tightness in the chest may be experienced. Sweat glands are stimulated to prevent overheating. The pupils of the eye dilate to admit more light and increase peripheral vision to scan for danger. Sensitivity to bright light, and visual disturbances may occur. The digestive system shuts down to conserve blood for the muscles. A dry mouth, nausea and constipation may result. Muscles tense to prepare for flight, but may cause spasms and trembling when action is not taken. This complex response was developed through evolution in many organisms and normally serves for survival and protection. As mentioned above, symptoms of sympathetic activation and symptoms of panic attack share many common symptoms. Therefore, panic attack may be viewed as an emergency response which occurs in a situation where it is not appropriate (Barlow & Craske, 1994).

Antidepressant and antipanic drugs are efficacious in treatment of PD

Another argument for the biological hypothesis of PD is that pharmacotherapy is efficacious in treatment of PD. Several classes of drugs are being used in PD patients, namely the benzodiazepines, tricyclic and heterocyclic antidepressants, monoamine oxidase (MAO) inhibitors, reversible MAO inhibitors and selective serotonin reuptake inhibitors (SSRI). Except for benzodiazepines, all anti-panic drugs are in fact antidepressants, and act on the aminergic systems (Taylor & Arnow, 1988). Benzodiazepines act on the GABA-receptors. If PD had no biological bases, its symptoms could not be alleviated by medication. Whatever the type of medication, efficacy of drugs to treat PD implies that the underlying mechanism of development or symptomatology is biological.

Frequency and intensity of PA varies during menstrual cycle and pregnancy

Clinical and scientific evidence exists demonstrating that gonadal hormones have a strong influence on PD, especially in terms of frequency and intensity of PA. Spontaneous panic attacks rarely start before puberty or after menopause, suggesting that, in women, occurrence of PA may be linked to production of female reproductive hormones (Klein et al., 1992). Premenstrual exacerbation of panic symptoms has been documented (Breier et al., 1986; Cameron et al., 1988). Several authors reported that women with LLPDD are more sensitive to panic-provocation procedures (Harrison et al., 1989; LeMellédo et al., 1996). In addition, panic rate in women with or without LLPDD increases when they are challenged during the luteal phase, the LLPDD patients having a higher rate (Sandberg et al., 1993; LeMellédo et al., 1996). This phenomenon is attributed to a drop in progesterone levels before the onset of menses, the women with largest progesterone fluctuation being most vulnerable (Halbreich et al., 1986).

Clinically, a marked decrease of panic has been observed during pregnancy and lactation, with postlactational exacerbation of symptoms. These changes most likely reflect increased levels of progesterone, estrogen and oxytocin during pregnancy or lactation (Klein, 1993). The fact that the condition of PD patients improves during this time is a strong argument for the biological view of PD. As Klein points out, pregnancy and childbirth present an increased vulnerability, marked by heightened threatening endogenous stimuli. According to cognitive theories, which postulate that PA result from catastrophic interpretation of physiological changes, such states should make patients more prone to panic. Apparently this is not the case (Klein, 1994).

Experimental procedures (challenge) reproduce panic attacks in laboratory

For nearly three decades, researchers have been using various procedures in order to reproduce the emotional, cognitive, physiological and neurochemical changes accompanying panic attacks. Among the first agents used to trigger anxiety-like symptoms were epinephrine and norepinephrine (Wearn & Sturgis, 1919; Lindemann, 1935; Lindemann & Finesinger, 1938). Cholinergic agents, such as cholinomimetic mecholyl and cholinesterase inhibitor physostigmime, were also used in several studies (Lindemann & Finesinger, 1938; Risch et al., 1981; Paul & Skolnick, 1981) One of the most researched panic-provoking pharmacological agents is sodium lactate (Pitts & McClure, 1967; Haslam, 1974; Appleby et al., 1981; Liebowitz et al., 1984). Voluntary hyperventilation and carbon dioxide have frequently been used to study the underlying mechanisms of panic attack (Van den Hout & Griez, 1984; Gorman et al., 1984; Papp et al., 1989). Caffeine challenge induces anxiety-like symptoms suggesting a possible implication of the adenosine system in panic anxiety (Charney et al., 1984a; Uhde, 1990; Boulenger et al., 1984). The administration of cholecystokinin tetrapeptide (CCK4) has also been used in several recent studies (Bradwejn and Koszycki, 1994a; Bradwejn & Koszycki, 1994b). Several other panic-provocation agents, such as yohimbine, isoproterenol, piperoxan act on the noradrenergic or adrenergic systems (Olpe et al., 1983, Charney et al., 1987; Charney et al., 1990; Pohl et al., 1990).

These procedures are a valuable tool for the experimental evaluation of neurochemical correlates of panic attack symptoms. They are capable of inducing experience that is phenomenologically similar to spontaneous panic attacks, as pointed out by panic patients. Therefore, a phenomenon which can be reproduced by pharmacological means must have biological bases.

Non-fearful and limited symptom panic attacks

Many panic patients report experiencing so called limited symptom panic attacks, which are characterized by presence of less than four symptoms, and, importantly, absence, or low levels of anxiety. Limited symptom panic is often seen in patients undergoing pharmaco or psychotherapy. Some patients experience panic attacks without fear, which may contain several physical symptoms without the emotional and cognitive components of PA . The mere existence of this phenomenon points to biological bases of PD.

PA may be triggered or aggravated by use of drugs

Many patients can trace the onset of panic attacks to the use of drugs, especially cocaine and amphetamines. Both of these drugs alter the noradrenergic function (Taylor & Arnow, 1988). The fact that drugs can trigger panic attacks and bring about the onset of panic disorder is yet another argument for the biological bases of PD.

Animal models of anxiety and PD

The existence and validity of animal models of anxiety and panic form another argument for the biological nature of PD. These models, mainly using rodents and non-human primates, parallel human anxiety. Despite their inherent limitations, animal models of anxiety have been repeatedly proven to be useful in testing of anti-anxiety and anti-panic drugs. They are used to study neurochemical, especially central, changes in anxiety states, taking advantage of techniques such as microdialysis, single neuron recording, electro-chemical stimulation of various brain regions etc. (File, 1990).

Specific brain regions are implicated in regulation of panic anxiety

The complex nature of symptoms of PA suggests that various brain regions would be implicated. Number of techniques have been used in order to provide an explanation for panic attacks, including brain imaging, staining, electrical and chemical stimulation as well as electrical recording. Brain imaging techniques are useful tools that can provide us with information about the brain regions with higher or lower oxygen or glucose metabolism (Huang et al., 1981, Raichle et al., 1976), cerebral blood flow (Herscovitch et al., 1983), cerebral blood volume (Grubb et al., 1978), BBB permeability (Herscovitch et al., 1987) and other indices indicating activated areas (Reiman, 1990). Several studies showed apparent region-specific modifications of cerebral blood flow during panic attack (Stewart et al., 1988; Reiman et al., 1986). Alterations of the permeability of the blood-brain barrier, which is directly regulated by afferents originating in the locus ceruleus, have been linked to the development and treatment of panic disorder (Preskorn et al., 1980; Raichle, 1983).

The limbic system, especially the amygdala, has long been considered to be directly implicated in anxiety and other emotions. Amygdala receives projections from frontal cortex, association cortex, temporal lobe, olfactory system and other parts of the limbic system. It sends its afferents to frontal and prefrontal cortex, orbitofrontal cortex, hypothalamus, hippocampus as well as brain stem nuclei, such as locus ceruleus and raphé nucleus. Amygdala and its central nucleus thus communicate with many brain regions, including those that control breathing, motor function, autonomic response, release of hormones as well as processing of interoceptive and external information (Carlson, 1992). Amygdala is thus in a good position to modulate autonomic responses related to anxiety and panic because of its connections with the brain stem and the reticular formation, both of which control vegetative functions.

Indeed, numerous studies have demonstrated an implication of limbic system, and amygdala in particular, in PD. Halgren et al. (1978) electrically stimulated the amygdala and hippocampus in humans which resulted in somatic and emotional symptoms of panic attack. In animals, Iwata et al. (1987) observed increases in heart rate and blood pressure, symptoms of sympathetic activation that are also present during a panic attack, after injections of excitatory amino acids into central nucleus of amygdala. Microinjections of benzodiazepines into amygdala had "anti-conflict" properties that are correlated with anxiolytic effects in humans (Hodges et al., 1987; Kuhar, 1986). In addition to this, microinjections of CCK8 (both sulfated and non-sulfated) into the amygdaloid nucleus produce fear-motivated behavior in rats, such as facilitation of extinction of active avoidance behavior and retention of passive avoidance (Fekete et al., 1984).

Locus ceruleus is a particularly important region related to anxiety. This region is a metencephalic nucleus located in the caudal pontine central grey. It contains 50% of all brain noradrenergic neurons and is composed almost exclusively of 12 000 noradrenergic neurons on each side of brain. It also produces a major portion of norepinephrine in the central nervous system. Collateral branches of axons of noradrenergic neurons project to most regions of the brain. Of those numerous projections, there are many that have been associated with panic disorder or panic attacks: with the limbic system, especially the amygdala, hippocampus, septum, and cingulate cortex, all cortices, brain stem, reticular formation, cerebellum and spinal cord (Cooper et al., 1991).

Evidence from lesion, electrical and chemical stimulation, and single-unit recording studies suggests that locus ceruleus seems to be implicated in the sleep-wake cycle, arousal, anxiety and fear (Redmond & Huang, 1979; Redmond et al., 1976). In addition, most agents that alleviate anxiety (benzodiazepines, alcohol, opiates, barbiturates) act also to lower the activity of locus ceruleus (Nybäck et al., 1975; Geyer & Lee, 1984; Huang, 1979). Locus ceruleus also contains benzodiazepine receptors, as well as receptors for endogenous opiates. During syndrome of withdrawal from benzodiazepines, opiates and alcohol, anxiety increases as does the activity of locus ceruleus, both lasting as long as the withdrawal symptoms persist.

Other important brain regions appear to be implicated in modulation of anxiety. Hypothalamus, pituitary gland, especially anterior pituitary gland are involved in synthesis and release of numerous stress-related hormones. Numerous brain stem regions, namely pons, medulla oblongata, cerebellum, reticular formation, periaqueductal gray matter, are also involved, especially in functions such as perception of somatic and sensory stimuli, fear-related reflexes, arousal, and neuro-vegetative functions. Cerebral cortex is implicated in anxiety control and development in terms of storage of memory, cognitive processes and control of motor movement (Carlson, 1992; Taylor & Arnow, 1988).

Genetic and family studies

Clinical experience with patients revealed that PD seems to run in families and have a genetic component. These findings led to epidemiological studies investigating the incidence of this disorder in families of the patients. Despite methodological differences and variations in the definitions of the disorder as well as the target population of the anxious patients, the results seem to be relatively consistent. Carey and Gottsman (1981) studied families of probands with anxiety disorders and found that 15% of the first-degree relatives also suffered from anxiety disorders. A more pertinent study by Crowe et al. (1983) focused on panic disorder. Around 25% of the first-degree relatives of PD patients received the same diagnosis, as compared to 2.3% of relatives of normal controls.

Studies with twins who grew up together can also provide us with a useful piece of information. In Slater and Shields' study (1969), monozygotic twins had concordance rate of 41% for anxiety states, whereas the concordance among dizygotic was only 4%. Torgersen (1983, 1990) investigated concordance rates for anxiety disorders with panic attacks and found that 31% of the monozygotic twins had a similar diagnosis compared to 0% of the dizygotic twins. When he narrowed down the comparison to PD with agoraphobia, the concordance rate between monozygotic twins was 15%. Even though the differences in concordance rates might appear important, they might be misleading. First of all, the number of subjects was small, such that, for example, the concordance rate of 31% in monozygotic twins was based on 4 out of 13 pairs of twins, which is obviously not enough to generalize. Secondly, the higher concordance in monozygotic twins could be potentially explained by other non-genetic factors. For instance, monozygotic twins may be treated differently by their parents, extended families and peers. They might have more profound identity crisis than the one that teenagers usually go through. Often they are dealt with as an entity rather than two separate individuals. In addition to this, they might tend to develop mutual dependency and have more experiences of separation anxiety, a state that seems to be related to agoraphobia and panic disorder.

Neurochemistry of PD

The evidence for neurochemical pathology in PD comes from numerous sources: challenge studies, effects of antipanic medication, biochemical comparisons of PD population with healthy subjects in terms of reactivity and basal levels, brain imagery and animal experiments. Several major hypotheses, explaining the neurochemical bases of PD, have been formed and supported by evidence.

One of the most intriguing hypotheses postulates an abnormality of the noradrenergic and adrenergic systems. Increased plasmatic and urinary concentrations of epinephrine (EPI) and norepinephrine (NE) in panic disorder patients have been shown in some but not all studies (Braune et al., 1994; Butler et al., 1992; Villacres et al., 1987; Nesse et al., 1985a; Appleby et al., 1981; Wyatt et al., 1971; Cameron et al., 1984). In addition, augmentations of plasma 3-methoxy-4-hydroxyphenylethylene (MHPG), a metabolite of NE, have been documented in patients with frequent and severe panic attacks (Charney et al., 1984b). Panic patients confronted with anxiogenic situations have increased plasma free MHPG and NE levels (Braune et al., 1994; Ko et al., 1983; Uhde et al., 1982; Nesse et al., 1985b). Stimulation of the noradrenergic system by alpha2-adrenoceptor antagonist yohimbine and beta-adrenoceptor agonist isoproterenol produces panic-like symptoms in PD patients and some healthy subjects (Charney et al., 1987; Charney et al., 1990; Pohl et al., 1990). Pathological changes in the alpha or/and beta-receptors have been demonstrated (Charney & Heninger, 1986; Rainey et al., 1984; Nesse et al., 1984; Pohl et al., 1985).

Another plausible hypothesis concerns serotonergic system, especially in terms of interaction with noradrenergic system (Zacharko et al., 1995). Raphé nucleus, a midbrain structure with high concentration of serotonergic neurons, projects to locus ceruleus, and has an inhibitory influence on the activity of noradrenergic neurons (Meltzer, 1987). Pharmacological agents that decrease serotonergic activity have anxiolytic effect in animals (Briley et al., 1990). Serotonin and its metabolite 5-HIAA are reduced in anxious dogs (Guttmacher et al., 1983). In humans, alleviation of symptoms is achieved by administration of selective serotonin reuptake inhibitors. Murphy & Pigott (1990) have presented evidence suggesting that the anxiolytic effects of benzodiazepines might also be related to serotonergic activity. In addition to it, panic disorder patients reported an exacerbation of symptoms when they received serotonin precursors tryptophan and 5-HTP, serotonin receptors' agonist m-chlorophenylpiperazine and flenfluramine, a drug that increases the synaptic availability of serotonin (Murphy & Pigott, 1990, Den Boer & Westenberg, 1990, Targum, 1990, Kahn & Van Praag, 1988). It is thus possible that an altered serotonergic transmission is one of the elements that are implicated in anxiety and panic.

Another major hypothesis for PD etiology involves benzodiazepine receptors and their natural ligands. The anxiolytic action of benzodiazepines is mediated through benzodiazepine receptor complex, potentiating the inhibitory effects of GABA (Lima, 1991; Paul & Skolnick, 1981; Skolnick & Paul, 1982). Sensitivity of central and peripheral benzodiazepine receptors have been shown to be modified by aversive life events and social variables (Trullas & Skolnick, 1993). Studies have demonstrated that animals with low exploratory behavior (anxious) have lower density of brain benzodiazepine receptors (Rago et al., 1991). Also, stimulation of benzodiazepine receptors by their inverse agonists, beta-carbolines, produces anxiety and panic-like symptoms in PD patients and healthy subjects (Zacharko et al., 1995).

Adenosine system also appears to be implicated in PD. Numerous studies have demonstrated that PD patients are hypersensitive to the effects of caffeine, adenosine antagonist, and often spontaneously reduce intake. In large doses, caffeine can produce panic-like symptoms in PD patients and healthy subjects, especially those with low regular consumption of caffeine (Boulenger et al., 1984; Uhde, 1990). Caffein-induced panic is typically accompanied by increases in plasma lactate, glucose and cortisol (Orlikov & Ryuzov, 1991).

Other neurotransmitters and neuromodulators appear to be implicated in PD. For example, dopamine-containing mesocorticolimbic system appears to be implicated in anticipation, conditioning and motivation, and contains neurons with high concentrations of various neuropeptides, including those associated with arousal (enkephalines), anxiogenesis (beta-carbolines) and anxiolysis (Zacharko et al., 1995). Most or some central dopaminergic systems respond to MAO inhibitors, stress and anxiogenic beta-carbolines (Cooper et al., 1991). Recently, numerous neuropeptides have been linked to mediation and control of anxiety. Cholecystokinin peptides, neuropeptide Y, beta-carbolines, enkephalines, substance P, corticotropin releasing factor might have modulatory effects on panic and anxiety, to name just a few. (Zacharko et al., 1995)

Nocturnal Panic Attacks

Nearly 70% of PD patients report experiencing a panic in their sleep at some point of their lives, and about one third experience recurrent sleep panic (Mellman & Uhde, 1989a; Stein et al., 1993). Sleep panic attacks appear to emerge from non-REM sleep, especially during the transition to early delta sleep (Mellman & Uhde, 1989b; Mellman & Uhde, 1989c). Therefore, sleep panic does not appear to be provoked by nightmares. Aside from nocturnal panic attacks, insomnia and restless sleep are among common complaints in PFD patients. Some studies suggest that PD patients display a moderate reduction in REM latency, decreased REM density, increased eye movement time, and report more frequent awakenings because breading discomfort (Stein et al., 1993; Mellman & Uhde, 1989b; Uhde et al., 1984).


Mental and Emotional Health

Please, support this site by visiting our sponsors!
Ad Placement By NarrowCast Media... Get Noticed!

Copyright© 1996, Plumeus Inc.