CHAPTER 26

Anxiety Disorders

Isabelle M. Rosso, Ph.D.

Dan J. Stein, M.D., Ph.D.

Scott L. Rauch, M.D.

Although anxiety has long held a central place in theories of psychopathology, it has only recently been appreciated that the anxiety disorders are the most prevalent category of mental illness in the United States (Kessler et al. 2010) and that they account for approximately a third of the country’s total mental health costs (Lepine 2002). Furthermore, although it has long been recognized that specific neurological lesions may lead to anxiety symptoms (Von Economo 1931), only in recent decades have advances in research allowed specific neuroanatomical hypotheses to be proposed for each of the anxiety disorders.

DSM-III (American Psychiatric Association 1980) provided significant impetus to research on anxiety disorders by replacing the category of “anxiety neurosis” with several different conditions and by providing each with operational diagnostic criteria. DSM-IV-TR (American Psychiatric Association 2000) anxiety disorders include panic disorder with and without agoraphobia, social phobia (social anxiety disorder), generalized anxiety disorder (GAD), posttraumatic stress disorder (PTSD), obsessive-compulsive disorder (OCD), substance-induced anxiety disorder, and anxiety disorder due to a general medical condition. The DSM-5 (American Psychiatric Association 2013) section on anxiety disorders no longer includes PTSD (which is found in the section on trauma- and stressor-related disorders) or OCD (which is included within the section on obsessive-compulsive and related disorders). For the purposes of this volume, we use the current DSM-5 classification and also retain inclusion of PTSD.

In each of the anxiety disorders, it is possible to discern a component comprising anxiety symptoms and a component comprising avoidance symptoms. In GAD, patients have anxiety about the future, and worry may serve as an avoidance behavior. In panic disorder, the anxiety symptoms are those of the panic attack, a discrete period of anxiety that develops rapidly, often spontaneously. The individual also may develop agoraphobia symptoms, or avoidance of those stimuli that appear to promote panic attacks. In social anxiety disorder, panic attacks develop only in the context of performance or other social situations in which the person fears embarrassment or humiliation. As a result of these fears, the person may avoid these situations. In PTSD, in the aftermath of a traumatic event, the person has intrusive experiences, hyperarousal symptoms, negative alterations in cognition and mood, and a range of avoidance and numbing symptoms.

In this chapter, we review these developments in our understanding of the anxiety disorders from a neuropsychiatric perspective. The neurochemistry and neuroanatomy of each of the main anxiety disorders and PTSD are considered first. Neurological disorders that may manifest with anxiety symptoms are then discussed, and future directions in the neuropsychiatry of anxiety disorders are considered briefly.

Primary Anxiety Disorders

In the following sections, we consider the neuropsychiatry of each of the major primary anxiety disorders (i.e., the “idiopathic” psychiatric disorders in which anxiety is the defining feature). Each section begins by sketching a simple neurochemical and neuroanatomical model of the relevant anxiety disorder. This sketch is then used as a framework for attempting a more complex integration of animal data, clinical biological research (e.g., pharmacological probe studies), and morphometric brain imaging studies. Although much remains to be learned about the neurobiology of the anxiety disorders, there is a growing consolidation of different avenues of information, with increasingly specific models now existing for each of the major anxiety disorders.

Generalized Anxiety Disorder

The term “generalized anxiety disorder” was first introduced in DSM-III, where it represented a refinement of the earlier concept of “anxiety neurosis.” In DSM-III, GAD was viewed as a residual diagnosis, to be made in the absence of other disorders. More recent editions of DSM have, however, increasingly emphasized the cognitive symptoms of GAD and have also emphasized that GAD is an independent entity that may be found alone or comorbid with other anxiety and mood conditions. GAD is the least common anxiety disorder in specialty anxiety clinics but the most common anxiety disorder in primary care practice (Kessler 2000).

Neuroanatomical models of GAD have not been well delineated to date. However, it may be speculated that GAD involves abnormalities of the amygdala and its connections with various prefrontal cortical regions, some of which may be shared across anxiety disorders and others which may reflect core GAD symptoms of excessive worrying and planning. In reviewing research relevant to this speculative model, we consider first neurochemical studies and then neuroanatomical findings.

Neurochemical Studies

Dysregulation of serotonin neurotransmitter systems is implicated in GAD. Indeed, there is substantial evidence that serotonergic compounds are effective in the pharmacotherapy for GAD; buspirone, a serotonin type 1A (5-HT_1A) receptor partial agonist, is effective in some studies, and growing evidence now shows the efficacy of the selective serotonin reuptake inhibitors (SSRIs) and serotonin-norepinephrine reuptake inhibitors (SNRIs) in this disorder. As a result, most current treatment guidelines recommend SSRIs and SNRIs as first-line pharmacologic agents in GAD (Koen and Stein 2011).

The norepinephrine sympathetic nervous system is sensitive to stress and anxiety, and it has been implicated in GAD (Nutt 2001). In clinical studies of GAD, increased plasma norepinephrine and 3-methoxy-4-hydroxy-phenylglycol (MHPG) and reduced platelet α₂-adrenergic peripheral receptor binding sites have been reported, although not all studies of static noradrenergic measures have produced consistent findings. Administration of more dynamic adrenergic probes has, however, indicated reduced adrenergic receptor sensitivity in GAD, perhaps an adaptation to high levels of circulating catecholamines (Nutt 2001). The locus coeruleus system may well play a regulatory role in GAD, even if it is not the sole dysfunctional neurochemical system in the disorder. Indeed, dual SNRIs have been shown to be effective in GAD. The locus coeruleus system projects to the amygdala and to other structures involved in anxiety responses, so that noradrenergic involvement is not inconsistent with the neuroanatomical model outlined earlier.

Involvement of the γ-aminobutyric acid (GABA)–benzodiazepine receptor complex in GAD is supported by several studies, and a review of this literature found reduced GABA_A receptors in temporocortical areas (Nikolaus et al. 2010). GABA is the brain’s predominant inhibitory neurotransmitter, and GABAergic pathways are widely distributed; nevertheless, the distribution of GABA and benzodiazepine receptors is particularly dense in limbic and paralimbic areas. Clinical studies have shown that benzodiazepines have efficacy comparable to the SSRIs and venlafaxine in the treatment of GAD (Koen and Stein 2011). At the same time, long-term use of benzodiazepines can be associated with withdrawal symptoms and rebound anxiety, such that most treatment guidelines do not recommend them as a first-line pharmacotherapy in GAD (Koen and Stein 2011).

Neuroanatomical Studies

Neuroimaging research on GAD remains at a relatively early stage. Nevertheless, findings are arguably consistent with involvement of limbic, paralimbic, and prefrontal regions. Patients with GAD show hypoactivation of certain prefrontal regulatory regions that are involved in spontaneous emotion regulation and conflict adaptation, including the ventral cingulate cortex and dorsomedial prefrontal cortex (Etkin 2010). In contrast, patients display hyperactivation of lateral prefrontal areas, which may reflect compensatory engagement of worrying (Etkin et al. 2010). Functional neuroimaging studies of amygdalar activity have been inconsistent, with some reports of amygdala hypoactivation during processing of threatening stimuli and other reports of amygdala hyperactivation of this region (Etkin 2010). In a recent meta-analysis of voxel-based morphometry studies, DSM-5 anxiety disorders, including GAD, were associated with reduced gray matter volume in right ventral anterior cingulate cortex and left inferior frontal gyrus (Shang et al. 2014).

Preliminary imaging data on receptor binding in GAD are also available. In a review of receptor binding studies in anxiety disorders, GAD was uniquely associated with reduced temporocortical GABA_A receptors, and GAD shared with all other anxiety disorders reductions in mesencephalic and cingulate serotonin 1A (5-HT_1A) receptors, striatal dopamine type 2 (D₂) receptors, and cortical GABA_A receptors (Nikolaus et al. 2010). This suggests a role for temporolimbic regions and the GABA-benzodiazepine receptor complex in mediating core GAD symptoms. Nevertheless, serotonergic neurons branch widely throughout the brain, affecting each of the main regions postulated to mediate anxiety symptoms (Figure 26–1).

FIGURE 26–1. Neuroanatomical model of generalized anxiety disorder (GAD).

There is evidence of hypofunction and reduced volumes of ventral anterior cingulate cortex (vACC) and inferior frontal gyrus (IFG), and hypofunction of dorsomedial prefrontal cortex (dmPFC). In contrast, areas of lateral prefrontal cortex (LPFC) are hyperactive in GAD. The amygdala (AMY) has been reported to be hypoactive or hyperactive across studies.

Panic Disorder and Agoraphobia

Panic disorder is a highly prevalent disorder, with rates fairly similar across different social and cultural settings. It is now well recognized that panic disorder is associated with significant morbidity (mood, anxiety, and substance use disorders) and also with severe impairments in occupational and social functioning. Indeed, a growing pharmacoeconomic literature has emphasized the personal and financial costs of panic disorder; this is a serious disorder that has a substantial negative effect on quality of life. DSM-5 has separated panic disorder from agoraphobia, such that each is an independent diagnosis with its own criteria. The co-occurrence of panic disorder and agoraphobia are now coded as two comorbid diagnoses, eschewing the DSM-IV diagnoses of panic disorder with agoraphobia, panic disorder without agoraphobia, and agoraphobia without a history of panic disorder. Moreover, panic attacks can be listed as a specifier to any and all DSM-5 disorders.

Over the past decade or two, models of panic disorder have become increasingly sophisticated (Gorman et al. 2004). The current neuroanatomical models of panic disorder (Figure 26–2) emphasize the following: 1) afferents from viscerosensory pathways to thalamus to the lateral nucleus of the amygdala, as well as from thalamus to cortical association areas to the lateral nucleus of the amygdala; 2) the extended amygdala, which is thought to play a central role in conditioned fear (Ciocchi et al. 2010); 3) the hippocampus, which is thought crucial for conditioning to the context of the fear (and so perhaps for phobic avoidance) (Alvares et al. 2012); and 4) efferent tracts from the amygdala to the hypothalamus and brain stem structures, which mediate many of the symptoms of panic. Thus, efferents of the central nucleus of the amygdala include the lateral nucleus (autonomic arousal and sympathetic discharge) and paraventricular nucleus (increased adrenocorticoid release) of the hypothalamus and the locus coeruleus (increased norepinephrine release), parabrachial nucleus (increased respiratory rate), and periaqueductal gray (defensive behaviors and postural freezing) in the brain stem. This kind of general outline can be used as a starting framework for considering the range of data relevant to the neurobiology of panic disorder.

FIGURE 26–2. Neuroanatomical model of panic disorder.

There is substantial evidence of hyperactivation of the amygdala (AMY) and insular cortex (INS), as well as preliminary evidence of reduced activity of ventral anterior cingulate cortex (vACC) and reduced volume of orbitofrontal gyrus (OFG) in panic disorder.

Neurochemical Studies

Early animal studies found that the locus coeruleus plays a key role in fear and anxiety, with both electrical and pharmacological stimulation resulting in fear responses. The locus coeruleus contains the highest concentration of noradrenergic-producing neurons in the brain. Viscerosensory input reaches the locus coeruleus via the nucleus tractus solitarius and the medullary nucleus paragigantocellularis, and the locus coeruleus sends efferents to a range of important structures, including the amygdala, hypothalamus, and brain stem periaqueductal gray (Nutt 2001).

Several clinical studies of panic disorder provide support for the role of the locus coeruleus; administration of yohimbine, for example, resulted in greater increases in MHPG in panic disorder patients than in control subjects without panic disorder. However, not all studies have replicated such findings, and studies of noradrenergic function in lactate-induced panic also have been inconsistent (Gorman et al. 2004), suggesting that additional neurochemical factors are important in the mediation of panic attacks.

Certainly, increasing evidence indicates that the serotonergic system plays a crucial role in panic disorder. Multiple lines of evidence support this; for example, several studies have found that m-chlorophenylpiperazine (m-CPP) administration leads to an acute exacerbation of panic symptoms in panic disorder patients (Klein et al. 1991; van der Wee et al. 2004). In addition, a good deal of evidence supports the efficacy of the SSRIs in panic disorder; fluoxetine, paroxetine, and sertraline all have received U.S. Food and Drug Administration (FDA) approval for use in panic disorder. They are as effective as and better tolerated than older agents including tricyclics and monoamine oxidase inhibitors (MAOIs) (Koen and Stein 2011). Benzodiazepines also are effective in treating panic disorder (Gorman et al. 2004; Koen and Stein 2011). Alprazolam, clonazepam, diazepam, and lorazepam are FDA approved to treat panic disorder (Koen and Stein 2011). However, their side-effect profile makes them a less preferred option to serotonin agents.

The serotonergic system interacts at several points with neuroanatomical structures thought to be important in panic disorder. First, serotonergic projections from the dorsal raphe nucleus generally inhibit the locus coeruleus, whereas projections from the locus coeruleus stimulate dorsal raphe nucleus serotonergic neurons and inhibit median raphe nucleus neurons. Furthermore, the dorsal raphe nucleus sends projections to prefrontal cortex, amygdala, hypothalamus, and periaqueductal gray among other structures. Thus, modulation of the serotonin system has the potential to influence the major regions of the panic disorder circuit, resulting in decreased noradrenergic activity, diminished release of corticotropin-releasing factor, and modification of defense and escape behaviors.

A consideration of the various afferents to the locus coeruleus and amygdala is relevant to considering the extensive literature on panicogenic stimuli. It has been argued that respiratory panicogens (e.g., carbon dioxide, lactate), baroreceptor stimulation, and circulating peptides (cholecystokinin) promote panic via a limbic visceroreceptor pathway. In contrast, panic attacks that are conditioned by visuospatial, auditory, or cognitive cues may be mediated by pathways from cortical association areas to the amygdala (Coplan and Lydiard 1998). Ultimately, it may be possible to determine particular genetic loci that are involved in contextual fear conditioning, allowing for an integration of the neurochemical, genetic, and environmental data on panic disorder (Gorman et al. 2004).

Neuroanatomical Studies

Preliminary studies in nonanxious control subjects reported activation of amygdala and periamygdaloid cortical areas during conditioned fear acquisition and extinction (Gorman et al. 2004). Furthermore, in patients with panic disorder, increasing evidence suggests temporal or amygdalar-hippocampal abnormalities (Uchida et al. 2008), as well as frontal abnormalities (Konishi et al. 2014). Insular cortex abnormalities also appear to be central to the pathophysiology of panic (Paulus and Stein 2006), perhaps mediating conditioning of fear to interoceptive cues. Although hypocapnia-induced vasoconstriction has made the results of certain imaging studies in panic disorder difficult to interpret, it is noteworthy that imaging data may predict response to panicogens (Kent et al. 2005).

Advances in brain imaging methods have begun to allow the integration of neuroanatomical and neurochemical data. Thus, a review of receptor binding studies across the anxiety disorders concluded that panic disorder is uniquely associated with reduced frontocortical GABA receptors and shares with other anxiety disorders reductions in striatal dopamine and midbrain 5-HT_1A receptors (Nikolaus et al. 2010).

Social Anxiety Disorder

Social anxiety disorder (formerly social phobia) is characterized by a fear of social situations in which the individual may be exposed to the scrutiny of others. These fears may be divided into those that concern social interaction situations (e.g., dating, meetings) and performance fears (e.g., talking, eating, or writing in public). These fears result in avoidance of social situations or endurance of these situations with considerable distress. Growing evidence indicates that social phobia is a chronic disorder, with substantial comorbidity (particularly of mood and substance use disorders) and significant morbidity. Patients with social anxiety disorder are more likely to be unmarried, to have weaker social networks, to fail to complete high school and college, and to be unemployed (Ballenger et al. 1998). One of the main changes in diagnostic criteria since publication of DSM-IV is that individuals need not recognize that their anxiety is excessive; instead, the anxiety must be judged to be out of proportion to the objective danger or actual threat in the situation. Moreover, DSM-5 has added a 6-month minimum duration criterion, which was restricted to individuals less than 18 years of age in DSM-IV.

Detailed neuroanatomical models of social anxiety disorder are emerging from the past two decades of research. We review below evidence that social anxiety shares certain brain changes with other anxiety disorders and has also been associated with brain alterations that appear more specific to this disorder.

Neurochemical Studies

Multiple lines of evidence support the role of serotonergic circuits in social anxiety disorder. Pharmacotherapy with a number of SSRIs is effective and well tolerated by patients, and paroxetine and sertraline are FDA-approved for treatment of this disorder. Prior to the appearance of serotonergic agents, MAOIs and benzodiazepines had also shown efficacy in social anxiety, although their unfavorable risk:benefit profile makes them less attractive (Koen and Stein 2011).

There is also evidence that the dopaminergic system is involved in social anxiety disorder (Stein et al. 2002). Timid mice have decreased cerebrospinal fluid dopamine levels, and introverted depressed patients also may have decreased cerebrospinal fluid dopamine levels. Social status in monkeys may be reflected in differences in dopamine type 2 (D2) striatal density. More persuasively, social anxiety may develop in the context of Parkinson’s disease or after the administration of neuroleptics.

Evidence indicating the hypothalamic-pituitarys-adrenal (HPA) axis may be dysfunctional in social anxiety disorder is inconsistent to date. The aggregate of findings points to hyper-responsiveness of the adrenal cortex in patients with social anxiety, although this is found more consistently following a stressor, while baseline HPA function appears similar in patients and normally functioning control subjects. In addition, there are a number of findings linking HPA axis abnormalities to early life stress in social anxiety disorder, suggesting that history of trauma may help parse the inconsistent findings in the literature (Faravelli et al. 2012).

Additional neurochemical systems deserve exploration in social anxiety disorder. Glutamate is a particularly widespread excitatory neurotransmitter in the brain, and a number of psychotropics act to alter glutamatergic neurotransmission. It is notable that D-cycloserine, a partial glutamatergic agonist, is useful in the augmentation of cognitive-behavioral therapy in social anxiety disorder and some other anxiety disorders. Various neuropeptide systems, including oxytocin, have also been explored in relation to social anxiety disorder and other anxiety disorders and may ultimately provide treatment targets.

Neuroanatomical Studies

Meta-analyses of neuroimaging studies have revealed that social anxiety disorder is associated with hyperactivity in the limbic fear circuitry, especially the amygdala and insular cortex, which is a shared pathology across many anxiety disorders (Etkin and Wager 2007; Hattingh et al. 2013) (Figure 26–3). Thus, patients with social anxiety disorder show hyperactivation of the amygdala and insula when exposed to both socially relevant fearful stimuli and “nonspecific” novel stimuli (Bruhl et al. 2014). Interestingly, subjects with behavioral inhibition, when studied as adults, also have heightened amygdala responses to novel fear-relevant stimuli (Schwartz et al. 2003). It is noteworthy that nonphobic control subjects with a particular variant in the serotonin transporter gene that is associated with anxiety traits, as well as subjects with social anxiety, have decreased volume or increased activity in amygdala or related circuitry (Bruhl et al. 2014; Furmark et al. 2004).

FIGURE 26–3. Neuroanatomical model of social anxiety disorder.

Hyperactivity of the amygdala (AMY) and insular cortex (INS) is consistently seen in social anxiety disorder. Thus far, more specific to social anxiety is evidence for hyperactive medial and lateral prefrontal cortex (PFC), posterior cingulate cortex (PCC), and areas of parietal-occipital cortex including supramarginal gyrus (SMG) and fusiform gyrus (FFG).

There is also burgeoning evidence for involvement of prefrontal and parietal-occipital cortices in the pathophysiology of social anxiety disorder. In response to tasks using socially relevant stimuli, patients with social anxiety disorder show hyperactivation of bilateral medial and ventrolateral prefrontal cortex (Brodmann area [BA] 10), as well as hyperactivation of the posterior cingulate cortex (BA 31), supramarginal gyrus (BA 40), and fusiform gyrus (BA 19) (Bruhl et al. 2014). Evidence of such widespread cortical hyperactivation, although still fairly preliminary, appears to set social anxiety disorder apart from other anxiety disorders where cortical hypoactivity is a more common finding.

Several molecular imaging studies provide additional data that are relevant to an integrated model of social anxiety disorder. Thus, involvement of the basal ganglia is indicated by evidence that striatal dopamine reuptake site densities are markedly lower in patients with social anxiety than in nonphobic control subjects, although this reduction is also seen in other anxiety disorders (Nikolaus et al. 2010). Other findings support the hypothesis that social anxiety disorder may be associated with a dysfunction of the striatal dopaminergic system (Schneier et al. 2000). These types of data may be consistent with a link between social anxiety and dysfunctional processing of positive or rewarding information (Hare et al. 2005) and may ultimately suggest novel approaches to pharmacotherapy for this condition.

Posttraumatic Stress Disorder

Trauma and stressor-related disorders begin, by definition, in the aftermath of exposure to a trauma or stressor. Four sets of subsequent symptoms characterize PTSD: reexperiencing intrusive phenomena (such as visual flashbacks), avoidance and numbing symptoms, alterations in arousal and reactivity, and negative alterations in mood and cognition hyperarousal. This latter category is new in DSM-5 and includes most of the DSM-IV numbing symptoms as well as some new or reconceptualized symptoms, including persistent negative emotional state. The cluster for alterations in arousal and reactivity retains most of the DSM-IV hyperarousal symptoms and adds irritable/aggressive behavior as well as reckless or self-destructive behavior. Of note, the stressor criterion has also changed, such that Criterion A (points 1–4) delineates more specific and restrictive definitions of what qualifies as a “traumatic” event, and Criterion A2 (subjective reaction) has been deleted.

It should be emphasized that the prevalence of exposure to trauma is significantly higher than the prevalence of PTSD, indicating that most trauma does not lead to this disorder. Indeed, an important development in the PTSD literature is a growing emphasis that this is not a “normal” reaction to an abnormal event (Yehuda and LeDoux 2007). Rather, PTSD has been established as a serious disorder that is associated with significant morbidity and mediated by neurobiological and psychological dysfunctions (Etkin and Wager 2007; Rauch et al. 2006; Yehuda and LeDoux 2007).

Features of current neuroanatomical models of PTSD (Figure 26–4) include the following: 1) Amygdalothalamic pathways are involved in the rapid, automatic (implicit) processing of incoming information. 2) Hyperactivation of the amygdala and insular cortex, which sends afferents to other regions involved in the anxiety response (e.g., hypothalamus, brain stem nuclei), occurs. 3) The hippocampus is reduced in volume and involved in (explicitly) remembering the context of traumatic memories. 4) Activity is decreased in certain frontal cortical areas, consistent with decreased verbalization during processing of trauma (e.g., deactivation of Broca’s area), failure of fear extinction (e.g., failure to recruit medial and ventral prefrontal areas), and an inability to override automatic amygdala processing.

FIGURE 26–4. Neuroanatomical model of posttraumatic stress disorder (PTSD).

Hyperactivity of the amygdala (AMY) and insular cortex (INS) are consistently seen in PTSD, as well as hypoactivity of the ventromedial prefrontal cortex (vmPFC) and ventral anterior cingulate cortex (vACC). The hippocampus (HIPPO) is reduced in volume.

Neurochemical Studies

A number of neurochemical findings in PTSD are consistent with sensitization of various neurotransmitter systems (Charney 2004). In particular, there is evidence of hyperactive noradrenergic function and dopaminergic sensitization. Such sensitization is also consistent with the role of environmental traumas in PTSD; dopamine agonists and environmental traumas act as cross-sensitizers of each other. Evidence indicates that the amygdala and related limbic regions may play a particularly important role in the final common pathway of such hyperactivation.

Also, growing evidence suggests the importance of the serotonin system in mediating PTSD symptoms. Clinical studies of abnormal paroxetine binding and exacerbations of symptoms in response to administration of m-CPP are certainly consistent with a role for serotonin in PTSD (Southwick et al. 1997). Furthermore, a number of serotonin reuptake inhibitors and venlafaxine have been found to be effective and safe for the treatment of PTSD (Koen and Stein 2011). These agents may act on amygdala circuits, helping to inhibit efferents to structures such as hypothalamus and brain stem nuclei, which mediate fear.

A third set of neurochemical findings in PTSD has focused on the HPA system. PTSD is characterized by increased baseline cortisol-releasing factor (CRF) and decreased plasma levels of cortisol, and these findings differ from those observed in other anxiety disorders and in depression (Yehuda and LeDoux 2007). It has been suggested that the joint occurrence of high CRF and low basal cortisol reflect a long-term physiological adaptation of the HPA axis to chronic stress. Moreover, there is considerable preclinical evidence that early life stress leads to short- and long-term alterations of HPA function. Specifically, an initial sensitization and high cortisol levels may occur during persistent and recurrent early traumatization, followed by a blunting of HPA axis responsivity as a longer-term adaptation to chronic stress, along with downregulation of CRF receptors.

One important implication of the HPA findings is the possibility that dysfunction in this system results in neuronal damage, particularly to the hippocampus (Rosso et al. 2017). Animal studies have documented hippocampal damage after exposure to either glucocorticoids or naturalistic psychosocial stressors. Parallel neurotoxicity in human PTSD could account for some of the cognitive impairments that are characteristic of this disorder, although the association between hippocampus integrity and glucocorticoid functioning has been more difficult to determine from human brain imaging studies of PTSD (Yehuda and LeDoux 2007).

Neuroanatomical Studies

A number of structural imaging studies are, in fact, consistent with the possibility of hippocampal dysfunction occurring in PTSD. A meta-analysis of magnetic resonance imaging studies, for example, emphasized the consistent finding of decreased hippocampal volume in PTSD secondary to adult or childhood trauma (Woon et al. 2010). In some studies, decreased volume has been associated with greater trauma exposure, increased symptom severity, or worse neuropsychological impairment. Nevertheless, evidence also shows that decreased hippocampal volume may precede the onset of PTSD and thus constitutes a risk factor for the development of this condition. In addition, there are now increasing data suggesting decreased volume in medial and ventral prefrontal cortex (Rauch et al. 2006).

Functional imaging studies have provided additional information in support of a neuroanatomical model of PTSD. Several studies in control subjects without PTSD have provided evidence for subcortical processing of masked emotional stimuli by the amygdala. Furthermore, a range of studies have found that PTSD patients exposed to audiotaped traumatic and neutral scripts had increases in neuronal activity in limbic and paralimbic areas compared with healthy control subjects (Etkin and Wager 2007). Also, areas of decreased activity may mediate symptoms; for example, decreased activity in Broca’s area during exposure to trauma in PTSD is consistent with patients’ inability to verbally process traumatic memories (Rauch et al. 2006). Moreover, in a meta-analysis of functional imaging studies of symptom provocation and negative emotional processing across multiple anxiety disorders, PTSD was associated with greater activity of the amygdala and insular cortex, as well as hypoactivation of the anterior cingulate cortex and ventromedial prefrontal cortex (Etkin and Wager 2007). Interestingly, the latter finding was specific to PTSD and not seen in the other anxiety disorders, which may be consistent with extinction deficits in PTSD (Milad et al. 2009).

Once again, modern techniques have allowed for the integration of neurochemical and neuroanatomical data. For example, positron emission tomography has been used in combat veterans with PTSD and healthy control subjects after administration of yohimbine (Bremner et al. 1997); this noradrenergic agent resulted in a significant increase in anxiety in the patients with PTSD, and these subjects also had a decrease in activity in several areas, including prefrontal, temporal, parietal, and orbitofrontal cortex.

Neurological Disorders With Anxiety Symptoms

Neurological conditions that affect a range of different neuroanatomical structures may be associated with anxiety symptoms or disorders (Muller et al. 2005). Given that temporolimbic regions, striatum, and prefrontal cortex all likely play an important role in the pathogenesis of certain anxiety disorders, we begin by reviewing the association between lesions in these areas and subsequent anxiety symptoms before moving on to disorders with more widespread pathology. This literature not only is clinically relevant but also raises valuable questions for further research.

Various lesions of the temporolimbic regions have been associated with the subsequent development of panic disorder. Temporal lobe seizures (Muller et al. 2005), tumors (Kellner et al. 1996–1997), and parahippocampal infarction (Maricle et al. 1991) all have been reported to manifest with panic attacks. The association seems particularly strong with right-side lesions. Conversely, removal of the amygdala results in placidity toward previously feared objects (Kluver and Bucy 1939) and deficits in fear conditioning (Muller et al. 2005).

This literature, taken together with clinical observations that panic disorder may be accompanied by dissociation and depersonalization and possibly by electroencephalographic abnormalities and temporal lobe abnormalities (see subsection “Panic Disorder and Agoraphobia”), as well as preliminary data that show panic disorder can respond to anticonvulsants, raises the question of whether partially overlapping mechanisms may be at work in both temporal lobe seizure disorder and panic disorder. Certainly, it has been suggested that electroencephalogram and anticonvulsant trials may be appropriate in patients with panic disorder refractory to conventional treatment (Koen and Stein 2011).

Anxiety symptoms may be seen in striatal disorders (Muller et al. 2005). In Huntington’s disease, for example, anxiety has been reported as a common prodromal symptom, with later development of several different anxiety disorders (Leroi and Michalon 1998). Anxiety symptoms and disorders are also common in Parkinson’s disease (Muller et al. 2005) and may correlate inversely with left striatal dopamine transporter availability (Erro et al. 2012). Such findings arguably indicate that further attention should be paid to the role of the dopaminergic system in anxiety disorders (Stein et al. 2002), although other neurotransmitter systems also may be important in mediating anxiety symptoms in Parkinson’s disease.

Anxiety symptoms and disorders can also be seen in a range of neurological disorders that affect multiple brain regions, including the frontal cortex. In multiple sclerosis, for example, anxiety symptoms may be found in up to 44% of subjects, and anxiety disorders are also not uncommon (Marrie et al. 2015). Whether this anxiety reflects a psychological reaction or reflects the deposition of demyelinating plaques remains somewhat unclear, but treatment of symptoms should not be ignored.

Similarly, anxiety symptoms have been noted to be common in Alzheimer’s disease and in other dementias including vascular and frontotemporal dementias (Regan and Varanelli 2013). The relation between regional pathology and anxiety symptoms in these conditions deserves further attention. Additional work on the management of anxiety in dementia is also needed.

Although the prevalence of depression after stroke has been well studied, fewer studies have focused on anxiety after stroke. However, in one study of 309 admissions to a stroke unit, GAD was present in 26.9% of the patients (Castillo et al. 1993). The authors reported that anxiety plus depression was associated with left cortical lesions, whereas anxiety alone was associated with right hemisphere lesions. Also, worry was associated with anterior and GAD with right posterior lesions. Longitudinal studies have found that GAD can persist for several years after the stroke (Morrison et al. 2005). Again, anxiety plus depression may be associated with left hemisphere lesions and anxiety alone with right hemisphere lesions.

Anxiety disorders also have been reported in the aftermath of traumatic brain injury (TBI). Anxiety symptoms are present in as many as 70% of victims of mild TBI, and anxiety disorders can be diagnosed in 29% of individuals across all severity levels of TBI (Moore et al. 2006). Prevalence rates for individual anxiety disorders vary widely across TBI studies. PTSD is the most commonly studied anxiety disorder following traumatic injury, with anywhere from 20% to 84% of PTSD patients have symptoms that also meet TBI criteria across studies (Moore et al. 2006). Of particular interest is the finding that PTSD can develop even when the patient has neurogenic amnesia for the traumatic event; this finding may suggest that implicit memories of trauma are sufficient for later PTSD to emerge.

Conclusion

Several lessons emerge from a review of the neuropsychiatry of anxiety disorders. First, the anxiety disorders are common and disabling disorders not only in general clinical settings but also in patients with neurological illnesses such as Alzheimer’s disease, stroke, and traumatic brain injury. Although the link between depression and neuropsychiatric disorders is increasingly recognized, the importance of anxiety disorders in the context of neurological illnesses has perhaps been relatively overlooked, paralleling their underdiagnosis and undertreatment in primary care settings. The anxiety disorders deserve to be carefully diagnosed, thoroughly assessed, and rigorously treated.

Second, both animal and clinical studies increasingly indicate that the amygdala and paralimbic structures play important roles in conditioned fear and in anxiety disorders. Amygdala lesions are classically associated with decreased fear responses, and conversely, limbic hyperactivation is characteristic of several different anxiety disorders. Paralimbic regions such as the anterior cingulate appear to play a key role at the interface of cognition and emotion. The apparent centrality of such systems to different anxiety disorders may account in part for their high comorbidity. Other limbic involvement may be specific to particular disorders (e.g., decreased hippocampal volume in PTSD or parahippocampal asymmetry in panic disorder).

Models of anxiety disorders increasingly integrate data from genetics, brain imaging, and treatment studies. Thus, particular genetic variants appear to be associated with increased activation of specific neuronal circuits during functional imaging, and effective pharmacotherapy and psychotherapy may act to normalize such circuitry. Serotonin reuptake inhibitors and cognitive-behavioral therapy are increasingly viewed as first-line treatments for anxiety disorders. Innervation of amygdala and paralimbic structures by serotonergic neurons may be crucial in explaining their efficacy. Further advances in our understanding of the neurobiological bases of fear conditioning and extinction may lead to new therapeutic interventions.

References

Alvares LdeO, Einarsson EO, Santana F, et al: Periodically reactivated context memory retains its precision and dependence on the hippocampus. Hippocampus 22(5):1092–1095, 2012 22120981

American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, 3rd Edition. Washington, DC, American Psychiatric Association, 1980

American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, 4th Edition, Text Revision. Washington, DC, American Psychiatric Association, 2000

American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, 5th Edition. Arlington, VA, American Psychiatric Association, 2013

Ballenger JC, Davidson JR, Lecrubier Y, et al: Consensus statement on social anxiety disorder from the International Consensus Group on Depression and Anxiety. J Clin Psychiatry 59 (suppl 17):54–60, 1998 9811431

Bremner JD, Innis RB, Ng CK, et al: Positron emission tomography measurement of cerebral metabolic correlates of yohimbine administration in combat-related posttraumatic stress disorder. Arch Gen Psychiatry 54(3):246–254, 1997 9075465

Bruhl AB, Delsignore A, Komossa K, et al: Neuroimaging in social anxiety disorder—a meta-analytic review resulting in a new neurofunctional model. Neurosci Biobehav Rev 47:260–280, 2014 25124509

Castillo CS, Starkstein SE, Fedoroff JP, et al: Generalized anxiety disorder after stroke. J Nerv Ment Dis 181(2):100–106, 1993 8426166

Charney DS: Psychobiological mechanisms of resilience and vulnerability: implications for successful adaptation to extreme stress. Am J Psychiatry 161(2):195–216, 2004 14754765

Ciocchi S, Herry C, Grenier F, et al: Encoding of conditioned fear in central amygdala inhibitory circuits. Nature 468(7321):277–282, 2010 21068837

Coplan JD, Lydiard RB: Brain circuits in panic disorder. Biol Psychiatry 44(12):1264–1276, 1998 9861469

Erro R, Pappatà S, Amboni M, et al: Anxiety is associated with striatal dopamine transporter availability in newly diagnosed untreated Parkinson’s disease patients. Parkinsonism Relat Disord 18(9):1034–1038, 2012 22789824

Etkin A: Functional neuroanatomy of anxiety: a neural circuit perspective. Curr Top Behav Neurosci 2:251–277, 2010 21309113

Etkin A, Wager TD: Functional neuroimaging of anxiety: a meta-analysis of emotional processing in PTSD, social anxiety disorder, and specific phobia. Am J Psychiatry 164(10):1476–1488, 2007 17898336

Etkin A, Prater KE, Hoeft F, et al: Failure of anterior cingulate activation and connectivity with the amygdala during implicit regulation of emotional processing in generalized anxiety disorder. Am J Psychiatry 167(5):545–554, 2010 20123913

Faravelli C, Lo Sauro C, Godini L, et al: Childhood stressful events, HPA axis and anxiety disorders. World J Psychiatry 2(1):13–25, 2012 24175164

Furmark T, Tillfors M, Garpenstrand H, et al: Serotonin transporter polymorphism related to amygdala excitability and symptom severity in patients with social phobia. Neurosci Lett 362(3):189–192, 2004 15158011

Gorman JM, Kent JM, Sullivan GM, et al: Neuroanatomical Hypothesis of Panic Disorder, Revised. Focus 2(3):426–439, 2004

Hare TA, Tottenham N, Davidson MC, et al: Contributions of amygdala and striatal activity in emotion regulation. Biol Psychiatry 57(6):624–632, 2005 15780849

Hattingh CJ, Ipser J, Tromp SA, et al: Functional magnetic resonance imaging during emotion recognition in social anxiety disorder: an activation likelihood meta-analysis. Front Hum Neurosci 6:347, 2013 23335892

Kellner M, Hirschmann M, Wiedemann K: Panic attacks caused by temporal tumors: an exemplary new case and a review. Depress Anxiety 4(5):243–245, 1996–1997 9167792

Kent JM, Coplan JD, Mawlawi O, et al: Prediction of panic response to a respiratory stimulant by reduced orbitofrontal cerebral blood flow in panic disorder. Am J Psychiatry 162(7):1379–1381, 2005 15994724

Kessler RC: The epidemiology of pure and comorbid generalized anxiety disorder: a review and evaluation of recent research. Acta Psychiatr Scand Suppl (406):7–13, 2000 11131470

Kessler RC, Ruscio AM, Shear K, et al: Epidemiology of anxiety disorders. Curr Top Behav Neurosci 2:21–35, 2010 21309104

Klein E, Zohar J, Graci MF, et al: Anxiogenic effects of m-CPP in patients with panic disorder: comparison to caffeine's anxiogenic effects. Biol Psychiatry 3(10):978–984, 1991 1756202..

Kluver H, Bucy PC: Preliminary analysis of functions of the temporal lobes in monkeys. Arch Neurol Psychiatry 42(6):979–1000, 1939

Koen N, Stein DJ: Pharmacotherapy of anxiety disorders: a critical review. Dialogues Clin Neurosci 13(4):423–437, 2011 22275848

Konishi J, Asami T, Hayano F, et al: Multiple white matter volume reductions in patients with panic disorder: relationships between orbitofrontal Gyrus volume and symptom severity and social dysfunction. PLoS One 9(3):e92862, 2014 24663245

Lepine JP: The epidemiology of anxiety disorders: prevalence and societal costs. J Clin Psychiatry 63 (suppl 14):4–8, 2002 12562112

Leroi I, Michalon M: Treatment of the psychiatric manifestations of Huntington’s disease: a review of the literature. Can J Psychiatry 43(9):933–940, 1998 9825166

Maricle RA, Sennhauser S, Burry M: Panic disorder associated with right parahippocampal infarction. J Nerv Ment Dis 179(6):374–375, 1991 2051154

Marrie RA, Reingold S, Cohen J, et al: The incidence and prevalence of psychiatric disorders in multiple sclerosis: a systematic review. Mult Scler 21(3):305–317, 2015 25583845

Milad MR, Pitman RK, Ellis CB, et al: Neurobiological basis of failure to recall extinction memory in posttraumatic stress disorder. Biol Psychiatry 66(12):1075–1082, 2009 19748076

Moore EL, Terryberry-Spohr L, Hope DA: Mild traumatic brain injury and anxiety sequelae: a review of the literature. Brain Inj 20(2):117–132, 2006 16421060

Morrison V, Pollard B, Johnston M, et al: Anxiety and depression 3 years following stroke: demographic, clinical, and psychological predictors. J Psychosom Res 59(4):209–213, 2005 16223623

Muller JE, Koen L, Stein DJ: Anxiety and medical disorders. Curr Psychiatry Rep 7(4):245–251, 2005 16098277

Nikolaus S, Antke C, Beu M, et al: Cortical GABA, striatal dopamine and midbrain serotonin as the key players in compulsive and anxiety disorders—results from in vivo imaging studies. Rev Neurosci 21(2):119–139, 2010 20614802

Nutt DJ: Neurobiological mechanisms in generalized anxiety disorder. J Clin Psychiatry 62 (suppl 11):22–27; discussion 28, 2001 11414547

Paulus MP, Stein MB: An insular view of anxiety. Biol Psychiatry 60(4):383–387, 2006 16780813

Rauch SL, Shin LM, Phelps EA: Neurocircuitry models of posttraumatic stress disorder and extinction: human neuroimaging research—past, present, and future. Biol Psychiatry 60(4):376–382, 2006 16919525

Regan B, Varanelli L: Adjustment, depression, and anxiety in mild cognitive impairment and early dementia: a systematic review of psychological intervention studies. Int Psychogeriatr 25(12):1963–1984, 2013 24125507

Rosso IM, Crowley DJ, Silveri MM, et al: Hippocampus glutamate and N-acetyl aspartate markers of excitotoxic neuronal compromise in posttraumatic stress disorder. Neuropsychopharmacology. .42(8):1698–1705, 201728195577

Schneier FR, Liebowitz MR, Abi-Dargham A, et al: Low dopamine D(2) receptor binding potential in social phobia. Am J Psychiatry 157(3):457–459, 2000 10698826

Schwartz CE, Wright CI, Shin LM, et al: Inhibited and uninhibited infants “grown up”: adult amygdalar response to novelty. Science 300(5627):1952–1953, 2003 12817151

Shang J, Fu Y, Ren Z, et al: The common traits of the ACC and PFC in anxiety disorders in the DSM-5: meta-analysis of voxel-based morphometry studies. PLoS One 9(3):e93432, 2014 24676455

Southwick SM, Krystal JH, Bremner JD, et al: Noradrenergic and serotonergic function in posttraumatic stress disorder. Arch Gen Psychiatry 54(8):749–758, 1997 9283511

Stein DJ, Westenberg HG, Liebowitz MR: Social anxiety disorder and generalized anxiety disorder: serotonergic and dopaminergic neurocircuitry. J Clin Psychiatry 63 (suppl 6):12–19, 2002 12027115

Uchida RR, Del-Ben CM, Busatto GF, et al: Regional gray matter abnormalities in panic disorder: a voxel-based morphometry study. Psychiatry Res 163(1):21–29, 2008 18417322

van der Wee NJ, Fiselier J, van Megen HJ, Westenberg HG: Behavioural effects of rapid intravenous administration of meta-chlorophenylpiperazine in patients with panic disorder and controls. J Clin Psychiatry 14(5):413–417, 2004 15336303

Von Economo C: Encephalitis Lethargica, Its Sequelae and Treatment. London, Oxford University Press, 1931

Woon FL, Sood S, Hedges DW: Hippocampal volume deficits associated with exposure to psychological trauma and posttraumatic stress disorder in adults: a meta-analysis. Prog Neuropsychopharmacol Biol Psychiatry 34(7):1181–1188, 2010 20600466

Yehuda R, LeDoux J: Response variation following trauma: a translational neuroscience approach to understanding PTSD. Neuron 56(1):19–32, 2007 17920012

________________

Dr. Stein is supported by the Medical Research Council of South Africa. Dr. Rosso is supported by the National Institute of Mental Health and the Dana Foundation. Dr. Rauch is partially supported by the National Institute of Mental Health and the United States Army Medical Research Acquisition Activity.