Depression and Psychosis in Neurological Practice

Principles of Differential Diagnosis

Emotional and cognitive processes are based on brain structure and physiology. Abnormal behavior can be attributable to the complex interplay of neural physiology, social influences, and physical environment ( ). Psychosis, mania, depression, disinhibition, obsessive compulsive behaviors, and anxiety all can occur as a result of neurological disease and can be virtually indistinguishable from the idiopathic forms ( ). Neurological conditions should be considered in the differential diagnosis of any disorder with psychiatric symptoms.

Neuropsychiatric abnormalities can be associated with altered functioning in anatomical regions. Any disease, toxin, drug, or process that affects a particular region can be expected to show changes in behavior mediated by the distributed network encompassing that region. The limbic system and the frontosubcortical circuits are most commonly implicated in neuropsychiatric symptoms. This neuroanatomical conceptual framework can provide useful information for localization and thus differential diagnosis. For example, the Klüver-Bucy syndrome—which consists of placidity, apathy, visual and auditory agnosia, hyperorality, and hypersexuality—occurs in processes that cause injury to the bilateral medial temporoamygdalar regions. A few of the most common causes of this syndrome include herpes encephalitis, traumatic brain injury (TBI), frontotemporal dementias (FTDs), and late-onset or severe Alzheimer disease (AD). Disinhibition, a particularly common neuropsychiatric symptom, may be observed in patients with brain trauma, cerebrovascular ischemia, demyelination, abscesses, or tumors as well as neurodegenerative disorders. Damage to any portion of the cortical and subcortical portions of the orbitofrontal-striatal-pallidal-thalamic circuit can result in disinhibition ( ).

Mood disorders, paranoia, disinhibition, and apathy derive in part from dysfunction in the limbic system and basal ganglia, which are phylogenetically more primitive ( ). In some cases, the behavioral changes represent a psychological response to the underlying disability; in others, neuropsychiatric abnormalities manifest as a result of intrinsic alterations of the neural network caused by the disease itself. For example, studies have shown that apathy in Parkinson disease (PD) is probably related to the underlying disease process rather than being a psychological reaction to disability or to depression and is closely associated with cognitive impairment ( ). Positron emission tomography (PET), single-photon emission computed tomography (SPECT), and functional magnetic resonance imaging (fMRI) studies suggest the involvement of similar regions of abnormality in acquired (secondary) forms of depression, mania, obsessive-compulsive disorder (OCD), and psychosis as in their primary psychiatric presentations ( ). Table 10.1 summarizes neuropsychiatric symptoms and their anatomical correlates. Additionally, the developmental phase during which a neurological illness occurs influences the frequency with which some neuropsychiatric syndromes are manifested. Adults with post-TBI sequelae tend to exhibit a higher rate of depression and anxiety. In contrast, post-TBI sequelae in children often involve attention deficits, hyperactivity, irritability, aggressiveness, and oppositional behavior ( ). When temporal lobe epilepsy or Huntington disease (HD) begins in adolescence, a higher incidence of psychosis is noted than when their onset occurs later in life. Earlier onset of multiple sclerosis (MS) and stroke are associated with a higher incidence of depression ( ).

Patients with AD, PD, HD, and FTDs can develop multiple coexisting symptoms such as irritability, agitation, impulse-control disorders, apathy, depression, delusions, and psychosis, many of which may be exacerbated by medications used to treat the underlying disorder ( Table 10.2 ). For example, in patients with PD, dopamine (DA) agonists such as pramipexole and ropinirole have been found to increase the risk of pathological gambling, compulsive shopping, hypersexuality, and other impulse-control disorders, sometimes referred to as dopamine dysregulation ( ). Management outcome can be influenced by multiple factors. For instance, the complex relationship between behavioral changes and the caregiver’s ability to cope play a role in illness management and nursing home placement ( ). For example, behavioral disturbances in patients with neurological illnesses are well described to be associated with caregiver distress and fatigue ( ).

Principles of Neuropsychiatric Evaluation

A number of important principles must be considered when patients are being evaluated and treated for behavioral disturbances.

• 1.

  The clinical history may offer clues to the index of suspicion for a secondary (neuropsychiatric) etiology versus an idiopathic presentation. For example, late-life initial onset of mania or depression is more commonly associated with central nervous system (CNS) pathology ( ).

• 2.

  A normal neurological examination does not exclude neurological conditions. Lesions in the limbic, paralimbic, and prefrontal regions may manifest with cognitive-affective-behavioral changes in the absence of elemental neurological abnormalities.

• 3.

  Normal routine laboratory testing, brain imaging, electroencephalography, and cerebrospinal fluid (CSF) analysis do not necessarily exclude diseases of neurological origin.

• 4.

  New neurological complaints or behavioral changes that are atypical for a coexisting primary psychiatric disorder should not be dismissed as being of psychiatric origin in a person with a preexisting psychiatric history.

• 5.

  The possibility of iatrogenically induced symptoms—such as lethargy with benzodiazepines, parkinsonism with neuroleptics, or hallucinations with dopaminergic medications—must be taken into account. Medication side effects can significantly complicate the clinical history and physical examination in both the acute and long-term setting. Medication side effects can also potentially be harbingers of underlying pathology or progression of illness. For example, marked parkinsonism occurring after neuroleptic exposure can be a feature of PD and dementia with Lewy bodies (DLB) ( ) before the underlying neurodegenerative condition becomes clinically apparent. PD patients may develop hallucinations as a side effect of dopaminergic medications ( ).

• 6.

  Treatments of primary psychiatric and neurological behavioral disturbances share common principles. A response to therapy does not constitute evidence for a primary psychiatric condition.

The medical evaluation of affective and psychotic symptoms must be individualized based on the patient’s family history, social environment (including social network), habits, risk factors, age, gender, clinical history, and examination findings. A careful review of the patient’s medical history and a general physical examination as well as a neurological examination ( ) should be performed to assess for possible neurological and medical causes. The most basic evaluation should include vital signs (blood pressure, pulse, respirations, and temperature) and a laboratory evaluation that minimally includes a complete blood cell count (CBC), electrolyte panel, serum glucose, blood urea nitrogen (BUN), creatinine, calcium, total protein and albumin as well as assessments of liver and thyroid function. Additional laboratory testing may be considered according to the clinical history and risk factors. These studies might include a toxicology screen, cobalamin (vitamin B ₁₂ ), homocysteine, methylmalonic acid, folate, vitamin D, human immunodeficiency virus (HIV) serology, rapid plasma reagin (RPR), antinuclear antibodies (ANAs), erythrocyte sedimentation rate (ESR), C-reactive protein (CRP), ceruloplasmin, heavy metal screen, ammonia, serum and CSF paraneoplastic panel, urine porphobilinogen, number of cytosine-adenine-guanine (CAG) repeats for HD, and other specialized rheumatologic, metabolic, and genetic tests. Consideration should also be given to checking (especially in the elderly) the patient’s oxygen saturation on room air. Neurological abnormalities suggested by the clinical history or identified on examination, especially those attributable to the CNS, should prompt further evaluation for neurological and medical causes of psychiatric illness. A clear consensus has not been reached as to when neuroimaging is indicated as part of the evaluation of new-onset depression in patients without focal neurological complaints and a normal neurological examination. This must be individualized based on clinical judgment. Treatment-resistant depression should prompt reassessment of the diagnosis and evaluation to rule out secondary causes of depressive illness, including cerebrovascular (small vessel) disease. A careful history to rule out a primary sleep disorder such as obstructive sleep apnea should be considered in the evaluation of refractory depressive symptoms ( ) or cognitive complaints. When new-onset atypical psychosis presents in the absence of identifiable infectious/inflammatory, metabolic, toxic, or other causes, we recommend that magnetic resonance imaging (MRI) of the brain be incorporated into the evaluation. In our experience, 5%–10% of such patients have MRI abnormalities that identify potential neurological contributions (particularly in those 65 years of age and older). The MRI will help to exclude lesions (e.g., demyelination, ischemic disease, neoplasm, congenital structural abnormalities, evidence of metabolic storage diseases) in limbic, paralimbic, and frontal regions that may not be clearly associated with neurological abnormalities on elemental examination ( ). An electroencephalogram (EEG) should be considered to evaluate for complex partial seizures if there is a history of intermittent, discrete, or abrupt episodes of psychiatric dysfunction (e.g., confusion, spells of lost time, psychotic symptoms), stereotypy of hallucinations, automatisms (e.g., lip smacking, repetitive movements) associated with episodes of psychiatric dysfunction (or confusion), or a suspicion of encephalopathy (or delirium). Sensitivity of the EEG for detecting seizure activity is highest when the patient has experienced the specific symptoms while undergoing the study. Selected cases may require 24-hour or prolonged EEG monitoring to capture a clinical event and thus to clarify whether a seizure disorder is present.

Cognitive-Affective-Behavioral Brain–Behavior Relationships

We begin with a brief overview of cortical functional anatomy related to perceptual, cognitive, affective, and behavioral processing. Thereafter a synopsis of frontal network functional anatomy will follow, describing the distinct prefrontosubcortical circuits subserving important cognitive-affective-behavioral domains.

The cerebral cortex can be subdivided into five major functional subtypes: primary sensorimotor, unimodal association, heteromodal association, paralimbic, and limbic ( Fig. 10.1 ). The primary sensory areas are the points of entry for sensory information into the cortical circuitry. The primary motor cortex conveys complex motor programs to motor neurons in the brainstem and spinal cord. Processing of sensory information occurs as information moves from primary sensory areas to adjacent unimodal association areas. The unimodal and heteromodal cortices are involved in perceptual processing and motor planning. The complexity of processing increases as information is then transmitted to heteromodal association areas, which receive input from more than one sensory modality. Examples of heteromodal association cortices include the prefrontal cortex, posterior parietal cortex, parts of the lateral temporal cortex, and portions of the parahippocampal gyrus. These cortical regions have a six-layered cytoarchitecture. Further cortical processing occurs in areas designated as paralimbic . These regions demonstrate a gradual transition of cortical architecture from the six-layered areas to the more primitive and simplified allocortex of limbic structures. The paralimbic regions, implicated in idiopathic and secondary neuropsychiatric symptoms, consist of the orbitofrontal cortex (OFC), cingulate gyrus, insula, temporal pole, and parahippocampal cortex. Cognitive, emotional, and visceral inputs merge in these regions. The limbic subdivision is composed of the hippocampus, amygdala, substantia innominata, prepiriform olfactory cortex, and septal area ( Fig. 10.2 ). Limbic structures are to a great extent reciprocally interconnected with the hypothalamus. Limbic regions are intimately involved with processing and regulation of emotion, memory, motivation, and autonomic and endocrine function. The highest level of cognitive processing occurs in regions referred to as transmodal areas . These are composed of heteromodal, paralimbic, and limbic regions, which are collectively linked, in parallel, to other transmodal regions. Interconnections among transmodal areas (e.g., Wernicke area, posterior parietal cortex, hippocampal-enterorhinal complex) enable the integration of distributed perceptual processing systems, resulting in perceptual recognition (i.e., of phenomena such as scenes and events becoming experiences and words taking on meaning) ( ).

Cortical Networks

Classically, five distinct cortical networks have been conceptualized as governing various aspects of cognitive functioning:

• 1.

  The language network, which includes transmodal regions or “epicenters” in the Broca and Wernicke areas located in the pars opercularis/triangular portions of the inferior frontal gyrus and posterior aspect of the superior temporal gyrus, respectively

• 2.

  Spatial awareness, based in transmodal regions in the frontal eye fields and posterior parietal cortex

• 3.

  The memory and emotional network, located in the hippocampal-enterorhinal region and amygdala

• 4.

  The executive function–working memory network, based in transmodal regions in the lateral prefrontal cortex and possibly the inferior parietal cortices

• 5.

  The face-object recognition network, based in the temporopolar and middle/ventral temporal cortices ( )

Lesions of transmodal cortical areas result in global impairments such as hemineglect, anosognosia, amnesia, and multimodal anomia. Disconnection of transmodal regions from a specific unimodal input will result in selective perceptual impairments such as category-specific anomias, prosopagnosia, pure word deafness, or pure word blindness.

The emergence of functional neuroimaging technologies—including task-based ( ) and resting-state functional connectivity analyses ( )—has over the past several decades allowed for the in vivo inspection of brain networks. Apart from the five networks already described, several additional networks have emerged as particularly important to the understanding of brain–behavior relationships in behavioral neurology and neuropsychiatry:

• 1.

  The default mode network (DMN)—which includes areas along the anterior and posterior cortical midline (medial prefrontal cortex, posterior cingulate cortex, precuneus), posterior inferior parietal lobules, and medial temporal lobe—is linked to self-referential processing ( ).

• 2.

  The salience network—which is anchored in the dorsal anterior cingulate cortex (ACC) and insular cortex—has strong subcortical and limbic connections and is linked with reactions to the external world and homeostasis ( ).

• 3.

  The parietofrontal mirror neuron system—which includes the parietal lobe and the premotor cortex plus the caudal part of the inferior frontal gyrus—is involved in the recognition of voluntary behavior in other people ( ).

The limbic mirror system, formed by the insula and the anterior mesial frontal cortex, is devoted to the recognition of affective behavior. DMN and parietofrontal mirror neuron system abnormalities have been linked to mentalization deficits including impairments of theory of mind, while the right anterior insula and ACC have been implicated in emotional and self-awareness ( ).

Frontosubcortical Networks

Five frontosubcortical circuits subserve cognition, emotion, behavior, and movement. Disruption of these networks at the cortical or subcortical level can be associated with similar neuropsychiatric symptoms ( ). Each of these circuits shares similar nonoverlapping components: (1) frontal cortex; (2) striatum (caudate, putamen, ventral striatum); (3) globus pallidus and substantia nigra; and (4) thalamus (which then projects back to frontal cortex) ( ) ( Fig. 10.3 ). Integrative connections also occur to and from other subcortical and distant cortical regions related to each circuit. Neurotransmitters such as DA, glutamate, γ-aminobutyric acid (GABA), acetylcholine, norepinephrine, and serotonin are involved in various aspects of neural transmission and modulation in these circuits. The frontosubcortical networks are named according to their site of origin or function. Somatic motor function is mediated by the motor circuit originating in the supplementary motor area. Oculomotor function is governed by the oculomotor circuit originating in the frontal eye fields. Three of the five circuits are intimately involved in cognitive, emotional, and behavioral functions: the dorsolateral prefrontal, the orbitofrontal, and the anterior cingulate circuits. Each circuit has both efferent and afferent connections with adjacent and distant cortical regions.

The dorsolateral prefrontal cortex (DLPFC)–subcortical circuit is principally involved in attentional and higher-order cognitive executive functions. These functions include the ability to shift sets, organize, and solve problems, as well as the abilities of cognitive control and working memory. Shifting sets is related to mental flexibility and consists of the ability to move between different concepts or motor plans or the ability to shift between different aspects of the same or related concept. Working memory is the online maintenance and manipulation of information. The DLPFC–subcortical circuit includes the dorsolateral head of the caudate, the lateral mediodorsal globus pallidus interna, and the parvocellular aspects of the mediodorsal and ventral anterior thalamic nuclei. Dysfunction in this circuit has been linked with environmental dependency syndromes (including utilization and imitation behavior), poor organization and planning, mental inflexibility, and working memory deficits. Executive dysfunction is also a principal component of subcortical dementias. Deficits identified in subcortical dementias include slowed information processing, memory retrieval deficits, mood and behavioral changes, gait disturbance, dysarthria, and other motor impairments. Vascular dementias, PD, and HD are a few examples of conditions that affect this circuit.

The OFC–subcortical circuit is implicated in socially appropriate and empathic behavior, value-based decision making, mental flexibility, response inhibition, and emotion regulation. It pairs thoughts, memories, and experiences with corresponding visceral and emotional states. The OFC has functional specificity along its anteroposterior and mediolateral axes. The medial OFC has been linked to reward processing and behavioral responses in the context of viscerosomatic evaluations, whereas more lateral regions mediate more external, sensory evaluations including decoding punishment. Anterior subregions process the reward value for more abstract and complex secondary reinforcing factors such as money, whereas more concrete factors such as touch and taste are encoded in the posterior areas. The posteromedial OFC is particularly implicated in evaluating the emotional significance of stimuli ( ). The OFC–subcortical connections include the ventromedial caudate, mediodorsal aspects of the globus pallidus interna, and the medial ventral anterior and inferomedial aspects of the magnocellular mediodorsal thalamus. OFC dysfunction, depicted in the classic personality change experienced by Phineas Gage following injury of his left medial prefrontal cortex by a metal rod in a construction accident, is associated with impulsivity, disinhibition, irritability, aggressive outbursts, socially inappropriate behavior, and mental inflexibility. Persons with bilateral OFC lesions may manifest “theory of mind” deficits. Theory of mind is a model of how a person understands and infers other people’s intentions, desires, mental states, and emotions ( ). Conditions that exhibit OFC and related neurocircuit impairment include schizophrenia ( ), depression ( ), OCD ( ), FTD ( ), and HD. Other conditions that may affect this circuit include closed head trauma, rupture of anterior communicating aneurysms, and subfrontal meningiomas.

The ACC and its subcortical connections are implicated in motivated behavior, conflict monitoring, cognitive control, and emotion regulation. Regions of the ACC located subgenually and rostral to the genu of the corpus callosum have reciprocal amygdalar connections and are implicated in the regulation of emotion. Dorsal ACC regions are interconnected to lateral and mediodorsal prefrontal regions and are involved in cognitive functions and the behavioral expression of emotional states ( ). An important function of the dorsal ACC is the ability to engage in aspects of cognitive control —the ability to pursue and regulate goal-oriented behavior. ACC–subcortical connections include the nucleus accumbens/ventromedial caudate, ventral globus pallidus, and ventral aspects of the magnocellular mediodorsal and ventral anterior thalamic nuclei. Deficit syndromes linked to the ACC–subcortical circuit include the spectrum of amotivational syndromes (apathy, abulia, akinetic mutism) and cognitive impairments including poor response inhibition, error detection, and goal-directed behavior. Some conditions that may affect this circuit include AD, FTD, PD, HD, head trauma, brain tumors, cerebral infarcts, and obstructive hydrocephalus.

Cerebrocerebellar Networks

The cerebellum is engaged in the regulation of cognition and emotion through a feed-forward and feedback loop. The cortex projects to pontine nuclei, which in turn project to the cerebellum. The cerebellum projects to the thalamus, which then projects back to the cortex. Cognitive processing tasks such as language, working memory, and spatial and executive tasks appear to activate the posterior cerebellar lobe. The posterior cerebellar vermis may function as a putative limbic cerebellum, modulating emotional processing ( ). Distractibility, executive and working memory problems, impaired judgment, reduced verbal fluency, disinhibition, irritability, anxiety, emotional lability or blunting, obsessive-compulsive behaviors, depression, and psychosis have been reported in association with cerebellar pathology in the context of the cognitive-affective cerebellar syndrome ( ).

Biology of Psychosis

Schizophrenia is a chronic disintegrative thought disorder where patients frequently experience auditory hallucinations and bizarre or paranoid delusions. Among several etiological hypotheses for schizophrenia, the neurodevelopmental model is one of the most prominent. This model generally posits that schizophrenia results from processes that begin long before clinical symptom onset and is caused by a combination of environmental and genetic factors ( ). Several postmortem and neuroimaging studies support this hypothesis with findings of brain developmental alterations such as agenesis of the corpus callosum, arachnoid cysts, and other abnormalities in a significant number of schizophrenic patients ( ). Environmental factors are associated with an increased risk for schizophrenia. These factors include being a first-generation immigrant or the child of a first-generation immigrant, urban living, drug use, head injury, prenatal infection, maternal malnutrition, obstetrical complications during delivery, and winter birth ( ). Genetic risks are clearly present but not well understood ( ). The majority of patients with schizophrenia lack a family history of the disorder. The population lifetime risk for schizophrenia is 1%; it is 10% for first-degree relatives and 4% for second-degree relatives. There is an approximately 50% concordance rate for monozygotic twins as compared with approximately 15% for dizygotic twins. Advancing paternal age increases risk in a linear fashion, which is consistent with the hypothesis that de novo mutations contribute to the genetic risk for schizophrenia. It is most likely that many different genes make small but important contributions to susceptibility. The disease typically manifests only when these genes are combined or certain adverse environmental factors are present. A number of susceptibility genes show an association with schizophrenia: catechol- O -methyl-transferase, neuroregulin 1, dysbindin, disrupted in schizophrenia 1 (DISC1), metabotropic glutamate receptor type 3 gene, and G27/G30 gene complex ( ). Research in twins and first-degree relatives of patients has shown that genes predisposing to schizophrenia and related disorders affect heritable traits related to the illness. Such traits include neurocognitive functioning, structural MRI brain volume measures, neurophysiological informational processing traits, and sensitivity to stress ( ). A small proportion of schizophrenia incidence may be explained by genomic structural variations known as copy number variants (CNVs). CNVs consist of inherited or de novo small duplications, deletions, or inversions in genes or regulatory regions. CNV deletions generally show higher penetrance (more severe phenotype) than duplications, and larger CNVs often have higher penetrance and/or more clinical features than smaller CNVs. These genomic structural variations contribute to normal variability, disease risk, and developmental anomalies; they also act as a major mutational mechanism in evolution. The most common CNV disorder, 22q11.2 deletion syndrome (velocardiofacial syndrome), has an established association with schizophrenia. Individuals with 22q11.2 deletions have a 20-fold increased risk for schizophrenia and constitute about 0.9%–1% of schizophrenia patients. When this syndrome is present, genetic counseling is helpful ( ). Studies are also identifying shared genetic risk for schizophrenia and autism spectrum disorders ( ).

A wide variety of neurological conditions, medications, and toxins are associated with psychosis. No consensus is available in the literature regarding the precise anatomical localization of various psychotic syndromes. Evidence from neurochemistry, cellular neuropathology, and neuroimaging studies supports that schizophrenia is a brain disease that affects multiple interacting neural circuits. The two best-known neurotransmitter models offered to explain the various manifestations of schizophrenia are the “dopamine hypothesis” ( ) and the “glutamate hypothesis.” Schizophrenia has been associated with frontal lobe dysfunction and abnormal regulation of subcortical DA and glutamate systems ( ).

Advances in structural and functional neuroimaging techniques over the past 30 years have greatly aided our understanding of neurocircuit alterations in schizophrenia. Structural studies have commonly identified diminished whole brain volume, increased ventricular size, and regional atrophy in hippocampal, prefrontal, superior temporal, and inferior parietal cortices in schizophrenic patients compared with control groups ( ). A reversal of or diminished hemispheric asymmetry has also been characterized. Functional neuroimaging studies have commonly identified decreased cerebral blood flow (CBF) and blood oxygen level–dependent (BOLD) hypoactivation of the prefrontal cortex (including the DLPFC) during cognitive task performance and temporal lobe dysfunction ( ). Schizophrenic patients with prominent negative symptoms have displayed reduced glucose utilization in the frontal lobes. A clinical and neurobiological overlap across schizophrenia, schizoaffective disorder, and bipolar disorder is also increasingly recognized ( ). Overall, functional imaging studies suggest that the DLPFC, OFC, ACC, ventral striatum, thalamus, temporal lobe subregions, and cerebellum are sites of prominent functional alterations. Several neurological conditions that may manifest psychosis (e.g., HD, PD, frontotemporal degenerations, stroke) are commonly also associated with frontal and subcortical dysfunction. For example, dorsolateral and mediofrontal hypoperfusion on functional imaging has been demonstrated in a subset of AD patients with delusions ( ).

Biology of Depression

The intersection of neurology and psychiatry is nowhere more evident than the remarkable comorbidity of psychiatric illness, especially depression, in many neurological disorders, with a 20%–60% prevalence rate of depression in patients with stroke, neurodegenerative diseases, MS, headache, HIV, TBI, epilepsy, chronic pain, obstructive sleep apnea, intracranial neoplasms, and motor neuron disease. Depression amplifies the physiological response to pain ( ), whereas pain-related symptoms and limitations frequently lead to the emergence of depressive symptoms. In a community-based study, almost 50% of adolescents with chronic daily headaches had at least one psychiatric disorder, most commonly major depression and panic. Women with migraine who have major depression are twice as likely as those with migraine alone to report having been sexually abused when they were children. If the abuse continued past age 12, women with migraine were five times more likely to report depression ( ). Despite the proliferation of antidepressant therapeutics, major depression is often a chronic and/or recurrent condition that remains difficult to treat. Up to 70% of patients taking antidepressants in a primary care setting may be poorly adherent, most often due to adverse side effects during both short- and long-term therapy.

Although the heritability of idiopathic depression based on twin studies is estimated to be between 40% and 50% ( ), the genetics of depression have thus far proven difficult to fully elucidate ( ). Depression is a polygenetic condition that does not adhere to simple Mendelian genetics, and genetic mechanisms implicated in depression suggest complex gene–environment interactions. An individual’s genetic makeup may lead to increased susceptibility for the development of depression in the context of adverse environmental (psychosocial) influences. Behavioral genetics research based on stress-diathesis models of depression demonstrates that the risk of depression after a stressful event is enhanced in populations carrying genetic risk factors and is diminished in populations lacking such risk factors. A gene’s contribution to depression may be missed in studies that do not account for environmental interactions and may be revealed only when studied within the context of environmental stressors specifically mediated by that gene ( ). Genotype–environment interactions are ubiquitous because genes not only affect the risk for depression by creating susceptibility to specific environmental stressors but may also predispose individuals to persistently place themselves in highly stressful environments. Approaches to the study of genetic influences in depression include association studies of candidate genes, genetic linkage studies of pedigrees with a strong family history of depression, and genome-wide association studies.

Association studies in depression have focused on monoaminergic candidate genes ( ). An intriguing interaction between polymorphisms in the promoter region of the serotonin transporter (5-HTT) gene and depression as well as an association between 5-HTT promoter region polymorphisms and depression-related neurocircuit activation patterns has emerged. The promoter activity of the 5-HTT gene is modified by sequence elements proximal to the 5′ regulatory region, termed the 5-HTT gene-linked polymorphic region (5-HTTLPR). The short “s” allele of the 5-HTTLPR is associated with lower transcription output of 5-HTT mRNA compared with the long “l” allele. A prospective longitudinal study has demonstrated that individuals with one or two copies of the short allele exhibited more depressive symptoms and suicidality following stressful life events in their early 20s compared with individuals homozygous for the long allele ( ). Genome-wide association studies in depression, including treatment-refractory depression (TRD), have largely failed to identify robust, reproducible findings ( ). This suggests that genome-wide association studies in depression have been underpowered to date.

Studies of epigenetic mechanisms in depression, though in their early stages, appear to hold promise in elucidating the mechanisms by which environmental factors affect gene expression. Epigenetics is the study of changes in gene activity caused by factors other than changes in the underlying nucleotide sequence. Whereas the genomic sequence defines the potential genetic repertoire of a given individual, the epi-genome delineates which genes in the repertoire are expressed (along with the degree of expression) ( ). As an example, DNA methylation is one of several epigenetic modifications that influence gene expression. In a pioneering animal study probing the impact of early life experiences on subsequent epigenetic programming, rat pups who experienced high rates of licking and grooming behaviors (positive influences) exhibited decreased methylation at the glucocorticoid receptor transcription factor binding site ( ). A postmortem human study examining epigenetic glucocorticoid receptor regulation revealed increased methylation in the neuron-specific glucocorticoid receptor and decreased glucocorticoid receptor mRNA in suicide victims with a history of childhood abuse compared with nonabused suicide victims and nonsuicide controls ( ).

At the cellular neurobiological level, the potential clinical relevance of neurogenesis in the adult mammalian brain represents a recent major breakthrough in depression studies. Imaging studies have demonstrated a 10%–20% decrease in the hippocampal volume of patients with chronic depression ( ). Cell proliferation studies using 5-bromo-2′-deoxyuridine injection to label dividing cells show that antidepressants also lead to increased cell numbers in the mammalian hippocampus. This effect is seen with chronic but not acute treatment; the time course of the effect mirrors the known time course of the therapeutic action of antidepressants in humans (approximately 2 weeks for initial effect, upward of 4–8 weeks for maximal benefit) ( ). Although a role for neurogenesis in the pathophysiology of depression appears to be a promising avenue of research, the relevance of animal studies described here with respect to humans remains controversial ( ).

From a systems-level perspective, amygdalar-hippocampal, ACC, OFC, DLPFC, and subcortical regions are implicated in the neurobiology of primary and acquired depression ( ). Increased basal and stimuli-driven amygdalar activity has been extensively characterized in depression ( ). In an early PET imaging study, depressed patients with a family history of depression demonstrated increased activation of the left amygdala; this pattern of amygdalar hyperactivation was also observed in remitted subjects with a family history of depression ( ). This suggests that enhanced amygdalar activity potentially represents a trait vulnerability biomarker for depressive illness. A number of studies have specifically linked enhanced amygdalar activity to the negative attentional bias of information processing in depression. Increased amygdalar metabolic activity has also been positively correlated with plasma cortisol levels ( ), suggesting a link between elevated amygdalar activity and dysfunction of the hypothalamic–pituitary–adrenal axis.

Dysfunction of the prefrontal cortex also plays an important role in the pathophysiology of depression. The subgenual ACC has been implicated in the modulation of negative mood states ( ). Several neuroimaging studies characterized elevated baseline subgenual activation in depression ( ), whereas other investigations have described reduced subgenual activations ( ). Mayberg and colleagues have suggested that depression can be potentially defined phenomenologically as “the tendency to enter into, and inability to disengage from, a negative mood state” ( ). Subgenual ACC dysfunction may play a critical role in the inability to effectively modulate mood states. In addition to the ACC, the OFC and DLPFC exhibit abnormalities in depression. Consistent with OFC lesions linked to increased depression risk, depression severity is inversely correlated with medial and posterolateral OFC activity in neuroimaging studies ( ). Reduced OFC activations may lead to amygdalar disinhibition in depression. Meanwhile, the DLPFC potentially exhibits a lateralized dysfunctional pattern in depression. Though not consistently identified, depressed patients have shown left DLPFC hypoactivity and right DLPFC hyperactivity ( ); left DLPFC hypoactivity was linked to negative emotional judgments whereas right DLPFC hyperactivity was associated with attentional deficits. Subcortically, decreased ventral striatum/nucleus accumbens activation has been linked with anhedonia ( ). In neurological disorders, damage to the prefrontal cortex from stroke or tumor or to the striatum from degenerative diseases such as PD and HD is associated with depression ( ). Functional imaging studies of subcortical disorders such as these reveal that hypometabolism in paralimbic regions, including the anterotemporal cortex and anterior cingulate, correlates with depression ( ). Depression in PD, HD, and epilepsy has been associated with reduced metabolic activity in the OFC and caudate nucleus.

Functional imaging studies of untreated depression have been extended to evaluate responses to pharmacological, cognitive-behavioral, and surgical treatments. Clinical improvement after treatment with selective serotonin reuptake inhibitors (SSRIs) such as fluoxetine correlates with increased activity on PET in brainstem and dorsal cortical regions including the prefrontal, parietal, anterior, and posterior cingulate areas and with decreased activity in limbic and striatal regions including the subgenual cingulate ( ), hippocampus, insula, and pallidum. These findings are consistent with the prevailing model for the involvement of a limbic-cortical-striatal-pallidal-thalamic circuit in major depression. The same group has shown that imaging can be used to identify patterns of metabolic activity predictive of treatment response. Hypometabolism of the rostral anterior cingulate characterized patients who failed to respond to antidepressants, whereas hypermetabolism characterized responders. used PET to search for neuroimaging profiles that might predict clinical response to anterior cingulotomy in patients with TRD. Responders displayed elevated preoperative metabolism in the left prefrontal cortex and the left thalamus. A combination of functional imaging and pharmacogenomic technologies might allow subsets of treatment responders to be classified and outcomes to be predicted more precisely than with either technology alone. Goldapple and coinvestigators (2004) used PET to study the clinical response of cognitive-behavioral therapy (CBT) in patients with unipolar depression; they found increases in the hippocampus and dorsal cingulate and decreases in the dorsal, ventral, and medial frontal cortex activity ( ). The authors speculate that the same limbic-cortical-striatal-pallidal-thalamic circuit is involved but that differences in the direction of metabolic changes may reflect different underlying mechanisms of action of CBT and SSRIs. Resting-state metabolism of the right anterior insula as determined by PET has also been identified as a potential treatment-selective biomarker in depression for CBT and SSRI treatment response ( ), although reliable neuroimaging biomarkers of treatment response in major depression remain ill defined ( ).

Clinical Symptoms and Signs Suggesting Neurological Disease

Many neurological conditions have associated psychiatric symptoms. Psychiatrists and neurologists must be intimately acquainted with features of the clinical history and examination that point to the need for further investigation. Box 10.3 outlines some key features that have historically suggested an underlying neurological condition. eBox 10.4 reviews some key areas of the review of systems that can be helpful when a patient is being assessed for neurological and medical causes of psychiatric symptoms. eTable 10.3 reviews abnormalities in the elemental neurological examination associated with diseases that can exhibit significant neuropsychiatric features.