The American Psychiatric Association (APA) has updated its Privacy Policy and Terms of Use, including with new information specifically addressed to individuals in the European Economic Area. As described in the Privacy Policy and Terms of Use, this website utilizes cookies, including for the purpose of offering an optimal online experience and services tailored to your preferences.

Please read the entire Privacy Policy and Terms of Use. By closing this message, browsing this website, continuing the navigation, or otherwise continuing to use the APA's websites, you confirm that you understand and accept the terms of the Privacy Policy and Terms of Use, including the utilization of cookies.

×
Clinical SynthesisFull Access

A Clinical Approach to Cognitive Impairment

Abstract

Cognitive impairment is ubiquitous in medicine and a common experience in everyday life. When severe, persistent, or progressive, it becomes a matter of clinical concern. Top causes of clinically significant cognitive impairment are neurodegenerative disease, stroke or atherosclerotic vascular disease of the brain, medication toxicity, and alcoholism. Several of these conditions are associated with aging of the population in industrialized nations, and they are comorbid with multiple other illnesses and polypharmacy. In evaluating such problems, history is paramount, including a detailed time course of symptoms and input from caregivers. Examination should include neurologic and mental status components and should be as thorough as possible. Ancillary testing should be strategic and address specific problems in the differential diagnosis. Appropriate referrals to other medical disciplines should be promptly arranged, with engagement of the primary care provider. The caregiver is often a hidden patient. The psychiatrist who works with cognitively impaired patients should stay abreast of developments in a rapidly evolving field and adopt a team approach to care with other clinical specialties and caregivers.

The Problem

Cognitive impairment is a common medical problem. There is hardly a disease in medicine that is not associated with some type of cognitive problem, if only in the realm of attention and executive performance. Cognitive failures are also common in everyday life: most people have experienced small lapses in concentration or ability to remember a name or a number or, especially when preoccupied, difficulty in recalling where they parked the car or left their keys. When such problems become severe, persistent, or progressive, they need to be clinically evaluated. It is these latter problems that are the focus of this article.

The problem of clinically significant cognitive impairment is associated with the aging of the population in the United States and other industrialized nations and increase in the prevalence of age-associated neurodegenerative and central nervous system (CNS) vascular disease. The epidemiological burden is certain to increase in the future; the population of 46 million Americans over 65 at present will surge to 100 million by 2060 (1). These trends are also important for the clinical evaluation of the patient: knowledge of epidemiology is a key factor in differential diagnosis. The consideration of a neurodegenerative disease would be unthinkable in a patient presenting with severe cognitive impairment when Alois Alzheimer wrote about Auguste Deter in 1901; yet, neurodegenerative disease would be on top of the differential diagnosis of a patient over 75 presenting with this problem in 2016. In the early 20th century, one of the first things in the clinician’s mind would be tertiary syphilis. Today, the top diagnosis—or at least a top rule-out—would be Alzheimer’s disease (AD). As for syphilis, consider for a moment how often a positive rapid plasma reagin (RPR) test occurs in the laboratory workup of a patient with dementia. The high prevalence of neurodegenerative or CNS vascular disease as cause of dementia also has implications for management. With the increasing cost of chronic care and shortfalls of trained caregivers, emphasis may shift toward solutions that help delay or prevent transfer to assisted living or nursing homes for chronic care, allow the patient to live at home, or assist (even incentivize) family members to provide care in the patient’s home environment.

This brief clinical guide is not meant to replace more comprehensive treatments of the subject, many of which will be used here as references. It is based on clinical experience from a large outpatient neuropsychiatric program and on 20 years of consulting for inpatients at the Sheppard and Enoch Pratt Hospital. Naturally, such patients tend to have a higher-than-average burden of psychiatric symptoms. I have placed special emphasis on the examination because, in my experience, psychiatric trainees have great difficulty in performing general neurological and cognitive aspects of the examination or incorporating the information they uncover in their diagnostic formulation. All concepts and procedures or probes mentioned here have been extensively tested and have proven helpful in the diagnosis of patients or in their treatment management.

Some Definitions

For cognitive impairment that is severe enough to require clinical attention, the term dementia is often used. The term is best reserved for cases that are severe enough to interfere with independent living, especially cases that feature progressive cognitive deficits. For example, it is inappropriate to use the term for attentional deficits or perceptual errors associated with high fever or lapses in working memory when someone is preoccupied with pressing life circumstances. On the other hand, the term is very appropriate and perhaps necessary to use for progressive neurodegenerative disease that affects a number of higher CNS functions, such as AD. A clinical problem often viewed as a minor form of dementia, at least in the DSM-5 classification, is that of mild cognitive impairment (MCI). The term applies to patients who show measurable, enduring impairments that do not cause major disruptions in everyday life. Although the term is often used in the context of memory loss, the same principle can apply to loss of language, executive function, psychomotor speed, and so on. Amnestic MCI will convert to AD in many cases; such development may be influenced by genetic factors (family history of dementia, presence of ApoE4 genotype), small hippocampal volume, and AD-type findings in the cerebrospinal fluid (CSF; see below). The concept of benign (normal) forgetfulness of the elderly population is useful for older patients with mild loss of episodic memory that does not impair daily functioning and is static with repeat evaluations. Distinction between MCI and benign forgetfulness is not always easy and depends on how successfully the individual can maintain a baseline level of daily functioning.

Central to the concept of dementia is a severe or progressive memory loss (amnesia). Although it is true that most common forms of dementia feature some degree of memory loss, there are types of cognitive decline that principally affect nonmnemonic faculties and impair the individual’s ability to function just as much as amnesia. Because of this, the clinician must be prepared to evaluate other cortical functions, such as language (the impairment of which is known as aphasia), complex perceptions of the world (the disturbance of which causes agnosia), constructional abilities and motor planning (the impairment of which is apraxia), or subcortical functions such as motor speed, motivation, and affect (deficits of these include psychomotor retardation and apathy). Complex frontal functions that underlie general planning, abstraction, and judgment (executive functions) are very often degraded along with focal cortical or subcortical functions, and the clinician should be proficient in extracting relevant information from the history and exam. All the above problems have been expertly reviewed by Mesulam in Harrison’s Neurology in Clinical Medicine (2). When patients with dementia become medically ill or take certain medications, or when daylight retreats in the evening, they may experience impairments in consciousness leading to an acute confusional state or delirium, often a negative prognostic sign. Other clinical problems accompanying dementia are sleep disorders; changes in thought form and content; affective changes; and problems with gaze, motor strength and dexterity, tone and flow of movement, and stance or gait.

Common Causes

It is impossible to review here all the disorders that cause cognitive impairment. Seeley and Miller have recently published a comprehensive and easy-to-use table of the most common of them, classified by neurodegenerative, atherosclerotic vascular, alcohol related, drug or toxin related, vitamin deficiency related, endocrine, infectious, trauma or other acquired injury related, and neoplastic etiology (3). Neurodegeneration, CNS atherosclerosis/stroke, medication toxicity, and alcoholism are on top of that list. Neurodegeneration is to a large extent age related, but it is not a natural part of the aging process. Although in most American and European studies the leading cause of dementia is AD followed by CNS vascular disease, the trend may be the reverse in other countries, such as Japan. In addition, the extent of vascular dementia in specific U.S. populations, such as African Americans, who may have a higher burden of vascular disease, is unknown. Also, atherosclerotic vascular findings are common in brain autopsies performed on AD patients (4), and this pattern is consistent with the increasing appreciation of CNS vascular disease in living older patients with dementia with the spreading use of magnetic resonance imaging (MRI). The relative contribution of vascular and neurodegenerative mechanisms in age-associated dementia is difficult to establish. Note that AD and vascular dementia share vascular risk factors.

The most common neurodegenerative dementias are AD; dementia with Lewy bodies; frontotemporal degeneration (FTD) and related disorders, such as progressive supranuclear palsy and corticobasal degeneration; Parkinson’s disease (PD); and prion diseases, such as Creutzfeldt-Jacob (CJD). AD, dementia with Lewy bodies, FTD, and other neurodegenerative diseases have both genetic and sporadic forms, of which the latter are much more common. The etiology of neurodegenerative diseases is a matter of active investigation. The field has closely followed developments in the neurosciences since the mid-1980s and has incorporated key paradigms and mechanisms—for example, the concepts of inappropriate, as contrasted to programmed or developmental, cell death, apoptosis, and autophagy. Current trends favor the destructive role of common neuronal proteins that become pathologically folded and aggregated, then recruit and thus “corrupt” normal protein, and then become propagated via synaptic contacts between neurons or other mechanisms. As these indigestible protein forms circulate across synapses, it is believed that host neurons in interconnected brain regions die and individuals develop symptoms. Each one of the main neurodegenerative diseases is featured by a signature protein or set of proteins: for example, AD is associated with amyloid beta (Aβ); FTD with tau, TDP-43, and FUS; dementia with Lewy bodies with α-synuclein; and CJD with PrPsc. Besides their importance for research and their role as possible targets for pharmacological interventions, these proteins are also targeted for specific diagnostic tests, beginning with AD (amyloid scans).

Vascular dementia is the CNS version of the disease that also causes coronary artery disease in the heart or peripheral vascular disease elsewhere; like its nonnervous correlates, it is associated either with large infarcts (multi-infarct dementia) or small-vessel disease. As mentioned earlier, the latter is increasingly appreciated with the now widespread use of MRI and often presents in the form of islands of vasogenic fluid–attenuated inversion recovery (FLAIR) signal, with or without corresponding T1 attenuation. The exact pathological nature of such signals is debated. Such findings are common also among normal individuals, but when these signals become confluent and occupy large areas of the central white matter of the hemispheres, the condition may be the cause of dementia (Binswanger’s disease). Confluent FLAIR signals are common among patients with memory complaints. A problem with the construct of microvascular dementia as an entity separate from AD is difficulty in assessing and quantitating such vascular changes under the microscope, contrasted with the ease of identifying and counting amyloid plaques and neurofibrillary tangles.

Traumatic brain injury (TBI) is one of the most common neurological diseases worldwide, with an estimated prevalence of 170 million. TBI is a common cause of cognitive impairment or dementia that often has frontal presentation and presents very differently from AD and other neurodegenerative diseases. Right-sided contusions are common in neuropsychiatric practice because they are typically associated with deterioration of mood and social comportment and with deficits in awareness (anosognosia). Besides aging and possibly genetics, moderate to severe TBI is one of the top causes of AD. It is also a cause of normal-pressure hydrocephalus, one of the reversible causes of dementia that can be treated with ventricular shunting. In the few cases of normal-pressure hydrocephalus I have seen in my clinic, shunting was effective for gait and perhaps urinary symptoms but much less so for cognitive impairments. On the other hand, repeat concussion in the course of contact and collision sports is associated with a trauma-induced neurodegenerative disease known as chronic traumatic encephalopathy.

Although the term dementia is often used in cases of chronic alcoholism with cognitive difficulties, the only established relationship between alcoholism and dementia is via chronic deficiency of B1 (thiamin), which is an effect of malnutrition rather than of alcohol per se. This infrequent condition is called Korsakoff’s syndrome and features severe anterograde amnesia, often leading to confusion and disorientation. Of course, the person with chronic alcoholism may have a whole host of transient cognitive difficulties for reasons related to intoxication, withdrawal, coexisting medical illness, and other complications. Although a significant percentage of alcoholics may have some degree of brain atrophy on structural neuroimaging, these changes tend to remit with sustained abstinence.

A number of occult malignancies (e.g., small-cell carcinoma, testicular tumors, and ovarian teratoma) may cause complex neuropsychiatric syndromes with confusion, amnesia, seizures, mood changes, or dementia. These syndromes are traditionally known as limbic encephalitis and belong to the larger group of paraneoplastic syndromes thought to result from autoimmune reactions triggered by neuronal antigens in tumor cells. A condition that often presents with fulminant psychiatric symptomatology and is likely to be seen by neuropsychiatrists is paraneoplastic encephalitis associated with NMDA receptor antibodies. This problem is usually seen in young women with ovarian teratoma. The neurology of paraneoplastic syndromes is an evolving and exciting field and has been expertly reviewed by Leypoldt and colleagues (5).

CJD and other prion diseases are rare causes of dementia, but the outpatient psychiatrist or neuropsychiatrist occasionally will be asked to evaluate patients with rapidly advancing dementia, focal cortical deficits, rigidity, and myoclonus that will unfortunately progress to death in about a year. These disorders have much in common with infectious and paraneoplastic encephalitis and require referrals to experts for recognition and definitive diagnosis, which is not always straightforward. The clinician interested in these conditions should consult recent reviews by Geschwind and others (6).

A number of conditions associated with cognitive impairment are treatable. These reversible causes of cognitive impairment include hypothyroidism, thiamin and B12 deficiency, normal-pressure hydrocephalus (see above), subdural hematoma, and the toxic effects of drugs, including prescription medications. In many of these cases, cognitive impairment is only an element of a larger neuropsychiatric picture and, in fact, may not be the main presenting problem. Nevertheless, keeping these conditions in mind prompts the clinician to order diagnostic laboratory tests and have the underlying problem treated. Depression in the elderly population is often considered a form of reversible cognitive impairment, although it is unclear whether the associated concept of depressive pseudodementia is a homogeneous category or is fully reversible in all cases. Delirium is sometimes mentioned as reversible dementia, but this notion has its drawbacks. First, changes in consciousness that are diagnostic in delirium are not common in dementia. Second, although many cases of delirium remit with the appropriate treatment of a general medical condition, delirium is often the early sign of an underlying dementing process and itself a negative prognostic sign for future cognitive and noncognitive functioning or survival.

The psychiatrist or neuropsychiatrist caring for patients with cognitive impairments would benefit enormously from consulting excellent reviews on the behavioral neuroanatomy of neocortex, the limbic system, and basal ganglia by Mesulam (7) and Pandya and colleagues (8). In a rapidly evolving field, a working knowledge of neuroanatomy can provide the conceptual framework within which the modern clinician can organize principles, scientific concepts, and clinical information, as well as follow diagnostic and treatment developments.

Common Patient Presentations

The problem of cognitive impairment is very broad and clinical presentations vary widely. However, several common patient profiles encountered in outpatient practice are worth illustrating here.

One common presentation is that of a patient who is usually older than 75, with recent episodes of progressive impairments in memory, often associated with difficulties in navigation and disorientation in the evening or in novel environments. Vascular risk factors such as hypertension, hyperlipidemia, or diabetes may be present. There are usually no focal neurological symptoms or signs or disturbances in gait or rigidity. This vignette is typical of AD.

Another presentation is the patient with an established history of vascular risk factors and often clinical vascular history, such as peripheral vascular disease, coronary artery disease, or stroke. This patient may seek consultation because of slow thinking, problems with motivation or low mood or excessive emotional responses, difficulties with occupational or instrumental activities, or problems with fine movements and gait. The neurological examination may or may not be focal. MRI often shows extensive white matter disease (FLAIR sequences), usually in the form of confluent high signal or lacunes in T1 sequences. Larger infarcts may or may not be present. Course varies and does not have to be progressive. This is the typical vascular dementia patient.

Another presentation is the combination of parkinsonism and dementia. One or the other may dominate the picture or may have started first, often several years apart. Detailed history may reveal the presence of REM sleep behavior disorder for many years prior to onset of symptoms; this condition is featured by absence of normal motor inhibition during REM sleep and the enactment of dream content that can be terrifying to the sleep companion. Some patients may have visual hallucinations or delusional thoughts for years prior to parkinsonism and a history of intolerance to antipsychotics, usually in the form of extreme rigidity that requires acute management; these patients also have fluctuating alertness with confusional episodes and frequent falls. Other patients have established history of paucity and slowness of movement and resting tremor that is then followed by progressive cognitive impairment and visual hallucinations, especially during treatment with L-dopa. Anxiety and mood disorders may be present in either of the two scenarios (9). These profiles are typical of Lewy body disorders (specifically, dementia with Lewy bodies and PD).

Other patients, typically younger, present with progressive language difficulties. Some patients speak little and struggle with words and sounds, but they tend to make sense; they may progress to mutism. Others speak fluently but with odd word choices and do not make sense; they do not understand the meaning of common nouns; they may also have difficulties recognizing faces or objects. Both groups of patients have progressive aphasia, the first of the nonfluent type and the second of the semantic type. They all have little or no awareness of the problem and may be inappropriate or disinhibited. Their memory is relatively intact. These are typical presentations of the progressive aphasic variants of FTD.

Another patient, also relatively young, may present with grossly inappropriate behaviors, sexual or otherwise, with callous disregard for the feelings of others, excessive eating or odd food choices, or expansive or incontinent mood. The person may have been seen by a general psychiatrist and received the diagnosis of hypomania or mania. Memory and language are relatively intact. This presentation is consistent with the behavioral variant of FTD.

Another patient may present with apathy and disinhibition, as in the behavioral variant of FTD, and then progress to freezing of vertical gaze, rigidity of the neck and torso, and dysarthria or problems with swallowing and a jerky gait. This picture is typical of progressive supranuclear palsy. In a related scenario, a patient may present with rigidity and apraxia of one hand or arm and often also with behavioral changes, anxiety, dysarthria, and a more global cognitive impairment. This is the patient with corticobasal degeneration.

In several cases, the presentation is linked to an index event in the patient’s history or is embedded in the course of an established neurological disorder. Such is the case of patients with a history of TBI or other acquired brain injury, such as hypoxia, or of sepsis. Moderate to severe TBI may present as frontal lobe dementia with cortical or subcortical features that run a static course through the years, but later they may worsen or convert to AD. Hypoxic ischemic encephalopathy is a particularly severe form of chronic dementia with poor prognosis. Sepsis is increasingly recognized as a cause of chronic cognitive impairment. Epilepsies are not commonly associated with dementia, except in some forms of childhood-onset epilepsy (e.g., Lennox-Gastaut syndrome) that are associated with intellectual disability. Some early-onset movement disorders, such as Huntington’s disease, are associated with dementia, whereas others, such as Wilson’s disease, are featured primarily by mood lability and psychosis.

Clinical Assessment

Clinical History

A careful, detailed clinical history, with input from caregivers, provides the main clues to the diagnosis and management of most neurological and psychiatric disorders; cognitive impairments are not an exception. If the clinician does not have a clear sense of direction after taking a detailed history, it is unlikely that the diagnosis will be straightforward; with such patients, there is often need for extensive laboratory investigation or referral to subspecialists. Detailed history includes an expanded narrative of present illness, beginning with open questions; a detailed history of medications, including prescription medications from all providers involved in the patient’s care as well as over-the-counter remedies; general medical history; developmental history regarding pregnancy, labor, and time of capturing of linguistic, motoric, and social milestones; history of learning impairments and childhood seizures; educational history, including school performance and educational attainment; and history of occupational and interpersonal adjustment.

Some aspects of medical history are especially relevant and should be elicited with targeted questions. These include known causes of cognitive impairment or dementia such as history of TBI, history of vascular disease and vascular risk factors, and history of hypoxia or sepsis. A history of major hospitalizations or surgeries and episodes of confusion or agitation related to these is important. History of substance abuse should include both alcohol and nonalcoholic substances, although alcoholism is by far the most relevant in the case of the cognitively impaired adult. Developmental and educational history and related achievement is extremely helpful in establishing baseline cognitive capacity and level of adaptation; such information will also allow the clinician to properly evaluate findings from bedside cognitive and mental status examination and data from specialized cognitive testing.

Medications and Polypharmacy

Medications, especially in the context of the all-too-common polypharmacy regimens, do cause cognitive impairments. Unfortunately, polypharmacy is becoming the norm in older patients. Known culprits are antihistamines; major anticholinergics such as tricyclic compounds; certain antipsychotic medications such as chlorpromazine, thioridazine, clozapine, and olanzapine; and benzodiazepines or other soporific GABA agonists. Anticholinergic compounds used for overactive bladder, such as the atropine analog tolterodine and others, do cross the blood-brain barrier and should be of concern. Many other commonly prescribed medications are known to block muscarinic receptors, at least in vitro, and thus exert anticholinergic effects. Tune and colleagues have published a table of atropine equivalence of such compounds that is worth consulting, especially when dealing with polypharmacy (10). Polypharmacy should always be considered as a factor in age-associated cognitive impairment regardless of pharmacodynamic specificity of individual agents; the clinician must establish, to the extent possible, associations between prescription history and symptoms.

Psychosocial Structure and Family History

Psychosocial structure and relationships are extremely important, if only to guide decisions on location of treatment and type of management. Family history of neuropsychiatric illness is naturally important; many conditions causing dementia have genetic foundations, with classical examples being autosomal-dominant conditions, such as Huntington’s disease and familial AD. The known pedigree should be screened for not only early-onset dementia and early-onset movement disorder but also history of psychotic or mood disorders. If there is no clear direction from the patient’s history, the clinician may discover important clues in an expanded family history that includes all medical conditions as well as causes of death of deceased members of the pedigree. Detailed review of systems is also important; in this regard, the assessment of the cognitively impaired patient does not differ from a visit to an internist’s office.

Examination

General

There is no such thing as a complete neuropsychiatric examination. The clinician should adjust the examination to the chief complaint and to important information gleaned from history. Here I will lay out items of examination that are relevant to the cognitively impaired adult patient across causes and presentations, anticipating that the clinician will choose those that best fit the needs of a particular case. It is useful to divide the exam into general medical, neurological, cognitive, and mental status components. The general medical exam should be selective and to a large extent guided by history and review of systems. Additional items of importance are general appearance and development, state of nutrition, and vital signs. When the patient complains of dizziness and there is evidence of tachycardia, the clinician should try to elicit orthostatic signs: the patient’s blood pressure and heart rate are taken in the sitting position and, after the patient stands and stays up for a few minutes, vitals are taken again. Patients taking commonly used antipsychotic agents that block α1 adrenergic receptors often show an increase in heart rate with minimal or no changes in blood pressure. This orthostatic postural tachycardia is a mild form of autonomic dysfunction and should be handled carefully, also based on clinical history: if the patient remains symptomatic with dizziness, there is risk of falling or other accidents. Other major findings such as fever, respiratory difficulties, anemia, and ankle edema, should be taken into consideration and properly referred for further investigation.

Neurologic Examination

The neurologic examination is useful because it may help with the diagnostic formulation, divulge problems requiring immediate attention by the general neurologist, or inform management choices (e.g., selection of medications). Also, the diagnosis of cognitive deficits such as agnosia or apraxia requires proving that the patient has no motor weakness, extrapyramidal impairment, ataxia, or sensory deficits, all of which are tested in the general neurological exam.

The video library of neurologic examination by Martin Samuels that accompanies the last two editions of Harrison’s Principles of Internal Medicine (11) is a superb resource used in teaching in the Sheppard and Enoch Pratt Neuropsychiatry Program. At the very least, the neurologic examination should include cranial nerves, the motor system, and stance-gait. With the exception of Romberg testing, sensory examination is often unreliable with the neuropsychiatric patient. I strongly believe that psychiatrists taking care of cognitively impaired patientsshould know how to perform neurologic examinations with great skill. In the TBI clinics of the Neuropsychiatry Program, none of the patients with history of moderate to severe injury has normal neurological exams; this is also true for many patients in the memory clinic.

Cranial nerves.

The examination of cranial nerves is important for several reasons: metastases of breast, lung, and prostate cancer that may cause cognitive deficits often involve the meninges; cranial nerve findings are often associated with disease in the frontal lobes or the limbic system that may point to specific causes of cognitive impairment; cranial nerve deficits, especially those related to vision and hearing, may compound cognitive dysfunction and increase the risk of confusion and accidents; findings from certain cranial nerves—for example, glossopharyngeal or hypoglossal (IX and XII)—may alert to aspiration risk with certain medications; and pachymeningitis associated with tertiary syphilis, a major cause of cognitive complaints in the past, is classically associated with cranial nerve impairments.

When it comes to testing special nerves, the olfactory nerve (I) is very important: anosmia is classically associated with contrecoup contusions in the ventral frontal regions and may be one of the earliest findings in PD. I recall one case in which anosmia was the main clue for suspecting traumatic frontal contusion in a patient with legal history of sexual aggression. The olfactory nerve can be easily examined with glass tubes or small bottles containing substances with distinctive odors, such as coffee, cinnamon, and lavender. The set of cranial nerves supporting visual functions (II, III, IV, VI, and VIII) can then be examined. The second nerve can be screened with a portable ophthalmoscope for fundoscopy (optic nerve, venous pulsations, fovea), Snellen chart for acuity (with the patient’s glasses on), and gross visual fields (various confrontation or extinction techniques with one or both eyes open). Cranial nerves III, IV, and VI are examined by testing the pupil (shape, size, direct and consensual response to light, and near response with convergence and meiosis), conjugate eye movements (saccades and pursuits and presence or absence of double vision), and vestibular-ocular reflexes. Examination of the trigeminal (V) nerve is optional. The examination of the facial nerve (VII) is important for the distinction between infranuclear (Bell) and supranuclear nerve palsy. The hearing part of nerve VIII can be examined by testing air conduction, difference between air and bone conduction (Rinne), and voice recognition. The glossopharyngeal (IX) and hypoglossal (XII) nerves can be examined by eliciting lingual, buccal, and guttural sounds and checking palate elevation and tongue appearance and strength. The accessory (XI) nerve can be examined by asking the patient to perform various neck movements and shoulder shrug.

Motor system.

The examination of the motor system should begin by asking the question, “How much movement?” The answer is based on careful observation of quantity and spontaneity of movement. Little or no movement (bradykinesia, akinesia) is associated with parkinsonism or hypokinetic-hypoverbal states, as in the case of severe frontal lobe disease or catatonia. Decreased blinking or little facial movement betrays the presence of mild extrapyramidal disease. Too much movement, which is not easy to discern from normal in borderline cases, is typical of chorea. Large-amplitude abnormal movements are called ballism and slow twisting movements are called athetosis. The latter is classically associated with tardive dyskinesia, but I have seen both ballism and athetosis in a patient with traumatic brain injury. Akathisia is a type of excessive movement that compels the patient to move, often causes substantial discomfort, and may not always involve leg movements. Tics are repetitive simple movements. Myoclonic jerking is an important sign that may suggest metabolic or important neurological disease; it is often seen in CJD, and I have seen it in patients with advanced AD. Examination of motor strength includes a screening maneuver such as the pronator drift test: asking the patient to extend hands palms up or, in the recumbent position, to keep legs raised in 45 degrees. In both cases, a downward drift is suggestive of weakness. A more detailed examination of motor strength uses the Medical Research Council system, ranging from 0, which corresponds to absolutely no movement (not even palpable mini-contractions), to 5, corresponding to robust movement against gravity and resistance. In the clinic, the Hoover leg maneuver or upper-extremity equivalents are commonly used when a factitious movement disorder is suspected. Next, the examination of muscle tone is extremely important in the evaluation of the cognitively impaired patient. Decreased tone (hypotonia) may be associated with cerebellar disease but is more typical with peripheral neuropathy. Increased tone presents as spasticity, rigidity, and paratonia (gegenhalten). Spasticity, a velocity-dependent stiffness generating a clasp-knife feeling is typical of supranuclear disease of the corticospinal tract. Lead-pipe rigidity with or without superimposed tremor (cogwheel) is independent of velocity and is suggestive of idiopathic or drug-induced parkinsonism. Paratonia is a very interesting form of rigidity that varies with the effort of the examiner and can be seen in factitious disease, but it is quite common among cognitively impaired patients, often associated with frontal pathology of a cortical or subcortical nature; some degree of paratonia may be encountered with older patients who are cognitively normal. A proper evaluation of tremor requires paying attention to its relationship to activity (resting tremor in parkinsonism or action tremor in a variety of conditions, including drug-induced tremor), rate (parkinsonic tremor is slow), amplitude (cerebellar tremor is especially wide), and regularity. The examination of reflexes is a very important part of the exam.

Among the various classes of myotatic, proprioceptive, nociceptive, antigravity, and release reflexes, the clinician who evaluates the cognitively impaired patient needs to be proficient in eliciting the five proprioceptive (muscle stretch) reflexes and the plantar reflex. Antigravity reflexes are usually encountered with very ill or comatose patients. The diagnostic value of release reflexes (grasp, snout, rooting) is probably exaggerated, but it may be helpful to test them in some conditions—for example, when assessing a patient with late-onset hyperthymia and severe disinhibition or other manifestations of FTD. When elicited with the appropriate technique, a great increase or absence of muscle stretch reflexes is always pathological and requires further neurological investigation. The clinician should also be proficient in testing the plantar reflex for the presence or absence of Babinski sign. This sign is unmistakable evidence of corticospinal tract lesion; in the case of the cognitively impaired patient, such a lesion often localizes in higher levels of the tract in frontal cortex or the subcortical white matter.

Stance and gait.

The examination of stance and gait includes standard testing of cerebellar function (finger-nose-finger, rapid alternating movements, mirror test), and spontaneous as well as stressed (walking on toes, heel, or heel-to-toe) gait. The finger-tapping test is often used in the clinic to differentiate among cerebellar, extrapyramidal, or other movement disorders; it is often abnormal in patients with even mild forms of frontal lobe injury (sticky tap). The clinician should be expert in recognizing the main types and variants of abnormal gait, specifically the ataxic cerebellar gait, gaits associated with pyramidal and extrapyramidal disease (spastic or parkinsonic), the frontal small-step gait (marche à petits pas), and the stomping gait of posterior column disease (which is different from steppage). Finally, from the sensory exam, the clinician must be proficient in eliciting the Romberg sign, often a sign of proprioceptive and posterior column dysfunction associated with diseases that also cause cognitive impairment, such as subacute combined degeneration (vitamin B12 deficiency), HIV myelopathy, and syphilis.

For older patients, several factors need to be taken into account before the clinician can draw conclusions about the significance of neurological findings. First, several neurological functions change with normal aging, and motor speed is one of them; also, up to a third of normal patients over 75 may have positive glabellar sign (inability to inhibit a blink) or limited upgaze or downgaze without having supranuclear palsy. Second and more important, some age-associated deterioration of gait is to be expected because of subtle long tract dysfunction, mild parkinsonism, low-grade neuropathy, microvascular disease in the frontal lobes, and other factors (12). Such problems are significantly exacerbated by drugs, especially CNS-acting medications, including antidepressants and mood stabilizers. The combination of the previous factors with cognitive impairment and loss of vision or hearing represents a substantial fall risk and requires extra caution in prescribing for older individuals.

Cognitive Examination

From the patient’s history the clinician already has a good sense of attention, memory, and language functions, as well as insight and judgment. Nevertheless, cognitive functions should be formally tested, starting with the assessment of wakefulness and alertness. The two are not identical, and a distinction is often necessary in psychiatric practice. When it comes to wakefulness, the patient may be either awake or drowsy (can awaken with loud voice); I have occasionally seen bed-bound obtunded patients with severe TBI, but these are exceptional scenarios. Alertness is the preparedness or ability of the patient to pay attention. I have seen patients who looked awake but were under the influence of opiates or large doses of psychotropic agents or had mild hypoglycemia; their alertness was clearly impaired. The delirious patient may be awake but not alert. Moreover, the manic patient or the patient in extreme fear or pain is in a state of hyperalertness that usually impairs attentional capacity. The theory and clinical examination of consciousness is becoming one of the most exciting fields in cognitive neuroscience, but the remarkable findings related to the subclinical alertness of patients in minimally conscious states are beyond the scope of this article (13).

Attention.

The cognitive exam then proceeds with testing of attention: focus, vigilance, and flexibility (divided attention). Normal attention is the sine qua non for all cognitive processes. There is no point in testing orientation, memory, or other cognitive functions if attention is impaired; the exam stops there! As an example, memory cannot be tested in delirious patients; this is a waste of the clinician’s time, generates meaningless data, and should be strongly discouraged. Focus (immediate attention) is often tested with five letters forward and three letters backward. Vigilance is tested in various ways—for example, continuous performance using strings of spoken letters; other commonly used bedside probes, such as having the patient count out loud by serial sevens may also test vigilance. Divided attention is a good measure of working memory and is often tested with the Trail Making Test B, in written or verbal form; the test can be made substantially more difficult by pushing toward higher number-letters.

Memory.

When testing memory, it is important to test both verbal and nonverbal (spatial) domains, because they engage different sides of the brain: verbal memory implicates the left hemisphere of most right-handed and half or more left-handed patients, whereas nonverbal memory is more dependent on right hemisphere circuits. Word-list recall is a good bedside test for verbal memory. The hidden-object test is useful for the bedside assessment of nonverbal memory. Performance on these tests is often noncongruent in focal disease of the hemispheres, as with TBI patients, and can help differentiate between left- and right-sided injuries.

Aspects of language.

It is important to test receptive and expressive language, repetition, and naming. More than one of these faculties can be tested with one probe. For example, complex or difficult sentence repetition, such as the sentences contained in Montréal Cognitive Assessment testing (MoCA) (14) or the “ifs ands or buts” of the Mini-Mental State Exam (MMSE) (15), test repetition and articulation in addition to comprehension. Fluency can be further tested with letter-word generation and confrontational naming with various probes, such as the Boston Naming Test or the animal picture sets included in MoCA. Apraxia is closely related to aphasia and is elicited by testing the patient’s ability to carry out simple learned activities: the use of wrong joints or hand-as-object in pantomimes of simple learned acts that does not improve after demonstrating the correct way is known as ideomotor apraxia and is evidence of anterior brain dysfunction. In contrast, the display of incorrect plans of relatively complex actions that improves with correction (ideational apraxia) is an indication of posterior dysfunction.

Visual-constructional skills, as in cube or intersecting pentagon copying, or constructional skills, such as clock face design, engage both posterior and anterior circuitry and are good tests of associative functions. They are very useful screening tests and are often impaired in patients with neurodegenerative disease or certain types of TBI (e.g., diffuse axonal injury).

Frontal functions may be further tested with the patient’s ability to inhibit using go–no go or Stroop interference conditions or with alternating pattern probing, as detailed by Weintraub (16). Abstraction and judgment are difficult to evaluate in the clinic, but probing with similarities or proverbs can be informative. If time allows, it is important to go beyond routine probes and try to test the patient’s upper limits. For example, variants of “one should not throw stones in glass houses” are often overlearned through schooling or common use, but variants of “good is the enemy of the best” (or its reverse) are rather unusual and require higher levels of abstraction and creative ability. Likewise, similarity between “train and plain” is a much easier task than that between “statue and poem.” Additional bedside probes for reasoning, abstraction, and mental flexibility are described in some detail by Weintraub (16) and Taylor (17) and may involve simpler versions of the Visual-Verbal Test, visual reasoning by identifying errors in pictures depicting logical absurdities, and so on. The clinician should always adjust the evaluation to the patient in question.

I have found that MoCA, especially with the use of multiple forms to obviate the effect of learning, suffices for the purpose of a bedside cognitive evaluation and may be superior to the popular MMSE for patients with frontal lobe syndromes or TBI; it may also be more sensitive, especially for milder forms of cognitive impairment. Many bedside tasks, and also tasks included in neuropsychological batteries, test more than one cognitive domain. As an example, Trails B probes both resistance to interference (divided attention) and the ability of the brain to temporarily hold information to allow the execution of brief plans requiring parallel streams of information (working memory, a cognitive function corresponding to the memory of a computer).

The clinician must be sensitive to the fact that dissecting mental status in individual components in not only artificial, but also awkward: the brain does not function that way in real life. Some otherwise normal people find it difficult to respond to some of the questions, either because they are surprised by novelty or because some questions may be too arcane. Samuels (11) uses the example of awkwardness of asking the patient to recall details of the otherwise simple, one-sentence story, “Tom and Bill went fishing and Bill caught three black bass.” On the other hand, many individuals are not aware of the formal delineation of seasons as required in the MMSE, and they certainly cannot be blamed for not knowing what county (MMSE) they are in if the clinic or the physician’s office is not in their neighborhood.

Mental Status Examination

Mental status.

The term mental status implies a somewhat arbitrary separation from the cognitive examination and includes some very important items conventionally delegated to the territory of general psychiatry. In important aspects of the exam (e.g., the assessment of mood), the examiner’s subjective response can serve as a diagnostic tool. Mental status plays a central role in the evaluation of the cognitively impaired patient, both because of the very high frequency of psychiatric symptoms in such patients and also the difficulty in appreciating the degree of cognitive impairment without a good mental status. In the beginning, the clinician should pay attention to psychomotor activity, including its increase (as in activation or agitation) and decrease (as in bradykinesia/bradyphrenia, apathy, and sedation).

Form and content of thought.

The form and content of thought need to be fully evaluated. Proper syntax and grammar, linearity of associations, and flow of thinking in response to conversation are important aspects of thought form. Two extreme forms of thought disorder are jargon talk and looseness of associations. The former can be encountered across the spectrum of mental disorders, in fluent aphasia as well as severe psychosis, whereas the latter is more typical of schizophrenia. Disorders of thought content are more common and may include overvalued ideas, obsessions, delusions, and confabulatory thinking. Confabulations usually denote neurological illness; in neuropsychiatric practice, common causes are frontal lesions, such as traumatic contusions, and less commonly, Korsakoff’s syndrome.

Speech evaluation.

Speech evaluation dovetails with evaluation of language in the cognitive part of the exam and includes evaluation of rhythm, volume, rate, and so on. Articulation, prosody (melody, intonation, accents, etc.) and pragmatics of language are also tested in this part of the examination. Dysarthria is a very common accompaniment of nonfluent aphasia. Dysprosody or aprosody is common in focal right-sided disease such as stroke or traumatic contusion; it can be one of the most dramatic clinical findings, sometimes rendering a grammatically and syntactically intact speech totally unintelligible.

Affect.

The evaluation of the individual’s affect—that is, his or her displayed emotional tone individual—is based on features such as range, intensity, quality, and appropriateness. The patient with frontal lobe apathy typically has a constricted, or “flat,” affect. The patient with schizophrenia may have inappropriate affect (e.g., expressing joy when grief is expected). Some patients with multiple sclerosis may have incontinent affect, which is a combination of excess range with shallowness and grossly inappropriate display of joy or sadness with sometimes explosive laughter or crying in response to a stimulus. In contrast, a typical manic patient may display excess range but intense affect that often transmits to the interviewer a sense of elation or excitement and dominates the interaction. The examination of affect is both interesting and rewarding and may offer important clues as to the nature of the underlying neuropsychiatric process. Mood is different from affect. Its quality is derived rather than observable and is extracted from the patient’s or caregivers’ narrative and the examiner’s reflection of the patient’s emotion. In most cases, affect and mood are mutually congruent. A discrepancy may indicate underlying neurological illness, typically involving the frontal lobe or large portions of the white matter (TBI, stroke, multiple sclerosis, etc.). A typical example is a TBI patient who routinely displays a smiling face despite chronic depression.

Destructive thoughts, plans, or intentions.

Destructive thoughts, plans, or intentions, either suicidal or homicidal, need to be explored and assessed with specific questions to the patient and caregivers. Suicidality is not uncommon in neurological illness and in the early stages of cognitive impairment. Homicidality, which I have occasionally seen in male patients with TBI, especially those taking certain medications, needs to be thoroughly investigated with input by caregivers and others, especially if directed toward specific people, and it may generate legal obligations for the clinician. Aggression and assaultiveness are extremely common in TBI patients and need to be explored with respect to precipitating factors and also the presence or absence of remorse.

Insight and judgment.

Insight and judgment are important aspects of the examination of the cognitively impaired patient. Lack of awareness of illness (anosognosia) is a severe deficit of insight unique to some cognitively impaired patients, often with right-side lesions, and may range from lack of appreciation of severity, especially the social or interpersonal consequences of the problem, to an outright denial of illness (in the case of TBI, denial that injury has ever taken place). Subtle problems of insight are difficult to appreciate in the first visit and require more in-depth interviewing—or better yet, familiarity—with the patient over repeat visits. Judgment may be related to everyday life or to need for treatment or adherence to contracted interventions and treatment plans. The latter has major implications for management, including the form and location of treatment. It is often extremely difficult to know the full extent of deficits in insight and judgment unless one has the opportunity to observe the patient in the appropriate interpersonal and social environment (the problem of ecological validity).

Ancillary Testing

At the end of the bedside examination, the clinician must decide whether to order additional testing to decide on the diagnosis and directions of management. Such testing may include measuring specific cognitive functions in the neuropsychological laboratory; blood testing for markers of disease, especially markers that may reveal treatable dementia; brain imaging (structural, metabolic, or molecular); or, rarely, investigations of the cerebrospinal fluid (CSF).

The use of the neuropsychological laboratory varies widely among clinicians. It can also be particularly stressful for some patients because it entails multiple hours of intense questioning and tasking. In my practice, neuropsychological testing is used with patients early in the course of what appears to be progressive cognitive impairment or in atypical presentations. I may also consult with the neuropsychologist when progression is subtle or when there is discrepancy between subjective impairment and bedside findings. I do not routinely use neuropsychological testing to monitor progression or to evaluate response to medication.

Blood testing includes standard tests such as complete blood count, thyroid function tests (at least the thyroid-stimulating hormone test), and vitamin B12 levels (if borderline, along with the B12 substrates methylmalonate and homocysteine), as well as a comprehensive metabolic panel. I do not routinely order red blood cell sedimentation rate, HIV testing, or endocrine (adrenal, parathyroid) workups unless there is suggestive history or findings from standard tests. Testing for syphilis (RPR) is optional and depends on historical context; a positive test would need to be confirmed with treponemal tests. In general, diagnosis of neurosyphilis is a complicated matter and requires consulting with experts. A positive HIV finding for the cognitively impaired patient is also a specialized matter in need of consultation. Apolipoprotein E (ApoE) genotyping adds to the certainty of diagnosis of AD, but I do not order it routinely.

Cranial imaging choices also vary among clinicians. Whereas a noncontrast brain CT scan is probably satisfactory in most cases, CT scan with contrast is a better choice in patients with history of malignancy. If cost is not a limiting factor, and certainly in patients with vascular history or findings or atypical presentations, MRI is a better exam. Both brain anatomy (T1 sequences) and white matter disease (T2, FLAIR) should be evaluated. Anterior versus posterior patterns of atrophy, the presence of atrophy of the medial temporal lobe or hippocampus, and the presence of white matter disease are some of the issues that can be clarified with MRI. Diffusion weighted scans may be especially helpful in prion disease (cortical ribboning, high signal in basal ganglia). I have occasionally used SPECT or fludeoxyglucose PET in equivocal cases, at least for a gross impression of anterior versus posterior blood flow/metabolic deficit patterns. PET is expensive and usually not reimbursed by many payers unless justified on the merits of the case. Molecular imaging for amyloid and, in near future, tau, may be useful ancillary tests. The presence of Aβ signal does not usually help with the diagnosis of AD, but absence of the signal makes the diagnosis unlikely.

The CSF should not be routinely tested in the cognitively impaired patient, although it is commonly used in unusual or rapidly evolving dementias; it is also increasingly used in some places as ancillary testing for the diagnosis of AD. CSF examination can substantially help with the diagnosis of neurosyphilis. The stress protein 14-3-3, a marker that had been touted as useful in the diagnosis of CJD, is neither specific not particularly sensitive; CSF is often normal for patients with prion disorders. Low Aβ42 and increased total or phosphorylated tau and a certain range of values of Aβ42/tau ratio appear to be specific and sensitive biomarkers of AD, but unfortunately values of these proteins differ substantially from one laboratory to another. EEG is useful in the diagnosis of CJD (triphasic or polyphasic sharp waves) and delirium (alpha slowing) and follow its progress and resolution.

Approach to Management

Despite the diverse presentations of cognitively impaired patients, there are some general rules in patient management that apply across the board. An important rule is sensitive handling of news that can be devastating to patients and often cause for panic or despair. How the clinician will handle information or break the news depends on many factors, including mental status and psychosocial structure. For patients with mild or early disease, I often use the expression, “It is too early to know, but it looks like the problem is caused by brain disease that needs to be followed up.” This statement is telling the truth, because in most cases the diagnosis is established only with follow-up visits; it is also not particularly threatening to the patient, who is given time to process, discuss with family members, and adjust. Excessive referrals should be avoided, and the clinician should have specific questions in mind when sending the patient to specialists. The general practitioner should be engaged in all cases.

When it comes to treatment, perhaps the most important rule is dealing with polypharmacy before one begins to prescribe cognitive enhancer medications. Alleviating polypharmacy is a collaborative task that requires buy-in by the general practitioner and other important specialists, such as cardiologists or rheumatologists. The efficacy of drugs approved by the U.S. Food and Drug Administration for AD, such as cholinesterase inhibitors and memantine, is mediocre at best, although some patients with significant vascular components or parkinsonism show robust responses.

Supporting the caregiver and preventing, to the extent possible, caregiver burnout is a top priority: connecting the caregiver with disease-specific associations such as the Alzheimer Association, planning for respite breaks, involving social work, or even referring to therapy are all part of the concept of treating the caregiver as a hidden patient.

In the case of progressive neurodegenerative disease, the clinician should also serve as a consultant for life decisions, such as dealing with advance directives and issues related to power of attorney, management of property, and living will. Caring for the cognitively impaired patient is a multifaceted task that is often too much for any single individual, and the seasoned clinician who works in specialized memory clinics is not an exception. Indeed, it is a medical challenge that tests one’s abilities as clinician, scientist, and empathic human being. Many of the conditions that cause cognitive impairment may not yet be curable, but this fact stresses even more the importance of caring for the patient and the patient’s family and sticking by them through a very difficult period in their lives.

Dr. Koliatsos is the Stulman Scholar in Clinical Neuropsychiatry and Director, Neuropsychiatry Program, Sheppard and Enoch Pratt Hospital, Towson, Maryland; professor of Pathology and Neurology and associate professor of Psychiatry and Behavioral Sciences, Johns Hopkins University School of Medicine, Baltimore; and clinical professor of Psychiatry, University of Maryland School of Medicine, Baltimore (e-mail: ).

The author reports no financial relationships with commercial interests.

References

1 A Profile of Older Americans: 2015. Washington, DC, US Department of Health and Human Services, Administration on Aging, 2015. http://www.aoa.acl.gov/Aging_Statistics/Profile/index.aspx. Accessed July 26, 2016Google Scholar

2 Mesulam M: Aphasia, memory loss, and other focal cerebral disorders; in Harrison’s Neurology in Clinical Medicine, 3rd ed. Edited by Hauser SL. New York, McGraw-Hill Education, 2013Google Scholar

3 Seeley WW, Miller BL: Alzheimer’s disease and other dementias; in Harrison’s Neurology in Clinical Medicine, 3rd ed. Edited by Hauser SL. New York, McGraw-Hill Education, 2013Google Scholar

4 Dolan H, Crain B, Troncoso J, et al.: Atherosclerosis, dementia, and Alzheimer disease in the Baltimore Longitudinal Study of Aging cohort. Ann Neurol 2010; 68:231–240. doi: 10.1002/ana.22055CrossrefGoogle Scholar

5 Leypoldt F, Armangue T, Dalmau J: Autoimmune encephalopathies. Ann N Y Acad Sci 2015; 1338:94–114. doi: 10.1111/nyas.12553CrossrefGoogle Scholar

6 Geschwind MD: Rapidly progressive dementia. Continuum 2016; 22(2 Dementia):510–537Google Scholar

7 Mesulam BN: Principles of Behavioral and Cognitive Neurology. Oxford, Oxford University Press, 2000Google Scholar

8 Pandya D, Seltzer B, Petrides M, et al.: Cerebral Cortex: Architecture, Connections, and the Dual Origin Concept. Oxford, Oxford University Press, 2015CrossrefGoogle Scholar

9 Zweig RM, Disbrow EA, Javalkar V: Cognitive and psychiatric disturbances in parkinsonian syndromes. Neurol Clin 2016; 34:235–246. doi: 10.1016/j.ncl.2015.08.010CrossrefGoogle Scholar

10 Tune L, Carr S, Hoag E, et al.: Anticholinergic effects of drugs commonly prescribed for the elderly: potential means for assessing risk of delirium. Am J Psychiatry 1992; 149:1393–1394CrossrefGoogle Scholar

11 Harrison’s Principles of Internal Medicine, 19th ed. Edited by Kasper DL, Fauci AS, Longo DL, et al.. New York, McGraw-Hill, 2015Google Scholar

12 Jenkyn LR, Reeves AG, Warren T, et al.: Neurologic signs in senescence. Arch Neurol 1985; 42:1154–1157CrossrefGoogle Scholar

13 Fernández-Espejo D, Owen AM: Detecting awareness after severe brain injury. Nat Rev Neurosci 2013; 14:801–809. doi: 10.1038/nrn3608CrossrefGoogle Scholar

14 Nasreddine ZS, Phillips NA, Bédirian V, et al.: The Montreal Cognitive Assessment, MoCA: a brief screening tool for mild cognitive impairment. J Am Geriatr Soc 2005; 53:695–699CrossrefGoogle Scholar

15 Folstein MF, Folstein SE, McHugh PR: “Mini-Mental State”: a practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 1975; 12:189–198CrossrefGoogle Scholar

16 Weintraub S: Neuropsychological assessment of mental state; in Principles of Behavioral and Cognitive Neurology. Oxford, Oxford University Press, 2000Google Scholar

17 Taylor MA: The Fundamentals of Clinical Neuropsychiatry. Oxford, Oxford University Press, 1999Google Scholar