Chapter 1. Approach to the Patient with Neurologic Disease

Neurology is regarded by many as one of the most difficult and exacting medical specialties. Students and residents who come to a neurology service for the first time may be intimidated by the complexity of the nervous system through their brief contact with neuroanatomy, neurophysiology, and neuropathology. The ritual they then witness of putting the patient through a series of maneuvers designed to evoke certain mysterious signs is hardly reassuring. In fact, the examination appears to conceal the intellectual processes by which neurologic diagnosis is made. Moreover, the students have had little or no experience with the many special tests used in neurologic diagnosis—such as lumbar puncture, EMG (electromyography), EEG (electroencephalography), CT (computed tomography), MRI (magnetic resonance imaging), and other imaging procedures—nor do they know how to interpret the results of such tests. Neurology textbooks only confirm their fears as they read the detailed accounts of the many unusual diseases of the nervous system.

The authors believe that many of the difficulties in comprehending neurology can be overcome by adhering to the basic principles of the clinical method. Even the experienced neurologist faced with a complex clinical problem depends on this basic approach.

The importance of the clinical method stands out more clearly in the study of neurologic disease than in certain other fields of medicine. In most cases, it consists of an orderly series of steps:

1. The symptoms and signs are secured with as much confidence as possible by history and physical examination.

2. The symptoms and physical signs considered relevant to the problem at hand are interpreted in terms of physiology and anatomy—i.e., one identifies the disorder(s) of function and the anatomic structure(s) that are implicated.

3. These analyses permit the physician to localize the disease process, i.e., to name the part or parts of the nervous system involved. This is the anatomic, or topographic diagnosis, which often allows the recognition of a characteristic clustering of symptoms and signs, constituting a syndrome. This step is called syndromic diagnosis and is sometimes conducted in parallel with anatomic diagnosis.

4. Expert diagnosticians often make successively more accurate estimates of the likely diagnosis, utilizing pieces of the history and findings on the examination to either further refine or exclude specific diseases. Flexibility of thought must be practiced so as to avoid the common pitfall of retaining an initially incorrect impression and selectively ignoring data that would bring it into question. It is perhaps not surprising that the method of successive estimations works well in that evidence from neuroscience reveals that this is the mechanism that the nervous system uses to process information.

5. From the anatomic or syndromic diagnosis and other specific medical data—particularly the mode of onset and speed of evolution of the illness, the involvement of nonneurologic organ systems, the relevant past and family medical histories, and the laboratory findings—one deduces the pathologic diagnosis and, when the mechanism and causation of the disease can be determined, the etiologic diagnosis. This may include the rapidly increasing number of molecular and genetic etiologies if they have been determined for a particular disorder.

6. Finally, the physician should assess the degree of disability and determine whether it is temporary or permanent (functional diagnosis); this is important in managing the patient's illness and judging the potential for restoration of function.

In recent decades, many of these steps have been eclipsed by imaging methods that allow precise localization of a lesion and furthermore often characterize the etiology of disease. Many of the elaborate parts of the examination that were intended to localize lesions are no longer necessary in daily clinical work. Nonetheless, insufficient appreciation of the history and examination and the resulting overdependence on imaging leads to diagnostic errors and has other detrimental consequences. A clinical approach is usually more efficient and far more economical than is resorting to scans. The loss of the personal impact by the physician that is created by listening to a story and observing responses to various maneuvers is regrettable. Images are also replete with spurious or unrelated findings, which elicit unnecessary further testing and needless worry on the part of the patient.

All of these steps are undertaken in the service of effective treatment, an ever-increasing prospect in neurology. As is emphasized repeatedly in later chapters, there is always a premium in the diagnostic process on the discovery of treatable diseases. Even when specific treatment is not available, accurate diagnosis may in its own right function as a therapy, as uncertainty about the cause of a neurologic illness may be more troubling to the patient than the disease itself.

Of course, the solution to a clinical problem need not always be schematized in this way. The clinical method offers a much wider choice in the order and manner by which information is collected and interpreted. In fact, in some cases, adherence to a formal scheme is not necessary at all. In relation to syndromic diagnosis, the clinical picture of Parkinson disease, for example, is usually so characteristic that the nature of the illness is at once apparent. In other cases it is not necessary to carry the clinical analysis beyond the stage of the anatomic diagnosis, which, in itself, may virtually indicate the cause of a disease. For example, when vertigo, cerebellar ataxia, unilateral Horner syndrome, paralysis of a vocal cord, and analgesia of the face occur with acute onset, the cause is an occlusion of the vertebral artery, because all the involved structures lie in the lateral medulla, within the territory of this artery. Thus, the anatomic diagnosis determines and limits the etiologic possibilities. If the signs point to disease of the peripheral nerves, it is usually not necessary to consider the causes of disease of the spinal cord. Some signs themselves are almost specific—e.g., opsoclonus for paraneoplastic cerebellar degeneration and Argyll Robertson pupils for neurosyphilitic or diabetic oculomotor neuropathy. Nonetheless, one is cautious in calling any single sign pathognomonic as exceptions are found regularly.

Ascertaining the cause of a clinical syndrome (etiologic diagnosis) requires knowledge of an entirely different order. Here one must be conversant with the clinical details, including the speed of onset, course, laboratory and imaging characteristics, and natural history of a multiplicity of diseases. Many of these facts are well known and form the substance of later chapters. When confronted with a constellation of clinical features that do not lend themselves to a simple or sequential analysis, one resorts to considering the broad classical division of diseases in all branches of medicine, as summarized in Table 1–1.

Irrespective of the intellectual process that one utilizes in solving a particular clinical problem, the fundamental steps in diagnosis always involve the accurate elicitation of symptoms and signs and their correct interpretation in terms of disordered function of the nervous system. Most often when there is uncertainty or disagreement as to diagnosis, it is found later that the symptoms or signs were incorrectly interpreted in the first place. Thus, if a complaint of dizziness is identified as vertigo instead of light-headedness or if partial continuous epilepsy is mistaken for a tremor or choreoathetosis, then the clinical method is derailed from the beginning. Repeated examinations may be necessary to establish the fundamental clinical findings beyond doubt. Hence the aphorism: A second examination is the most helpful diagnostic test in a difficult neurologic case.

To offer the physician the broadest perspective on the relative frequency of neurologic diseases, estimates of their approximate prevalence in the United States, taken from several sources, including the NIH, are given in Table 1–2. Donaghy and colleagues have provided a similar but more extensive listing of the incidence of various neurologic diseases that are likely to be seen by a general physician practicing in the United Kingdom. They note stroke as far and away the most commonly encountered condition; those that follow in frequency are listed in Table 1–3. More focused surveys, such as the one conducted by Hirtz and colleagues, give similar rates of prevalence, with migraine, epilepsy, and multiple sclerosis being the most common neurologic disease in the general population (121, 7.1, and 0.9 per 1,000 persons in a year); stroke, traumatic brain injury, and spinal injury occurring in 183, 101, and 4.5 per 100,000 per year; and Alzheimer disease, Parkinson disease, and amyotrophic lateral sclerosis (ALS) among older individuals at rates of 67, 9.5, and 1.6 per 100,000 yearly. Data such as these assist in guiding societal resources to the cure of various conditions, but they are somewhat less helpful in leading the physician to the correct diagnosis except insofar as they emphasize the oft-stated dictum that "common conditions occur commonly" and therefore should be considered a priori to be more likely diagnoses (see further discussion under "Shortcomings of the Clinical Method").

In neurology, perhaps more than any other specialty, the physician is dependent upon the cooperation of the patient for a reliable history, especially for a description of those symptoms that are unaccompanied by observable signs of disease. If the symptoms are in the sensory sphere, only the patient can tell what he sees, hears, or feels. The first step in the clinical encounter is to enlist the patient's trust and cooperation and make him realize the importance of the history and examination procedure.

The practice of making notes at the bedside or in the office is recommended. Of course, no matter how reliable the history appears to be, verification of the patient's account by a knowledgeable and objective informant is always desirable.

The following points about taking the neurologic history deserve further comment:

1. Special care must be taken to avoid suggesting to the patient the symptoms that one seeks. Errors and inconsistencies in the recorded history are as often the fault of the physician as of the patient. The patient should be discouraged from framing his symptom(s) in terms of a diagnosis that he may have heard; rather, he should be urged to give a description of the symptom—being asked, for example, to choose a word that best describes his pain and to describe precisely what he means by a particular term such as dizziness, imbalance, or vertigo. The patient who is given to highly circumstantial and rambling accounts can be kept on the subject of his illness by directive questions that draw out essential points.

2. The setting in which the illness occurred, its mode of onset and evolution, and its course are of paramount importance. One must attempt to learn precisely how each symptom began and progressed. Often the nature of the disease process can be decided from these data alone, such as in stroke. If such information cannot be supplied by the patient or his family, it may be necessary to judge the course of the illness by what the patient was able to do at different times (e.g., how far he could walk, when he could no longer negotiate stairs or carry on his usual work) or by changes in the clinical findings between successive examinations.

3. In general, one tends to be careless in estimating the mental capacities of patients. Attempts are sometimes made to take histories from patients who are cognitively impaired or so confused that they have no idea why they are in a doctor's office or a hospital. Asking the patient to give his own interpretation of the possible meaning of symptoms may sometimes expose unnatural concern, anxiety, suspiciousness, or even delusional thinking. Young physicians and students also have a natural tendency to "normalize" the patient, often collaborating with a hopeful family in the misperception that no real problem exists. This attempt at sympathy does not serve the patient and may delay the diagnosis of a potentially treatable disease.

The neurologic examination begins with observations of the patient while the history is being obtained. The manner in which the patient tells the story of his illness may betray confusion or incoherence in thinking, impairment of memory or judgment, or difficulty in comprehending or expressing ideas. A common error is to pass lightly over inconsistencies in history and inaccuracies about dates and symptoms, only to discover later that these flaws in memory were the essential features of the illness. A more extensive examination of attention, memory, cognitive ability, and language is undertaken if the history or the manner in which it is given indicates the problem lies in those spheres. Otherwise, asking the date and place, repeating words, and simple arithmetic are adequate screening procedures.

One then proceeds from an examination of the cranial nerves including the optic discs, neck, and trunk to the testing of motor, reflex, and sensory functions of the upper and lower limbs. This is followed by an assessment of the function of sphincters and the autonomic nervous system if appropriate and testing for meningeal irritation by examining the suppleness of the neck and spine. Gait and station (standing position) are observed before or after the rest of the examination.

When an abnormal finding is detected, whether cognitive, motor, or sensory, it becomes necessary to analyze the problem in a more elaborate fashion. Details of these sensitive examinations are addressed in appropriate chapters of the book (motor: Chaps. 3, 4, and 5; sensory: Chaps. 8 and 9; and cognitive and language disorders: Chaps. 22 and 23) and cursorily, below.

The neurologic examination is ideally performed and recorded in a relatively uniform manner in order to avoid omissions and facilitate the subsequent analysis of records. Some variation in the order of examination from physician to physician is understandable, but each examiner should establish a consistent pattern. Even when it is impractical to perform the examination in the customary way, as in patients who are unable to cooperate because of age or cognitive deficiency, it is good practice to record the findings in an accustomed and sequential fashion. If certain portions are not performed, this omission should be stated so that those reading the description at a later time are not left wondering whether an abnormality was not previously detected. Some aspects of the complete examination that were performed routinely by neurologists in former years are now infrequently included because they provide limited or duplicative information—among these are tests of olfaction and superficial reflexes but each finding may have a place in special circumstances or to corroborate another sign.

The thoroughness of the neurologic examination must also be governed by the type of clinical problem presented by the patient. To spend a half hour or more testing cerebral, cerebellar, cranial nerve, and sensorimotor function in a patient seeking treatment for a simple compression palsy of an ulnar nerve is pointless and uneconomical. The examination must also be modified according to the condition of the patient. Obviously, many parts of the examination cannot be carried out in a comatose patient; also, infants and small children, as well as patients with psychiatric disease, must be examined in special ways.

Portions of the general physical examination that may be particularly informative in the patient with neurologic disease should be included. For example, examination of the heart rate and blood pressure, as well as carotid and cardiac auscultation, are essential in a patient with stroke. Likewise, the skin can reveal a number of conditions that pertain to congenital, metabolic, and infectious causes of neurologic disease.

Numerous guides to the examination of the nervous system are available (see the references at the end of this chapter). For a full account of these methods, the reader is referred to several of the monographs on the subject, including those of Bickerstaff and Spillane, Campbell (DeJong's Neurological Examination), and of the staff members of the Mayo Clinic, each of which approaches the subject from a somewhat different point of view. An inordinately large number of tests of neurologic function have been devised, and it is not proposed to review all of them here. Some are described in subsequent chapters dealing with disorders of mentation, cranial nerves, and motor, sensory, and autonomic functions. Many tests are of doubtful value or are repetitions of simpler tests and thus should not be taught to students of neurology. Merely to perform all of them on one patient would require several hours and, in most instances, would not make the examiner any the wiser. The danger with all clinical tests is to regard them as indicators of a particular disease rather than as ways of uncovering disordered functioning of the nervous system. The following approaches are relatively simple and provide the most useful information.

Testing of Higher Cortical Functions

These functions are tested in detail if the patient's history or behavior has provided a reason to suspect some defect. Broadly speaking, the mental status examination has two main components, although the separation is somewhat artificial: the psychiatric aspects, which incorporate affect, mood, and normality of thought processes and content, and the cognitive aspects, which include the level of consciousness, awareness (attention), language, memory, visuospatial, and other executive abilities.

Questions are first directed toward determining the patient's orientation in time and place and insight into his current medical problem. Attention, speed of response, ability to give relevant answers to simple questions, and the capacity for sustained and coherent mental effort all lend themselves to straightforward observation. There are many useful bedside tests of attention, concentration, memory, and clarity of thinking including repetition of a series of digits in forward and reverse order, serial subtraction of 3s or 7s from 100, and recall of three items of information or a short story after an interval of 3 min. More detailed examination procedures appear in Chaps. 20, 21, 22, and 23. The patient's account of his recent illness, dates of hospitalization, and day-to-day recollection of recent incidents are excellent tests of memory; the narration of the illness and the patient's choice of words (vocabulary) and syntax provide information about language ability and coherence of thinking.

If there is any suggestion of a speech or language disorder, the nature of the patient's spontaneous speech should be noted. In addition, the accuracy of reading, writing, and spelling, executing spoken commands, repeating words and phrases spoken by the examiner, naming objects and parts of objects, and solving simple logical problems should be assessed.

The ability to carry out commanded tasks (praxis) has great salience in the evaluation of several aspects of cortical function. Bisecting a line, drawing a clock or the floor plan of one's home or a map of one's country, and copying figures are useful tests of visuospatial perception and are indicated in cases of suspected cerebral disease. The testing of language, cognition, and other aspects of higher cerebral function are considered in Chaps. 21, 22, and 23.

Testing of Cranial Nerves

The function of the cranial nerves must be investigated more fully in patients who have neurologic symptoms than in those who do not. If one suspects a lesion in the anterior cranial fossa, the sense of smell should be tested in each nostril; then it should be determined whether odors can be discriminated. Visual fields can be outlined by confrontation testing, ideally by testing each eye separately. If an abnormality is suspected, it should be checked on a perimeter and scotomas sought on the tangent screen or, more accurately, by computerized perimetry. Pupil size and reactivity to light, direct, consensual, and during convergence, the position of the eyelids, and the range of ocular movements should next be observed. Details of these tests and their interpretations are given in Chaps. 12, 13, and 14.

Sensation over the face is tested with a pin and wisp of cotton. Also, the presence or absence of the corneal reflexes, direct and consensually, may be determined.

Facial movements should be observed as the patient speaks and smiles, for a slight weakness may be more evident in these circumstances than on movements to command.

The auditory meati and tympanic membranes should be inspected with an otoscope. A high-frequency (512 Hz) tuning fork held next to the ear and on the mastoid discloses hearing loss and distinguishes middle-ear (conductive) from neural deafness. Audiograms and other special tests of auditory and vestibular function are needed if there is any suspicion of disease of the vestibulocochlear nerve or of the cochlear and labyrinthine end organs (see Chap. 15). The vocal cords must be inspected with special instruments in cases of suspected medullary or vagus nerve disease, especially when there is hoarseness. Voluntary pharyngeal elevation and elicited reflexes are meaningful if there is an asymmetrical response; bilateral absence of the gag reflex is seldom significant. Inspection of the tongue, both protruded and at rest, is helpful; atrophy and fasciculations may be seen and weakness detected. Slight deviation of the protruded tongue as a solitary finding can usually be disregarded, but a major deviation represents under action of the hypoglossal nerve and muscle on that side. The pronunciation of words should be noted. The jaw jerk and the snout, buccal, and sucking reflexes should be sought, particularly if there is a question of dysphagia, dysarthria, or dysphonia.

Testing of Motor Function

In the assessment of motor function, the most informative aspects are observations of the speed and strength of movements and of muscle bulk, tone, and coordination and these are considered in the context of the state of tendon reflexes. The maintenance of the supinated arms against gravity is a useful test; the weak arm, tiring first, soon begins to sag, or, in the case of a corticospinal lesion, to resume the more natural pronated position ("pronator drift"). The strength of the legs can be similarly tested with the patient prone and the knees flexed and observing downward drift of the weakened leg. In the supine position at rest, weakness due to an upper motor neuron lesion causes external rotation of the hip.

It is essential to have the limbs exposed and to inspect them for atrophy and fasciculations. Abnormalities of movement and posture as well as tremors may be revealed by observing the limbs at rest and in motion (see Chaps. 4, 5, and 6). This is accomplished by watching the patient maintain the arms outstretched in the prone and supine positions; perform simple tasks, such as alternately touching his nose and the examiner's finger; make rapid alternating movements that necessitate sudden acceleration and deceleration and changes in direction, such as tapping one hand on the other while alternating pronation and supination of the forearm; rapidly touch the thumb to each fingertip; and accomplish simple tasks such as buttoning clothes, opening a safety pin, or handling common tools. Estimates of the strength of leg muscles with the patient in bed are often unreliable; there may seem to be little or no weakness even though the patient cannot arise from a chair or from a kneeling position without help. Running the heel down the front of the shin, alternately touching the examiner's finger with the toe and the opposite knee with the heel, and rhythmically tapping the heel on the shin are the only tests of coordination that need be carried out in bed.

Testing of Reflexes

Testing of the biceps, triceps, supinator-brachioradialis, patellar, Achilles, and cutaneous abdominal and plantar reflexes permits an adequate sampling of reflex activity of the spinal cord. Elicitation of muscle stretch (tendon) reflexes requires that the involved muscles be relaxed; underactive or barely elicitable reflexes can be facilitated by voluntary contraction of other muscles (Jendrassik maneuver).

The plantar response poses some difficulty because several different reactions besides the Babinski response can be evoked by stimulating the sole of the foot along its outer border from heel to toes. These are (1) the normal quick, high-level avoidance response that causes the foot and leg to withdraw; (2) the pathologic slower, spinal flexor nocifensive (protective) reflex (flexion of knee and hip and dorsiflexion of toes and foot, "triple flexion"). Dorsiflexion of the large toe and fanning of the other toes as part of the latter reflex is the well-known Babinski sign (see Chap. 3); (3) plantar grasp reflexes; and (4) support reactions in infants. Avoidance and withdrawal responses interfere with the interpretation of the Babinski sign and can sometimes be overcome by utilizing one of several alternative stimuli (e.g., squeezing the calf or Achilles tendon, flicking the fourth toe, downward scraping of the shin, lifting the straight leg, and others) or by having the patient scrape his own sole. An absence of the superficial cutaneous reflexes of the abdominal, cremasteric, and other muscles are useful ancillary tests for detecting corticospinal lesions, particularly when unilateral.

Testing of Sensory Function

Because this part of the examination is attainable only through the subjective responses of the patient, it requires considerable patient cooperation. For the same reason, it is subject to overinterpretation and suggestibility. Usually, sensory testing is reserved for the end of the examination and, if the findings are to be reliable, should not be prolonged for more than a few minutes. Each test should be explained briefly; too much discussion with a meticulous, introspective patient encourages the reporting of meaningless minor variations of stimulus intensity.

It is not necessary to examine all areas of the skin surface. A quick survey of the face, neck, arms, trunk, and legs with a pin takes only a few seconds. Usually one is seeking differences between the two sides of the body (it is better to ask whether stimuli on opposite sides of the body feel the same than to ask if they feel different), a level below which sensation is lost, or a zone of relative or absolute analgesia (loss of pain sensibility) or anesthesia (loss of touch sensibility). Regions of sensory deficit can then be tested more carefully and mapped. Moving the stimulus from an area of diminished sensation into a normal area is recommended because it enhances the perception of a difference. The finding of a zone of heightened sensation ("hyperesthesia") calls attention to a disturbance of superficial sensation.

The sense of vibration may be tested by comparing the thresholds at which the patient and examiner lose perception at comparable bony prominences. We suggest recording the number of seconds for which the examiner appreciates vibration at the malleolus, toe, or finger after the patient reports that the fork has stopped buzzing.

Variations in sensory findings from one examination to another reflect differences in technique of examination as well as inconsistencies in the responses of the patient. Sensory testing is considered in greater detail in Chaps. 8 and 9.

Testing of Gait and Stance

The examination is completed by observing the patient arise from a chair, stand and walk. An abnormality of stance or gait may be the most prominent or only neurologic abnormality, as in certain cases of cerebellar or frontal lobe disorder; and an impairment of posture and highly automatic adaptive movements in walking may provide the most definite diagnostic clues in the early stages of diseases such as Parkinson disease. Having the patient walk tandem or on the sides of the soles may bring out a lack of balance or dystonic postures in the hands and trunk. Hopping or standing on one foot may also betray a lack of balance or weakness. Standing with feet together and eyes closed will bring out disequilibrium due to sensory loss (Romberg test) that is usually attributable to a disorder of the large diameter sensory fibers in the nerves and posterior columns of the spinal cord. Disorders of gait are discussed in Chap. 7.

In this situation, brevity is desirable but any test that is undertaken should be done carefully and recorded. Accurate recording of negative data may be useful in relation to some future illness that requires examination. As indicated in Table 1–4, the patient's orientation, insight, judgment, and the integrity of language function are readily assessed in the course of taking the history. With respect to the cranial nerves, the size of the pupils and their reaction to light, ocular movements, visual and auditory acuity, and movements of the face, palate, and tongue should be tested. Observing the bare outstretched arms for atrophy, weakness ("pronator drift"), tremor, or abnormal movements; checking the strength of hand grip and dorsiflexion at the wrist; inquiring about sensory disturbances; and eliciting the biceps, brachioradialis, and triceps reflexes are usually sufficient for the upper limbs. Inspection of the legs while the feet, toes, knees, and hips are actively flexed and extended; elicitation of the patellar, Achilles, and plantar reflexes; testing of vibration and position sense in the fingers and toes; and assessment of coordination by having the patient alternately touch his nose and the examiner's finger and run his heel up and down the front of the opposite leg, and observation of walking complete the essential parts of the neurologic examination.

This entire procedure adds only a few minutes to the physical examination but the routine performance of these few simple tests provides clues to the presence of disease of which the patient is not aware. For example, the finding of absent Achilles reflexes and diminished vibratory sense in the feet and legs alerts the physician to the possibility of diabetic or nutritional neuropathy, even when the patient does not report symptoms.

Although subject to obvious limitations, careful examination of the stuporous or comatose patient yields considerable information concerning the function of the nervous system. It is remarkable that, with the exception of cognitive function, almost all parts of the nervous system, including the cranial nerves, can be evaluated in the comatose patient. The demonstration of signs of focal cerebral or brainstem disease or of meningeal irritation is useful in the differential diagnosis of diseases that cause stupor and coma. The adaptation of the neurologic examination to the comatose patient is described in Chap. 17.

One is compelled in the examination of psychiatric patients to rely less on the cooperation of the patient and to be unusually critical of their statements and opinions. The depressed patient, for example, may perceive impaired memory or weakness when actually there is neither amnesia nor reduced power, or the sociopath or hysteric may feign paralysis. The opposite is as often true: Psychotic patients may make accurate observations of their symptoms, only to have them ignored because of their mental state. It is well to keep in mind that patients with even the most extreme psychiatric disease are subject to all of the neurologic conditions typical of others their age.

If the patient will speak and cooperate even to a slight degree, much may be learned about the functional integrity of different parts of the nervous system. By the manner in which the patient expresses ideas and responds to spoken or written requests, it is possible to determine whether there are hallucinations or delusions, defective memory, or other recognizable symptoms of brain disease merely by watching and listening to the patient. Ocular movements and visual fields can be tested with fair accuracy by observing the patient's response to a moving stimulus or threat in the visual fields. Cranial nerve, motor, and reflex functions are tested in the usual manner, but it must be remembered that the neurologic examination is never complete unless the patient will speak and cooperate in testing. On occasion, mute and resistive patients judged to be psychotic prove to have some widespread cerebral disease such as hypoxic or hypoglycemic encephalopathy, a brain tumor, a vascular lesion, or extensive demyelinative lesions.

The reader is referred to the special methods of examination described by André-Thomas and colleagues, Volpe and the staff members of the Mayo Clinic, which are listed in the references and described in Chap. 28. Many of these volumes address the developmental aspects of the child's nervous system, and although some signs may be difficult to obtain because of the age of the patient, they still stand as the best explications of the child's neurologic examination.

The general medical examination often reveals evidence of an underlying systemic disease that has secondarily affected the nervous system. In fact, many of the most serious neurologic problems are of this type. Two common examples will suffice: adenopathy or a lung infiltrate implicates neoplasia or sarcoidosis as the cause of multiple cranial nerve palsies, and the presence of low-grade fever, anemia, a heart murmur, and splenomegaly in a patient with unexplained stroke points to a diagnosis of bacterial endocarditis with embolic occlusion of cerebral arteries. The examination of a patient with stroke is incomplete without a search for hypertension, carotid bruits, heart murmurs, and irregular heart rhythm.

Once the technique of obtaining reliable clinical data is mastered, students and residents may find themselves handicapped in the interpretation of the findings by a lack of knowledge of the basic sciences of neurology. For this reason, each of the later chapters dealing with the motor system, sensation, special senses, consciousness, memory, and language is introduced by a review of the anatomic and physiologic facts that are necessary for understanding the associated clinical disorders.

At a minimum, physicians should know the anatomy of the corticospinal tract; motor unit (anterior horn cell, nerve, and muscle); basal ganglionic and cerebellar motor connections; main sensory pathways; cranial nerves; hypothalamus and pituitary; reticular formation of brainstem and thalamus; limbic system; areas of cerebral cortex and their major connections; visual, auditory, and autonomic systems; and cerebrospinal fluid pathways. A working knowledge of neurophysiology should include an understanding of neural excitability and nerve impulse propagation, neuromuscular transmission, and contractile process of muscle; spinal reflex activity; central neurotransmission; processes of neuronal excitation, inhibition, and release; and cortical activation and seizure production. The genetics and molecular biology of neurologic disease have assumed increasing importance in the past few decades. The practitioner should be familiar with the terminology of mendelian and mitochondrial genetics and the main aberrations in the genetic code that give rise to neurologic disease.

The wide availability of imaging offers the possibility of localization and etiologic diagnosis with limited input from the traditional clinical method. At a minimum, the educated neurologist must therefore be very familiar with the optimal imaging technique to disclose each of the multitudes of clinical diseases encountered in practice, the imaging appearance of each, and the risk and pitfalls of imaging.

From a practical diagnostic and therapeutic point of view, we believe the neurologist is helped by a knowledge of pathologic anatomy—i.e., the neuropathologic changes that are produced by disease processes such as infarction, hemorrhage, demyelination, physical trauma, compression, inflammation, neoplasm, and infection, to name the more common ones. Experience with the gross and microscopic appearances of these disease processes greatly enhances one's ability to explain their clinical effects. The ability to visualize the abnormalities of disease on nerve and muscle, brain and spinal cord, meninges, and blood vessels gives one a strong sense of which clinical features to expect of a particular disease and which features are untenable or inconsistent with a particular diagnosis. An additional advantage of being exposed to neuropathology is, of course, that the clinician is able to intelligently evaluate pathologic changes and reports of material obtained by biopsy. For many conditions there is a parallel representation of neuropathology through various imaging techniques. This allows the clinician to deduce the pathology from the imaging appearance.

From the foregoing description of the clinical method, it is evident that the use of laboratory aids, including imaging in the diagnosis of diseases of the nervous system is ideally preceded by rigorous clinical examination. As in all of medicine, laboratory study can be planned intelligently only on the basis of clinical information. To reverse this process is wasteful of medical resources and prone to the discovery of irrelevant information, and in some cases can expose a patient to unnecessary risk.

In the prevention of neurologic disease, however, the clinical method in itself is inadequate; thus, of necessity, one resorts to two other approaches, namely, the use of genetic information and laboratory screening tests. Biochemical screening tests are applicable to an entire population and permit the identification of neurologic diseases in individuals, mainly infants and children, who have yet to show their first symptom; in some diseases, treatment can be instituted before the nervous system has suffered damage. Similarly in adults, screening for atherosclerosis and its underlying metabolic causes is profitable in certain populations as a way of preventing stroke. Genetic information enables the neurologist to arrive at the diagnosis of certain illnesses and to identify patients and relatives at risk of developing certain diseases.

The laboratory methods that are available for neurologic diagnosis are discussed in the next chapter and in Chap. 45, on clinical electrophysiology. The relevant principles of genetic and laboratory screening methods for the prediction of disease are presented in the discussion of the disease to which they are applicable.

If one adheres to the clinical method, neurologic diagnosis is greatly simplified. In most cases one can reach an anatomic diagnosis. However, even after the most assiduous application of the clinical method and laboratory procedures, there are numerous patients whose diseases elude diagnosis. In such circumstances we have often been aided by the following perspectives:

As mentioned earlier, when the main sign has been misinterpreted—if a tremor has been taken for ataxia or fatigue for weakness—the clinical method is derailed from the start. Focus the clinical analysis on the principal symptom and signs and avoid being distracted by minor signs and uncertain clinical data.

As the lessons of cognitive psychology have been applied to error analysis in medical diagnosis, several heuristics (rules of thumb) have been identified as both necessary to the diagnostic process and as major pitfalls for the unwary clinician. The advantage of awareness of these heuristics is the opportunity to incorporate corrective strategies when shortcuts are employed. Investigators such as Redelmeier have given the following categories of cognitive mistakes that are common in arriving at a diagnosis:

• The framing effect reflects excessive weighting of specific initial data in the presentation of the problem.
• Anchoring heuristic, in which an initial impression cannot be subsequently adjusted to incorporate new data.
• Availability heuristic, in which experience with recent cases has an undue impact on the diagnosis of the case at hand.
• Representative heuristic refers to the lack of appreciation of the frequency of disease in the population under consideration, a restatement of Bayes theorem.
• Blind obedience, in which there is undue deference to authority or to the results of a laboratory test.

With our colleague Vickery, we have reviewed the workings of these heuristics in neurological diagnosis. Common to all of these cognitive errors is the tendency to come to early closure in diagnosis. Often this is the result of premature fixation on some item in the history or examination, closing the mind to alternative diagnostic considerations (the anchoring effect). The first diagnostic formulation should be regarded as only a testable hypothesis, subject to modification when new items of information are secured (anchoring heuristic). Should the disease be in a stage of transition, time will allow the full picture to emerge and the diagnosis to be clarified.

When several of the main features of a disease in its typical form are lacking, an alternative diagnosis should always be entertained. In general, however, one is more likely to encounter rare manifestations of common diseases than the typical manifestations of rare diseases (a paraphrasing of the representative heuristic).

It is advantageous to base diagnosis on clinical experience with the dominant symptoms and signs and not on statistical analyses of the frequency of clinical phenomena. Nonetheless, implicit in all diagnostic methods is an assessment of the likely causes of a sign or syndrome in the context of the patient's broad demographic characteristics including their sex, age, race, ethnicity, and the geographical circumstances. Moreover, as mentioned earlier, neurologists place a premium on finding treatable illnesses, even if the odds do not favor its presence.

As pointed out by Chimowitz, students tend to err in failing to recognize a disease they have not seen, and experienced clinicians may fail to appreciate a rare variant of a common disease. There is no doubt that some clinicians are more adept than others at solving difficult clinical problems. Their talent is not intuitive, as sometimes is presumed, but is attributable to having paid close attention to the details of their experience with many diseases and having catalogued them for future reference. The unusual case is recorded in memory and can be resurrected when another one like it is encountered. To achieve expert performance in all areas, cognitive, musical, and athletic, a prolonged period of focused attention to the subject and to personal experience is required.

Among medical specialties, neurology has traditionally occupied a somewhat anomalous position, in the past being thought of by many as little more than an intellectual exercise concerned with making diagnoses of untreatable diseases. This view of our profession is fallacious, as we have pointed out in a recent review of 200 years of neurological progress (Ropper). There are a growing number of diseases, many medical and others surgical, for which specific therapy is now available; through advances in neuroscience, their number is steadily increasing. Among the most sweeping changes, now that many infectious diseases of the nervous system are being addressed, have been entirely novel medications for stroke, multiple sclerosis, Parkinson disease, migraine, neuropathy, brain tumor and epilepsy. These therapies and the dosages, timing, and manner of administration of particular drugs are considered in later chapters in relation to the description of individual diseases. The neurologist must also be familiar with the proper application of surgical treatment when it is an integral part of the amelioration or cure of disease, as it is for brain tumor, degenerative and neoplastic diseases of the spine, cerebral aneurysm, extracranial arterial stenosis, and some congenital disease of the brain and spinal cord.

There are, in addition, many diseases in which neurologic function can be restored to a varying degree by appropriate rehabilitation measures or by the judicious use of therapeutic agents. Claims for the effectiveness of a particular therapy based on statistical analysis of large-scale clinical studies must be treated circumspectly. Was the study well conceived as reflected in a clearly stated hypothesis and outcome criteria; was there adherence to the plans for randomization and admission of cases into the study; were the statistical methods appropriate; and were the controls truly comparable? It has been our experience that the original results must be accepted with caution and it is prudent to wait until further studies confirm the benefits that have been claimed.

There are, of course, many instances in which evidence is not available or is not applicable to difficult individual therapeutic decisions. This is in part true because small albeit statistically significant effects may be of little consequence when applied to an individual patient. It goes without saying that data derived from trials must be used in the context of a patient's overall physical and mental condition and age. Furthermore, for many neurologic conditions there is, at the moment, inadequate evidence on which to base treatment. Here, the patient requires a skilled physician to make judgments based on partial or insufficient data. Even deciding actively to wait before committing to an intervention displays wisdom and adheres to the dictum, "first, do no harm". Even when no effective treatment is possible, neurologic diagnosis is more than an intellectual pastime. The first step in the scientific study of any disease process is its identification in the living patient.

In closing this introductory chapter, a comment regarding the extraordinary burden of diseases of the nervous system throughout the world is appropriate. It is not just that conditions such as brain and spinal cord trauma, stroke, epilepsy, mental retardation, mental diseases, and dementia are ubiquitous and account for the majority of illness, second only in some parts of the world to infectious disease, but that these are highly disabling and often chronic in nature, altering in a fundamental way the lives of the affected individuals. Furthermore, more so than in other fields, the promise of cure or amelioration by new techniques such as molecular biology, genetic therapy, and brain–computer interfaces has excited vast interest, for which reason aspects of the current scientific insights are included in appropriate sections.