CHAPTER 32

The Exteroceptive Sensations

Exteroceptive sensations originate in peripheral receptors in response to external stimuli and changes in the environment. There are four main types of general somatic sensation: pain, thermal or temperature sense, light touch or touch-pressure, and position sense or proprioception. Some include pruritic (itch) as a separate modality, and there is growing evidence for a modality that conveys the positive affective (pleasant) properties of touch.

Pain and Temperature Sensation

Anatomy and Physiology

Impulses carrying superficial pain sensation arise in nociceptors—free or branched nerve endings in the skin and mucous membranes. Some nociceptors respond to specific types of stimuli, whereas others are polymodal. Thermoreceptors for heat and cold sensation are free nerve endings in the dermis. Warm and cold stimuli activate different fibers. Pain and thermal sensation are carried along small myelinated A-delta and unmyelinated C nerve fibers to the dorsal root ganglion (DRG), where the first cell body is situated (Figure 32.1). The impulses in response to moderate heat or cold travel primarily over A-delta and some C fibers. The response to the pain associated with the extremes of temperature is conveyed along C fibers. Axons from small and intermediate size neurons in the DRG traverse the lateral division of the dorsal root to enter the dorsolateral fasciculus of the spinal cord (Lissauer’s tract), where they ramify longitudinally for one or two segments. The axons leave Lissauer’s tract, enter the posterior gray horn, and synapse in laminae I to V. Second-order neurons for the spinothalamic system lie primarily in laminae I, II, and V (see Chapter 24). The other related posterior horn cells are interneurons in the pain pathway. The posterior horn contains a variety of neurotransmitters; pain impulses are thought to be mediated primarily by substance P and glutamate. Activity in the spinothalamic tract (ST) neurons of the posterior horn is modulated by descending pathways. Stimulation of certain brain regions inhibits the response of ST cells to noxious stimuli. Descending influences are known to arise from the nucleus raphe magnus, periaqueductal gray, brainstem reticular formation, periventricular gray, ventral posterior lateral (VPL) thalamic nucleus, and parietal cortex and travel primarily in the corticospinal tract and dorsolateral funiculus. These pathways are important in pain control mechanisms.

FIGURE 32.1 Diagram of the spinal cord and dorsal root showing the peripheral receptors and terminations of fibers within the spinal cord.

The majority of axons originating from second-order spinothalamic neurons cross the midline in the anterior white commissure and gather into the anterior and lateral STs; a small proportion of fibers ascend ipsilaterally. Fibers crossing in the anterior white commissure are affected early in syringomyelia. In the past, anatomists thought the anterior ST carried crude touch and the lateral ST pain and temperature; current evidence suggests all these modalities are carried in both tracts, so the lateral and anterior STs are now sometimes lumped together as the anterolateral or ventrolateral system (ALS) or simply the spinothalamic tract or system. For clinical purposes, it remains useful to consider the pain and temperature pathways in the ST as a distinct system. The ST ascends in an anterolateral position, just medial to the anterior spinocerebellar tract (Figure 32.2). Intermingled with the fibers of the ST are ascending spinoreticulothalamic fibers, which contribute to the ALS. The ST is somatotopically organized, and the distribution of fibers is clinically relevant. Lowermost, sacral and lumbar, fibers entering first are displaced progressively more laterally by subsequently entering fibers. As the tract ascends, the sacral fibers come to lie most lateral and superficial, nearer to the surface of the cord (Figure 32.3), with cervical fibers most medial. There is also a slight rotation so that the sacral fibers also come to lie somewhat more posterior as the tract ascends. At midbrain levels, lower extremity and sacral fibers are posterior, and those from the upper limb and trunk are more anterior. Because the sacral fibers lie most laterally, an intramedullary spinal cord lesion, such as a neoplasm, may produce “sacral sparing,” preservation of sensation in a saddle distribution in the face of sensory loss otherwise present below a certain spinal level. Conversely, a compressive lesion pressing on the upper spinal cord may preferentially involve the sacral spinothalamic fibers, causing sacral dysfunction first. Fibers carrying deep pain are in general thought to lie nearer the midline than those carrying superficial pain. The spinoreticulothalamic fibers in the ALS subserve diffuse, poorly localized pain from deep and visceral structures. They may also be involved in the affective aspects of pain.

FIGURE 32.2 The lateral spinothalamic tract.

FIGURE 32.3 Diagram of cross section of the cervical region of the spinal cord showing the arrangement of fibers in the spinothalamic and pyramidal tracts and dorsal columns. Heavy dots indicate fibers carrying temperature sensation, crosses indicate fibers carrying pain sensation, and fine dots indicate fibers carrying tactile impulses. C, T, L, and S indicate fibers from or destined for cervical, thoracic, lumbar, and sacral levels of the spinal cord.

In the medulla, the ST lies peripherally, dorsolateral to the inferior olivary nuclei; in the pons, it is lateral to the medial lemniscus (ML) and medial to the middle cerebellar peduncle; in the mesencephalon, it is peripheral, dorsal to the ML and just dorsolateral to the red nucleus. It passes near the colliculi and enters the diencephalon just medial to the brachium of the inferior colliculus.

Pain and temperature fibers from the face enter the pons through the gasserian ganglion and then descend in the spinal tract of the trigeminal nerve to varying levels, where they synapse on neurons in the adjacent nucleus of the spinal tract (see Chapter 15). These second-order neurons decussate and form the trigeminothalamic tract, which runs near the ascending spinothalamic and lemniscal fibers (Figure 15.2). The other cranial nerves carrying exteroceptive pain sensation have ganglia comparable to the DRG and pathways corresponding to the trigeminothalamic tract. These are discussed in the chapters on the individual cranial nerves.

In the upper lateral midbrain, all the somatosensory fibers begin to converge. The ST fibers are joined in the rostral brainstem by the laterally migrating fibers from the ML and by ascending trigeminothalamic fibers so that ultimately all the fibers subserving somatosensory function run together as they approach the thalamus. The tracts enter the ventrobasal and ventral posterior nucleus of the thalamus together; body sensation fibers terminate in the VPL nucleus and facial sensation fibers in the ventral posterior medial (VPM) nucleus. There is detailed somatotopic organization within VPL and VPM. From the thalamus, fibers run in the thalamic radiations through the posterior limb of the internal capsule to the primary somesthetic cortex in the postcentral gyrus for conscious recognition. The primary somesthetic cortex communicates with the parietal sensory association cortex and with other cortical areas. Thalamocortical fibers also project to the superior bank of the sylvian fissure.

In the thalamoparietal radiations, fibers carrying lower-extremity sensation curve medially to the superior medial surface of the hemisphere adjacent to the medial longitudinal fissure; those from the upper body go to the midportion of the surface of the parietal lobe; those from the face terminate on the lateral, inferior portion of the postcentral gyrus (Figure 6.7). Fibers of the spinoreticulothalamic tract carry nociceptive information in the ALS. There are synapses in the brainstem reticular formation and medial part of the thalamus. Spinoreticulothalamic fibers terminate in the intralaminar thalamic nuclei. The thalamic neurons that mediate pain project both to the parietal lobe and to the limbic cortex. Projections from the intralaminar nuclei terminate in the hypothalamus and limbic system and probably mediate the affective and autonomic responses to pain. Evidence from both humans and nonhuman primates suggests that the posterior insula and medial operculum is an important cortical target of the spinothalamic system, and that operculo-insular (parasylvian) pain may exist as a distinct entity.

Descending pathways serve to modulate pain. Fibers from the frontal cortex and hypothalamus project to the midbrain periaqueductal gray. The descending pain modulation pathway then descends in the dorsal part of the lateral funiculus to the posterior horn. Descending fibers from the locus caeruleus, the raphe nuclei, and other brainstem areas also modulate the pain response. These descending pathways are important in endogenous pain control and opiate analgesia.

Clinical Examination

There are many methods for testing superficial pain sensation. A simple and commonly used method, as reliable as any, is to use a common safety pin bent at right angles so its clasp may serve as a handle. The instrument should be sharp enough to create a mildly painful sensation, but not so sharp as to draw blood. A hypodermic needle is far too sharp unless its point has been well blunted against some hard surface. A broken wooden applicator stick is often used and is usually satisfactory provided the shards are sharp. Adequately sharp ends can be obtained by holding the stick at the very ends while breaking it. Disposable sterile devices, sharp on one end and dull on the other, are commercially available. Although it is not necessary for the stimulating instrument to be sterile, whatever is used must be discarded after use on a single patient to avoid the risk of transmitting disease from accidental skin puncture. There is no place in modern neurology for reusable sharp instruments such as the Wartenberg wheel, but disposable pinwheels are available. Various sensory testing devices have been used experimentally. Instruments for evaluating sensation quantitatively are available commercially ( Video 32.1).

A helpful trick is to hold the pin or shaft of the applicator stick lightly between thumb and fingertip and let the shaft slide between fingertip and thumb tip with each stimulation. This helps insure more consistent stimulus intensity than putting a fingertip on the end of the instrument and trying to control the force with the hand or wrist. Experience teaches how to gauge the intensity of the applied stimulus and the expected reaction to it. The clinical evaluation of superficial pain, temperature, and touch sensation shows a reasonably good correlation with quantitative assessment.

It is best to do the examination with the patient’s eyes closed. The patient should be asked to judge whether the stimulus feels as sharp on one side as on the other, and whether it is as sharp distally as proximally. Always suggest that the stimuli should be the same, as by language such as, “Does this feel about the same as that?” Avoid such language as “Does this feel any different?” or “Which feels sharper?” Suggesting there should be a difference encourages some patients to overanalyze and predisposes them to spurious findings and a tedious, often unreliable examination. A commonly used technique is asking the patient to compare one side to the other in monetary or percentage terms, for example, “If this (stimulating the apparently normal side) side is a dollar’s worth (or 100%), how much is this (stimulating the apparently abnormal side) worth?” The overanalytical but neurologically normal patient often responds with an estimate on the order of “95 cents,” whereas the patient with real, clinically significant sensory loss is more apt to respond with “5 cents” or “25 cents.” Delivering alternately sharp and dull stimuli, as with the sharp and blunt ends of a safety pin and instructing the patient to reply “sharp” or “dull” may not elicit subtle sensory loss only detectable in comparison with an uninvolved area. Slight changes can sometimes be demonstrated in a cooperative patient by asking her to indicate the alterations in sensation when a pinpoint is drawn lightly over the skin. A cooperative patient with a discrete distribution of sensory loss may be able to map out the involved area quite nicely if instructed how to proceed and left alone for a short time with tools and a marking instrument. The affected area can then be compared with a figure showing sensory distributions (see Web Link 32.1). As noted in Chapter 31, if the presenting complaint is a discrete area of altered sensation, begin testing in a presumably normal area, then proceed to the involved area, and map out the extent of the abnormality.

The latent time in the response to stimulation is eliminated and the delineation more accurate if the examination proceeds from areas of lesser sensitivity to those of greater sensitivity rather than the reverse. If there is subjective hypalgesia, move from areas of decreased sensation to those of normal sensation; if there is hyperalgesia, proceed from the normal to the hyperalgesic area. There may be a definite line of demarcation between the areas of normal and abnormal sensation, a gradual change, or at times a zone of hyperesthesia between them. It is occasionally useful to move from the normal to the numb area. In myelopathy, a spinal sensory level that is the same going from rostral to caudal as from caudal to rostral suggests a very focal and destructive lesion; when the two levels are far apart, the lesion is usually less severe.

If testing is done too rapidly, the area of sensory change may be misjudged. Applying the stimuli too close together may produce spatial summation; stimulating too rapidly may produce temporal summation. Either of these may lead to spurious findings. If stimulation is too rapid, or if conduction is delayed, a given response may refer to a previous stimulation. Stimuli should be applied at irregular intervals to avoid patient anticipation. If the patient knows when to expect a stimulus, a seemingly normal response can occur even from an anesthetic area. Include control stimuli from time to time, especially if the patient is comparing sharp and dull (e.g., using the dull end of the pin while asking if it is sharp), to be sure the patient has understood the instructions and is paying attention.

Temperature testing is difficult. Temperature sensation may be tested with test tubes containing warm and cool water, or by using various objects with different thermal conductivity. Ideally, for testing cold, the stimuli should be 5°C to 10°C (41°F to 50°F), and for warmth°C to 45°C (104°F to 113°F). The extremes of free-flowing tap water are usually about 10°C and 40°C. Temperatures much lower or higher than these elicit pain rather than temperature sensations. Normally, it is possible to detect a difference of about 1°C in the range around 30°C. The tubes must be dry, as dampness may be interpreted as cold. The tines of a tuning fork are naturally cool and work well for giving a quick impression of the ability to appreciate coolness. The tines quickly warm with repeated skin contact; applying the tines alternately and waving the fork in the air between stimuli helps prevent this warming. Holding the tines under cold running tap water may also be helpful. Some examiners warm one tine deliberately by rubbing and then test the ability to discriminate between the warm side and the cool side of the fork. This technique has limited practicality because the cool side warms so rapidly with skin contact. The latency for detecting temperature is longer than for other sensory modalities, and the application of the stimulus may need to be extended.

In temperature testing, it is usually sufficient to determine whether the patient can distinguish hot and cold stimuli. In peripheral neuropathy, a gradient sometimes occurs where the patient begins sensing cold or hot sensation more normally. To detect more subtle impairment in temperature sensibility, the Mayo Clinic developed disks of variable temperature conductance (steel, copper, glass, plastic) to determine if patients could discern subtle temperature variations.

In most instances, heat and cold sensibility are equally impaired. Rarely, one modality may be involved more than the other; the area of impaired heat sensibility is usually the larger. Pain and temperature sensibility are usually involved equally with lesions of the sensory system, and it is rarely necessary to test both. Temperature testing is more difficult and less often done. Testing temperature may be useful when the patient does not tolerate pinprick stimuli, has confusing or inconsistent responses to pain testing, or to help map an area of sensory loss. In some instances, the deficit is more consistent with temperature testing than with pinprick. Temperature testing may not be very reliable in patients with circulatory insufficiency or vasoconstriction causing acral coolness.

Quantitative sensory testing (QST) uses computerized neurophysiologic methods to examine sensation. It provides very accurately measured stimuli of various types and uses strict paradigms for recording responses. Temperature sensation is tested by delivering pulses of hot and cold and determining the threshold for detection. Extremes of temperature assess pain. Testing is expensive, is time consuming, and must be administered by trained technicians. There is good correlation between QST and clinical methods. QST is very useful for longitudinal research studies but of little clinical utility.

Tactile Sensation

Anatomy and Physiology

Cutaneous receptors that mediate light touch or general tactile sensibility include free nerve endings, Merkel cell endings, and encapsulated endings such as Meissner’s and pacinian corpuscles and Ruffini endings. All the encapsulated receptors function as mechanoreceptors with afferent nerve fibers in the group II and III range. Pacinian corpuscles are large, lamellated structures located subcutaneously in the palmar, plantar and digital skin, genitalia, and other sensitive areas; they function as rapidly adapting mechanoreceptors. They are especially responsive to vibration, most notably in the 40 to 1,000 Hz frequency range. Meissner tactile corpuscles are found primarily in thick hairless skin, such as the hand, foot, and lips, and are most highly developed in the finger pads. They also respond to vibration in the low-frequency range (10 to 400 Hz) and are maximally sensitive at 100 to 200 Hz. Merkel cell receptors are also slowly adapting mechanoreceptors that respond to low-frequency vibration. Ruffini endings are slowly adapting mechanoreceptors located in hairy as well as glabrous skin, in joint capsules, tendon insertions, and elsewhere. They are particularly responsive to stretching or indentation of the skin.

Light touch sensation is conveyed over large and small myelinated peripheral nerve fibers to unipolar DRG cells. The neurons subserving fine discriminative touch are the largest cells in the DRG. Tactile sensation follows several different pathways within the central nervous system. The central processes enter the spinal cord via the medial division of the posterior roots and bifurcate into ascending and descending fibers (Figure 32.4). Fibers carrying fine discriminatory and localized tactile sensibility then, without synapsing, turn upward in the ipsilateral posterior column. Fibers carrying crude touch synapse within several segments of their point of entry, and the axons of the neurons of the next order cross to the opposite ALS. Other tactile fibers have a synapse in the posterior horn and ascend in the dorsolateral funiculus to the lateral cervical nucleus at C1-C2, where axons of the next order neurons decussate and join the ML. In the posterior columns, fibers from the lumbosacral region aggregate near the midline, and fibers from successively more rostral regions aggregate in a progressively more lateral position, producing somatotopic lamination that is the reverse of the STs (Figure 32.3). In the STs, the sacral fibers are most lateral; in the posterior columns, the lowest fibers are most medial. All the fibers below about T8 are grouped together in the fasciculus gracilis; analogous fibers above T8 form the fasciculus cuneatus.

FIGURE 32.4 The tactile pathways.

ALS fibers transmit light touch and light pressure sensations, without accurate localization. The posterior column fibers are concerned with highly discriminatory and accurately localized sensibility, including spatial and two-point discrimination. Because of the overlap and duplication of function, and because of the multisynaptic pathways for general tactile sensation, tactile sensibility is the sensory modality least likely to be completely abolished with lesions of the spinal cord, and disturbances of it may fail to give localizing information. A myelopathy severe enough to abolish light touch will often render the patient nonambulatory. However, loss of light touch is common in peripheral neuropathy, and identifying the severity and extent of loss of touch sensation is extremely useful.

Axons in the gracile and cuneate fasciculi synapse with second-order neurons in the gracile and cuneate nuclei at the cervicomedullary junction. The second-order neurons sweep anteriorly as internal arcuate fibers, cross the midline, and accumulate in the ML. Within the medulla, the ML is a vertical band of fibers situated along the median raphe; in the pons, the tract becomes more horizontal and shifts to a ventral position; and in the mesencephalon, the tract migrates to lie far laterally in an oblique position. Somatotopic organization is maintained in the ML. In the medulla, the fibers from the nucleus gracilis lie ventrally and those from the nucleus cuneatus dorsally (homunculus erect). As the ML ascends the brainstem, it moves from a vertical, paramidline position gradually to a horizontal position (homunculus sits, then lies down). In the pons, fibers from the nucleus gracilis lie laterally and those from cuneatus medially. In the midbrain, the fibers from the nucleus gracilis lie dorsolaterally (homunculus in Trendelenburg). The lemniscal fibers are joined by analogous fibers subserving facial sensation that have decussated after synapsing in the trigeminal principal sensory nucleus in the pons. These fibers all terminate in the thalamus, from which the thalamocortical radiations project to the somatosensory cortex. The distribution of the tactile impulses within the thalamic nuclei and their radiation to the parietal cortex in general follow that for pain and temperature impulses.

Clinical Examination

Chapters 32 to 35 discuss sensory examination techniques. Interpretation and sensory localization are covered in Chapter 36.

There are many methods available for evaluating tactile sensation. Light touch can be tested with a wisp of cotton, tissue paper, a feather, a soft brush, light stroking of the hairs, or even using a very light touch of the fingertip. Creating a wisp of cotton on the end of a cotton-tipped applicator stick serves the purpose well. Some appreciation of light touch may be obtained by noting the responses to the blunt end of the stimulus used to test pinprick. More detailed and quantitative evaluation can be accomplished using Semmes-Weinstein filaments, an aesthesiometer, or von Frey hairs. These methods employ filaments of different thicknesses to deliver stimuli of varying, graded intensity.

For routine testing, simple methods using cotton, tissue, or fingertip suffice. It is enough to determine whether the patient recognizes and roughly localizes light touch stimuli and differentiates intensities. The stimulus should not be heavy enough to produce pressure on subcutaneous tissues. Ask the patient to say “now” or “yes” on feeling the stimulus or to name or point to the area stimulated. Allowance must be made for the thicker skin on the palms and soles and the especially sensitive skin in the fossae. Similar stimuli are used for evaluating discriminatory sensory functions such as tactile localization and two-point discrimination. It is best to avoid hairy skin because the sensory stimulation due to hair motion may be confused with the test stimulus; hairy skin is exceptionally sensitive to touch. Two-point discrimination is considered both a delicate tactile modality and a more complex sensation requiring cortical interpretation; it can be reduced in both peripheral and cortical lesions.

Semmes-Weinstein monofilaments are now used rather than von Frey hairs, after use in leprosy clinics showed that patients who could not detect the 10-g filament on the plantar foot surface were likely to develop ulcers (Figure 32.5). This concept was then used in diabetic neuropathy. Detailed bedside QST is tedious.

FIGURE 32.5 Examination of protective foot sensation using Semmes-Weinstein monofilaments.

Using painless and noninvasive reflectance in vivo confocal microscopy of skin, investigators are able to visualize and quantitate Meissner’s corpuscles (MC) in dermal papillae. Comparing the density of MC may prove very useful for noninvasive detection and monitoring of patients with sensory neuropathy. Epidermal nerve fiber layer assessment on skin biopsy has been used to evaluate patients with small fiber neuropathies and is rapidly becoming a useful clinical tool to assess the degree of small fiber loss in patients with normal electrophysiologic studies and the suggestion of predominant small fiber involvement on the neurologic examination. Other detectable changes in neuropathy include distortion of MC structure, focal thinning, or loss of myelin and short myelin internodes.

Web Links

Web Link 32.1. Peripheral nerve sensory distributions. http://www.neuroguide.com/nerveindex.html

Bibliography

Bourne S, Machado AG, Nagel SJ. Basic anatomy and physiology of pain pathways. Neurosurg Clin N Am 2014;25:629–638.

Cohen MS, Wall EJ, Brown RA, et al. 1990 AcroMed Award in basic science. Cauda equina anatomy. II: Extrathecal nerve roots and dorsal root ganglia. Spine 1990;15:1248–1251.

Defrin R, Ohry A, Blumen N, et al. Sensory determinants of thermal pain. Brain 2002;125:501–510.

Dyck PJ. Enumerating Meissner corpuscles: future gold standard of large fiber sensorimotor polyneuropathy? Neurology 2007;69:2116–2118.

Friehs GM, Schrottner O, Pendl G. Evidence for segregated pain and temperature conduction within the spinothalamic tract. J Neurosurg 1995;83:8–12.

Fuller G. Neurological Examination Made Easy. 5th ed. New York: Churchill Livingstone–Elsevier, 2013.

Garcia-Larrea L. Insights gained into pain processing from patients with focal brain lesions. Neurosci Lett 2012;520:188–191.

Garcia-Larrea L. The posterior insular-opercular region and the search of a primary cortex for pain. Neurophysiol Clin 2012; 42:299–313.

Gilman S. Clinical Examination of the Nervous System. New York: McGraw-Hill, 2000.

Gilman S, Newman SW. Manter and Gatz’s Essentials of Clinical Neuroanatomy and Neurophysiology. 10th ed. Philadelphia: FA Davis, 2003.

Herrmann DN, Boger JN, Jansen C, et al. In vivo confocal microscopy of Meissner corpuscles as a measure of sensory neuropathy. Neurology 2007;69:2121–2127.

Ikoma A, Rukwied R, Stander S, et al. Neurophysiology of pruritus: interaction of itch and pain. Arch Dermatol 2003;139:1475–1478.

Kandel ER. Principles of Neural Science. 5th ed. New York: McGraw-Hill, 2013.

Kennedy WR, Selim MM, Brink TS, et al. A new device to quantify tactile sensation in neuropathy. Neurology 2011;76:1642–1649.

Kiernan JA, Rajakumar N. Barr’s the Human Nervous System: An Anatomical Viewpoint. 10th ed. Philadelphia: Wolters Kluwer/Lippincott Williams & Wilkins, 2014.

Kikuchi S, Sato K, Konno S, et al. Anatomic and radiographic study of dorsal root ganglia. Spine 1994;19:6–11.

Krumova EK, Geber C, Westermann A, et al. Neuropathic pain: is quantitative sensory testing helpful? Curr Diab Rep 2012;12:393–402.

Massey EW, Pleet AB, Scherokman BJ. Diagnostic Tests in Neurology: A Photographic Guide to Bedside Techniques. Chicago: Year Book Medical Publishers, Inc., 1985.

McGlone F, Reilly D. The cutaneous sensory system. Neurosci Biobehav Rev 2010;34:148–159.

Nathan PW, Smith MC, Cook AW. Sensory effects in man of lesions of the posterior columns and of some other afferent pathways. Brain 1986;109:1003–1041.

Nathan PW, Smith M, Deacon P. The crossing of the spinothalamic tract. Brain 2001;124:793–803.

Nolano M, Provitera V, Crisci C, et al. Quantification of myelinated endings and mechanoreceptors in human digital skin. Ann Neurol 2003;54(2):197–205.

Noseworthy JH, Murray TJ, Lee SHA. Risks of the neurologist’s pin. N Engl J Med 1979;301:1288.

Posnick JC, Zimbler AG, Grossman JA. Normal cutaneous sensibility of the face. Plast Reconstr Surg 1990;86:429–433.

Ropper A, Samuels M, Klein J. Adams and Victor’s Principles of Neurology. 10th ed. New York: McGraw-Hill Medical, 2014.

Ross RT. How to Examine the Nervous System. 4th ed. Totowa: Humana Press, 2006.

Shy ME, Frohman EM, So YT, et al. Quantitative sensory testing: report of the Therapeutics and Technology Assessment Subcommittee of the American Academy of Neurology. Neurology 2003;60:898–904.

Siao P, Cros DP. Quantitative sensory testing. Phys Med Rehabil Clin N Am 2003;14:261–286.

Standring S, ed. Gray’s Anatomy: The Anatomical Basis of Clinical Practice. 41st ed. New York: Elsevier Limited, 2016.

Weibers DO, Dale AJD, Kokmen E, et al., eds. Mayo Clinic Examinations in Neurology. 7th ed. St. Louis: Mosby, 1998.

Wolf JK. Segmental Neurology. Baltimore: University Park Press, 1981.