ANATOMYNumbness, pain, and/or weakness involving one or both arms are common reasons for referral to the neuromuscular clinician. These symptoms may be due to radiculopathy, brachial plexopathy, or one or more mononeuropathies. Some systemic etiologies for these focal neuropathic disorders have been discussed in preceding chapters (e.g., Lyme disease, vasculitis, and diabetes mellitus). This chapter will focus mainly on radiculopathies secondary to compression (e.g., degenerative joint disease and herniated discs), brachial plexitis, traumatic plexopathies, and focal mononeuropathies related to compression or entrapment. Before discussing the evaluation and management of these disorders, a review of the normal anatomy would be helpful.
ANATOMYRecall that there are seven cervical vertebrae, the first of which, the atlas, articulates with the skull’s occipital condyles. The orientation of this joint allows primarily for flexion/extension movements. The second cervical vertebra, the axis, has a superiorly directed bony prominence, the dens, which articulates with the atlas and allows for rotational movements of the head and neck. The third through seventh cervical vertebrae are composed of the vertebral bodies themselves as well as short pedicles giving rise to laminae, which end in comparatively short and often bifid spinous processes. The transverse processes arise near the junctional zone of the pedicle and lamina. Between the transverse processes at each vertebral level lies a sulcus for the spinal nerves.
The spinal nerves are composed of a dorsal root and a ventral root (Fig. 23-1). The dorsal root consists of sensory fibers emanating from the dorsal root ganglia that lie outside the spinal cord. These dorsal root fibers enter the posterolateral aspect of the spinal cord and into the dorsal horn. Along the anterior aspect of the spinal cord, two or as many as 12 individual rootlets arising from anterior horn cells, fila radicularia, fuse to form the ventral root. Just distal to the dorsal root ganglion, the ventral and dorsal roots merge to form the spinal nerve. In the cervical region, there are eight cervical spinal roots on each side but only seven cervical vertebrae (Fig. 23-2). The first cervical spine nerve arises between the skull and atlas. As a result, each numbered cervical nerve root is related to the bony level immediately inferior to it down to the T1 vertebra. For example, the fifth cervical nerve root exits the spinal column just superior to the fifth cervical vertebrae. The eighth cervical nerve root exits the spinal column superior to the first thoracic vertebra.
Figure 23-1. The spinal cord is depicted with multiple ventral and dorsal rootlets joining to form the mixed spinal nerve root. Communications between the sympathetic ganglia and the spinal nerves are appreciated, and the gray and white rami are seen as well. (Reproduced with permission from Ferrante MA. Brachial plexopathies: classification, causes, and consequences. Muscle Nerve. 2004;30(5):547–568.)
Figure 23-2. A sagittal section of the adult spinal column is depicted with the spinal cord demarcated by individual neural segments. Note the anatomic discrepancy between the termination of the spinal cord and vertebral column. The disparity between the spinal cord’s neural segment and associated bony level with respect to spinal nerve exit is also shown. (Modified with permission from Haymaker W, Woodhall B. Peripheral Nerve Injuries: Principles of Diagnosis. Philadelphia, PA: WB Saunders; 1953.)
At the intervertebral foramina, the spinal nerves are joined by the gray rami from the cervical sympathetic chain ganglia (Fig. 23-1). The superior cervical ganglion communicates with C1–4 spinal roots, the middle cervical ganglion with the C5 and C6 spinal nerves, and the inferior cervical ganglion with C8 and T1 spinal roots. Importantly, the sympathetic nerves to head and neck arise from the first thoracic segment. Thus, injuries to the T1 nerve root may result in ipsilateral Horner syndrome (miosis, ptosis, and anhidrosis). Just distal to the entry point of the gray rami, the cervical spinal nerves branch to form an anterior and posterior primary ramus. The nerve fibers in the posterior primary ramus innervate the paraspinal muscles, while the anterior primary rami of C5–T1 cervical spinal nerves form the brachial plexus (Fig. 23-3).
Figure 23-3. Diagrammatic representation of the brachial plexus (trunks, cords, and divisions) as well as its terminal nerves are depicted. A, anterior division; P, posterior division; n, nerve. (Modified with permission from Dumitru D, Zwarts MJ. Brachial plexopathies and proximal mononeuropathies. A, anterior division; P, posterior division; n, nerve. In: Dumitru D, Amato AA, Zwarts MJ, eds. Electrodiagnostic Medicine. 2nd ed. Philadelphia: Hanley & Belfus; 2002.)
A dermatome refers to the cutaneous region supplied by a specific spinal nerve root segment (Fig. 23-4). Notably, there is some overlap of the cutaneous innervation by individual spinal nerves. The motor fibers emanating from the anterior horn cells, which course through the ventral root, spinal root, brachial plexus, and finally individual nerves, innervate specific muscle groups. Most muscles are supplied by motor nerves arising from at least two spinal cord segments (e.g., the deltoid muscle is innervated by motor fibers within the C5 and C6 spinal roots).
Figure 23-4. Dermatomal representation. (Modified with permission from Keegan JJ, Garret FD. Segmental distribution of the cutaneous nerves in the limbs of man. Anat Rec. 1948;102(4):409–437.)
The brachial plexus is composed of three trunks (upper, middle, and lower), with two divisions (anterior and posterior) per trunk. Subsequently, the trunks divide into three cords (medial, lateral, and posterior), and from these arise the multiple terminal nerves innervating the arm (Table 23-1, Fig. 23-4).1–3 More specifically, the anterior primary rami of C5 and C6 fuse to form the upper trunk; the anterior primary ramus of C7 continues as the middle trunk, while the anterior rami of C8 and T1 join to form the lower trunk. Of note, in approximately 62% of anatomic dissections of the brachial plexus, the C4 spinal nerve contributes to the upper trunk.1,4 In this situation, the brachial plexus is said to be a “prefixed plexus,” in which all of the spinal nerve contributions usually are shifted up one level. As a result, the contribution from the T1 spinal segment to the lower trunk of the brachial plexus may be minimal. In contrast, in approximately 7% of anatomic dissections of the brachial plexus, C5 contributes minimally to the brachial plexus, a so-called “postfixed plexus.”1 In such cases, the spinal nerve contributions may be shifted down by one level; therefore, C7 contributes to the upper trunk while the lower trunk might receive nerves from the T2 spinal segment. However, the frequency of contributions of C4 and T2 to the brachial plexus is controversial, based on surgical explorations in patients following trauma.1,5
TABLE 23-1. INNERVATION OF THE MUSCLES IN THE UPPER LIMB
The anterior divisions of the upper and middle trunks fuse to form the lateral cord, while the anterior division of the lower trunk continues as the medial cord. The three posterior divisions of the upper, middle, and lower trunks join forming the posterior cord. The designations medial, lateral, and posterior cords refer to their respective anatomic positions relative to the axillary artery. The cords constitute the longest subsections of the brachial plexus.6 There is some anatomic variation and communication between nerve fibers running between the different cords.1,4For example, some nerve fibers may exit the lateral cord and join the medial cord. Thus, the ulnar nerve may have contributions from the C7 spinal nerve.
The terminal nerves arise from the brachial plexus and may be purely sensory, motor, or mixed sensorimotor (Fig. 23-3). The dorsal scapular nerve, long thoracic nerve, and a branch to the phrenic nerve arise directly from the spinal roots. The only two terminal nerves arising from the trunks are the subclavian and suprascapular nerves, and these both leave from the upper trunk. No terminal nerves come directly from the middle and lower trunk. The upper and lower subscapular and thoracodorsal nerves depart from the posterior cord, while the posterior cord terminates as the axillary and radial nerves. From the proximal aspect of the medial cord arises a single motor branch innervating the pectoral muscle, the medial pectoral nerve. The purely sensory medial brachial and medial antebrachial cutaneous nerves originate from the distal aspect of the medial cord. The medial cord terminates by sending a medial branch to the median nerve with the remnant continuing as the ulnar nerve. The lateral pectoral nerve comes off the proximal portion of the lateral cord. The lateral cord terminates as the musculocutaneous nerve and a lateral branch that joins a branch from the medial cord to form the median nerve. Individual terminal nerves are discussed in more detail below.
Although not a nerve arising from the brachial plexus, the spinal accessory nerve or cranial nerve XI courses through the neck and shoulder region and is often affected in brachial plexus injuries. The nerve consists of a bulbar or accessory component that arises from the medulla and a spinal portion that arises from the anterior horn cells in the cervical cord down to C6. The nerves from the bulbar origin supply the soft palate and contribute to the recurrent laryngeal and possibly parasympathetic fibers, which then merge into the vagal nerve to the heart. The spinal component ascends between the ligamentum denticulatum and posterior spinal nerve roots, enters the cranium through the foramen magnum, and then exits the skull via the jugular foramen. The nerve descends posterior to the digastric and stylohyoid muscles to the sternocleidomastoid muscle, which it innervates, and terminates in the trapezius muscle, which it also supplies.
The phrenic nerve is derived primarily from the C4 spinal nerve, but C3 and C5 roots may also contribute (Fig. 23-3). The phrenic nerve crosses the anterior scalene and enters the thorax, where it innervates the diaphragm.
The dorsal scapular nerve usually arises directly from the C5 spinal nerve shortly after it exits the intervertebral foramen (Fig. 23-5). The nerve courses between the middle and posterior scalene musculature and innervates the major and minor rhomboid muscles and the levator scapulae.
Figure 23-5. The posterior aspect of the thorax is shown, with the dorsal scapular and suprascapular nerves coursing to their respective muscles. The suprascapular nerve passes beneath the suprascapular notch (not depicted) as well as around the spinoglenoid notch, which are two potential areas of compromise. (Modified with permission from Haymaker W, Woodhall B. Peripheral Nerve Injuries: Principles of Diagnosis. Philadelphia, PA: WB Saunders; 1953.)
Branches arising from the C5, C6, and C7 spinal nerves join forming the long thoracic nerve. The nerve descends to the lateral chest wall, where it innervates the serratus anterior muscle.
The suprascapular nerve arises from the upper trunk shortly after it is formed (Fig. 23-3). The nerve descends posteriorly between the omohyoid and the trapezius muscles. In the posterior shoulder, it courses through the suprascapular notch under the scapula’s superior transverse ligament to innervate the supraspinatus muscle, then through the spinoglenoid notch to innervate the infraspinatus muscle (Fig. 23-5).
This is a small nerve that arises from the C5 root or upper trunk, which innervates the small subclavius muscle that runs between the clavicle and first rib.
The medial pectoral nerve arises from the medial trunk (Fig. 23-3). This nerve innervates both the pectoralis major and the pectoralis minor muscles. The major spinal contributions to this nerve are C8 and T1.
The lateral pectoral nerve innervates the pectoralis major. It usually comes from the lateral cord, but occasionally arises from the anterior division of the upper and middle trunks just prior to the formation of the lateral cord. This anatomic variation may explain the observation that in plexus injuries affecting the medial and lateral cords resulting in a flail arm, the strength of the pectoralis major muscle may be relatively preserved. The major spinal contributions of this nerve are C5–7.
The upper and lower subscapular nerves originate from the posterior cord in the axilla. The upper subscapular nerve innervates the subscapularis muscle, while the lower subscapular nerve supplies subscapularis and the teres major muscle. The major spinal contributions to this nerve are from C5 and C6.
The thoracodorsal nerve, also known as the middle subscapular nerve, comes off the posterior cord and innervates the latissimus dorsi muscle. This nerve can also arise in some cases from the radial and axillary nerves.4 The major spinal nerves contributing to the thoracodorsal nerve are C5–7, particularly C7.
The medial cutaneous nerve of the arm (medial brachial cutaneous nerve) originates from the medial cord and supplies sensation to the medial aspect of the arm. Its primary contribution comes from the C8 and T1 spinal nerves.
The medial cutaneous nerve of the forearm (medial antebrachial cutaneous nerve) usually projects from the medial cord, but it may arise from the medial cutaneous nerve of the arm.4 The nerve supplies sensation from the medial forearm and also originates from the C8 and T1 spinal nerves.
The lateral cord terminates as a bifurcation resulting in the musculocutaneous nerve and a lateral branch that combines with a branch from the medial cord to form the median nerve (Fig. 23-6). In about 5% of individuals, the musculocutaneous nerve originates from the anterior division of the upper trunk, in which case the lateral root to the median nerve arises from the middle trunk only.4 The major spinal nerves contributing to the musculocutaneous nerve are C5 and C6. In addition, C7 contributes to this nerve in at least half but less than two-thirds of cadavers examined.4 The musculocutaneous nerve innervates the coracobrachialis, biceps brachii, and brachialis muscles. It terminates as the lateral cutaneous nerve of the forearm, supplying sensation to the lateral aspect of the volar surface of the forearm.
Figure 23-6. The musculocutaneous nerve is the termination of the lateral cord and supplies the coracobrachialis, biceps brachii, and brachialis muscles. It terminates as the lateral antebrachial cutaneous nerve, which splits into two cutaneous branches to supply the radial aspect of the forearm. (Modified with permission from Haymaker W, Woodhall B. Peripheral Nerve Injuries: Principles of Diagnosis. Philadelphia, PA: WB Saunders; 1953.)
The axillary nerve contains portions of the spinal nerves arising from C5 and C6 and is one of the two terminal branches of the posterior cord (Fig. 23-7). The nerve usually originates near the subscapularis muscle posterior to the pectoralis minor muscle and then traverses the quadrangular or quadrilateral space formed inferiorly by teres major, laterally by the long head of the triceps brachii, medially by the humerus, and superiorly by the teres minor. Upon exiting this space, the axillary nerve innervates the teres minor and deltoid muscles. The axillary nerve also sends cutaneous branches that supply sensation to the lateral aspect of the proximal arm overlying the deltoid muscle.
Figure 23-7. One of the terminal branches of the posterior cord is the axillary nerve. It supplies both the teres minor and the deltoid muscles as well as providing cutaneous sensation to the skin overlying the deltoid muscle (upper lateral cutaneous nerve of the arm). (Modified with permission from Haymaker W, Woodhall B. Peripheral Nerve Injuries: Principles of Diagnosis. Philadelphia, PA: WB Saunders; 1953.)
The radial nerve contains contributions from mainly C5–8 (as well as T1 in approximately 10% of individuals) and, in essence, is a continuation of the posterior cord after the axillary nerve branches off (Fig. 23-8).1,7 While still in the axillary region, a posterior cutaneous nerve branches off the radial nerve to provide sensation to the posterior aspect of the upper arm to the level of the elbow. In the proximal arm, the radial nerve travels medial to the humerus and descends between the medial and long heads of the triceps muscle along the spiral groove. In the proximal arm, the radial nerve innervates the long, medial, and lateral heads of the triceps brachii and the anconeus muscles. Upon leaving the spiral groove in the mid- to distal aspect of the arm, the radial nerve courses down to the lateral aspect of the arm and innervates the brachioradialis and extensor carpi radialis longus as well as a small branch to the brachialis muscle, the latter receiving its main contribution from the musculocutaneous nerve. An additional branch, the posterior antebrachial cutaneous nerve, separates from the radial nerve in the mid-arm region and descends to supply sensation to the posterior aspect of the forearm.
Figure 23-8. The course and muscular innervation of the radial nerve is depicted. In the axilla and proximal arm, the triceps muscle is innervated, and the three sensory branches originate. Sensory disturbances can help localize a lesion at or proximal to the spiral groove. (Modified with permission from Haymaker W, Woodhall B. Peripheral Nerve Injuries: Principles of Diagnosis. Philadelphia, PA: WB Saunders; 1953.)
In the elbow region, the radial nerve splits to form the purely sensory superficial radial nerve and the purely motor posterior interosseous nerve. In this area is the so-called radial tunnel bound by the radius, the capsule of the radiocapitellar joint, the brachialis and biceps brachii tendons (forming the medial walls), and the brachioradialis, extensor carpi radialis, and extensor carpi ulnaris muscles (forming the lateral and anterior walls). The radial tunnel ends at the fibrous band around the superficial head of the supinator muscle, which is known as the arcade of Fröhse. The superficial radial nerve travels on the undersurface of the brachioradialis, outside the radial tunnel, into the forearm. Around the mid-forearm, the nerve moves more superficially and travels along the extensor aspect of the distal forearm. After the superficial radial nerve passes the wrist, it supplies sensation to the lateral, extensor surface of the hand and fingers, analogous to the median distribution on the palmar surface (except the distal aspects of the fingertips on the dorsal surface which are supplied by the median nerve). The posterior interosseous nerve traverses the radial tunnel and then descends under the arcade of Fröhse. The posterior interosseous nerve continues down the extensor aspect of the forearm. Along the way it innervates the supinator, extensor digitorum communis, extensor carpi radialis brevis, extensor digiti minimi, extensor carpi ulnaris, abductor pollicis longus, extensor pollicis longus and brevis, and the extensor indicis proprius.
The median nerve is formed by the fusion of branches from the lateral and medial cords (Fig. 23-9). The main spinal nerve contributions to the median nerve are C6–T1. Motor fibers arise from C6–T1 spinal segments, while sensory fibers are derived primarily from the C6 and C7 segments. Occasionally, C5 can also contribute to the median nerve.1 The sensory fibers travel through the upper and middle trunks to the lateral cord into the median nerve, while the motor fibers pass through all the trunks as well as the medial and lateral cords.
Figure 23-9. There are no muscular or cutaneous branches arising from the median nerve in the axillary region or arm. The first branch originating from the median nerve is to the pronator teres in the proximal forearm. (Modified with permission from Haymaker W, Woodhall B. Peripheral Nerve Injuries: Principles of Diagnosis. Philadelphia, PA: WB Saunders; 1953.)
The median nerve descends in the anterior compartment of the arm on the medial side to the antecubital fossa region. Past the elbow, the median nerve courses through the two heads of the pronator teres muscle and then between the flexor digitorum superficialis and profundus muscles to the wrist. In the forearm, the median nerve innervates the pronator teres, flexor carpi radialis, palmaris longus, and flexor digitorum superficialis muscles. In the upper to mid-forearm level, the anterior interosseous nerve branches from the main median nerve. This is a pure motor nerve that supplies the flexor digitorum profundus 1 and 2, flexor pollicis longus, and pronator quadratus muscle. The main median nerve trunk continues distally down the forearm to the wrist. Just before entering the carpal tunnel, the palmar cutaneous branch arises to supply sensation over the thenar eminence. The nerve then enters the carpal tunnel bounded by the carpal bones with the transverse ligament serving as the roof. Also within the carpal tunnel lie the nine flexor tendons to the fingers. Within or just distal to the carpal tunnel, the recurrent branch of the median nerve arises and innervates the abductor pollicis brevis, opponens pollicis, and the superficial head of the flexor pollicis brevis. The terminal branches of the median nerve supply the first and second lumbrical muscles, while the digital branches provide sensation to the volar aspects (and the tips of the dorsal aspects) of the thumb, index, and middle fingers, and the lateral half the ring finger.
The ulnar nerve arises at the termination of the medial cord distal to the medial cutaneous nerves of the arm and forearm and the medial branch of the median nerve (Fig. 23-10). The spinal nerve contributions are mainly C8 and T1, but C7 fibers may also be present in 43–92% of cases, as suggested by brachial plexus dissections.1,4 The C7 contribution derives from a branch of the lateral cord and innervates the flexor carpi ulnaris muscle. The ulnar nerve descends anterior to the teres major and latissimus dorsi muscles into the arm. Then the nerve travels down the posterior compartment of the upper arm to the ulnar groove at the elbow. The ulnar groove is formed by the medial epicondyle of the humerus and the olecranon process of the ulna, with the ulnar collateral ligament serving as the floor. Approximately 1.0–2.5 cm distal to the ulnar groove, the nerve traverses under a fibrous aponeurotic arch connecting the humeral and ulnar heads of the flexor carpi ulnaris muscle. The area encompassing the ulnar groove and aponeurotic arch is commonly referred to as the cubital tunnel. Of note, the ulnar nerve yields no branches in the arm proximal to the elbow.
Figure 23-10. The ulnar nerve does not have any motor or cutaneous branches in the arm. ADM, abductor digiti minimi. The cutaneous branches of the medial cutaneous nerves of the arm and forearm are depicted. (Modified with permission from Haymaker W, Woodhall B. Peripheral Nerve Injuries: Principles of Diagnosis. Philadelphia, PA: WB Saunders; 1953.)
Distal to the elbow, the ulnar nerve travels between the flexor carpi ulnaris and flexor digitorum profundus muscles descending to the wrist. In the forearm, it innervates the flexor carpi ulnaris and the flexor digitorum profundus III and IV muscles. The dorsal ulnar cutaneous nerve originates in the mid or distal forearm to provide sensation to the dorsum of the medial aspect of the hand and fourth and fifth digits. Just before entering Guyon’s canal at the wrist, the palmar branch arises to provide sensation to the hypothenar eminence and motor innervation to the palmaris brevis muscle. The remaining components of the ulnar nerve travel into Guyon’s canal, formed by the hook of the hamate bone (on the radial aspect), the pisiform bone (on the ulnar aspect), the pisohamate ligament (serves as the floor), and the transverse carpal ligament (serves as the roof). Within or just distal to Guyon’s canal, the ulnar nerve splits into its terminal branches. A superficial terminal branch supplies sensation to the palmar aspect of the little finger and half of the ring finger, plus some of the distal aspects of these digits dorsally. A deep motor branch innervates the hypothenar muscles and then turns and continues across the hand to innervate the third and fourth lumbricals, interossei, adductor pollicis, and deep head of the flexor pollicis brevis muscle.
PATHOPHYSIOLOGY OF RADICULOPATHIES, PLEXOPATHIES, AND MONONEUROPATHIESBefore discussing the approach to patients with focal nerve lesions in the arm, it is important to understand the pathophysiologic basis of these neuropathies. Clinicians need to be aware of the mechanisms of nerve injury so that they can plan, time, and interpret the electrodiagnostic evaluation in order to offer the most accurate prognoses and treatment options. The pathophysiologic bases of nerve injury are limited: demyelination, conduction block, or axonal degeneration. The method by which an individual nerve is injured (e.g., gun shot wound to upper arm, prolonged hyperextension of the arm during surgery, and falling asleep on arm) often provides insight into the underlying pathophysiology.
The term “neuropraxia,” also known as first-degree injury, refers to neuronal dysfunction due to transient conduction block.1,8–10 In regard to focal peripheral nerve lesions, neuropraxia may arise from ischemia or demyelination. Compression of a nerve can result in segmental ischemia, which if of only short duration, results in a rapidly reversible physiologic conduction block lasting minutes or perhaps a few hours. However, experimental studies suggest that pressure related to compression on the nerve can result in distortion of the underlying nerve segment with paranodal and then segmental demyelination.11 Neuropraxia due to demyelination may resolve after several weeks following remyelination of the nerve segment. Thus, prognosis in lesions associated with only conduction block resulting from mechanical (as opposed to immune-mediated or radiation) mechanisms without secondary axonal loss is excellent.
Axonotmesis or second-degree injury refers to nerve injuries in which the axon is interrupted but the epineurium is intact.1,8–10 Following this type of nerve injury, the axon distal to the lesion, now separated from its cell body, will degenerate over the next 7–10 days. Subsequently, regenerating nerve sprouts emerge from the proximal stump of the sectioned nerve to attempt reinnervation of previously denervated tissues (e.g., muscle or cutaneous skin). Because the endoneurium is preserved, there is a greater likelihood that the regenerating axons can grow back and reinnervate denervated tissues than in neurotmesis described below. Axons grow back at a rate of 1 mm/d, so restoration of function can take many months to over a year, depending on the site of the lesion and length of the nerve.
Neurotmesis refers to severe, often penetrating nerve injuries, in which the axon and the supporting epineurium are interrupted (i.e., nerve transaction).1,8–10 Present technology precludes distinction between axonotmesis and neurotmesis without exploratory surgery and direct inspection of the nerve. Because the endoneurium is also interrupted, it is more difficult for regenerating nerve sprouts to reinnervate the target tissues. Scarring secondary to the disruption of the overlying connective tissue can also impede reinnervation. Regenerating nerves may become entwined with the scar tissue creating a neuroma. Thus, the prognosis for spontaneous recovery following this type of lesion is poor.
APPROACH TO PATIENTSAs with other neuromuscular disorders, the most important step is trying to localize the site of the lesion based on the history and physical examination. Following this, electrodiagnostic studies are performed to confirm the localization or try to localize the exact site of the lesion more accurately, if not apparent by the clinical examination. Often radiologic studies are done to further assist in identifying the site of the lesion and the possible cause. We begin the discussion of the approach of such patients with a review of electrodiagnostic studies that can be helpful.
The evaluation of the arm for possible cervical radiculopathy, brachial plexopathy, or mononeuropathy(ies) requires performing sensory, motor, and mixed sensorimotor nerve conduction studies (NCS) along with electromyography (EMG) (Table 23-2). This text is not meant to be a “how-to book” on EMG and NCS, and for this we refer the reader to several excellent reference books regarding electrodiagnostic medicine (more details can also be obtained in Chapter 2 of this book).6,12–15 However, clinicians taking care of patients with neuromuscular disorders need to be aware of the utility and limitations of these studies. The electrodiagnostic studies also need to be tailored to the individual patients depending on their symptoms and signs and as the results of the ongoing EMG and NCS are being analyzed.
TABLE 23-2. NERVE CONDUCTION STUDIES
Evaluating the sensory nerve action potentials (SNAPs) is important in distinguishing a radiculopathy from a more distal process. The lesion in most radiculopathies is proximal to the dorsal root ganglia (DRG). Because the cell bodies and distal axons are intact in cervical radiculopathies, the SNAPs should be normal. In contrast, in brachial plexopathies and mononeuropathies (in nerves with sensory fibers), in which the lesion is distal to the dorsal root ganglion, one would expect to see reduced amplitudes of SNAPs in the distribution of the affected nerve, provided there is significant axonal injury. In cases of a demyelinating lesion or conduction block (as the case in neuropraxic injuries), the SNAP distal to the site of the lesion is usually normal. When the injury to the plexus or peripheral nerve is axonal in nature, one also needs to remember that it takes several days from the time of the injury for Wallerian degeneration of the axons to occur distally. Thus, it takes approximately 7–10 days for the SNAPs to disappear, even if the nerve is completely severed. After this period of time there is sufficient degeneration of axons to begin to distinguish postganglionic axonal loss from conduction block/demyelination or a preganglionic lesion. However, an abnormal SNAP does not imply that the spinal root is normal. For example, in traumatic brachial plexopathies, avulsion of nerve root may occur concurrently with injury to nerve distal to the DRG.
It is very important to compare the SNAPs in the affected arm to the analogous nerves in the contralateral arm. It is possible that the SNAP amplitude(s) in an affected arm may still fall “within normal limits” for that electrodiagnostic laboratory even if injured. An asymptomatic limb provides better normative data for that individual than does normative data derived from populations. We, like most electrodiagnosticians, consider SNAP amplitude(s) less than half of that obtained from the analogous nerve in an asymptomatic limb to be abnormal. This of course is not helpful if the symptoms are bilateral.
The specific sensory studies performed are again dependent on the possibilities for the site of the lesion (Table 23-3).16 If one is evaluating a patient with sensory symptoms affecting the thumb, the possibilities include a C6 radiculopathy, upper trunk, lateral cord, median or radial mononeuropathies, or injury to the digital branches of these nerves. The sensory studies most helpful would be a median SNAP from the thumb or index finger and the superficial radial SNAP, and perhaps the lateral antebrachial cutaneous SNAP. In addition, the median and ulnar mixed nerve palmar studies are helpful to look for carpal tunnel syndrome. As with all electrodiagnostic evaluations, it is important to define the boundaries of abnormality by identifying a normal response in a nerve that is not felt to be clinically affected (e.g., an ulnar SNAP in this situation). If the patient has symptoms involving little finger, then the investigator needs to conduct studies to differentiate a C8/T1 radiculopathy, lower trunk, medial cord, and ulnar neuropathy from one another.
TABLE 23-3. CAUSES OF RADICULOPATHY
Herniated nucleus proposus
Degenerative joint disease
Rheumatoid arthritis
Trauma
Vertebral body compression fracture
Pott’s disease
Compression by extradural mass (e.g., meningioma, metastatic tumor, hematoma, abscess)
Primary nerve tumor (e.g., neurofibroma, Schwanomma, neurioma)
Carcinomatous meningitis
Perineural spread of tumor (e.g., prostate cancer)
Acute inflammatory demyelinating polyradiculopathy
Chronic inflammatory demyelinating polyradiculopathy
Sarcoidosis
Amyloidoma
Diabetic radiculopathy
Infection (Lyme disease, Herpes Zoster, Cytomegalovirus, Syphilis, Strongyloides)
Evaluation of a motor nerve is performed by stimulating the nerves at several locations and recording the compound muscle action potential (CMAP) from accessible muscle bellies, as discussed in Chapter 2 (Table 23-3).16 As with sensory studies, the majority of accessible motor stimulation sites are remote from the proximally located lesions associated with a radiculopathy or plexopathy. Thus, it is technically very difficult to assess for a demyelinating/conduction block lesion in these proximal sites. Further, most electrodiagnostic laboratories routinely perform median, ulnar, and radial motor conductions recording from muscles that innervated primarily by the C8–T1 segments. These motor studies are most useful in lower trunk and medial cord injuries as well as median, ulnar, or radial neuropathies. They are of limited value in pathology affecting the C5–7 segments or the elements of the brachial plexus through which these fibers descend. Motor conduction studies can assist in localizing the site and nature of the lesion (e.g., axonal or demyelinating/conduction block) involving testable nerves. Again, following an axonal lesion, one needs to wait about 7–10 days until Wallerian degeneration has occurred and CMAP amplitudes become reliably reduced. Furthermore, in demyelination/conduction block, one could identify this type of lesion only by stimulating the nerve proximal and distal to the site of the lesion. Axillary CMAPs recorded from the deltoid or musculocutaneous CMAPs recorded from the biceps brachii are technically more difficult to perform and interpret and are limited by patient tolerance of the intensity of stimulus often required to achieve a supramaximal CMAP in deeply positioned nerves. In such cases, it would be important to study the contralateral asymptomatic side as a comparison. Radiculopathies are not usually associated with an abnormal CMAP, unless there has been severe end-stage neurogenic atrophy of the muscle. This is a product of most muscles being innervated by more than one nerve root, and by the difficulty of identifying conduction block at the root level. Thus, detecting reduced CMAP amplitudes in cervical radiculopathies is uncommon unless it is severe or there are multiple roots affected.
F-wave studies have limited value in the evaluation of most radiculopathies and entrapment neuropathies. The reason for this is clear if one understands the pathogenesis of most of these focal neuropathies and the limitations inherent in F-wave assessment. The length of a possible compressive/demyelinating lesion is small in most radiculopathies and even mononeuropathies due to entrapment/compression (e.g., ulnar neuropathy at the elbow or median neuropathy at the wrist). Remember, the F-wave latency takes into account the time for the stimulus to travel antegrade through the motor nerve, stimulate a pool of anterior horn cells, and then travel back down the motor axon to stimulate the muscle. Thus, even if there were focal slowing across a small site of demyelination, this may be obscured by the normal conduction to and from the spine across the majority of the nerve. Further, most criteria regarding F-waves use the shortest latencies of multiple responses to define if the study is abnormal. Thus, one only needs to have one normal axon for the F-wave study to be deemed normal. Finally, amplitude measurement, the most important parameter obtained in NCS, cannot be reliably assessed in F-wave studies. However, if one is looking for a large proximal demyelinating lesion, as can be seen in some focal forms of chronic inflammatory demyelinating neuropathy, then F-waves are of some value.
The only reliable H-reflex in the arm is from the flexor carpi radialis muscle following median nerve stimulation at the antecubital fossa.17 This study is not routinely performed, as it usually does not assist in localizing the lesion apart from what is gained from the clinical examination, routine motor and sensory NCS, and the EMG. Nevertheless, the H-reflex for the flexor carpi radialis muscle may be abnormal in C6 or C7 radiculopathies, upper or middle trunk plexopathies, lateral cord lesions, or proximal median neuropathies. The major value of the H-reflex in the upper extremities occurs when H-reflexes are recognized during routine NCS. This implicates the presence of pyramidal tract pathology affecting that limb and may shift the investigation of the patient’s complaints from the peripheral to central nervous system.
Somatosensory-evoked potentials have limited utility in radiculopathies and most neuropathies for much the same reason as discussed with F-waves. However, somatosensory-evoked potentials may be of value in assessment of brachial plexopathies because routine sensory NCS will not pick up a demyelination or conduction in the plexus.18–21
The EMG examination is essential in the evaluation of patients for radiculopathy, plexopathy, and mononeuropathy. In combination with the clinical examination and carefully performed motor and sensory NCS, EMG of muscles supplied by different spinal roots, trunks, divisions, and cords of the brachial plexus, and different terminal nerves solidifies the localization of the site of the lesion. As discussed in Chapter 2, with EMG we assess the presence of abnormal insertional and spontaneous activity, the morphology of motor unit action potentials (MUAPs), and the recruitment properties of these units. Abnormal insertional or spontaneous activity in the form of positive sharp waves or fibrillation potentials implies membrane instability, which in neurogenic processes is typically due to axonal degeneration. Irritation of the nerve with or without axonal degeneration may result in fasciculation potentials, complex repetitive discharges, or myokymic discharges. The detection of myokymic discharges in a patient with history of cancer and radiation, who is now presenting with focal deficits in an arm, would strongly suggest radiation-induced injury to the roots or plexus as opposed to tumor infiltration. The demonstration of abnormal spontaneous activity in the paraspinal muscles suggests that there is at least some injury to the anterior horn cells or spinal nerves but also does not exclude an injury more distally (e.g., double crush).
Importantly, fibrillation potentials and positive sharp waves may not be present for up to 1 week in a paraspinal muscle and 3 weeks in limb muscles following axonal injury to a nerve root. However, voluntary recruitment of MUAPs is affected immediately. Thus, any injury to the nerve that results in a significant loss of muscle strength should be accompanied by reduced recruitment (e.g., fast-firing) of MUAPs. In neuropraxic injuries in which there is demyelination or conduction block without axonal degeneration, fibrillation potentials and positive sharp waves are not seen and the only abnormality apparent on the EMG is reduced recruitment of MUAPs. Reduced recruitment in the absence of abnormal spontaneous activity, NCS abnormalities, and morphological change in MUAPs more than 3 weeks subsequent to symptom onset implicates conduction block.
Following an axonal injury and muscle fiber reinnervation, fibrillation potentials and positive sharp waves are no longer evident. Reinnervation is more complete in muscles closer to the site of axonal injury (e.g., paraspinal muscles in a radiculopathy). If reinnervation occurs by successful axonal regrowth, then reestablishment of a near-normal number of motor units and innervation ratio, motor unit number and morphology may appear normal. In contrast, if reinnervation takes place via collateral sprouting, the motor units of an effected muscle will remain chronically reduced in number and increased in size, even if strength is reestablished. Muscle groups more distal to the site of the lesion (e.g., hand intrinsic muscles in a cervical radiculopathy) may be less likely to be completely reinnervated, and thus fibrillation potentials and positive sharp waves may persist indefinitely.
Another important point is that because of fascicular arrangement of axons running through various segments of the nerve trunk from the spine to the target muscle, an incomplete nerve injury may not necessarily demonstrate an abnormality in every muscle innervated by an affected spinal nerve root, trunk, cord, or terminal nerve.
Imaging studies such as a myelogram or magnetic resonance imaging (MRI) of the cervical spinal and brachial plexus are extremely valuable and complement the clinical examination and electrodiagnostic medicine study. MRI has, for the most part, replaced myelogram and computerized axial tomographic (CT) scans except in individuals in whom MRI is contraindicated (e.g., those with magnetic implants) for evaluation of radiculopathies. CT scans, particularly with contrast within the subarachnoid space can be useful22 but high-resolution MRI is much more sensitive for radiculopathies, plexopathies, and focal neuropathies (Figs. 23-11 to 23-13).23–29 Several studies have investigated the utility of ultrasound in focal neuropathies.30–36
Figure 23-11. MRI of brachial plexus. Normal sagittal anatomy. (A) Roots C5–T1 just lateral to the intervertebral foramina, T1 is located below and C8 above the first rib (R1). (B) Subclavian artery (SA) and the roots C7, C8, and T1 are seen within the interscalene triangle between the anterior scalene muscle (ASM) and middle scalene muscle (MSM). The subclavian vein (SV) is positioned between the ASM and the clavicle (c). (C) Just lateral to the interscalene triangle the three trunks are formed, the superior (ST), the middle (MT), and inferior trunk (IT). (D) The divisions (D) are formed at the level where the brachial plexus crosses the clavicle. (E) Around the axillary artery (AA) the three cords are located, the lateral (LC) most anterior, the posterior (PC) most superior, and the medial (MC) most posterior. AV, axillary vein. (Reproduced with permission from van Es HW, Bollen TL, van Heesewijk HP. MRI of the brachial plexus: a pictorial review. Eur J Radiol. 2010;74(2):391–402.)
Figure 23-12. MRI of brachial plexus. Normal coronal anatomy. (A) Most posterior image with the horizontal course of the T1 nerve root (long arrow), very close to the lung apex. Short arrow points to the stellate ganglion. (B) Image just anterior to (A) with the C8 nerve roots (arrows). (C) T2-STIR image at the same level as (B) shows the slightly increased signal intensity of the normal C8 nerve roots (arrows). (D) Arrow points to the C7 nerve root. MSM, middle scalene muscle. (E) The cords (white arrow) are seen as linear structures above the axillary artery (AA). The dorsal scapular artery (DSA) courses between the trunks of the brachial plexus, black arrow points to the superior trunk. ASM, anterior scalene muscle. (Reproduced with permission from van Es HW, Bollen TL, van Heesewijk HP. MRI of the brachial plexus: a pictorial review. Eur J Radiol. 2010;74(2):391–402.)
Figure 23-13. MRI of cervical spine. Traumatic nerve root avulsion. (A) Axial balanced fast field echo (FFE) image demonstrates a traumatic pseudomeningocele (arrow). (B) Axial T1-weighted image with intravenous gadolinium shows the enhancement of an avulsed nerve root (arrow). (Reproduced with permission from van Es HW, Bollen TL, van Heesewijk HP. MRI of the brachial plexus: a pictorial review. Eur J Radiol. 2010;74(2):391–402.)
SPECIFIC DISORDERSRecall the disparity between the number of cervical vertebrae (seven) and nerve roots (eight) (Fig. 23-2). As a result, each numbered cervical nerve root is related to the bony level immediately inferior to it. For example, the C5 spinal root exits the spinal column between the fourth and fifth cervical vertebra, and it is vulnerable to compression from a herniated disc (herniated nucleus pulposus or HNP) between C4 and C5. The C6 spinal root exits the spinal column between the fifth and the sixth cervical vertebrae and may be injured from an HNP between C5 and C6. In the same manner, an HNP between C6 and C7 levels may damage the C7 root, while an HNP at the C7 and C8 vertebrae may impinge the C8 nerve root. The T1 spinal nerve exists between the eighth cervical and first thoracic vertebrae and may be damaged by an HNP at this level.
Most cervical radiculopathies involve the C5–8 spinal nerve roots (C7 occurring in 31–81%, C6 in 19–25%, C8 in 4–10%, and C5 in 2–10%).13,37–41 Causes of cervical radiculopathy are multiple (Table 23-3), and most commonly involve compression of nerve root by an HNP or osteophytes in the case of degenerative spine disease. Individuals with a cervical radiculopathy typically present with neck or posterior shoulder pain in the scapular region that radiates down the affected arm. Turning the head toward the painful arm, particularly with neck extension, can narrow the neural foramen further compressing the nerve root and thus exacerbating the pain, as can downward pressure on the affected individual’s head. The patient may have weakness in the distribution of the affected myotome and sensory loss in the dermatome that is involved. The deep tendon reflexes of affected segment may also be reduced. Because there is much overlap in the territories supplied by individual spinal roots, symptoms and signs can be similar to a plexopathy or focal neuropathy. Therefore, as previously discussed, EMG and NCS combined with imaging studies are extremely valuable in localization. Imaging studies are also important to assess structural etiology (e.g., HNP, osteophyte impinging on root, tumor of the nerve or extrinsic tumor/mass compression of the nerve, or inflammatory process). Further, nerve root avulsion may accompany nearly 80% of severe brachial plexopathies due to trauma.
People with a C5 radiculopathy may have weakness of shoulder abduction, external rotation, elbow flexion, and supination of the wrist along with sensory loss in the shoulder region although sensory signs and symptoms may be absent in many cases. The biceps brachii and brachioradialis deep tendon reflexes may be asymmetrically reduced compared to the unaffected limb. Routine median and ulnar motor and sensory NCS are normal, as these do not carry any fibers emanating from the C5 spinal root. Electrodiagnostic localization is dependent on the EMG examination (Table 23-1). Abnormalities in the mid-cervical paraspinal, supraspinatus, infraspinatus, deltoid, biceps brachii, supinator, and brachioradialis muscles are seen in C5 radiculopathies. However, these muscles are also innervated by C6. The rhomboids are primarily innervated by C5, so abnormalities in this group strongly support a C5 radiculopathy. Further, if one sees membrane instability in the triceps brachii, pronator teres, extensor carpi radialis, or flexor carpi radialis that are innervated by C6, but not by C5, the above findings are more consistent with a C6 nerve root lesion or multiple root involvement.
A common differential diagnostic consideration in a patient with C5 radiculopathy is rotator cuff injury, which may also produce shoulder pain and weakness in arm abduction and external rotation. Clinical distinction can be made be reproduction of discomfort by passive movement of the shoulder, rather than at the neck, and by the preservation of biceps strength and reflex in most rotator cuff injuries. Other differential diagnostic considerations not related to trauma or degenerative spine disease include brachial plexopathy, diseases that may present as multifocal neuropathies, and in the absence of pain or paresthesia, motor neuron disease.
TABLE 23-4. BRACHIAL PLEXUS CLASSIFICATION: NATURE OF INJURY
Individuals with a C6 radiculopathy can present in a similar manner to that described above with a C5 radiculopathy. However, weakness may also involve extension of the elbow (triceps), pronation, and extension of the wrist (extensor carpi radialis). Patients often complain of paresthesia localized to the thumb. In a patient with suspected C6 radiculopathy, one needs to consider an upper trunk or lateral cord lesion, median neuropathy, or radial neuropathy. Therefore, at the very least, we usually perform median and radial SNAPs and a median CMAP. In addition, we obtain lateral antebrachial cutaneous SNAPs, if we are suspicious of a brachial plexopathy affecting the upper trunk. As discussed, these NCS should be normal in a C6 radiculopathy, but the flexor carpi radialis H-reflex may be abnormal. However, localization hinges on the EMG study (Table 23-1). There is significant overlapping of findings in C6 and C5 as well as C7 radiculopathies. Needle EMG may demonstrate abnormalities in the mid to low cervical paraspinals, supraspinatus, infraspinatus, deltoid, biceps brachii, triceps brachii, pronator teres, brachioradialis, supinator, extensor carpi radialis, and flexor carpi radialis muscles. Rotator cuff injury, multifocal neuropathies, and brachial plexus neuritis are likely to be the most common differential diagnostic considerations unrelated to trauma or degenerative joint disease of the cervical spine.
People with a C7 radiculopathy often have pain or sensory symptoms radiating down the arm into the middle digit. Weakness of elbow extension and wrist flexion or finger extension may be evident along with a diminished triceps reflex. In patients with suspected C7 radiculopathy, a median CMAP, median SNAP to the third digit, and radial SNAP can be done to assess for a more distal lesion. But again, one should expect routine NCS to be normal in a C7 radiculopathy except for perhaps the flexor carpi radialis H-reflex. On EMG, abnormalities may be detected particularly in the triceps brachii, anconeus, pronator teres, flexor carpi radialis, extensor digitorum communis, and less commonly the extensor digitorum indicis, extensor pollicis longus and brevis, and flexor pollicis longus muscles (Table 23-1). The differential diagnosis of C7 radiculopathy is limited and is probably mimicked most closely by radial mononeuropathies. Most radial neuropathies occur at or distal to the spiral groove, thus resulting in sparing of triceps and prominent weakness of wrist extension. C7 root lesions, on the other hand, typically affect the triceps but rarely produce severe weakness of wrist extension.
It is often very difficult to distinguish a C8 from a T1 radiculopathy, so these are discussed together. Individuals who are affected have sensory disturbance affecting the medial aspect of the hand and forearm along with hand weakness. The differential diagnosis included a lower trunk plexopathy such as neurogenic thoracic outlet syndrome, a medial cord lesion, an ulnar neuropathy, and in the absence of pain and sensory symptoms, motor neuron disease. In cases of ulnar neuropathy, the site of the lesion may be at the wrist, elbow, or elsewhere. The NCS are very helpful in terms of localization. It is important to perform ulnar and median CMAPs and SNAPs as well as a medial antebrachial cutaneous SNAP to exclude a plexus or root lesion unless unequivocal focal abnormalities of the ulnar motor conduction studies can be demonstrated (Table 23-3). The SNAPs should be normal in a radiculopathy, but in a lower trunk or medial cord injury the ulnar and medial antebrachial cutaneous SNAP amplitude may be reduced. A reduction in the median CMAP amplitude with a normal median SNAP would further support a lower trunk or medial cord injury (Table 23-2). On EMG, one may see abnormalities in any of the median- or ulnar-innervated muscles innervated by C8 and T1 spinal roots (Table 23-1). The thenar eminence may be predominantly innervated by T1, so the median CMAP amplitude may be disproportionately reduced compared to the ulnar CMAP in a T1 radiculopathy. Most of the radial-innervated muscles supplying the fingers originate from the C7 and C8 spinal segments but not T1. Therefore, EMG of these muscle groups can help distinguish a C8 from a T1 radiculopathy. In the case of a lower trunk lesion, muscles innervating radial muscles via the C8 nerve root will be affected as well as intrinsic hand muscles, whereas this will not be the case in medial cord lesions.
Most cervical radiculopathies involve only one root, but approximately 12–30% may involve multiple levels.38,40 If this is the case, CMAP amplitudes are more likely to be reduced although SNAPS will remain spared. The presence of EMG abnormalities suggesting a polyradiculopathy must raise the suspicion of other diseases, particularly motor neuron disease. In such cases, it is important to study the lower extremity, thoracic paraspinals, and even selected cranial nerves (e.g., the tongue and sternocleidomastoid).
For the sake of completeness, we will briefly discuss thoracic radiculopathies although these are relatively uncommon. HNPs in the thoracic region account for only 0.22–5.3% of all disc protrusions.42–45 Approximately 75% of symptomatic thoracic radiculopathies occur between T8 and T12, with most occurring between T11 and T12. Central and centrolateral HNPs can compress the spinal cord, leading to symptoms and signs of a myelopathy. Patients may present circumferential chest or abdominal pain and/or paresthesias, leg pain or weakness, or bowel or bladder difficulties (e.g., constipation, urinary retention, and incontinence). At the T11–12 region, the conus medullaris or cauda equina may be affected with ensuing bowel/bladder and lower extremity deficits.
Trauma is the most common cause of a herniated thoracic disc accounting for 14–63% of cases.46,47 Degenerative changes of the spine account for a minority of cases. Other structural causes for thoracic radiculopathies that need to be considered include compression due to metastatic disease, vertebral collapse, Pott’s disease, and primary nerve tumors. Perhaps, the most common etiology of thoracic radiculopathy is diabetes mellitus (e.g., diabetic radiculoneuropathy). Additional nonstructural causes of thoracic radiculopathies include Lyme disease, herpes zoster, cytomegalovirus, sarcoidosis, and carcinomatous meningitis. Of note, thoracic disc herniations on imaging are far more common than causally related clinical syndromes, and clinicians need to be cautious before attributing nonspecific clinical symptoms to an imaging abnormality.
The electrodiagnostic evaluation of thoracic radiculopathies is limited. NCS are not helpful. EMG may demonstrate abnormal insertional and spontaneous activity in the thoracic paraspinal muscles. Care must be taken not to insert the needle too far so as to avoid a pneumothorax. EMG of abdominal muscles may be of value, as one can also assess for MUAP morphology and recruitment abnormalities.
Treatment is dependent on the etiology of the radiculopathy. For the sake of discussion, we will focus here on treatment of radiculopathies related to HNPs or spondylotic disease, as radiculoneuropathies related to other entities (e.g., Lyme disease, diabetes) are discussed in other chapters in this book. It is difficult to make evidence-based medical decisions regarding the best therapeutic approach to patients with cervical radiculopathies due to the lack of well-designed, prospective, controlled, and blinded trials. That said, it is important to realize that the natural history of radiculopathies related to compression from HNPs is favorable. In a large series of patients with cervical radiculopathy followed for up to 5 years, 90% were asymptomatic or had only slight pain at last follow-up; however, 26% of all patients underwent surgery.48
Treatment of acute radiculopathy is focused on relief of pain. Injections of corticosteroids or local anesthetic agents have been employed, but again there is no strong medical evidence that these are beneficial. Likewise, there is no proven efficacy for bed rest, cervical traction, corticosteroids, nonsteroidal anti-inflammatory drugs, or muscle relaxants. Commonly used medications for neuropathic pain such as antiepileptic agents or tricyclic antidepressants can be tried, as can short courses of narcotics. In patients with intractable pain or those with significant weakness, decompressive surgery may be warranted. Small trials assessing the effects of surgery versus conservative management for cervical spondylotic radiculopathy or myelopathy have shown little or no difference in the long-term, though there was quicker pain relief in the surgical group.49,50
Brachial plexopathies can be classified on the basis of the nature of the injury (i.e., an open or closed brachial plexopathy), the anatomic location of the lesion, or the mechanism of injury (Table 23-2).1–3 MRI can be helpful in identifying both the site and cause of the lesion because anatomical resolution of the roots, trunks, divisions, and cords is very well depicted due to the inherent contrast differences between the nerves and surrounding fat (Figs. 23-11 and 23-4).23 However, MRI is expensive, and usually we can make an accurate diagnosis on the basis of clinical examination, supplemented when necessary by electrodiagnostic studies. Therefore, we will begin by reviewing clinical and electrodiagnostic features that one may expect to see with lesions affecting various trunks and cords of the brachial plexus.
Individuals with upper trunk lesions have weakness in the deltoid and biceps brachii muscles (Fig. 23-3). Therefore, they commonly complain of difficulty lifting their arms. Sensory loss involves the lateral arm and forearm down to the lateral aspect of the hand and fingers. The biceps brachii and brachioradialis reflexes are typically reduced. Injuries in isolation are relatively common when compared to isolated middle or lower trunk lesions.
EMG and NCS are useful in differentiating an upper trunk lesion from a C5 or C6 radiculopathy. Remember that upper trunk lesions are distinguished from C5–6 injuries in that the posterior primary rami are spared, as are the nerve branches to the rhomboid and serratus anterior muscles. Also, trunk lesions are distal to the dorsal root ganglion. Therefore, if the nature of upper trunk injury is axonal damage and not just neuropraxia (i.e., conduction block and/or demyelination), the radial, median recording from the thumb and possibly the index finger, and lateral antibrachial cutaneous SNAPs may have reduced amplitudes, particularly when compared to the asymptomatic contralateral arm (Table 23-2). These SNAPs would be normal in a cervical radiculopathy or in a neuropraxic or demyelinating process affecting the trunk. Routine median and ulnar CMAPs are not particularly helpful other than excluding involvement of other trunks or nerves. A musculocutaneous CMAP can be done by recording from the biceps brachii, but this is usually not useful in distinguishing a C5 or C6 radiculopathy from an upper trunk lesion. However, the EMG can be localizing in combination with the sensory studies. Recall that the posterior primary rami to the paraspinal muscles, the dorsal scapular nerve to the rhomboid, and long thoracic nerve to the serratus anterior come off the cervical roots before the formation of the upper trunk. EMG of these muscles may show evidence of denervation in a C5 or C6 radiculopathy, but would be spared if the lesion only involved the upper trunk. One should do an extensive EMG to ensure that abnormalities are restricted to muscles innervated by the upper trunk, with sparing of muscles innervated by the middle and lower trunk (Table 23-1).
Isolated middle trunk lesions are extremely rare and middle trunk lesions most often occur in combination with other plexus lesions (Fig. 23-3). Symptoms and signs would resemble a C7 radiculopathy. Affected people may experience weakness of elbow, wrist, and finger extension and sensory loss or pain in the posterior forearm and the dorsal and palmar aspect of the middle finger. The triceps reflex may be reduced.
Provided the injury is axonal in nature, a diminished amplitude of the median SNAP to the third digit may be evident as the cutaneous fibers that supply this finger usually traverse the middle trunk (Table 23-2). Also, the radial CMAP recorded from the extensor indicis proprius may have reduced amplitude, if there is sufficient axon loss. However, the EMG is most important in delineating the extent of motor involvement. Remember that the middle trunk contains the C7 spinal nerves and after passing through the middle trunk these diverge and traverse the posterior and lateral cords (Table 23-1, Fig. 23-3). Thus, most muscles innervated by the radial nerve with the exception of the brachioradialis would be affected in addition to some median-innervated forearm muscles (e.g., those muscles with C6 and C7 and lateral cord innervation). Additionally, EMG abnormalities may be appreciated in the pectoralis major, latissimus dorsi, and teres major muscles, as these muscles are, in part, innervated by the C7 spinal nerves and middle trunk via the medial and posterior cords. However, the serratus anterior muscle, which has C7 innervation in common but not the middle trunk, would be spared with a middle trunk lesion (recall the long thoracic nerve branches directly off the roots). There are no nerve branches arising directly from the middle trunk, and so it can be difficult to distinguish a lesion involving the middle trunk from those affecting portions of the lateral and posterior cords.
Lesions affecting the lower trunk have symptoms similar to a C8/T1 radiculopathies, medial cord plexopathies, and ulnar neuropathies (Fig. 23-3). Affected individuals have sensory loss of the medial aspect of the forearm and hand along with weakness of ulnar-, median-, and radial-nerve–innervated wrist/hand muscles. Involvement of radial-nerve and posterior cord–innervated C8/T1 muscles puts the lesion more proximal than the medial cord.
NCS are valuable in localizing the lesion (Table 23-2). With axonal lesions, one would expect to see reduced amplitudes of ulnar and medial antebrachial cutaneous SNAPs in both lower trunk and medial cord lesions, but not in a C8 or T1 radiculopathy. A reduction in the amplitude of the median and ulnar CMAPs may be seen in a severe radiculopathy, lower trunk, or medial cord axonopathies. EMG should show signs of denervation in radial-, median-, and ulnar-innervated distal arm muscles, as the nerves supplying these muscles all course through the lower trunk (Table 23-1). However, the lower cervical paraspinal muscles should be spared in brachial plexus lesions. Compared to other plexus injuries, the prognosis for recovery is comparatively poor because of the long distance a regenerating nerve must cover to reinnervate the muscles in the distal arm.1
The nerves originating from the posterior cord include the thoracodorsal, the upper and lower subscapular, axillary, and radial nerves (Fig. 23-3). Depending on where the lesion is in the cord, individuals who are affected may have weakness of shoulder abduction, shoulder extension, supination of the wrist, and elbow/wrist/finger extension along with sensory disturbance in shoulder area, posterior arm and forearm, and the dorsum of the hand.
Provided the lesion is axonal in nature, both the superficial radial and posterior cutaneous nerve of the forearm SNAP amplitudes may be reduced in a posterior cord lesion (Table 23-2). However, the sensory studies of the lateral antebrachial cutaneous nerves, which also courses through the upper trunk but the lateral cord as opposed to the posterior cord, would be normal. A radial CMAP recording from the extensor indicis proprius would be expected to show reduced amplitude, when there has been significant axonal loss involving the C7 and C8 spinal nerve fibers coursing through the posterior cord. EMG would be expected to demonstrate abnormalities of the latissimus dorsi, teres major (though difficult to study), deltoid, and the radial-innervated muscles (Table 23-1).
The lateral cord is the continuation of the anterior division of the upper trunk (Fig. 23-3). Individuals with a lateral cord lesion may experience weakness of shoulder flexion and abduction, elbow flexion and pronation, and wrist flexion. In addition, sensory disturbance would involve the lateral, volar aspect of the upper arm and forearm along with the lateral and palmar aspect of the hand and fingers. The biceps brachii reflex should be reduced with sparing of the brachioradialis reflex.
Median SNAPs to the first three digits, superficial radial nerve to the thumb, and the lateral antebrachial cutaneous nerve should be studied to help localize pathology to the lateral cord (Table 23-3). With axonal lesions to the lateral cord, the median and lateral antebrachial SNAPs are expected to have decreased amplitude, but the radial SNAP should be normal as this arises from the posterior cord. A musculocutaneous CMAP recording from the biceps brachii muscles may have reduced amplitude. EMG can demonstrate evidence of denervation in the biceps brachii, pronator teres, flexor carpi radialis muscles, and perhaps the infraclavicular and midsternal fibers of the pectoralis major (Table 23-1). These findings coupled with normal EMG of the cervical paraspinal, supraspinatus, infraspinatus, deltoid, and triceps muscles localize the lesion distal to the upper trunk and out of the territory of the posterior cord.
The medial cord is a continuation of the anterior division of the lower trunk (Fig. 23-3). A medial cord injury clinically resembles a lower trunk lesion except that radial-nerve–innervated C8/T1wrist/hand muscles would be spared. Remember that nerves to radial-innervated muscles in the forearm course through the lower trunk and then enter the posterior cord rather than the medial cord. Therefore, the ulnar-, median-, and radial-innervated muscles to the digits are affected in a lower trunk lesion; only the ulnar- and median-innervated muscles will be abnormal with medial cord damage.
The medial antebrachial cutaneous and ulnar SNAPs may be reduced in amplitude provided that there is axonal injury, but this does not help distinguish a medial cord from a lower trunk lesion (Table 23-2). The medial antebrachial cutaneous response is disproportionately reduced in comparison to the ulnar SNAP in neurogenic thoracic outlet syndrome, however, presumably due to a larger component from the T1 spinal nerve. Conversely, the ulnar SNAP is typically more affected from poststernotomy plexopathies as the C8 spinal nerve appears to be the primary structure injured. Decreased amplitudes of median and ulnar CMAPs recording from thenar and hypothenar muscles do not discriminate between injury to the lower trunk and medial cord. One way to try to differentiate a medial cord from a lower trunk lesion would be by assessing the radial CMAP recorded from the extensor indicis proprius muscles. EMG is more helpful in distinguishing between a lower trunk and a medial cord injury. Again, if EMG demonstrates signs of denervation in C8/T1 radial- as well as median- and ulnar-innervated musculature, then a lower trunk as opposed to medial cord injury should be considered (Table 23-1).
In the following section, we will go into more detail about the common types of brachial plexopathy.
Immune-mediated brachial plexus neuropathy (IBPN) goes by various terminologies, including acute brachial plexitis or neuritis, neuralgic amyotrophy, and Parsonage–Turner syndrome.2,3,51–55 IBPN usually presents with an acute, often nocturnal, onset of severe pain in the shoulder region. The pain is frequently described like a hot poker jammed into the upper arm. Sometimes the pain involves the forearm or may be restricted to this segment of the arm (as seen in individuals with anterior interosseous syndrome as a forme fruste of IBPN). The pain is often exacerbated by movement of the arm. The intense pain usually lasts several days to a few weeks, but a dull ache can persist for 3 years or more.55 Individuals who are affected may not appreciate weakness of the arm early in the course because the pain limits movement. However, as the pain dissipates, weakness and often, a lesser degree sensory loss are appreciated. Attacks can occasionally recur.55 This disorder is usually clinically unilateral, but the opposite arm may be affected occasionally to a lesser degree.
Clinical findings are dependent on the distribution of involvement (e.g., specific trunks, divisions, cords, or terminal nerves). Occasional mild abnormalities of the cerebrospinal fluid (increased protein or pleocytosis) are found indicative of presumed inflammatory process also extending to the roots.53 One large study suggested that 36% of patients recovered most functions within the first year, 75% by the second year’s end, and 89% by the end of the third year.53 However, another large study of 246 cases found that approximately two-thirds of patients still had persistent pain and weakness after 3 years and <8% had a full recovery according to the patients.55 Mild paresis was still evident in 69%, with severe weakness in 3%. Proximally located muscles are more likely to regain strength than the more distal hand muscles.
The most common pattern of IBPN involves the upper trunk or a single or multiple mononeuropathies primarily involving the suprascapular, long thoracic, or axillary nerves.1,54–58 Additionally, the phrenic59,60 and anterior interosseous55,61–63 nerves may be concomitantly affected. Any of these nerves may also be affected in isolation as a forme fruste of IBPN. In most IBPNs, the paraspinal muscles are normal on EMG, suggesting that the lesion is distal to the root/spinal nerve level, but occasionally signs of active denervation are apparent, suggesting root involvement. Rarely, multiple cranial nerves (IX, X, XI, and XII) may be involved.64 In this regard, an isolated spinal accessory neuropathy presenting as acute unilateral suboccipital and neck pain and weakness of the trapezius muscle may also represent a forme fruste of IBPN.65
The pathogenic basis of IBPN is unknown but presumed to be immunologic. Circumstantial evidence of an inflammatory basis is that IBPN may develop following immune system provocation by infection, vaccination, bone transplantation, or following treatment with immune-modulating agents (e.g., interferons, interleukin-2, and tumor necrosis-alpha blockers).53,66–70 In addition, some series have reported antibodies directed against peripheral nerve myelin and soluble terminal complement complexes.71 Biopsies of the brachial plexus are not typically performed for IBPN, but there are few descriptions of such biopsies revealing perivascular epineurial and endoneurial inflammatory cell infiltrates. The antigen(s) of which the autoimmune attack is directed is not known, but the electrodiagnostic abnormalities suggest a primary insult against the axons of the nerves as opposed to the myelin.
The electrodiagnostic findings are dependent on the site(s) of involvement and can be rather multifocal.1–3,51,56 The upper trunk is primarily involved in most patients. Thus, it is not surprising that median and ulnar motors studies are abnormal in only about 15% of patients with IBPN.56 Median, lateral antebrachial cutaneous, and radial SNAPs are more likely to be abnormal. Also, CMAPs recorded from the deltoid and biceps muscles can demonstrate abnormalities. Other laboratory abnormalities include slightly increased cerebrospinal fluid protein with or without mild pleocytosis in a little over 10% of patients.53,55 MRI scan of the plexus may demonstrate increased T2 signal suggestive of inflammation or edema.28,29,55
We often treat patients presenting acutely who continue to have severe pain with a short course of corticosteroids (e.g., prednisone 50 mg daily tapering by 10 mg every 4–5 days), although there are very few evidence-based studies that have demonstrated any efficacy. However, in our anecdotal experience, corticosteroids seem to help alleviate the pain, which can be useful in allowing the patient to proceed with physical therapy. They do not appear to expedite return of strength. If the pain has already resolved by the time we see them, we do not treat with corticosteroids. The mainstay of treatment is physical and occupational therapy to prevent contractures in an immobilized arm, improve function, and maintain strength in unaffected muscles.
Rarely, a painful or painless brachial plexopathy may be the sole manifestation of an asymmetric form of chronic inflammatory demyelinating neuropathy, multifocal acquired motor and sensory demyelinating neuropathy, or multifocal motor neuropathy.72 Diagnosis of these entities requires demonstration of conduction block or focal slowing localized to the brachial plexus, which is often technically difficult. The importance of identifying multifocal acquired motor and sensory demyelinating neuropathy or multifocal motor neuropathy is their potential responsiveness to immunotherapy. See Chapter 14 on “Chronic Inflammatory Demyelinating Polyneuropathy and Related Disorders” for more details.
The annual incidence of obstetrically related plexus injuries ranges between 0.38 and 2.0 per 1,000 live births.1–3,73–79 Three types of brachial plexus injury complicate childbirth: (1) diffuse plexopathy, (2) upper trunk plexopathy (Erb palsy), or (3) lower trunk plexopathy (Klumpke paralysis). The plexus can be damaged during childbirth due to traction on the arm and thereby the nerves. Increased risk is associated with heavy birth weight of the infant, mothers with short stature, breech presentation, long and difficult labor, and heavily sedated mothers (resulting in diminished muscle tone during delivery).1,74,75,80–82 In addition, forceful downward traction applied to the head after the fetal third rotation is a risk factor of obstetric brachial plexus palsy in vaginal deliveries in cephalic presentation.83
Erb palsy, the most common type of obstetric paralysis, results from stretch of the nerves of the upper trunk of the brachial plexus.18,74,76 Severe traction injury may also lead to avulsion of the C5 or C6 spinal nerves. Traction of the upper trunk can occur with shoulder dystocia in a vertex presentation or difficulty delivering the aftercoming head in a breech presentation. Upper trunk lesion leads to weakness of the supraspinatus, infraspinatus, deltoid, biceps brachii, teres minor, brachioradialis, extensor carpi radialis longus/brevis, and supinator muscles. An infant who is affected typically lies with their arm adducted and internally rotated (unopposed pull of the sternal portion of the pectoralis major and latissimus dorsi muscles), elbow extended and forearm pronated (unopposed triceps and pronator teres/quadratus muscles), and wrist/fingers flexed (weak wrist extensors—the so-called “waiter’s tip position”).1 Diaphragmatic or serratus anterior weakness suggests the possibility of root avulsion, as the nerves to these muscles arise proximal to the upper trunk.
Rarely, the lower trunk or C8 or T1 roots are injured during childbirth (Klumpke paralysis).78 These usually occur in the setting of face presentation and hyperextension of the neck but can also complicate breech deliveries with hyperabduction of the arm. Infants will have good proximal arm strength, but weakness of hand muscles is evident. Finally, the entire plexus can also be affected to varying degrees.84,85
Radiological imaging is essential to assess for the possibility of associated humeral or clavicular fractures as well as diaphragmatic paralysis. In addition, MRI should be done to assess for nerve root avulsion.2,86,87
Electrodiagnostic studies are useful to determine the site and severity of injury and prognosis and to decide about the appropriateness and timing of any operative intervention.1,2,74,75,80 Abnormalities in SNAPs and CMAPs may be evident in 7–10 days. Electrodiagnostic studies are typically performed 4–6 weeks following delivery, as it can take this long for active signs of denervation to be evident on EMG. However, detection of voluntary MUAPs at any time, even before the 4–6-week period, demonstrates that there is at least partial continuity between the anterior horn cells and the target muscle SNAPs are typically more vulnerable to injury than CMAPs. A pattern of where SNAP amplitude(s) are low, but CMAP amplitude(s) are disproportionately reduced suggests pathology both proximal and distal to the level of the dorsal root ganglia and raises the possibility of avulsion. Prognosis is better, if the nerve is not completely severed. If SNAPs or CMAPs are low or absent and there is initially no MUAP on EMG, serial studies can be performed every 6–8 weeks to assess for evidence of reinnervation.
The natural history is not well defined, but patients with upper trunk lesions often have significant improvement within 3 months. Those with lower trunk lesions are more likely to have a more prolonged course and incomplete recovery. Unfortunately, there is no chance for regeneration of the nerves following a root avulsion. Reconstructive surgical procedures may be employed in order to help restore elbow flexion and shoulder abduction in patients with severe axonal injury.60,88–90
The term “thoracic outlet syndrome” has been ascribed to disorders, including those attributed to compromise of blood vessels between the base of the neck and the axilla.1–3 Our discussion is limited to the rare neurogenic form of thoracic outlet syndrome, which, in essence, is a lower trunk plexopathy. Most cases of alleged thoracic outlet syndrome are unassociated with any objective clinical, electrophysiological, or imaging evidence of vascular or nerve compromise. Most individuals with true neurogenic thoracic outlet syndrome are women with a prominent C7 transverse process or true cervical rib that can be appreciated on plain films of the cervical spine (Fig. 23-14). These cases are often associated with a sharp fibrous band extending from the tip of the elongated C7 transverse process or cervical rib to the first thoracic rib. This band usually cannot be visualized on imaging studies, including MRI scans. Its presence is however suggested by demonstration of the bony anomalies described above. The proximal aspect of the lower trunk becomes angulated or stretched as it passes over this fibrous band. Because the T1 fibers lie below the C8 fibers, these are usually distorted and thus are more likely to be damaged. Thus, affected individuals have muscle atrophy and weakness that is often greater in the thenar muscles, which have more T1 innervation than the hypothenar muscles, which have more C8 innervation. In addition, patients complain of numbness, paresthesia, and pain along the medial aspects of the arm, forearm, and hand. Electrodiagnostic studies demonstrate that the median CMAP and medial antebrachial cutaneous SNAP amplitudes are reduced to a greater extent than the ulnar SNAP and CMAP, because the former studies primarily assess T1 fibers, while ulnar studies primarily assess C8 fibers.2,58,91–94 Neurogenic thoracic outlet syndrome is typically treated by surgical resection of the taut band. In our experience, surgery may arrest progression and relieve pain but would uncommonly restore bulk or strength of hand muscles.
Figure 23-14. Neurogenic thoracic outlet syndrome. Atrophy of the right thenar eminence and first dorsal interosseous muscles are evident (A). Plain cervical spine films demonstrate small cervical ribs (arrows) bilaterally on AP view (B) and oblique view (C). (Reproduced with permission from of Steven A. Greenberg, MD. Reproduced with permission from Greenberg SA, Amato AA. EMG Pearls. Philadelphia, PA: Hanley & Belfus; 2004.)
Neoplasms involving the brachial plexus may be primary nerve tumors, local cancers expanding into the plexus (e.g., Pancoast lung tumor or lymphoma), and metastatic tumors.95,96 Primary brachial plexus tumors are less common than the secondary tumors and include schwannomas, neurinomas, and neurofibromas.25,42,97–99 These primary tumors may present as mass lesions in the supraclavicular fossa region or axilla. Pain and paresthesias are early symptoms, while motor and sensory losses occur later as the tumor may initially distort the nerve fibers but do not result in conduction block, demyelination, or axon loss right away.
Schwannomas are commonly benign and well encapsulated, affect the proximal segments of the plexus, and may be surgically removed with minimal damage to the nearby nerve fibers (Fig. 23-15).97,99,100 However, malignant schwannomas do rarely occur.101 Neurofibromas are the most common form of peripheral nerve tumor and are typically benign. However, when seen in the context of neurofibromatosis, these are often multiple and affect a larger portion of the brachial plexus.97 Additionally, neurofibromas interdigitate more with nerve fibers within the nerve fascicle and are more commonly associated with neurological deficits than schwannomas. Further, it is difficult to remove these surgically without damaging the affected nerve. These tumors can also convert to a more malignant form, particularly in neurofibromatosis.
Figure 23-15. MRI of brachial plexus. Schwannoma of the superior trunk. (A) Sagittal T1-weighted image, arrows point to the tumor which is located in the superior trunk just lateral to the interscalene triangle and above the subclavian artery (SA). MSM, middle scalene muscle. (B) Coronal T1-weighted image with intravenous gadolinium shows the enhancing tumor (arrow). (Reproduced with permission from van Es HW, Bollen TL, van Heesewijk HP. MRI of the brachial plexus: a pictorial review. Eur J Radiol. 2010;74(2):391–402.)
Secondary tumors affecting the brachial plexus are more common and are always malignant. These may arise from local tumors expanding into the plexus. For example, a Pancoast tumor of the upper lobe of the lung may invade or compress the lower trunk, while a primary lymphoma arising from the cervical or axillary lymph nodes may also infiltrate the plexus.95,96 Pancoast tumors typically present as an insidious onset of pain in the upper arm, sensory disturbance in the medial aspect of the forearm and hand, and weakness and atrophy of the intrinsic hand muscles along with an ipsilateral Horner syndrome. Chest CT scans or MRI can demonstrate extension of the tumor into the plexus. Metastatic involvement of the brachial plexus may occur with spread of breast cancer into the axillary lymph nodes and the nearby nerves (Fig. 23-16). Pain is usually the presenting manifestation due to spread of the cancer into the plexus and is accompanied by widespread paresthesias. Weakness and sensory loss conform to the distribution of the affected nerves. Likewise, electrodiagnostic abnormalities are dependent on the nerves that are involved as previously discussed.14,21
Figure 23-16. MRI of brachial plexus. Metastatic plexopathy of breast carcinoma. (A) Sagittal T1-weighted image shows a mass at the level of the divisions of the brachial plexus (long arrows). Note the normal neighboring nerves of the brachial plexus (short arrows). SA, subclavian artery; SV, subclavian vein. (B) Coronal T1-weighted image with intravenous gadolinium demonstrates the enhancement of the metastasis (arrow). (Reproduced with permission from van Es HW, Bollen TL, van Heesewijk HP. MRI of the brachial plexus: a pictorial review. Eur J Radiol. 2010;74(2):391–402.)
The treatment for various malignancies (e.g., lung, breast, and lymphoma) often involves radiation therapy, the field of which may include parts of the brachial plexus. It can be difficult in such situations to determine if a new brachial plexopathy is related to tumor within the plexus or from radiation-induced nerve damage. Radiation can be associated with microvascular abnormalities and fibrosis of surrounding tissues, which can damage the axons and the Schwann cells.1,102 Radiation-induced plexopathy can develop months or years following therapy and is dose dependent.7,96,103
Tumor invasion is usually painful and more commonly affects the lower trunk, while radiation injury is often painless and affects the upper trunk.96 Imaging studies such as MRI and CT scans are useful but can be insensitive in detecting microscopic invasion of the plexus. EMG can be informative, if myokymic discharges are appreciated, as this finding strongly suggests radiation-induced damage. However, absence of myokymic discharges does not rule out radiation as the cause of the plexopathy.
This condition refers to paresis of the arms occurring in soldiers or civilians wearing heavy backpacks or rucksacks strapped around the shoulders.1–3,104,105 Motor and sensory losses most typically are in the distribution of the upper trunk but can be more widespread. The injury is usually neuropraxic in nature, although secondary axonal degeneration may occur. If one sees electrophysiological features of multifocal demyelination at common compression sites (e.g., at the carpal tunnel, across the elbow, and across the fibular head) then hereditary neuropathy with liability to pressure palsy (HNPP) needs to be considered.
The most common surgical procedures associated with brachial plexopathy as a complication are those that involve median sternotomies (e.g., open heart surgeries and thoracotomies). Brachial plexopathies occur in as many as 5.0% of patients following a median sternotomy and typically affects the spinal nerve of the C8 root.1–3,106–108 Thus, individuals manifest with sensory disturbance affecting the medial aspect of forearm and hand along with weakness of the intrinsic hand muscles, as discussed previously. Because of the location of the sensory symptoms, these lesions are often incorrectly blamed on ulnar neuropathies resulting from poor intraoperative elbow positioning or padding. The mechanism of this plexopathy is felt to be related to the stretch of the spinal nerve of the C8 root. These injuries are usually neuropraxic in nature, so most individuals who are affected recover in a few months.107,109 However, some patients with significant axon loss may have a longer and incomplete recovery. Neurophysiological features are those previously discussed for lower trunk lesions.
Burners and stingers refer to brachial plexus injuries caused by impact to shoulder region usually in the course of contact sports (e.g., football).1,2 Usually, the affected athlete notes severe pain and sensory disturbance in the arms without any motor loss. The symptoms typically resolve after a few minutes. The mechanism is unclear, but the rapid recovery in most cases suggests a neuropraxic injury to the cervical roots or plexus, particularly the upper trunk.
Hereditary neuralgic amyotrophy (HNA) is an autosomal-dominant disorder characterized by recurrent attacks of pain, weakness, and sensory loss in the distribution of the brachial plexus, often beginning in childhood.55,110 The clinical and electrophysiological features of HNA resemble those of IBPN. HNA should be considered in patients with recurrent attacks of brachial plexitis, even though the nonhereditary, idiopathic cases can recur.55 In addition, HNPPs can present as painless brachial plexopathy. This may be one etiology of backpack palsy that was discussed in a previous section. In contrast to HNA, HNPP is a generalized or multifocal process, which is demyelinating in nature. HNA can be caused by mutations in the gene encoding septin 9, (SEPT9), while HNPP is usually caused by deletions in chromosome 17p11.2, resulting in a loss of function of peripheral myelin protein 22 (PMP-22). See Chapter 11 regarding “Charcot–Marie–Tooth Disease and Related Disorders” for more details.
The treatment of traumatic brachial plexopathies and timing of any surgical intervention are dependent on the type and severity of the injury, the location, and the time frame.90 Most closed injuries result in neuropraxis or axonotmesis that may recover spontaneously. As a result, they are initially treated conservatively with physical and occupational therapy. Patients are followed closely with serial clinical and electrodiagnostic assessments to assess for recovery. If patients show no signs of recovery after 2–3 months in upper trunk lesions or 4–5 months for middle or lower trunk lesions, then surgical intervention should be considered.90 Injuries associated with high-energy trauma or those associated with near-total paralysis may be observed for a shorter period of time (3 weeks to 3 months) prior to surgery.111 Injuries associated with sharp penetrating trauma are more likely associated with severing of nerves and should be repaired within 72 hours, if possible.17,89,97,112 Worsening neurological function, hematoma formation, concomitant bone or vascular injuries, and compartment syndrome are other indications for more acute surgical intervention.90 Various surgical techniques including neurolysis, nerve grafting, neurotization, and free muscle transfer are performed in order to assist in regaining shoulder abduction and elbow flexion and some use of the hand function.111,113–115
TERMINAL NERVE LESIONSIn this section, we discuss mononeuropathies of the upper limb mainly due to trauma, compression, or entrapment, or those that are idiopathic in nature. Any of these nerves may be affected alone or in combination with other nerve lesions in other settings such as vasculitis (isolated or systemic), infection (e.g., Lyme disease, leprosy, HIV, cytomegalovirus, and hepatitis), immune-mediated demyelination (e.g., multifocal motor neuropathy and multifocal acquired demyelinating motor and sensory neuropathy), and other inflammatory neuropathies (e.g., perineuritis and sarcoidosis), as discussed in other chapters in this book.
As discussed previously, the spinal accessory nerve does not arise from the brachial plexus. Since it is often damaged with trauma to the neck and shoulder region with or without brachial plexus involvement, we discuss spinal accessory neuropathy in this chapter. Lymph node biopsy and other surgical procedures in the posterior triangle are very common etiologies for spinal accessory neuropathies. The nerve can also be involved in IBPN. Injury of the nerve is often painful, presumably due to the mechanical effects from the dropped shoulder it produces. The shoulder drop is best observed from behind the patient. An accessory nerve palsy often results in scapular winging as well and a reduced capability of flexing the arm fully at the shoulder in the sagittal plane. (Fig. 23-17). Winging from a spinal accessory nerve lesion is distinguished from winging from rhomboid and serratus anterior weakness by a number of observations and provocative maneuvers. Winging from trapezius weakness is accentuated by resisted external rotation of the arm at the shoulder. This occurs as the trapezius normally acts to hold the entire medial border of the scapula against the chest wall to provide the resistance necessary for effective external rotation. The winging typically affects the entire medial border of the scapula equally so the inferior angle and posterior angle tend to be at near-equivalent distances from both the spine and chest wall, maintaining the medial scapular border in a vertical orientation. Trapezius weakness can also be detected and distinguished from serratus anterior weakness by the patient’s inability to flex the arm at the shoulder in the prone position. This maneuver results in compensatory lumbar hyperlordosis and producing the triangle sign (the three sides of the triangle being the table, anterior chest wall, and undersurface of the arm with the axilla being the apex. 116,117 Most lesions are distal to the innervation of the sternocleidomastoid muscles; however, proximal damage may result in weakness of turning the head to the contralateral side. CMAPs recorded from the trapezius muscle may demonstrate reduced amplitude compared to the contralateral side, but electrodiagnosis usually relies on demonstrating denervation changes in this muscle.
Figure 23-17. Spinal accessory neuropathy. Winging of the left scapula is appreciated and is brought out by abduction of the shoulder. (Reproduced with permission from Steven A. Greenberg, MD. Reproduced with permission from Greenberg SA, Amato AA. EMG Pearls. Philadelphia, PA: Hanley & Belfus; 2004.)
The dorsal scapular arises mainly from C5 spinal root but may have contributions from the C4 segment. The nerve innervates the rhomboid major and minor along with the levator scapula, which assist in retraction (draw medial border closer to rib cage and midline), elevation, and medial inferior angle rotation of the scapula. Therefore, damage to the dorsal scapular nerve leads to scapular winging, with the inferior angle rotated laterally. Elevation of the arm overhead will accentuate the scapular winging. It is very unusual to have an isolated dorsal scapular nerve injury. NCS are not particularly helpful. Electrodiagnostic confirmation requires demonstration of EMG abnormalities isolated to the rhomboid and levator scapula muscles.
The long thoracic nerve originates from the fusion of branches from the C5, C6, and often C7 spinal roots, and it innervates the serratus anterior muscle (Fig. 23-3). This muscle stabilizes the scapula and helps hold it tight against the chest wall during movement of the shoulder girdle. In addition, it assists in rotating the scapula laterally to allow for full elevation of the arm as the glenohumeral joint provides for only 90 degrees of arm flexion and abduction at the shoulder. A long thoracic neuropathy manifests as scapular winging, a reduction in the ability to elevate the arm in a sagittal and coronal plane, and with reduced strength in pushing activities. The whole scapula is winged. As the muscle originates from the bottom half of the scapula, the inferior angle of the scapula tends to be more affected than the superior angle resulting in the inferior angle to be rotated toward the spine and to be farther off the chest wall than the superior angle. (Fig. 23-18). This winging is accentuated by having the patient flex the arm forward at the shoulder against resistance.
Figure 23-18. Long thoracic neuropathy. Winging of the right scapula is appreciated and is enhanced by having the patient flex the arm forward at the shoulder. There is also atrophy of the infraspinatus secondary to a superimposed suprascapular nerve injury. (Reproduced with permission from Steven A. Greenberg, MD. Reproduced with permission from Greenberg SA, Amato AA. EMG Pearls. Philadelphia, PA: Hanley & Belfus; 2004.)
The long thoracic nerve may be damaged from trauma118–120 or during surgical procedures, particularly mastectomies and thoracotomies (Table 23-5).121,122 Most often, we see long thoracic neuropathies either isolated or in combination with other neuropathies in the setting of IBPN.53 Motor NCS of the long thoracic nerve is not typically performed, and electrodiagnostic confirmation of a long thoracic neuropathy requires demonstration of EMG abnormalities isolated to the serratus anterior muscle. Needle EMG of this muscle should be done cautiously due to risk of pneumothorax.
TABLE 23-5. CONDITIONS ASSOCIATED WITH PROXIMAL LONG THORACIC NEUROPATHY
Trauma
Surgical injury (postthoracotomy, radical mastectomy, axillary surgery, rib resection)
Immune-mediated brachial plexus neuropathy
Long thoracic neuropathies are usually managed conservatively depending on etiology. Open injuries due to trauma may require surgery. Otherwise, in most instances we start with physical and occupational therapy along with bracing. Scapulothoracic stabilization braces can be used to help keep the shoulder abutted against the thorax. If the shoulder function does not improve over time, surgery can be considered to stabilize the scapula.123,124
The suprascapular nerve arises from the upper trunk and innervates the supraspinatus and infraspinatus muscles (Fig. 23-5). The supraspinatus muscle assists in the initial aspects of shoulder abduction, while the infraspinatus muscle is used to externally rotate the arm at the shoulder. Thus, these movements are limited, depending on the location of the suprascapular nerve injury.
The nerve may be damaged with trauma to the shoulder region, particularly if there is a dislocation or fracture of the shoulder.125–129 The nerve may be injured at the suprascapular notch affecting both muscles, or rarely in the spinoglenoid notch with weakness confined to the infraspinatus.130,131 More commonly, this suprascapular nerve is affected in the setting of IBPN and involvement may be isolated to this nerve.53,54
Motor conduction studies to this nerve are technically limiting, so electrodiagnosis relies on EMG demonstration of denervation changes in the supraspinatus and infraspinatus muscle, if the lesion is proximal to the suprascapular notch or limited to the infraspinatus muscle, if the lesion occurs in the region of the spinoglenoid.
Management is dependent on the etiology of the neuropathy. Surgery is warranted for open lesions related to trauma, otherwise conservative therapy with pain control is recommended. Local injections of corticosteroids can be tried if the cause is felt to be related to compression of the nerve in the suprascapular or supraglenoid notch, and some even advocate surgery; however, entrapment of the nerves at this site remains a controversial etiology.132,133
The medial and lateral pectoral nerves are discussed together as both innervate the pectoralis minor and major muscles (Fig. 23-3). The large pectoralis major muscle assists in internal rotation, anterior flexion, and adduction of the arm at the shoulder, while the pectoralis minor assists in scapula stabilization during arm extension at the shoulder. These nerves may be damaged rarely, usually during surgical procedures in the anterior chest and axillary region. Again, motor conduction studies of these nerves are not routinely performed and electrodiagnostic confirmation requires demonstration of EMG abnormalities in the pectoralis minor and major muscles.
Injury to the subscapular nerve has not been described in detail and rarely occur in isolation but may be involved in more generalized plexopathy. As the lower subscapular nerve innervates the teres major muscle, damage to this nerve may result in weakness of internal rotation and adduction of the arm at the shoulder. There are no motor NCS for this nerve, and needle EMG of the muscle is difficult given its deep location.
The thoracodorsal nerve arises form the posterior cord and innervates the latissimus dorsi muscle (Fig. 23-3). Weakness of this muscle results in impaired ability to adduct, internally rotate, and extend the arm at the shoulder. Slight winging of the inferior margin of the scapula may be observed when the patient is asked to place the dorsum of the hand of the affected arm on the buttock.10,121
The nerve is usually affected in association with posterior cord or more proximal brachial plexus injuries. NCS are not routinely done on this nerve, but EMG of the latissimus dorsi muscle is easy to perform and helps in localizing the lesion to C5–7 nerve fibers at or proximal to the posterior cord.
The musculocutaneous nerve represents a continuation of the lateral cord and innervates the coracobrachialis, biceps brachii, and to some extent the brachialis (Fig. 23-6). After innervating these muscles, it terminates as the lateral antebrachial cutaneous nerve to supply sensation to the lateral aspect of the forearm from the elbow to the wrist. Damage to the musculocutaneous nerve may therefore result in sensory loss in this distribution and weakness of elbow flexion accompanied by a reduced deep tendon reflex of the biceps brachii. The musculocutaneous nerve may be damaged by anterior dislocations of the shoulder and prolonged hyperextension of the arm, secondary to weight lifting (perhaps compressed within hypertrophic muscle) (Table 23-6).8,121,134–136 It is also often affected in IBPN.53
TABLE 23-6. CONDITIONS ASSOCIATED WITH MUSCULOCUTANEOUS NEUROPATHY
Trauma (fracture or dislocation of shoulder, fracture of humerus, missile injuries, stab wounds, blunt force injuries)
Injection injury
Immune-mediated brachial plexus neuropathy
Soft tissue or peripheral nerve tumor
Ischemia (e.g., vasculitis)
Multifocal motor neuropathy or multifocal acquired demyelinating motor and sensory neuropathy
Compression within hypertrophied biceps brachii muscle after vigorous exercise
Compression by sharp free margin of biceps aponeurosis
The lateral antebrachial cutaneous SNAP is easy to obtain and would be expected to be reduced in axonal lesions affecting the musculocutaneous nerve (Table 23-2). This is nonlocalizing in and of itself as the SNAP could also be reduced with lateral cord or upper trunk lesions; however, it would be normal in C6 radiculopathy. A musculocutaneous CMAP can be obtained by stimulating the brachial plexus in the supraclavicular fossa and recording from the biceps brachii. It can provide valuable prognostic information. A normal CMAP recording from biceps and stimulating in the axilla occurring 10 days after injury in a patient who cannot activate the biceps strongly implicates conduction block and suggests a rapid and excellent return of function. Comparing the CMAP amplitude to the opposite side can provide an estimate of the degree of axon loss. EMG may show denervation abnormalities in the coracobrachialis, biceps brachii, and brachialis muscles (Table 23-1). Again abnormalities in the supraspinatus, deltoid, biceps brachii, and pronator teres muscles, but not in serratus anterior, rhomboids, or paraspinal regions, would imply an upper trunk injury, while denervation changes in the latter three regions would suggest a radiculopathy or anterior horn cell disease. On the other hand, only finding abnormalities in the biceps brachii and pronator teres, sparing deltoid, is more consistent with a lateral cord injury.
Initial management depends on the etiology of the neuropathy. Those caused by severe trauma may require surgical treatment. However, in most cases a conservative approach is warranted.
The axillary nerve originates from the posterior cord and innervates the teres minor and deltoid muscle (Fig. 23-7). In addition, the lateral cutaneous nerve of the arm arises from the axillary nerve. Thus, axillary neuropathies may manifest with weakness of abduction of the arm and sensory loss in the region of skin overlying the deltoid muscle.
Axillary neuropathies may occur in the setting of IBPN, trauma to the shoulder, fractures of the upper humerus, or stretch injury (Table 23-7).8,9,121,137–139 Axillary CMAPs may be recorded from the deltoid muscle following supraclavicular stimulation of the brachial plexus to see if there is asymmetrical loss of amplitude on the affected site or a disconnect between the amount of movement and the size of the CMAP. A superficial radial SNAP would be expected to be normal in an axillary neuropathy and can help distinguish an axillary neuropathy from a posterior cord lesion or upper trunk lesion (Table 23-2). Furthermore, EMG should show evidence of denervation in the deltoid and teres minor muscles with sparing of radial-innervated muscles in an isolated axillary neuropathy (Table 23-1). In addition, a normal EMG of the supraspinatus, infraspinatus, rhomboids, biceps brachii, pronator teres, and brachioradialis suggests that the lesion is distal to the C5/C6 roots or upper trunk when combined with denervation of the deltoid.
TABLE 23-7. CONDITIONS ASSOCIATED WITH AXILLARY NEUROPATHY
Trauma (e.g., fracture or dislocation of shoulder, fracture of humerus, missile injuries, stab wounds, blunt force injuries)
Stretch injury (e.g., hyperabduction during sleep, surgery)
Injection injury
Immune-mediated brachial plexus neuropathy
Soft tissue or peripheral nerve tumor
Ischemia (e.g., vasculitis)
Multifocal motor neuropathy or multifocal acquired demyelinating motor and sensory neuropathy
Axillary neuropathies related to penetrating injuries should be surgically explored. Otherwise, these are managed conservatively with pain management and PT/OT. If there is no improvement within 6 months, surgical treatment and grafting can be considered.140
The radial nerve is one of the major terminations of the posterior cord and is composed of fibers from spinal segments C5–8 and occasionally contains T1 fibers (Fig. 23-8). The radial nerve is quite long and provides innervation to upper arm and forearm muscles as well as for cutaneous sensation of large aspects of the arm. The clinical and electrodiagnostic features of radial neuropathies depend on the site of the lesion. The superficial radial SNAP should be abnormal, if there is significant axonal nerve injury, except with posterior interosseous nerve damage, as this is a purely motor nerve. Radial CMAP to the radial-innervated muscles such as the extensor indicis proprius should be performed with short incremental stimulation of the radial nerve through the spinal groove to assess for focal conduction block or slowing across this site. With significant axonal injury, the CMAP amplitude should be reduced regardless of stimulation site. EMG is more helpful in localizing the site of the lesion with axonal injury. Evidence of active denervation in the form of fibrillation potentials and positive sharp waves would be expected in an axonal nerve injury, provided there has been substantial time for Wallerian degeneration to occur. In a pure neuropraxic injury, only reduced recruitment of MUAPs would be appreciated on EMG, although many predominantly demyelinating injuries may have some element of axon loss. A few fibrillation potentials do not preclude a good recovery as they may originate from a very small number of injured axons.
Damage to the nerve in the axilla or proximal arm is uncommon but can result from compression (e.g., crutches, intoxicated patients who fall asleep with outstretched arm pressed against a hard surface, missile injuries, and other trauma to the axilla) (Table 23-8).8,9,121 Of course, a radial neuropathy can also occur in the setting of a more widespread multifocal process (e.g., vasculitis, IBPN). Proximal radial nerve injuries can result in weakness of elbow, wrist, and finger extension as well as supination of the forearm. In addition, sensory disturbance may be evident in the posterior aspect of the forearm and back of the hand and fingers. Provided there is sufficient axon loss, the superficial radial SNAP and radial CMAP recorded from the extensor indicis proprius may have reduced amplitudes (Table 23-2). EMG should demonstrate signs of denervation in the triceps as well as more distal radial-innervated forearm muscles (Table 23-1).
TABLE 23-8. CONDITIONS ASSOCIATED WITH PROXIMAL RADIAL NEUROPATHY
Trauma
Fracture of humerus
Improper use of crutches (e.g., compression in axilla)
Stretch injury (e.g., hyperabduction of arm during surgery, sleep)
Saturday night palsy (external compression by arm being compressed against firm edge at the spiral groove—usually in intoxicated individuals)
Other external compression (partner falling asleep on arm)
Immune-mediated brachial plexus neuropathy
Soft tissue or peripheral nerve tumor
Ischemia (e.g., A-V fistulas, vasculitis)
Multifocal motor neuropathy or multifocal acquired demyelinating motor and sensory neuropathy
Radial neuropathy in the arm distal to the branches innervating the triceps arises from various mechanisms. One of the most common radial neuropathies is the so-called “Saturday night palsy” and is usually the result of prolonged compression of the radial nerve in the spiral groove in an individual who is intoxicated. Proximal radial nerve lesions have also been speculated to be the result of anomalous muscle compression or damage secondary to triceps muscle contraction.141,142 On clinical examination, one would expect to find weakness of the radial-innervated muscles distal to the triceps in addition to sensory loss in the posterior aspect of the forearm and back of the hand and fingers. Again, a superficial radial SNAP and a radial CMAP may have reduced amplitudes, if the injury is axonal. EMG should demonstrate signs of denervation of radial-innervated forearm muscles, unless it is caused by a pure conduction block, with sparing of the more proximal triceps muscles.
Proximal radial neuropathies caused by penetrating trauma should be surgically explored and treated with end-to-end anastomosis or grafting. Closed traumas, including humeral fractures, are often due to neuropraxia and recover gradually on their own. A trial of conservative therapy is employed prior to any surgery. Proximal radial neuropathies related to pressure or stretch injuries (e.g., Saturday night palsy) or IBPN are also treated conservatively. Finger and wrist splints, pain control, and physical and occupational therapy are employed.
Damage to the posterior interosseous nerve will result in weakness of wrist and finger extensors with sparing of sensation. The posterior interosseous nerve can be damaged from multiple mechanisms (Table 23-9). Although some have speculated that the nerve can be entrapped within the supinator muscle (arcade of Fröhse), this is quite rare in our opinion. Many such cases probably represent a forme fruste of an IBPN or another immune-mediated neuropathy (e.g., multifocal motor neuropathy) (Fig. 23-19). On NCS, the superficial radial SNAP should be normal, but the radial CMAP recorded from the extensor indicis proprius may reveal a reduction in amplitude, provided there is significant axon loss (Table 23-2). EMG should demonstrate signs of denervation in muscles innervated by the posterior interosseous nerve.
TABLE 23-9. CONDITIONS ASSOCIATED WITH POSTERIOR INTEROSSEOUS NEUROPATHY
Immune-mediated brachial plexus neuropathy
Trauma
Compression by tumors, ganglion cysts, lipoma, bursitis
Compression by the arcade of Fröhse
Compression by facial bands connecting the brachialis to the brachioradialis muscle at the radial head
Compression by edge or fibrous bands within the supinator muscle
Compression by a bifid extensor carpi radialis brevis muscle
Rheumatoid arthritis
Soft tissue or peripheral nerve tumor
Ischemia (e.g., A-V fistulas, vasculitis)
Multifocal motor neuropathy or multifocal acquired demyelinating motor and sensory neuropathy
Figure 23-19. MRI of (T2) forearms in a patient with multifocal motor neuropathy affecting the left radial nerve demonstrates focal enlargement and enhancement of the radial nerve (arrows) in the forearm on the left side. (Reproduced with permission from Steven A. Greenberg, MD. Reproduced with permission from Greenberg SA, Amato AA. EMG Pearls. Philadelphia, PA: Hanley & Belfus; 2004.)
Unless the posterior interosseous neuropathy is related to open trauma, it is managed conservatively as discussed with proximal radial neuropathies. Rare cases of the so-called radial tunnel syndrome with compression of the posterior interosseous nerve may improve with surgery.143 Again, we feel that such entrapment is quite rare and the existence is controversial.
The superficial radial nerve is a pure sensory branch of the radial nerve that provides sensation to the dorsum of the hand. It can be damaged by various means (Table 23-10). In particular, compression by tight bands, watches, and handcuffs can lead to a superficial radial neuropathy. The superficial radial SNAP is usually decreased in amplitude, while motor studies and EMG would be normal. This type of neuropathy is usually due to neuropraxia and improves spontaneously. Cases related to laceration or other trauma may require surgery.
TABLE 23-10. CONDITIONS ASSOCIATED WITH SUPERFICIAL RADIAL NEUROPATHY
External compression (handcuffs, tight wrist bands, casts)
De Quervain tenosynovitis
Trauma
Soft tissue or peripheral nerve tumor
As previously discussed, the median nerve contains fibers originating from spinal segments C6–T1, which then course through all three trunks and the medial and lateral cords. The median nerve is formed by the merging of branches from the medial and lateral cords (Fig. 23-9). Axons from spinal segments C5–7 that course through the upper and middle trunks and lateral cords are responsible for providing cutaneous sensation to the palmar aspect of the hand and digits 1–3 and usually the lateral half of digit 4. In addition, these segments also innervate several forearm muscles, primarily the pronator teres and flexor carpi radialis. On the other hand, C8 and T1 nerve fibers course through the lower trunk and medial cord and innervate muscles controlling finger movements and provide no sensory input.
Proximal median neuropathies in the axilla, upper arm, and forearm may result from misuse of crutches, missile injuries, and laceration of the nerve by trauma (Table 23-11) (Fig. 23-20).8,9,144,145 Compression of the nerve can also occur due to an awkward sleeping position—often in individuals who are intoxicated. Ischemic damage to the median nerve can occur as a complication of nerve ischemia due to arterial diversion resulting from creation of shunts of fistulas for renal dialysis.146 The median nerve can be affected as well in the setting of IBPN. Proximal median neuropathies have been reported to be caused by compression by the ligament of Struthers, but this is controversial.147–150 Compression by the lacertus fibrosus or bicipital aponeurosis at the elbow have also been implicated as possible etiologies.151
TABLE 23-11. CONDITIONS ASSOCIATED WITH PROXIMAL MEDIAN NEUROPATHY
Improper use of crutches (e.g., compression in axilla)
Trauma (e.g., dislocation of shoulder, fracture of humerus, missile injuries, stab wounds, tourniquets)
Compression by ligament of Struthers
Pronator teres syndrome
Thickened lacertus fibrosus
Fibrous arch of the flexor digitorum superficialis
Tendinous band or hypertrophied pronator teres muscle
Sleep palsies
Compartment syndrome
Ischemia (e.g., A-V fistulas, vasculitis)
Immune-mediated brachial plexus neuropathy
Soft tissue or peripheral nerve tumor
Multifocal motor neuropathy or multifocal acquired demyelinating motor and sensory neuropathy
Figure 23-20. Ultrasound images depicting transection of the median nerve in the forearm. (A) Sagittal view showing the distal nerve stump (arrow), proximal stump (arrowhead), and the transection (line). Cross-sectional views at the level of the distal nerve (B), site of transection (C), and proximal nerve (D). The median nerve is seen in images (B) and (D), but it is not present in image (C) at the site of transection. (Reproduced with permission from Cartwright MS, Chloros GD, Walker FO, Wiesler ER, William W, Campbell WW. Diagnostic ultrasound for nerve transaction. Muscle Nerve. 2007;796–799.)
Individuals with proximal median neuropathies present with weakness of the median-innervated forearm and hand muscles and reduced sensation in the palmar aspect of the hand, digits 1–3, and the lateral aspect of digit 4. In our experience, it is not uncommon for proximal median neuropathies to clinically manifest as predominantly anterior interosseous syndromes, even though electrodiagnosis suggests that the entire nerve is affected. In these cases, median SNAPs to any of these digits would be expected to show reduced amplitudes again, provided there is sufficient axonal injury. The distal latency or conduction velocity of the median SNAP would be expected to be normal or only slightly impaired compared to the loss of amplitude. Similarly, the median CMAP amplitude recorded from the abductor pollicis studies may be reduced. It is important in these proximal median neuropathies to look for evidence of slowing of CV, temporal dispersion, or focal conduction block (discussed in Chapters 2 and 12). EMG would be expected to demonstrate abnormalities in median-innervated muscles in the forearm and hand (Table 23-1).
A controversial entity is the so-called pronator teres syndrome. In this disorder, the median nerve is thought to be compressed where it passes under the fibrous arch connecting the two heads of the pronator teres muscle. The major clinical manifestation is pain and tenderness in the volar aspect of the forearm and paresthesias in the distribution of the median nerve. These symptoms are exacerbated by having the patient actively trying to pronate the forearm against resistance. We remain rather skeptical of this diagnosis, as there is usually no objective clinical or electrodiagnostical evidence of median nerve injury.
Proximal median neuropathies carry a poor prognosis if there is significant axonal degeneration. The reason is the long distance the nerve must grow in order for complete reinnervation to occur. As long as there are some voluntary MUAPs in the forearm and hand muscles, there is potential for recovery.
The proximal median neuropathies are usually treated conservatively unless trauma is involved. Decompression surgeries have not been adequately studied in a scientific fashion, owing in part to the rarity of proximal median compressive neuropathies.
The anterior interosseous nerve can be damaged from multiple mechanisms (Table 23-12). Most commonly, in our experience, an anterior interosseous neuropathy arises either in conjunction with or as a forme fruste of an IBPN. As mentioned, proximal median neuropathies may masquerade as anterior interosseous syndrome. As the anterior interosseous nerve is a pure motor nerve, patients do not have sensory loss. However, severe pain in the forearm for several days or weeks is typical in cases related to IBPN. Individuals have weakness in the flexor digitorum profundus I and II, flexor hallucis longus, and pronator quadratus muscles. This leads to difficulty with pinching maneuvers or forming the letter “O” with their thumb and index or middle fingers, as they have weakness of flexion of the distal aspects of these digits. Most cases should be managed conservatively. However, if there is no improvement in function after 4–6 months, surgical exploration to assess for compression can be considered.139,152
TABLE 23-12. CAUSES OF ANTERIOR INTEROSSEOUS NEUROPATHY
Immune-mediated brachial plexus neuropathy
Trauma
Fibrous band within the pronator teres
Compartment syndrome
Soft tissue or peripheral nerve tumor
Ischemia (e.g., A-V fistulas, vasculitis)
Multifocal motor neuropathy
Median neuropathy at the wrist or carpal tunnel syndrome (CTS) is the most common mononeuropathy. There are multiple causes of median neuropathy at the wrist, although the vast majority are thought to be related to tenosynovitis of the flexor tendons which also occupy the carpal tunnel along with the median nerve (Table 23-13).8,9 Some clinicians restrict the term “CTS” only to those median neuropathies at the wrist caused by tenosynovitis. People with median neuropathy at the wrist usually complain of intermittent numbness and tingling of their fingers particularly at night or in other situations where the carpal tunnel is narrowed by wrist extension or extension, for example, holding a steering wheel, telephone, or hairdryer. Sometimes the numbness and tingling as well as the pain patients describe extend beyond the territory of the median nerve (e.g., these may describe sensory symptoms in the little finger and aching in the forearm as well). The symptoms may be exacerbated by repetitive activity. However, the discomfort often occurs at rest. The painful paresthesias may be briefly alleviated by shaking the hands, the so-called ‘flick sign.”153
TABLE 23-13. CONDITIONS ASSOCIATED WITH MEDIAN NEUROPATHY AT THE WRIST
Idiopathic
Flexor tenosynovitis
Degenerative joint disease
Rheumatoid arthritis
Sarcoidosis
Space occupying lesions (e.g., ganglion cysts, lipomas, hemangiomas, giant cell tumors, osteomas)
Trauma (e.g., Colles’ fracture, dislocation/fracture of carpal bones)
Pregnancy
Endocrine (e.g., hypothyroid, acromegaly, diabetes mellitusa)
Amyloidosis (familial and primary)
Hereditary neuropathy with liability to pressure palsies
Soft tissue or peripheral nerve tumor
aIt is unclear if individuals with typical generalized diabetic polyneuropathy may be predisposed to focal mononeuropathies related to compression.
Clinical examination is frequently normal when the nerve is predominantly irritated, not injured. When axon loss occurs, patients may develop constant numbness and the examination may reveal loss of sensation in the median nerve distribution to the fingers. Motor function is generally spared, unless the injury is severe at which point weakness of atrophy of the thenar muscles is appreciated. Tapping over the wrist may elicit increased paresthesias in the fingers (Tinel sign), but this is not very specific. Having the persons maintain their wrists in the flexed posture may also exacerbate the discomfort in the fingers (Phalen sign), and this is more specific for CTS.
Sonography of the median nerve at the wrist may demonstrate flattening of the median nerve at the wrist (Fig. 23-21).30,31,36,154 A recent meta-analysis reported that the sensitivity and specificity of ultrasound in the diagnosis of CTS are 77.6% and 86.8%, respectively.154 MRI of the wrist may show reduction of the cross-sectional diameter of the carpal tunnel.25,26 Swelling of the tendons, boney and cystic lesions, as well as compression of the nerve may be visualized by MRI. The major drawback of MRI is that it is very expensive. Despite the potential benefits, imaging of the carpal tunnel is not routinely applied by most clinicians.
Figure 23-21. Sonograms of a patient with symptomatic right carpal tunnel syndrome. The right median nerve had a markedly increased cross-sectional area (CSA) of 25 mm2 at the distal wrist crease (normal <12 mm2) (A), and 5 mm2 in the forearm (B), as outlined by the green dashes resulting in an increased wrist/forearm CSA ratio of 5 (normal <1.5).
There are various NCS that can be performed to confirm the clinical impression of a median neuropathy at the wrist. It should be recognized that NCS like all tests are imperfect, in part because it tests for nerve injury, not irritation. It is estimated that approximately 10% of patients with histories highly suggestive of CTS will have normal NCS. In addition to performing median sensory and motor studies, it is essential to also include motor and sensory studies of other nerves (e.g., ulnar or radial) to ensure that the neuropathy is not more generalized. The studies should also be tailored according to the individual’s symptoms. If a patient complains of sensory disturbance mainly in the third digit, then a median SNAP to the third digit should be performed as opposed to doing the median SNAP to the thumb or second digit. Median SNAPs are more sensitive than CMAPs in detecting abnormalities associated with CTS. Mixed compound nerve action potentials (CNAPs), which are obtained by stimulating the median and ulnar mixed nerves in the palm and recording over the respective nerves at the wrist, are often even more sensitive as they are usually performed across a shorter distance. Significantly prolonged distal latencies of the median palmar mixed CNAP compared to the ulnar study would support the clinical impression of CTS. In addition to stimulating at the wrist and recording at the digit, it is sometimes useful to do a mid-palm stimulation (half-way between site of wrist stimulation and the recording electrodes) when performing the median SNAPs. If one sees a prolonged distal latency/slow CV and reduced amplitude following wrist stimulation, this should be compared to latency/CV and amplitude after stimulation in the palm to see if there was more focal slowing or conduction block across the wrist. This is particularly valuable in people who have a coexisting generalized polyneuropathy, in order to see if there is a superimposed median neuropathy at the wrist.
The earliest electrodiagnostic abnormalities on NCS in CTS are prolonged distal latencies or slowing of the median SNAP or palmar mixed CNAP across the palm. Subsequently, there may be a reduction in SNAP or mixed CNAP amplitude due to either axon loss or conduction block. Subsequently, distal latencies of the median CMAP become prolonged. Amplitudes of the median CMAP are usually affected much later in the course. In severe cases of CTS, median SNAPs and median CMAPs recorded from the abductor pollicis brevis may be unobtainable. Thus, from an NCS standpoint, one cannot localize the site of the median neuropathy, as a lesion may be anywhere from the hand to the origin of the median nerve in the plexus. In such cases, it is useful to perform median CMAP to the second lumbrical and ulnar CMAP to the second interosseous muscles while stimulating the median and ulnar nerves, respectively, at the wrist. The reason is that the median CMAP from the second lumbrical is often less affected in CTS than the CMAP from the APB. Therefore, a CMAP from this muscle may be obtained when one from the APB cannot. Thus, a prolonged distal latency and reduced amplitude of the median CMAP (second lumbrical) compared to the ulnar CMAP (second interosseous) may be appreciated and confirm the localization of the lesion to the wrist. EMG is often done to further assess the degree of axonal damage and assess the localization of the lesion. Often the EMG is normal in mild CTS. Reduced recruitment of normal appearing MUAPs suggests conduction block. Given its chronic nature, enlarged MUAPs in the APB commonly occur. Signs of active denervation (e.g., fibrillation potentials) are less common but may become evident in rapidly progressive or severe cases.
The treatment of CTS has been the subject of several recent reviews.155–158 Treatments of CTS include modification of activities, splinting of the wrist, corticosteroid injections, nonsteroidal anti-inflammatory drugs, diuretics, and surgery. To complicate matters, there are various surgical techniques that can be employed as well (standard open surgery with exploration and release versus minimally invasive endoscopic approach). Unfortunately, most studies of CTS have lacked scientific rigor, and thus recommendations for the best therapeutic approach are debatable.
Twenty to seventy percent of patients with CTS treated nonsurgically improve to some extent.155,159–161 Corticosteroid injections into the carpal tunnel have become an increasingly used alternative to surgery. Risks include cutaneous atrophy, depigmentation, and inadvertent puncture of the median nerve, blood vessels, or tendons within the carpal tunnel.155 Median nerve injury and tendon rupture are the most severe complications, but each occurs in <0.1% of injections. Approximately 30% of patients have no or only mild improvement following local corticosteroid injection, while 70% have a very good response (complete relief or only minor residual symptoms).155,162 A study comparing corticosteroid injection versus surgery demonstrated similar short-term efficacy of both treatments, but the relapse rate was common in the injection group and rare in the surgical group after 1 year.101 In this regard, there are no good studies assessing the safety and efficacy of repeated corticosteroid injections. A recent randomized trial comparing surgery to conservative management with wrist splints and nonsteroidal medications demonstrated a modest benefit of unclear clinical significance with surgery.161
The average success rate from surgery is approximately 75% (range 27–100%), but 8% of patients actually worsen.155 Failure rates may relate to patients being operated on who do not actually have CTS. In this regard, improvement following surgery is noted in only half of patients who had normal electrodiagnostic testing prior surgery, while success rates are much higher in those with NCS that were abnormal. Another common cause of failed surgery is incomplete division of the transverse carpal ligament, perhaps owing to poor choice of incision and inadequate exposure.155 Another reason for failed surgery is such end-stage denervation resulting from delayed treatment. There may be a <50% success rate for surgery in patients with marked thenar atrophy and weakness, no recordable median CMAPs and SNAPS, and active denervation on EMG.155,162 In our opinion, CTS surgery in this population should only be considered for reasons of pain relief, not with the expectation that strength or sensation will return in any meaningful way.
Complications of surgery occur in 1–2% of cases and include injury to the recurrent motor and cutaneous branches of the median nerve, lesions of the main trunk of the median nerve, the main trunk and deep motor branch of the ulnar nerve, postoperative hematoma, wound infection, scarring, and complex regional pain syndrome type II.155
With the above caveats, we initially try conservative management having patients wear neutral angle wrist splints, particularly those individuals with only sensory abnormalities on NCS. In patients with objective motor deficits, we still try a short trial of wrist splints along with corticosteroid injections. However, we refer them for surgery if there is no benefit after a couple of months. We usually do not recommend surgery when NCS are normal.
The ulnar nerve is the anatomic continuation of the medial cord and contains nerve fibers originating from the C8 and T1 spinal roots, which course through the lower trunk and then median cord (Fig. 23-10). As previously discussed, there may also be a contribution from C7 in some individuals. This is a long nerve and lesions may occur anywhere along its course. Therefore, the clinical and electrophysiological findings are dependent on the site and nature of the lesion.
Similar mechanisms that cause proximal median and radial neuropathies can cause a proximal ulnar neuropathy (Table 23-14).8,9 There are no ulnar-innervated muscles in the upper arm; therefore, any proximal lesion will clinically resemble those caused by more common ulnar neuropathy at the elbow (discussed in next section). Ulnar neuropathies in the upper arm related to open trauma usually require surgical repair.
TABLE 23-14. CONDITIONS ASSOCIATED WITH PROXIMAL ULNAR NEUROPATHY
Trauma
Compression during sleep
Soft tissue or peripheral nerve tumor
Ischemia (e.g., A-V fistulas, vasculitis)
Multifocal motor neuropathy or multifocal acquired demyelinating motor and sensory neuropathy
Provided there is significant axonal loss, the ulnar and dorsal ulnar SNAPs would be expected to have reduced amplitudes, while the medial antebrachial cutaneous SNAP should be normal. The ulnar CMAP may demonstrate reduced amplitude without focal slowing or conduction block across the elbow. EMG should show abnormalities confined to ulnar-innervated muscles in the hand and forearm (Table 23-1). However, these SNAPs and EMG alterations do not distinguish between a proximal ulnar neuropathy in the upper arm and one across the elbow. Electrophysiologically, the only way one can localize an ulnar neuropathy to the proximal upper arm is by demonstrating focal conduction block or slowing of ulnar CV between axillary and above the elbow stimulation sites.
This is the second most common mononeuropathy aside from CTS, and it is usually the result of compression of the nerve at this level. As the nerve is superficial around the ulnar groove, it is more susceptible to extrinsic compression (e.g., from leaning on the elbow). There are numerous intrinsic mechanisms by which the ulnar nerve may be injured in this region (Table 23-15).8,9 The term “tardy ulnar palsy” is applied to ulnar neuropathies that occur on a delayed basis following bone injuries at the elbow. It is speculated that the nerve may become stretched or compressed by exuberant callus formation or altered angle of the elbow joint. The nerve may also become entrapped or compressed by the humeroulnar aponeurotic retinaculum or by other anatomic structural variants in and around the ulnar groove and cubital tunnel. Also, the nerve can occasionally prolapse out of the ulnar groove although this happens in many normal individuals and should not be assumed to be pathological.
TABLE 23-15. CONDITIONS ASSOCIATED WITH ULNAR NEUROPATHY AT THE ELBOW
Tardy ulnar palsy (due to deformities of elbow related to previous fractures of humerus or other trauma to the joint)
Subluxation of the ulnar nerve
Compression by arcade of Struthers (medial intramuscular septum)
Compression by aponeurotic band between heads of flexor carpi ulnaris
Compression by ligament/band (retrocondylar)
Trauma
Soft tissue tumor or masses
Leprosy
Diabetes mellitusa
aIt is unclear if individuals with typical generalized diabetic polyneuropathy may be predisposed to focal mononeuropathies related to compression.
Regardless of the etiology, the clinical signs and symptoms of an ulnar neuropathy at the elbow are similar. Individuals who are affected often complain of discomfort and perhaps tenderness in the medial elbow. They will typically describe numbness and tingling in the medial aspect of the hand and the fifth digit along with the medial half of the fourth digit (both palmar and dorsal aspects of these fingers). It is important to realize that the vast majority of ulnar neuropathies present with sensory symptoms. Compressive or entrapment ulnar neuropathies presenting with purely motor signs and symptoms are extremely rare in our experience.
Tapping the nerve in the elbow often exacerbates these symptoms (e.g., positive Tinel sign). Weakness may involve any or all of the ulnar-innervated muscles in the hand and forearm. This can lead to decreased muscle grip and spreading out or bringing fingers closer together. It is not uncommon for an ulnar neuropathy at the elbow to produce detectable hand but not forearm muscle weakness. We have found assessing for weakness of the flexion of little finger at the distal interphalangeal joint to be the most reliable means by which to detect ulnar forearm muscle weakness when it is present. With axonal degeneration, atrophy of the hypothenar and interossei muscles may be seen (most notably appreciated in the first dorsal interosseous).
Imaging studies such as MRI and, in particular, ultrasound have been used to assist in diagnosis (Fig. 23-22).32,33 Ulnar and dorsal ulnar cutaneous SNAP amplitudes should be reduced if there is significant axonal damage at or near the elbow. However, if there is only a neuropraxic or demyelinating lesion in the elbow these SNAPs are typically normal, as these do not assess slowing of conduction across the elbow. Localization is dependent on the ulnar motor conductions and EMG.6,163–165 We usually perform ulnar motor conductions recording from both the first dorsal interosseous and the abductor digiti minimi muscles with stimulation sites at the wrist, below the elbow, and above the elbow. Because of the fascicular arrangement of nerves destined to innervate these muscles, one may find abnormalities in one but not the other muscle. Most of the lesions in the elbow initially lead to demyelination in this segment. One would expect to see normal distal latencies, amplitudes, and CV between the wrist and below-elbow stimulation sites. However, slowing of CV may be appreciated between the below- and above-elbow sites.164 As the size of the demyelinating lesion may be small, the shorter the distance between the below- and above-elbow sites, the more likely one will be able to demonstrate focal slowing (we try to keep the distance at most 8–10 cm). In addition, conduction block may be appreciated between these sites of stimulation. To further localize where in the elbow the nerve is damaged and to increase the sensitivity if routine ulnar CMAPs across the elbow are normal, one can do inching studies. Perhaps, a more appropriate term is “centimetering” as the nerve is stimulated every 1–2 cm, beginning 5 cm below to 5 cm above the elbow.6,163,165 In most patients with focal demyelinating lesions in this location detected with this technique, both an abrupt latency shift and change in CMAP morphology can be reproducibly demonstrated. Assessing for slowing of the ulnar CNAPs across the elbow may be informative in some patients.166 If secondary axonal degeneration occurs then focal slowing of conduction velocity or conduction block may no longer be apparent. In such cases, the ulnar CMAP does not help in differentiating an ulnar neuropathy at the wrist from a more proximal lesion. The dorsal ulnar cutaneous SNAP is thus important, because an abnormality in this study implies a lesion proximal to the wrist as do EMG abnormalities in ulnar-innervated forearm muscles (e.g., flexor carpi ulnar and flexor digitorum III and IV). Unfortunately, in a significant number of cases of ulnar mononeuropathy, predominantly those without a demyelinating component, the site of the lesion cannot be precisely localized. Again, in neuropraxic or demyelinating lesions, the EMG may just demonstrate reduced recruitment of MUAPs. One particular pitfall in assessment of potential ulnar neuropathies is the demonstration of a greater than 20% drop in ulnar CMAP amplitude comparing below elbow to wrist stimulation. Although this could implicate a demyelinating ulnar neuropathy in the forearm, a Martin-Gruber anastomosis provides a far more common possibility and needs to be excluded.
Figure 23-22. Ultrasound of ulnar nerve reveals marked increase in area above the elbow of 24 mm2 (normal ≤10 mm2) and normal area at the wrist of 5 mm2 (normal ≤6 mm2) resulting in an increased elbow/wrist ratio of 4.8 (normal <2).
There is a lack of randomized, prospective studies aimed at assessing the efficacy of various treatments of the more common ulnar neuropathy at the elbow.167 Most studies have been retrospective and subject to bias. Individuals with intermittent sensory symptoms may respond to conservative measures.168 Nonsurgical measures include elbow pads, avoidance of leaning on the elbow, splinting the elbow in extension at night, and nonsteroidal anti-inflammatory drugs. Surgical procedures may be more beneficial in patients who have motor signs and symptoms, but not everyone improves. There are various surgical approaches (e.g., simple decompression, medial epicondylectomy, and nerve transposition), and there does not seem to be any significant differences in clinical outcomes,169 but none has been rigorously studied in a scientific fashion. There may be increased risks with nerve transposition including infarction due to devascularization of the nerve and increased scarring. We are particularly reluctant to recommend this procedure to diabetics with their increased risk of microvasculopathy.
The ulnar nerve can be damaged at various locations within the wrist or hand and by different mechanisms (Table 23-16). One of the most common etiologies is a compression by a ganglion cyst, which can easily be seen on MRI (Fig. 23-23) or ultrasound of the hand.35 The ulnar nerve can be damaged in one of four sites within the hand, and the clinical and electrophysiological findings are dependent on the site and nature of the lesion (Fig. 23-24).
TABLE 23-16. CONDITIONS ASSOCIATED WITH ULNAR NEUROPATHY AT THE WRIST
External compression (e.g., bicyclist)
Space occupying lesions (e.g., ganglion cysts, lipoma, nerve sheet tumors)
Trauma (fracture to metacarpals, pisiform, hamate, dislocation of distal ulna, laceration
Degenerative arthritis
Rheumatoid arthritis
Diabetes mellitusa
aIt is unclear if individuals with typical generalized diabetic polyneuropathy may be predisposed to focal mononeuropathies related to compression.
Figure 23-23. MRI of the wrist in a patient with ulnar neuropathy. MRI reveals a ganglionic cyst (arrow) adjacent to the hamate in Guyon’s canal that is displacing the ulnar nerve and artery (arrowhead). (Reproduced with permission from of Steven A. Greenberg, MD. Reproduced with permission from: Greenberg SA, Amato AA. EMG Pearls. Philadelphia, PA: Hanley & Belfus, Inc; 2004.)
Figure 23-24. There are four main areas in which the ulnar nerve can be damaged at the wrist, and each leads to different clinical and electrophysiological abnormalities as discussed in the text. (Modified with permission from Stewart JD. Focal Peripheral Neuropathies. New York, NY: Elsevier; 1987.)
1. The entire nerve may be damaged just proximal to or within Guyon’s canal. This type of lesion affects the superficial sensory and deep motor branches of the distal ulnar nerve, resulting in sensory loss of the volar aspect of the fifth digit and usually the medial half of the fifth digit and weakness of all ulnar-innervated hand muscles. In contrast to more proximal ulnar lesions (e.g., ulnar neuropathy at the elbow), the dorsal ulnar cutaneous nerve is spared; thus, individuals have normal sensation of the dorsum of the ulnar aspect of the hand. In addition, there is normal strength of the flexor carpi ulnaris and flexor digitorum profundus III and IV. The ulnar SNAP may demonstrate prolonged distal latency or reduced amplitude depending on the degree of axon loss, while the dorsal ulnar cutaneous SNAP should be normal. The ulnar CMAP recorded from both the abductor digiti minimi and the first dorsal interosseous may show prolonged distal latencies or reduced amplitudes, again dependent on the nature and severity of the lesion. EMG may demonstrate evidence of denervation in the first dorsal interosseous and abductor digiti minimi but the flexor carpi ulnaris and flexor digitorum profundus III and IV should be normal.
2. The nerve may be compressed just outside Guyon’s canal such that only the superficial sensory branch is affected. In this case, sensation is decreased but all motor functions are spared. Electrodiagnostic studies would only show abnormalities of the ulnar SNAP.
3. The nerve may be damaged distal to the take off of the superficial sensory branch and affect only the deep motor branch. In such cases, sensation is spared, but motor function affecting any or all of the ulnar hand intrinsic muscles in the hand may be affected. The ulnar SNAPs would be normal, but ulnar CMAPs to both the first dorsal interosseous and the abductor digiti minimi should be abnormal, as would EMG of these muscles.
4. Finally, the nerve may be compressed distal to the branch innervating the hypothenar eminence. Thus, only the interossei and adductor pollicis muscles are affected. A lesion may be even more distal such that only the adductor pollicis or perhaps the first dorsal interosseous are abnormal. Ulnar SNAPs and ulnar CMAPs to the abductor digiti minimi would be normal. Only the ulnar CMAP from the dorsal interosseous would show abnormalities. Likewise, on EMG the abductor digiti minimi would be spared and denervation may be appreciated only in the dorsal interossei and adductor pollicis muscles.
Ulnar neuropathy in the hand due to external compression (e.g., bicyclist) may be treated conservatively. If caused by a fracture of the hamate or pisiform bones, exploratory surgery with decompression and neurolysis are often required. Also, ulnar neuropathies in the hand related to open trauma or internal compression (e.g., ganglion cyst) are usually managed with surgery.
CONCLUSIONThe multiple etiologies of focal neuropathies affecting the upper extremity (including radiculopathies, plexopathies, and mononeuropathies) can be quite daunting even to the experienced clinician. The key to the evaluation of patients with these types of focal neuropathies begins with localization. This is accomplished by a good physical examination and electrodiagnostic study. Other studies such as various modes of radiological imaging can assist in localizing and identifying an etiology. Prognosis is dependent on the cause and nature of the neuropathy. Neuropraxic lesions secondary to minor compression or stretch usually of the nerve tend to recover well; however, those associated with severe axonal degeneration take much longer to regenerate and recover. Treatment likewise is dependent on the nature of the nerve injury. Most focal neuropathies are initially managed conservatively, but severe nerve injuries may require surgical intervention.
1. Dumitru D, Zwarts MJ. Brachial plexopathies and proximal mononeuropathies. In: Dumitru D, Amato AA, Zwarts MJ, eds. Electrodiagnostic Medicine. 2nd ed. Philadelphia, PA: Hanley & Belfus; 2002:777–836.
2. Ferrante MA. Brachial plexopathies: Classification, causes, and consequences. Muscle Nerve. 2004;30(5):547–568.
3. Wilbourn AJ. Plexopathies. Neurol Clin. 2007;25(1):139–171.
4. Kerr AT. The brachial plexus of nerves in man, the variations in its formation and branches. Am J Anat. 1918;23:285–395.
5. Millesi H. Brachial plexus injuries—management and results. Clin Plast Surg. 1984;11:115–120.
6. Greenberg SA, Amato AA. EMG Pearls. Philadelphia, PA: Hanley & Belfus; 2004.
7. Haymaker W, Lindgreen M. Nerve disturbances following exposure to ionizing radiation. In: Vinken PJ, Bruyn GW, eds. Handbook of Clinical Neurology. Vol 7. Amsterdam: North Holland; 1970:338.
8. Dawson DM, Hallett M, Wilbourn AJ, Campbell WW. Entrapment Neuropathies. 3rd ed. Boston, MA: Little, Brown and Company; 1999.
9. Stewart JD. Focal Peripheral Neuropathies. New York, NY: Elsevier; 1987.
10. Sunderland S. Nerve and Nerve Injuries. Edinburgh: Churchill-Livingstone; 1978.
11. Ochoa J. Nerve fiber pathology in acute and chronic compression. In: Omer GE, Spinner M, eds. Management of Peripheral Nerve Problems. Philadelphia, PA: Saunders; 1980.
12. Aminoff MJ. Clinical electromyography. In: Aminoff MJ, ed. Electrodiagnosis in Clinical Neurology. 2nd ed. New York, NY: Churchill Livingstone; 1986:231–263.
13. Dumitru D, Amato AA, Swartz MJ. Electrodiagnostic Medicine, 2nd ed. Philadelphia, PA: Hanley & Belfus; 2002.
14. Kimura J. Electrodiagnosis of Diseases of Nerve and Muscle: Principles and Practice. 3rd ed. Philadelphia, PA: FA Davis; 2001.
15. Oh SJ. Clinical Electromyography. Nerve Conduction Studies. 3rd ed. Baltimore, MD: William & Wilkins; 2004.
16. Dumitru D, Amato AA, Swartz MJ. Nerve conduction studies. In: Dumitru D, Amato AA, Swartz MJ, eds. Electrodiagnostic Medicine. 2nd ed. Philadelphia, PA: Hanley & Belfus; 2002:159–223.
17. Schimseider RJ, de Visser BW, Kemp B. The flexor carpi radialis H reflex in lesions of the sixth and seventh cervical roots. J Neurol Neurosurg Psychiatry. 1985;48:445–449.
18. Eisen A, Hoirch M. Electrodiagnostic evaluation of radiculopathies and plexopathies using somatosensory evoked potentials. Electroencephalogr Clin Neurophysiol. 1982;36(suppl):349–357.
19. Jones SJ, Parry W, Landi A. Diagnosis of brachial plexus traction lesions by sensory nerve action potentials and somatosensory evoked potentials. Injury. 1981;12:376–382.
20. Jones SJ. Diagnostic value of peripheral and spinal somatosensory evoked potentials in traction lesions of the brachial plexus. Clin Plast Surg. 1984;11:167–172.
21. Lederman RJ, Wilbourn AJ. Brachial plexopathy: Recurrent cancer or radiation? Neurology. 1984;34:1331–1335.
22. Rapoport S, Blair DN, McCarthy SM, Desser TS, Hammers LW, Sostman HD. Brachial plexus: Correlation of MR imaging with CT and pathologic findings. Radiology. 1988;167:161–165.
23. van Es HW, Bollen TL, van Heesewijk HP. MRI of the brachial plexus: A pictorial review. Eur J Radiol. 2010;74:391–402.
24. Daily AT, Tsuruda JS, Filler AG, Maravilla KR, Goodkin R, Kliot M. Magnetic resonance neurography of peripheral nerve regeneration. Lancet. 1997;350:1221–1222.
25. Filler AG, Maravilla KR, Tsuruda JS. MR neurography and muscle MR for image diagnosis of disorders affecting the peripheral nerves and musculature. Neurol Clin. 2004;22:643–682.
26. Grant GA, Britz GW, Gookdkin R, Jarvik JG, Maravilla K, Kliot M. The utility of magnetic resonance imaging in evaluating peripheral nerve disorders. Muscle Nerve. 2002;25:314–331.
27. Gupta RK, Mehta VS, Banerji AK, Jain RK. MR evaluation of brachial plexus lesions. Neuroradiology. 1989;31:377–381.
28. Sherrier RH, Sostman JD. Magnetic resonance imaging of the brachial plexus. J Thoracic Imaging. 1993;8:27–33.
29. Wittenberg KH, Adkins MC. MR imaging of non-traumatic brachial plexopathies: Frequency and spectrum of findings. Radiographics. 2000;20:1023–1032.
30. Altinok T, Baysal O, Karakas HM, et al. Ultrasonographic assessment of mild and moderate idiopathic carpal tunnel syndrome. Clin Radiol. 2004;59(10):916–925.
31. Bayrak IK, Bayrak AO, Tilki HE, Nural MS, Sunter T. Ultrasonography in carpal tunnel syndrome: Comparison with electrophysiological stage and motor unit number estimate. Muscle Nerve. 2007;35:344–348.
32. Beekman R, Wokke JH, Schoemaker MC, Lee ML, Visser LH. Ulnar neuropathy at the elbow: Follow-up and prognostic factors determining outcome. Neurology. 2004;63(9):1675–1680.
33. Beekman R, Schoemaker MC, van der Plas JP, et al. The diagnostic value of high-resolution sonography in ulnar neuropathy at the elbow. Neurology. 2004;62:767–773.
34. Cartwright MS, Chloros GD, Walker FO, Wiesler ER, William W, Campbell WW. Diagnostic ultrasound for nerve transaction. Muscle Nerve. 2007;35(6):796–799.
35. Jacob A, Moorthy TK, Thomas SV, Sarada C. Compression of the deep motor branch of the ulnar nerve: An unusual cause of pure motor neuropathy and hand wasting. Arch Neurol. 2005;62(5):826–827.
36. Ziswiler HR, Reichenbach S, Vogelin E, Bachmann LM, Villiger PM, Juni P. Diagnostic value of sonography in patients with suspected carpal tunnel syndrome: A prospective study. Arthritis Rheum. 2005;52(1):304–311.
37. Kelsey JL, Githens PB, Walter SD, et al. An epidemiological study of acute prolapsed cervical intervertebral disc. J Bone Joint Surg. 1984;66A:907–914.
38. Lunsford LD, Bissonette DJ, Janetta PJ, Sheptak PE, Zorub DS. Anterior surgery for cervical disc disease. J Neurosurg. 1980;53:1–11.
39. Martins AN. Anterior cervical discectomy with and without interbody bone graft. J Neurosurg. 1976;44:290–295.
40. Negrin P, Lelli S, Fardin P. Contribution of electromyography to the diagnosis, treatment, and prognosis of cervical disc disease: A study of 114 patients. Electromyogr Clin Neurophysiol. 1991;31:173–179.
41. Scoville WB, Dohrmann GJ, Corkill G. Late results of cervical disc surgery. J Neurosurg. 1976;45:203–210.
42. Abbott KH, Retter RH. Protrusions of thoracic intervertebral disks. Neurology. 1956;6:1–10.
43. Arce CA, Dohrmann GJ. Herniated thoracic discs. Neurol Clin. 1985;3:383–392.
44. Arseni C, Nash F. Thoracic intervertebral disc protrusion: A clinical study. J Neurosurg. 1960;17:418–430.
45. Otani K, Yoshida M, Fujii E, Nakai S, Shibasaki K. Thoracic disc herniation: Surgical treatment. Spine. 1988;13:1262–1267.
46. Benson MK, Byrnes DP. The clinical syndromes and surgical treatment of thoracic intervertebral disc prolapse. J Bone Joint Surg. 1975;57B:471–477.
47. Bohlman HH, Zdeblick TA. Anterior excision of herniated thoracic discs. J Bone Joint Surg. 1988;70A:1038–1047.
48. Radhakrishnan K, Litchy WJ, O’Fallon WM, Kurland LT. Epidemiology of cervical radiculopathy: A population-based study from Rochester, Minnesota, 1976 through 1990. Brain. 1994;117:325–335.
49. Persson LC, Carlson CA, Carlson JY. Long-lasting cervical radiculopathy pain managed with surgery, physiotherapy, or a cervical collar. Spine. 1997;22:751–758.
50. Nikolaidis I, Fouyas IP, Sandercock PA, Statham PF. Surgery for cervical radiculopathy or myelopathy. Cochrane Database Syst Rev. 2010;1:CD001466.
51. England JD, Sumner AJ. Neuralgic amyotrophy: An increasingly diverse entity. Muscle Nerve. 1987;10:60–68.
52. Parsonage MJ, Turner AJ. Neuralgic amyotrophy. The shoulder-girdle syndrome. Lancet. 1948;1:973–978.
53. Tsairis P, Dyck PJ, Mulder DW. Natural history of brachial plexus neuropathy: Report on 99 cases. Arch Neurol. 1972; 27:109–117.
54. Turner AJ, Parsonage MJ. Neuralgic amyotrophy (paralytic brachial neuritis): With special reference to prognosis. Lancet. 1957;1:209–212.
55. van Alfen N, van Engelen BG. The clinical spectrum of neuralgic amyotrophy in 246 cases. Brain. 2006;129:438–450.
56. Cwik VA, Wilbourn AJ, Rorick M. Acute brachial neuropathy: Detailed EMG findings in a large series. Muscle Nerve. 1990;13:859.
57. Flaggman PD, Kelly JJ. Brachial plexus neuropathy: An electrophysiologic evaluation. Arch Neurol. 1980;37:160–164.
58. Weikers NJ, Mattson RH. Acute paralytic brachial neuritis: A clinical and electrodiagnostic study. Neurology. 1969;19:1153–1158.
59. Cape CA, Fincham RW. Paralytic brachial neuritis with diaphragmatic paralysis. Neurology. 1965;15:191–193.
60. Walsh NE, Dumitru D, Kalantri A, Roman A. Brachial neuritis involving the bilateral phrenic nerves. Arch Phys Med Rehabil. 1987;68:46–48.
61. Kiloh L, Nevin S. Isolated neuritis of the anterior interosseous nerve. Br Med J. 1952;1:850–851.
62. Renneis GD, Ochoa J. Neuralgic amyotrophy manifesting as anterior interosseous nerve palsy. Muscle Nerve. 1980;3:160–164.
63. Smith BE, Herbst B. Anterior interosseous nerve palsy. Arch Neurol. 1974;30:330–331.
64. Pierre PA, Laterre CE, van den Bergh PY. Neuralgic amyotrophy with involvement of cranial nerves IX, X, XI and XII. Muscle Nerve. 1990;13:704–707.
65. Eisen A, Bertrand G. Isolated accessory nerve palsy of spontaneous origin. A clinical and electromyographic study. Arch Neurol. 1972;27:496–502.
66. Bernsen PL, Wong Chung RE, Vinergoets HM, Janssen JT. Bilateral neuralgic amyotrophy induced by interferon treatment. Arch Neurol. 1988;45(4):449–451.
67. Cruz-Martinez A, Barrio JM, Arpa J. Neuralgic amyotrophy: Variable expression in 40 patients. J Peripher Nerv Sys. 2002;7:198–204.
68. Kiwit JC. Neuralgic amyotrophy after administration of tetanus toxoid. J Neurol Neurosurg Psychiatry. 1984;47:320.
69. Loh FL, Herskovitz S, Berger AR, Swerdlow ML. Brachial plexopathy associated with interleukin-2 therapy. Neurology. 1992;42:462–463.
70. Weintraub MI, Chia DT. Paralytic brachial neuritis after swine flu vaccination. Arch Neurol. 1977;34:518.
71. Vriesendorp FJ, Dmytrnko GS, Dietrich T, Koski CL. Anti-peripheral nerve myelin antibodies and terminal activation products of complement in serum of patients with acute brachial plexus neuropathy. Arch Neurol. 1993;50:1993.
72. Amato AA, Jackson CE, Kim JY, Worley KL. Chronic relapsing brachial plexus neuropathy with persistent conduction block. Muscle Nerve. 1997;20:1303–1307.
73. Adler JB, Patterson RL. Erb’s palsy: Long term results of treatment in eighty-eight cases. J Bone Joint Surg. 1967;49A:1052–1064.
74. Eng GD. Brachial plexus palsy in newborn infants. Pediatrics. 1971;48:18–28.
75. Eng GB, Koch B, Smokvina MD. Brachial plexus palsy in neonates and children. Arch Phys Med Rehabil. 1978;59:458–464.
76. Erb W. Ueber eine eigenthumliche localisation von lahmengen im plexus brachialis. Verhandl d Naturhist-Med Heidelberg. 1874;2:130–137.
77. Greenwald AG, Shute PC, Shiveley JL. Brachial plexus birth palsy: A 10 years report on the incidence and prognosis. J Pediatr Orthop. 1984;4:689–692.
78. Klumpke A. Contribution a l’etude des paralysies radiculaires du plexus brachial. Paralysies radiculaires totales. Paralysies radiculaires inferieures. De la participation des filest sympathiques oculo-pupillaires dans ces paralysies. Rev Med. 1885;5:591.
79. Specht EE. Brachial plexus palsy in newborn: Incidence and prognosis. Clin Orthop. 1975;110:32–34.
80. Johnson EW, Alexander MA, Koenig WC. Infantile Erb’s palsy (Smellie’s palsy). Arch Phys Med Rehabil. 1977;58:175–178.
81. McFarland LV, Raskin M, Daling JR, Benedetti TJ. Erb/Duchenne’s palsy: A consequence of fetal macrosomia and method of delivery. Obstet Gynecol. 1986;68:784–788.
82. Meyer RD. Treatment of adult and obstetrical brachial plexus injuries. Orthopedics. 1986;9:899–903.
83. Mollberg M, Wennergren M, Bager B, Ladfors L, Hagberg H. Obstetric brachial plexus palsy: A prospective study on risk factors related to manual assistance during the second stage of labor. Acta Obstet Gynecol Scand. 2007;86(2):198–204.
84. Boome RS, Kaye JC. Obstetric traction injuries of the brachial plexus. J Bone Joint Surg. 1988;70B:571–576.
85. Rossi LN, Vassella F, Mumenthaler M. Obstetrical lesions of the brachial plexus: Natural history in 34 personal cases. Eur Neurol. 1982;21:1–7.
86. Kneeland JB, Kellman GM, Middleton WD, et al. Diagnosis of disease of the supraclavicular region by use of MR imaging. Am J Roentgenol. 1987;148:1149–1151.
87. Popovich MJ, Taylor FC, Helmer E. MR imaging of birth-related brachial plexus avulsion. AJNR AM J Neuroradiol. 1989;10(suppl 5):S98.
88. Kline DG. Surgical repair of peripheral nerve injury. Muscle Nerve. 1990;13:843–852.
89. Kline DG, Hudson AR. Nerve Injuries. Philadelphia, PA: WB Saunders; 1995:611.
90. Spinner RJ, Kline DG. Surgery for peripheral nerve and brachial plexus injuries or other nerve lesions. Muscle Nerve. 2000;23:680–695.
91. Aminoff MJ, Olney RK, Parry GJ, Raskin NH. Relative utility of different electrophysiologic techniques in the evaluation of brachial plexopathies. Neurology. 1988;38:546–550.
92. Cuetter AC, Bartoszek DM. The thoracic outlet syndrome: Controversies, over diagnosis, over treatment, and recommendations for management. Muscle Nerve. 1989;12:410–419.
93. Gilliat RW, Le Quesne PM, Logue V, Sumner AJ. Wasting of the hand associated with a cervical rib or band. J Neurol Neurosurg Psychiatry. 1970;33:615–624.
94. Gilliatt RW, Willison RG, Dietz V, Williams IR. Peripheral nerve conduction in patients with a cervical rib and band. Ann Neurol. 1978;4:124–129.
95. Jaeckle KA. Nerve plexus metastases. Neurol Clin. 1991;9:857–866.
96. Kori SH, Foley KM, Posner JB. Brachial plexus lesions in patients with cancer: 100 cases. Neurology. 1981;31:45–50.
97. Lusk MD, Kline DG, Garcia CA. Tumors of the brachial plexus. Neurosurgery. 1987;21:439–453.
98. Richardson RR, Siqueira EB, Oi S, Nunez C. Neurogenic tumors of the brachial plexus: Report of two cases. Neurosurgery. 1979;4:66–70.
99. Sell PJ, Semple JC. Primary nerve tumours of the brachial plexus. Br J Surg. 1987;74:73–74.
100. Godwin JT. Encapsulated neurilemoma (schwannoma) of the brachial plexus: Report of 11 cases. Cancer. 1952;5:708–720.
101. Ly-Pen D, Andreu JL, de Blas G, Sanchez-Olaso A, Millan I. Surgical decompression versus local steroid injection in carpal tunnel syndrome: A one-year, prospective, randomized, open, controlled clinical trial. Arthritis Rheum. 2005;52:612–619.
102. Harper CM, Thomas JE, Cascino TL, Litchy WJ. Distinction between neoplastic and radiation-induced brachial plexopathy, with emphasis on the role of EMG. Neurology. 1989;39:502–506.
103. Thomas JE, Cascino TL, Earle JD. Differential diagnosis between radiation and tumor plexopathy of the pelvis. Neurology. 1985;35:1–7.
104. Corkill G, Lieberman JS, Taylor RG. Pack-palsy in backpackers. West J Med. 1980;132:569–572.
105. Daube JR. Rucksack paralysis. J Am Med Assoc. 1969;208:2447–2452.
106. Graham JF, Pye IF, McQueen IN. Brachial plexus injury after median sternotomy. J Neurol Neurosurg Psychiatry. 1981; 44:621–625.
107. Hanson MR, Breuer AC, Furland AJ, et al. Mechanism and frequency of brachial plexus injury in open-heart surgery: A prospective study. Ann Thorac Surg. 1983;36:675–679.
108. Morin JE, Long R, Elleker MG, Eisen AA, Wynands E, Ralphs-Thibodeau S. Upper extremity neuropathies following median sternotomy. Ann Thorac Surg. 1982;34:181–185.
109. Seyfer AE, Grammer NY, Bogumill GP, Provost JM, Chandry U. Upper extremity neuropathies after cardiac surgery. J Hand Surg. 1985;10:16–19.
110. Chance PF, Lensch MW, Lipe H, Brown RH Sr, Brown RH Jr, Bird TD. Hereditary neuralgic amyotrophy and hereditary neuropathy with liability to pressure palsies: Two distinct genetic disorders. Neurology. 1994;44:2253–2257.
111. Hentz VR. Is microsurgical treatment of brachial plexus palsy better than conventional treatment? Hand Clin. 2007;23(1):83–89.
112. Hentz VR. Brachial plexus injuries. In: Omer GE Jr, Spinner M, Van Beek AL, eds. Management of Peripheral Nerve Problems. 2nd ed. Philadelphia, PA: WB Saunders; 1998:445–453.
113. Shin AY, Spinner RJ, Steinmann SP, Bishop AT. Adult traumatic brachial plexus injuries. J Am Acad Orthop Surg. 2005;13(6): 382–396.
114. Terzis JK, Kostas I, Soucacos PN. Restoration of shoulder function with nerve transfers in traumatic brachial plexus palsy patients. Microsurgery. 2006;26(4):316–324.
115. Terzis JK, Kostopoulos VK. The surgical treatment of brachial plexus injuries in adults. Plast Reconstr Surg. 2007;119(4):73e–92e.
116. Levy O, Relwani JG, Mullett H, Haddo O, Even T. The active elevation lag sign and the triangle sign: New clinical signs of trapezius palsy. J Shoulder Elbow Surg. 2009;18:573–576.
117. Chan PK, Hems TE. Clinical signs of accessory nerve palsy. J Trauma. 2005;60:1142–1144.
118. Goodman CE, Kenrick MM, Blum MV. Long thoracic nerve palsy: A follow-up study. Arch Phys Med Rehabil. 1976;56:352–355.
119. Johnson JT, Kendall HO. Isolated paralysis of the serratus anterior muscle. J Bone Joint Surg. 1955;37A:567–574.
120. Kaplan PE. Electrodiagnostic confirmation of long thoracic nerve palsy. J Neurol Neurosurg Psychiatry. 1980;43:50–52.
121. Haymaker W, Woodhall B. Peripheral Nerve Injuries: Principles of Diagnosis. Philadelphia, PA: WB Saunders; 1953.
122. Petrera JE, Trojaborg W. Conduction studies of the long thoracic nerve in serratus anterior palsy of different etiology. Neurology. 1984;34:1033–1037.
123. Warner JJ, Navarro RA. Serratus anterior dysfunction: Recognition and treatment. Clin Orthop. 1998;349:139.
124. Wiater JM, Flatow EL. Long thoracic nerve injury. Clin Orthop. 1999;368:17.
125. Clein LJ. Suprascapular entrapment neuropathy. J Neurosurg. 1975;43:337–342.
126. Edeland HG, Zachrisson BE. Fracture of the scapular notch associated with lesion of the suprascapular nerve. Acta Orthop Scand. 1975;46:758–763.
127. Swafford AR, Lichtman DH. Suprascapular nerve entrapment: A case report. J Hand Surg. 1982;7:57–60.
128. Toon TN, Bravois M, Guillen M. Suprascapular nerve injury following trauma to the shoulder. J Trauma. 1981;21:652–655.
129. Zoltan JD. Injury to the suprascapular nerve associated with anterior dislocation of the shoulder: Case report and review of the literature. J Trauma. 1979;19:203–206.
130. Aiello I, Serra G, Traina GC, Tugnoli V. Entrapment of the suprascapular nerve of the spinoglenoid notch. Ann Neurol. 1982;12:314–316.
131. Carlson JA, Baruah JK. Suprascapular nerve entrapment at spinoglenoid notch. Neurology. 1985;35:78.
132. Antoniadis G, Richeter HP, Rath S. Suprascapular nerve entrapment: Experience with 28 cases. J Neurosurg. 1996;85:1020.
133. Antoniou J, Tae SK, Williams GR, Bird S, Ramsey ML, Iannotti JP. Suprascapular neuropathy: Variability in the diagnosis, treatment, and outcome. Clin Orthop Relat Res. 2001;386:131–138.
134. Braddom RL, Wolfe C. Musculocutaneous nerve injury after heavy exercise. Arch Phys Med Rehabil. 1978;59:290–293.
135. Kim SM, Goodrich JA. Isolated proximal musculocutaneous nerve palsy: Case report. Arch Phys Med Rehabil. 1984;65:735–736.
136. Liveson JA. Nerve lesions associated with shoulder dislocation: An electrodiagnostic study of 11 cases. J Neurol Neurosurg Psychiatry. 1984;47:742–744.
137. Aita JF. An unusual compressive neuropathy. Arch Neurol. 1984;41:341.
138. Berry H, Bril V. Axillary nerve palsy following blunt trauma to the shoulder region: A clinical and electrophysiological review. J Neurol Neurosurg Psychiatry. 1982;45:1027–1032.
139. Kirby JF, Kraft GH. Entrapment neuropathy of anterior branch of axillary nerve: Report of case. Arch Phys Med Rehabil. 1972;53:338–340.
140. Steinnman SP, Moran EA. Axillary nerve injury: Diagnosis and treatment. J Am Acad Orthop Surg. 2003;9:328.
141. Kameda Y. An anomalous muscle (accessory scapularis-teres-latissimus muscle) in the axilla penetrating the brachial plexus in man. Acta Anat. 1976;96:513–533.
142. Lotem N, Fried A, Levy M, Solzi P, Najenson T, Nathan H. Radial palsy following muscular effort. A nerve compression syndrome possibly related to a fibrous arch of the lateral head of the triceps. J Bone Joint Surg. 1971;53B:500–506.
143. Kim DH, Murovic JA, Kim YY, Kline DG. Surgical treatment and outcomes in 45 cases of posterior interosseous nerve entrapments and injuries. J Neurosurg. 2006;104(5):766–777.
144. Boswick JA, Stromberg WB. Isolated injury to the median nerve above the elbow. J Bone Joint Surg. 1967;49A:653–658.
145. Staal A, Van voorthuisen AE, Van Dijk LM. Neurological complications following arterial catheterization by the axillary approach. Br J Radiol. 1966;39:115–116.
146. Bolton CF, Driedger AA, Lindsay RM. Ischaemic neuropathy in uraemic patients caused by bovine arteriovenous shunt. J Neurol Neurosurg Psychiatry. 1979;42:810–814.
147. Al-Naib I. Humeral supracondylar spur and Struther’s ligament: A rare cause of neurovascular entrapment in the upper limb. Int Orthop. 1994;18:393.
148. Aydinlioglu A, Cirik B, Akpina F, Tosun N, Dogan A. Bilateral median nerve compression at the level of Struthers ligament. J Neurosurg. 2000;92:693.
149. Bilbe T, Yalaman O, Bilge S, Cokneşeli B, Barut S. Entrapment of the median nerve at the level of the ligament of Struthers. Neurosurgery. 1990;27:787–789.
150. Schrader PA, Reina CR. Struther’s ligament neuropathy in a juvenile. Orthopedics. 1994;17:723.
151. Spinner RJ, Carmichael SW, Spinner M. Partial median nerve entrapment in the distal arm because of an accessory bicipital aponeurosis. J Hand Surg Am. 1991;16:236.
152. Schantz K, Riegels-Nielsen P. The anterior interosseous nerve syndrome. J Hand Surg Br. 1992;17:510.
153. Pyrse-Phillips WE. Validation of a diagnostic sign in carpal tunnel syndrome. N Engl J Med. 1993;329:2013–2018.
154. Fowler JR, Gaughan JP, Ilyas AM. The sensitivity and specificity of ultrasound for the diagnosis of carpal tunnel syndrome: A meta-analysis. Clin Orthop Relat Res. 2011;469(4):1089–1094.
155. Bland JDP. Treatment of carpal tunnel syndrome. Muscle Nerve. 2007;36:167–171.
156. Marshall S, Tardif G, Ashworth N. Local corticosteroid injection for carpal tunnel syndrome. Cochrane Database Syst Rev. 2007;18(2):CD001554.
157. O’Connor D, Marshall S, Massy-Westropp N. Non-surgical treatment (other than steroid injection) for carpal tunnel syndrome. Cochrane Database Syst Rev. 2003;(1):CD003219.
158. Scholten RJ, Gerritsen AA, Uitdehaag BM, van Geldere D, de Vet HC, Bouter LM. Surgical treatment options for carpal tunnel syndrome. Cochrane Database Syst Rev. 2004;(4):CD003905.
159. Padua L, Padua R, Aprile I, Pasqualetti P, Tonali P. Multi-perspective follow-up of untreated carpal tunnel syndrome: A multicenter study. Neurology. 2001;56:1459–1466.
160. Verdugo RJ, Salinas RS, Castilo J, Cea JG. Surgical versus non-surgical treatment for carpal tunnel syndrome. Cochrane Database Syst Rev. 2003;3:CD0011552.
161. Jarvik JG, Comstock BA, Kliot M, et al. Surgery versus non-surgical therapy for carpal tunnel syndrome: A randomised parallel-group trial. Lancet. 2009;374:1074–1081.
162. Bland JD. Do nerve conductions predict the outcome of carpal tunnel decompression. Muscle Nerve. 2001;24:935–940.
163. American Association of Electrodiagnostic Medicine, American Academy of Physical Medicine and Rehabilitation, American Academy of Neurology. Practice parameter for electrodiagnostic studies in ulnar neuropathy at the elbow: Summary statement. Arch Phys Med Rehabil. 1999;80:357–359.
164. Kincaid JC, Phillips LH, Daube JR. The evaluation of suspected ulnar neuropathy at the elbow. Arch Neurol. 1986;43:44–47.
165. Visser LH, Beekman R, Franssen H. Short-segment nerve conduction studies in ulnar neuropathy at the elbow. Muscle Nerve. 2005;31(3):331–338.
166. Raynor EM, Shefner JM, Preston DC, Logigian EL. Sensory and mixed nerve conduction studies in the evaluation of ulnar neuropathy at the elbow. Muscle Nerve. 1994;17:785–792.
167. Mowlawi A, Andrews K, Lille S, Verhulst S, Zook EG, Milner S. The management of cubital tunnel syndrome: A meta-analysis of clinical studies. Plast Reconstr Surg. 2000;106:327–334.
168. Dellon AL, Hament W, Gittelshorn A. Nonoperative management of cubital tunnel syndrome: An 8-year prospective study. Neurology. 1993;43:1673–1677.
169. Macadam SA, Gandhi R, Bezuhly M, Lefaivre KA. Simple decompression versus anterior subcutaneous and submuscular transposition of the ulnar nerve for cubital tunnel syndrome: A meta-analysis. J Hand Surg Am. 2008;33:1314.e1–e12.
