Electromyographers occasionally are called on to evaluate the proximal nerves in the shoulder and arm. Isolated lesions of these nerves, including the suprascapular, axillary, musculocutaneous, long thoracic, and spinal accessory, are far less frequent than the common entrapment and compressive neuropathies of the median, ulnar, and radial nerves. The electrophysiologic evaluation of proximal neuropathies in the shoulder and arm relies principally on needle electromyography (EMG). Nerve conduction studies of these nerves are limited and are complicated by technical factors. In addition, nearly all lesions in these nerves are axonal loss, and cannot be localized by focal slowing or conduction block. Similar to other mononeuropathies, the goals of the electrophysiologic study are to localize the lesion as accurately as possible, to exclude a more widespread lesion or proximal radiculopathy, and to assess the underlying severity.
The suprascapular nerve comes off the upper trunk of the brachial plexus, receiving innervation from both the C5 and C6 roots. The nerve runs posteriorly under the trapezius, passing through the suprascapular notch of the scapula to enter the supraspinous fossa ( Figure 31–1 ). The suprascapular notch is U shaped, located along the superior border of the scapula, and covered by the transverse scapular ligament . The suprascapular nerve first supplies motor fibers to the supraspinatus muscle, a shoulder abductor, before proceeding laterally to supply deep sensory fibers to the glenoacromial and acromioclavicular joints, and the coracoacromial ligament. It then wraps around the spinoglenoid notch of the scapular spine under the spinoglenoid ligament to enter the infraspinous fossa, where it supplies motor fibers to the infraspinatus muscle, an external rotator of the shoulder. The suprascapular nerve usually carries no cutaneous sensory fibers, although rare anomalous innervations have been reported. In these rare cases, the suprascapular nerve carries cutaneous sensation to the proximal lateral arm, the area usually supplied by the axillary nerve.
Suprascapular entrapment most commonly occurs at the suprascapular notch, under the transverse scapular ligament. Less frequently, the nerve can also be entrapped distally at the spinoglenoid notch. The suprascapular nerve is relatively immobile both at its origin at the upper trunk and at the suprascapular notch. Because both the shoulder and scapula are quite mobile, movement, especially repetitive movement, results in stretch and nerve injury ( Figure 31–2 ). Also, like most of the major proximal upper extremity nerves, the suprascapular nerve is often prominently involved in neuralgic amyotrophy (see Chapter 30 ).
Rare cases of suprascapular nerve entrapment have been reported secondary to a variety of mass lesions, including ganglion cysts, sarcomas, and metastatic carcinomas. Ganglion cysts are especially common at the spinoglenoid notch. In addition, certain activities, positions, and professions are associated with suprascapular entrapment. For example, weight lifting has been implicated in several reports as a provocative factor in suprascapular entrapment, likely as a consequence of repetitive movement of the scapula, especially during lifts that involve shoulder abduction and protraction. Suprascapular neuropathy has also been reported as a consequence of positioning during surgical procedures, when patients are placed in a knee-chest position with the scapula protracted. Of interest, several professions put patients at risk for suprascapular entrapment. These include professional volleyball players, baseball pitchers, and dancers. In these professions, the clinical and electrophysiologic findings most often suggest a distal lesion at the spinoglenoid notch.
In addition, suprascapular neuropathy, which is sometimes confused clinically with a rotator cuff injury, may also accompany a rotator cuff injury. One might initially assume that both have a common traumatic etiology. However, a suprascapular neuropathy may actually occur as a result of a rotator cuff tear , usually a large and full thickness tear. Following a rotator cuff tear, there may be medial retraction of the tendons to the supraspinatus and infraspinatus muscles. This may result in increased tension on the suprascapular nerve both at the suprascapular notch and the spinoglenoid notch ( Figure 31–3 ).
Symptoms and signs depend on the site of nerve entrapment. At the most common site of entrapment, the suprascapular notch, shoulder pain may be prominent. Indeed, there is anatomic and clinical evidence that the suprascapular nerve supplies the majority of deep sensory fibers (including pain fibers) to the shoulder joint. The pain typically is described as deep and boring, occurring along the superior aspect of the scapula and radiating to the shoulder, but usually not more distally. The pain may be exacerbated by shoulder movements, especially adduction of the extended arm. This movement results in protraction of the scapula, which increases the nerve tethering between the upper trunk and the suprascapular notch. Occasionally, the suprascapular notch may be tender to palpation. Weakness involves shoulder abduction (supraspinatus) and external rotation (infraspinatus). Impairment of these motions may or may not be noticed by the patient, because both functions are subserved by other muscles as well. Atrophy may be recognized, especially over the infraspinatus muscle, which is only partially covered by the trapezius muscle ( Figure 31–4 ).
If the entrapment occurs more distally at the spinoglenoid notch, the syndrome is limited to atrophy and weakness of the infraspinatus muscle. Pain usually is absent because the deep sensory fibers to the shoulder joint have exited more proximally.
Several conditions may be confused with suprascapular neuropathy, including cervical radiculopathy, rotator cuff injury and other orthopedic conditions, and neuralgic amyotrophy. In contrast to suprascapular neuropathy, a C5–C6 radiculopathy may have radiating pain from the neck into the shoulder and arm, associated with sensory abnormalities in the lateral arm, forearm, and thumb. Often, the biceps and brachioradialis tendon reflexes are depressed or absent. Higher cervical radiculopathies (e.g., C3 or C4) may have a similar pain distribution to suprascapular neuropathy but are not associated with significant weakness of the shoulder or arm.
Local orthopedic conditions may be difficult to differentiate clinically from suprascapular neuropathy. Although weakness should not be present, pain often prevents full muscle activation. Exacerbation of pain by palpation (other than at the suprascapular notch) or by passive shoulder movement (other than protraction of the shoulder) would be unusual for suprascapular entrapment.
Lastly, neuralgic amyotrophy often presents with severe proximal arm and shoulder pain and later weakness (see Chapter 30 ). In some cases, the suprascapular nerve may be primarily involved. However, close clinical and electrophysiologic evaluation usually reveals evidence of more widespread involvement of other nerves.
The goal of electrodiagnosis is to demonstrate abnormalities of the suprascapular-innervated muscles and exclude cervical radiculopathy, brachial plexopathy, or involvement of other proximal nerves. Because the suprascapular nerve has no cutaneous distribution, there is no corresponding sensory nerve to be recorded. However, as the suprascapular nerve originates from the upper trunk of the brachial plexus, studies of the sensory nerves that pass through the upper trunk should be performed to help exclude a more widespread plexus lesion. These studies should include the lateral antebrachial cutaneous nerve and the median and radial sensory nerves, especially when recording from the thumb. Often, comparison with the contralateral asymptomatic side can be useful in identifying a mild abnormality, even if the studies are normal on the symptomatic side. Any abnormality present in these sensory studies suggests a more widespread brachial plexopathy. Of course, an abnormality found in the median sensory nerve may indicate a superimposed median neuropathy at the wrist, which may need to be studied further.
Motor conduction studies can be performed, stimulating Erb’s point and recording with a monopolar needle electrode in either the supraspinatus or infraspinatus muscle, or both, simultaneously ( Table 31–1 ). A surface recording electrode should not be used to record from the spinati muscles, especially the supraspinatus, because they are covered by the trapezius. A surface reference electrode is placed distally over the shoulder joint. Compound muscle action potential (CMAP) amplitude and latency are measured. Comparing amplitude side to side can give an estimate of the amount of axonal loss present. However, these studies generally do not increase the yield over conventional EMG in terms of localizing the lesion. Typically, the pathophysiology of these entrapment neuropathies is axonal loss. Thus, although motor nerve conduction studies may show reduced amplitudes and slightly prolonged latencies, there really is no information gained over needle EMG, which more easily demonstrates axonal loss. When Erb’s point stimulation is performed, high stimulating currents often are required, and supramaximal stimulation can be difficult to ensure.
|Nerve||Muscle||Latency (ms) Upper Limit Normal||Distances (cm) *|
† The axillary and musculocutaneous nerves can also be stimulated in the axilla, with typical distal motor latencies ≤3.3 ms. Both axillary and Erb’s point stimulation often are technically difficult. In patients with symptoms limited to one side, comparing latencies and amplitudes from side to side is always preferable to using normal value tables.
During needle EMG, both the supraspinatus and infraspinatus muscles should be sampled. Care must be taken to ensure that the EMG needle is not in the more superficial trapezius muscle, by checking that no motor unit action potentials (MUAPs) are activated with a shoulder shrug. In lesions at the suprascapular notch, both the supraspinatus and infraspinatus are abnormal. With spinoglenoid lesions, however, only the infraspinatus is involved. If either of these muscles is abnormal, it is essential to sample other C5–C6 innervated muscles (e.g., deltoid, biceps, brachioradialis), as well as the cervical paraspinal muscles, to exclude a cervical radiculopathy or more widespread brachial plexus lesion.
Along with the radial nerve, the axillary nerve originates from the posterior cord of the brachial plexus ( Figure 31–5 ). The axillary nerve is composed primarily of C5–C6 fibers, running through the upper trunk and posterior cord of the plexus. The nerve leaves the axilla through the quadrilateral space , which is formed by the humerus and the teres minor, teres major and long head of the triceps muscles ( Figure 31–6 ). Posteriorly in the quadrilateral space, it often divides into two major trunks. The posterior trunk always supplies the teres minor before terminating as the superior lateral brachial cutaneous nerve (i.e., axillary sensory nerve). The teres minor aids in external rotation of the shoulder while the deltoid is principally a shoulder abductor. The axillary sensory nerve supplies an oval-shaped area over the lateral shoulder. The anterior trunk travels deep to the fascia of the deltoid and always supplies the middle and anterior heads of the deltoid as well as a deep sensory branch to the shoulder joint. The posterior head of the deltoid is most commonly supplied by the posterior trunk, but some variations exist wherein it is supplied by the anterior trunk alone, and in others by a combination of the anterior and posterior trunks.
Axillary neuropathies typically result from trauma, especially dislocation of the shoulder and fracture of the humerus. Less commonly, athletes participating in contact sports have developed axillary neuropathies as a result of injury, typically a direct blow to the anterolateral deltoid area. Similar to suprascapular neuropathy, axillary neuropathies have been reported in professional volleyball players. Rare cases of entrapment in the quadrilateral space have been reported but are exceptional. Quadrilateral space syndrome results from compression of the axillary nerve and posterior humeral circumflex artery.
Patients with axillary neuropathies have a well-defined circular area of numbness over the lateral shoulder, along with partial weakness of shoulder abduction and external rotation ( Figure 31–7 ). The degree of weakness varies from patient to patient. The weakness is only partial, because other muscles also contribute to shoulder abduction (i.e., the supraspinatus) and external rotation (i.e., the infraspinatus).
The major goal of electrodiagnosis is to demonstrate abnormalities of axillary-innervated muscles and rule out cervical radiculopathy, brachial plexopathy, or involvement of other proximal nerves. Unfortunately, there is no routine sensory nerve conduction study for the axillary nerve. However, because the axillary nerve originates from the posterior cord and upper trunk, sensory nerves that run through the posterior cord or upper trunk of the brachial plexus should be studied. These include the radial and lateral antebrachial cutaneous sensory nerves and the median sensory nerve, especially when recording the thumb. To detect mild abnormalities, comparison with the contralateral asymptomatic nerve is suggested, even if the studies are normal on the symptomatic side. Abnormalities of any of these sensory studies suggest a more widespread brachial plexopathy.
Axillary motor nerve conduction studies can be performed, stimulating the axilla and Erb’s point and recording with a monopolar needle or surface electrode over the deltoid ( Table 31–1 ). A surface reference electrode is placed distally over the deltoid tendon. To calculate a conduction velocity, distances must be measured with calipers. CMAP amplitude can be compared both from side to side, to assess the amount of axonal loss, and between the axilla and Erb’s point on the symptomatic side, to look for conduction block. These studies can be technically difficult to perform, however, especially obtaining supramaximal stimulation, and are best used to assess axonal loss by comparing the symptomatic side to the asymptomatic side. Because these usually are axonal loss lesions, motor studies generally do not increase the yield of localizing the lesion beyond what is obtained from routine needle EMG.
In axillary neuropathies, needle EMG is used to demonstrate denervation, reinnervation, or both in the two axillary innervated muscles, the deltoid and the teres minor. All three heads of the deltoid are easily accessible to needle EMG; the teres minor is more difficult to study. If abnormalities are found in any of these muscles, it is essential to sample other muscles innervated by the upper trunk and posterior cord of the brachial plexus to ensure that the abnormalities found in the axillary-innervated muscles are not part of a more widespread brachial plexus lesion or cervical radiculopathy. Muscles that are important to check include the biceps, supraspinatus, infraspinatus, triceps, and brachioradialis. In addition, the cervical paraspinals should be sampled to help exclude a C5–C6 radiculopathy.
The musculocutaneous nerve arises directly from the lateral cord of the brachial plexus ( Figure 31–8 ). In the upper arm, it pierces the coracobrachialis muscle to run in the fascia between the biceps and brachialis muscles. It innervates all three of these elbow flexor muscles, including the biceps, brachialis, and coracobrachialis. The brachialis muscle also commonly receives some innervation from the radial nerve nearby, although clinically, this is of little or no importance. In the region of the elbow, the musculocutaneous nerve runs deep to the brachial fascia, over the brachialis muscle. Past the elbow, its terminal extension continues as a pure sensory nerve, known as the musculocutaneous sensory or lateral antebrachial cutaneous sensory nerve . In the forearm, the nerve becomes subcutaneous and separates into two terminal divisions (anterior and posterior) to supply sensation to the lateral half of the forearm.
Isolated musculocutaneous neuropathies are rare. Patients with nontraumatic musculocutaneous neuropathies have been reported due to strenuous physical activity (e.g., weight lifting, rowing, throwing a football), surgery, and pressure during sleep. There is a single report of a musculocutaneous neuropathy associated with repetitive carrying of items on the shoulder with the arm curled around the object (labeled the “carpet carrier’s palsy”). Similarly, there is one report of osteochondroma of the humerus compressing the musculocutaneous nerve.
More commonly, musculocutaneous neuropathies occur as part of more widespread traumatic lesions of the shoulder and upper arm, especially fractures of the proximal humerus. Clinically, musculocutaneous neuropathies result in weakness of elbow flexion, an absent biceps reflex, and sensory loss in the lateral forearm.
More common is entrapment of the distal musculocutaneous sensory nerve. This occurs at the elbow, where the nerve can become entrapped between the biceps tendon or fascia and the brachialis muscle. Characteristically, patients report worsening pain or paresthesias, or both, when the arm is pronated and extended, a position that increases the pressure on the nerve at the elbow site. A hyperextension injury of the elbow, such as may occur during sports-related activities such as tennis, also may cause musculocutaneous sensory neuropathy. Examination in these cases shows isolated altered sensation in the lateral forearm, with normal muscle strength and reflexes. There may be tenderness to palpation over the nerve at the elbow.
The aim of the electrophysiologic exam is to demonstrate isolated involvement of the musculocutaneous nerve and to exclude a brachial plexopathy, cervical radiculopathy, or involvement of other proximal nerves. The most important nerve conduction study to perform is the lateral antebrachial cutaneous sensory study. This sensory potential can be easily elicited by stimulating just lateral to the biceps tendon at the elbow and recording over the nerve 12 cm distally, on a line connecting the stimulation point to the radial pulse. Comparison with the contralateral side is useful in cases where symptoms are limited to one side. Musculocutaneous neuropathies, both distal and proximal, result in abnormal lateral antebrachial cutaneous sensory nerve action potentials (SNAPs). When an abnormal potential is found, it is important to check other sensory potentials, especially those that pass through either the lateral cord or the upper trunk of the brachial plexus (e.g., median and radial SNAPs). Abnormalities found in these nerves suggest a more widespread brachial plexopathy. As noted earlier, comparison with the asymptomatic side is helpful, especially if the studies are at the lower limits of normal.
Similar to axillary motor studies, proximal motor nerve conduction studies can be performed stimulating the axilla and Erb’s point and recording with either a monopolar needle or surface electrode over the biceps ( Table 31–1 ). A surface reference electrode is placed distally over the biceps tendon. The CMAP amplitude can be compared both from side to side, to assess the amount of axonal loss, and between the axilla and Erb’s point, to look for a conduction block. A conduction velocity can be calculated but requires calipers to measure the distance accurately. In contrast to the sensory studies, these motor studies are more technically difficult, especially obtaining supramaximal stimulation, and are best used to assess the degree of axonal loss by comparing the symptomatic side with the asymptomatic side. Similar to axillary and suprascapular neuropathies, musculocutaneous neuropathies usually are axonal loss lesions. Accordingly, motor studies generally do not increase the yield of localization over performing the needle EMG alone.
In distal musculocutaneous neuropathies at the elbow, the needle EMG is normal. In proximal lesions, EMG demonstrates denervation or reinnervation, or both, with decreased recruitment of motor unit action potentials (MUAPs) in the biceps. The brachialis and coracobrachialis can also be sampled but are more difficult than the biceps and offer no additional information. If abnormalities are found in the biceps, it is essential to sample other upper trunk and lateral cord innervated muscles to ensure that the abnormalities found are not part of a more widespread brachial plexus lesion or cervical radiculopathy, especially if the lateral antebrachial cutaneous SNAP is normal. Important muscles to check include the pronator teres and flexor carpi radialis (lateral cord) and deltoid, brachioradialis, supraspinatus, and infraspinatus (upper trunk). In addition, the cervical paraspinals need to be sampled to help exclude a C5–C6 radiculopathy.