The median nerve arises from the medial and lateral cords of the brachial plexus and has contributions from C5-T1 nerve roots (1–8). It descends lateral to the brachial artery and between the biceps brachii and brachialis muscles. The median nerve then traverses the antecubital fossa and runs deep to the bicipital aponeurosis and anterior to the brachialis muscle. It then passes between the superficial and deep heads of the pronator teres (PT). In the forearm, the median nerve travels between the flexor digitorum superficialis and profundus muscle bellies. The most common median nerve neuropathy is carpal tunnel syndrome (CTS), where the median nerve is compressed under the flexor retinaculum (4–6). Proximal median nerve (PMN) neuropathies denote lesions of the median nerve at locations proximal to the carpal tunnel, which include the forearm, elbow, upper arm, and brachial plexus. They are often underdiagnosed or misdiagnosed as CTS due to the similarity of symptoms and overlapping pathologies, which may co-exist in the same patient (4, 9).

Several proximal median nerve entrapment sites have been identified (from proximal to distal): (1) beneath the ligament of Struthers, which extends from the supracondylar process of the distal humerus; (2) bicipital aponeurosis (lacertus fibrosus); (3) proximal insertion of the PT humeral head; (4) passage between the two heads of the PT; (5) dysplasia of the deep PT head; (6) common arcade between the PT and flexor digitorum superficialis (FDS); (7) fibrous arch/arcade of the FDS; and (8) Gantzer muscle (1, 5, 7, 9–14). At several entrapment sites, mechanical compression causing a compromised arterial supply to the median nerve may lead to PMN neuropathies (10). At the level of the axilla, the nerve is vulnerable to compression in the medial fascial compartment (15).

Both PT and anterior interosseous nerve (AIN) syndromes involve compression/entrapment of the median nerve proximal to the carpal tunnel (4, 7, 14, 16). In 1951, Seyffarth described a case of median nerve compression in the proximal forearm and coined the term “pronator syndrome” (17). PT syndrome is considered to be a compressive/entrapment neuropathy of the median nerve in the proximal forearm with symptoms and signs that are similar to those of CTS (1, 10, 11, 18–20). AIN syndrome is a pure motor palsy resulting from compression/entrapment of the AIN close to its origin, most often by the deep and superficial heads of the PT or fibrous arcade of the FDS (10, 16).

The median nerve may be injured by either iatrogenic (excessive traction or use of retractors during surgery) or non-iatrogenic causes (direct penetrating trauma by a gunshot wound, knife wound, or broken glass) (3). EDX and US studies are invaluable in identifying the lesion site and distinguishing between the different underlying pathologies (16). These studies, especially US, are valuable to ascertain the etiology of the PMN neuropathies, which can potentially direct medical or surgical management (5).

A large series of patients with PMN neuropathies that includes EDX and US data is lacking in the literature. In this report, we describe 62 patients with PMN neuropathies, confirmed by EDX studies, 52 (83.9%) of whom also had US studies. The clinical, EDX, and US findings are presented. The etiologies and mechanisms of PMN neuropathies based on the specific anatomical zone of involvement are discussed. The significance of US in complementing the EDX studies for localization and determining the underlying cause is also highlighted.

The differential diagnoses of PMN neuropathy include CTS, thoracic outlet syndrome (TOS), other brachial plexopathies, and cervical radiculopathy (1). EDX studies are the primary modality for confirming anatomical localization of the site of PMN and determining the underlying pathology and its severity (16). Slowing of motor and sensory conduction, along with alterations in the morphology and amplitude of CMAP and SNAP are metrics used in this assessment (6, 20). EDX studies are useful in differentiating PMN neuropathy from CTS, although they are usually inconclusive or negative in clinically diagnosed PT syndrome (1, 16, 19). The needle EMG should include muscles innervated through the entire length of the median nerve so that the topography of abnormal muscles can be mapped (14). Denervation/reinnervation changes are identified in the median nerve-innervated muscles at and distal to the lesion site (10).

When the lesion is in Zone 1 (involving fascicles in the brachial plexus that form the median nerve and the median nerve itself up to the branch to the PT), precise localization by EDX studies can be challenging. The addition of US can provide further insight. High-resolution US provides high soft-tissue and spatial resolution, cost-effectiveness, safety, ready accessibility, easy portability, and real-time and dynamic nerve imaging in a timely manner (2, 5, 8, 12). US complements EDX studies in providing clues to localizing the lesion and insight into the underlying pathology in patients with PMN neuropathy. There are several scenarios where US is invaluable for the localization of median nerve neuropathy. In Zone 1, nerve conduction studies (NCS) and needle EMG do not always provide adequate findings for localization. Denervation in the PT (the most proximal muscle innervated by the median nerve) makes it difficult to differentiate between lesions at the site of the branch to the PT and more proximal lesions in the upper arm and axilla. US can locate the lesion in most circumstances. Similarly in Zone 4 lesions (in the forearm distal to the origin of the AIN) with denervation of the APB, differentiation from entrapment at the carpal tunnel is difficult. When motor conduction is slow proximal to the wrist, it is not necessarily a conclusive finding as retrograde slowing of motor conduction can occur in longstanding entrapment at the carpal tunnel (25). Precise localization may also be challenging when only the sensory fascicles are affected. While an inching study across the carpal tunnel can be helpful, detecting lesions in the distal forearm may be problematic. US can be useful in detecting such lesions in the forearm. When no CMAPs are detected over the APB and second lumbrical and no SNAPs are recordable over the digits, US is crucial for localization.

US is also useful in identifying the underlying pathology in two specific situations: (1) detection of lesions such as a schwannoma or other tumors/cysts compressing the nerve and (2) distinction between neurotmesis and axonotmesis in cases of nerve injuries. After Wallerian degeneration occurs distal to the site of injury, no conduction can be documented distal to the injury (unlike in a conduction block). While documentation of neurotmesis dictates quick surgical repair, in cases of axonotmesis, one could wait to see if reinnervation occurs within the expected period, which also depends on the length of the injured segment. EMG cannot make the distinction between total axonotmesis and neurotmesis, while US is able to in the majority of circumstances.

Fascicular lesions are particularly important in cases of inflammatory changes. A fascicular somatotopia can be seen early on in the nerve, which may explain the pattern of muscle paresis and sensory disturbance of more proximally located lesions (26). The fascicular anatomy of the median nerve has been well-described, and precise localization by nerve conduction and extensive needle EMG may fail. In PaTS, fascicular involvement proximal to the location suggested by the clinical picture has been reported (26), and our patients did not undergo MR neurography to localize the lesion.

The majority of the studies of PMN neuropathy have focused on entrapment/injury at the elbow and proximal forearm (13, 14, 27). In the study by Sos et al., 53 patients (55 cases, including 2 bilateral) with median nerve entrapment syndrome in the elbow and proximal forearm underwent surgical release of the median nerve and AIN around the elbow and/or proximal forearm. The most common compressed structure was isolated proximal PT insertion (14). The mean age at diagnosis was 56 years with a male-to-female ratio of 1:1 (29 male patients and 26 female patients). The symptom onset was gradual in 32 (58%) patients and acute in 23 (42%) patients. EMG studies demonstrated reduced motor conduction velocity in 38 (56%) patients, conduction block in 5 (9%) patients, and median nerve territory EMG abnormalities in 47 (85%) patients. Three patients had no EMG abnormalities. In Olehnik and colleagues’ study of 36 patients (39 limbs) who underwent surgical decompression of the median nerve in the proximal forearm, the site of compression was the FDS in 22 patients, PT in 13 patients, and both in 4 patients (13). The average age was 39 years, and 29 (80%) were female. Of the 37 limbs that underwent motor and sensory nerve conduction tests, 12 showed an abnormal nerve conduction velocity from the elbow to the wrist. Motor and sensory conduction abnormalities were detected in 12 and 2 limbs, respectively. In Gross and Jones’ study of 17 patients with PMN neuropathy who underwent EMG, the cause of neuropathy was trauma in 5, overuse of the PT in 3 patients, post-infection in 2 patients, secondary to a congenital lesion in 1, and undetermined in 6 (27). The main branch of the median nerve at or proximal to the PT was identified in 14 patients and the AIN in 3. The EMG demonstrated that median nerve compression by the PT caused denervation of the PT and distal muscles. EMG was unable to differentiate a median nerve lesion at the PT from a more proximal lesion (27). Our study concurs with the study by Sos et al. with respect to a similar mean age of patients (56.7 years). However, our study features a higher percentage of male patients (58.1%), acute symptom onset (95.2%), and iatrogenic etiology (48.4%). Our study was also the only one that presented US findings that revealed 42.3% of patients with an increased CSA of the median nerve and 30.2% of patients with a hyperechoic median nerve. All patients in our study underwent EDX studies, which showed low amplitude or no CMAP in 82.2% of the patients and abnormal/no SNAP in 79.0% of the patients.

Traumatic neuropathies were referred for EMG at our neurodiagnostic center after several weeks to months, and, therefore, we could classify them into only conduction block or axonal injury. US was helpful in showing neuroma in continuity versus neurotmesis. However, we were not successful in detecting hourglass fascicular lesions with US in our study.

Neurologists should be aware of the four localization zones of the PMN neuropathy with their distinctive etiologies and utilize additional investigations to confirm the diagnosis and formulate appropriate management. EDX and US studies are important techniques that complement each other in determining the specific localization zone for the PMN neuropathy and providing valuable clues to the underlying pathology in patients with PMN neuropathy. The classification into the four localization zones should be based on the final localization. The suspected lesion may initially be established by the clinical findings, with subsequent details offered by the EDX studies, and final exact localization with the assistance of the US studies. This tiered approach emphasizes the importance of the different modalities. Further studies are warranted to determine the contribution of inflammatory changes.