Look for this and other related articles in Instructional Course Lectures, Volume 57, which will be published by the American Academy of Orthopaedic Surgeons in March 2008:"Compression of the Ulnar Nerve at the Elbow: Cubital Tunnel Syndrome," by Harris Gellman, MDCubital tunnel syndrome is the most well-recognized compression/traction neuropathy occurring around the elbow. Compression of the median and radial nerves about the elbow and forearm are less often encountered yet must be understood by any physician who treats upper extremity pathology.
Look for this and other related articles in Instructional Course Lectures, Volume 57, which will be published by the American Academy of Orthopaedic Surgeons in March 2008:"Compression of the Ulnar Nerve at the Elbow: Cubital Tunnel Syndrome," by Harris Gellman, MD
"Compression of the Ulnar Nerve at the Elbow: Cubital Tunnel Syndrome," by Harris Gellman, MD
Nontraumatic compression of the median and radial nerves at the elbow may each present in one of two forms: as a motor palsy or as a pain syndrome. Patients with anterior interosseous nerve syndrome present with hand weakness, whereas patients with pronator syndrome complain of pain and paresthesias that can be easily confused with carpal tunnel syndrome. Patients with posterior interosseous nerve syndrome present with hand weakness, whereas patients with radial tunnel syndrome complain of lateral elbow and forearm pain that can be easily confused with lateral epicondylitis. A discussion of each condition will include information on presentation, evaluation, the approach to management, and the results of treatment, with particular emphasis on controversial issues such as differential diagnosis, surgical indications, and surgical approaches.
Anatomy
The median nerve is formed distal to the axilla from portions of the medial and lateral cords of the brachial plexus, composed of fibers from C6, C7, C8, and T1. The lateral cord provides mostly sensory axons from C6 and C7, with the medial cord contributing the bulk of motor input through C8 and T1. The terminal divisions of the medial and lateral cords coalesce in a "Y" shape to become the median nerve, which then courses lateral and superficial to the brachial artery. The median nerve is parallel and anterior to the medial intermuscular septum. At the middle of the brachium, the median nerve crosses over the brachial artery to lie just medial to the artery. The nerve and artery pass under the lacertus fibrosus (bicipital aponeurosis) and enter the antecubital region medial to the biceps tendon and anterior to the brachialis (Fig. 1-A). The median nerve then passes beneath the humeral (superficial) head of the pronator teres and subsequently passes between the humero-ulnar and radial portions of the flexor digitorum superficialis, under the proximal arch of this muscle belly (Fig. 1-B). The nerve courses further down the forearm under cover of the flexor digitorum superficialis and lies over the flexor digitorum profundus along the lateral side of this muscle.
The last major branch of the median nerve in the forearm, the anterior interosseous nerve, departs from the median nerve approximately 4 cm distal to the medial epicondyle of the humerus, passing under fibrous tissue originating from either the flexor digitorum superficialis or its pronator teres, runs along the interosseous membrane between the flexor digitorum profundus and flexor pollicis longus, and finally terminates in the distal part of the forearm, deep to the pronator quadratus, innervating the flexor digitorum profundus, flexor pollicis longus, and pronator quadratus muscles along the way. Anatomical dissections have demonstrated the innervation of the flexor pollicis longus and flexor digitorum profundus through multiple nerve branches along the medial and lateral edges of these muscles, respectively1,2. The autonomous zone of sensory innervation for the median nerve includes the palmar surfaces of the index and long fingers and their dorsal aspects past the level of the distal interphalangeal joint.
When considering the normal anatomy of the median and anterior interosseous nerves, one must also keep in mind the potential anomalous structures and innervation patterns of the upper extremity. The presence of a Martin-Gruber anastomosis in the forearm may change the clinical presentation of pronator syndrome. Additionally, anatomical variations as the anterior interosseous nerve courses distally in the antebrachium may affect sites of compression of the nerve. In the study by Beaton and Anson, the nerve passed between the deep and superficial heads of the pronator teres in 82% of the limbs, deep to the ulnar head in 7%, and through the superficial head of the pronator in 2%; in the remaining 9%, the deep head was absent3. Additionally, there have been descriptions of the nerve passing anterior to the superficial head. Underneath the deep head of the pronator, one may find Gantzer's muscle (the accessory head of the flexor pollicis longus). In a recent investigation, al-Qattan4 found the muscle in 52% of twenty-five specimens; the muscle arose from the medial epicondyle in approximately 85% of the specimens in which it was present and from the medial epicondyle and coronoid in the remaining 15%. A supracondyloid process of the distal part of the humerus (present in approximately 1% of individuals of European descent) may indicate the presence of a ligament of Struthers, which can cause compression of the median nerve as it passes through the foramen formed by the fibrous band5.
Anterior Interosseous Nerve Syndrome and Pronator Syndrome
Compression of the median nerve at the elbow can present in one of two forms: anterior interosseous nerve syndrome and pronator syndrome. Anterior interosseous nerve syndrome presents as a motor palsy, whereas pronator syndrome is associated with pain and paresthesias that can be easily confused with carpal tunnel syndrome. While anterior interosseous nerve palsy may occur after fracture or surgery about the elbow, only nontraumatic anterior interosseous nerve syndrome will be addressed here.
Anterior Interosseous Nerve Syndrome
The earliest reports of anterior interosseous nerve syndrome appeared in a 1948 report by Parsonage and Turner that described anterior interosseous nerve palsy associated with paralytic lesions about the shoulder6. Subsequently, Kiloh and Nevin reported cases of spontaneous neuritis of the anterior interosseous nerve with isolated paralysis of the flexor pollicis longus and median-innervated flexor digitorum profundus7. (Although Parsonage and Turner had described this lesion years earlier, they had not thought that it was anatomically possible to have an isolated lesion of the anterior interosseous nerve.)
Sites of isolated anterior interosseous nerve compression include the fibrous arch of the flexor digitorum superficialis and the pronator teres, under which the nerve travels early in its course. A patient who has anterior interosseous nerve syndrome will complain of weakness of pinch, which can affect activities such as writing and picking up small objects. A careful history may reveal an antecedent episode of spontaneous forearm pain followed by progressive weakness in the hand. Repetitive elbow flexion or forearm pronation have been proposed as dynamic causes of compression of the anterior interosseous nerve in the proximal part of the forearm. A history of transient shoulder pain (often following an immunization or viral illness) preceding the development of upper extremity weakness or paresthesias should alert one to the possibility of brachial neuritis (Parsonage-Turner syndrome). While a variety of cutaneous and motor nerves can be affected, anterior interosseous nerve palsy has been described in a high percentage of patients with brachial neuritis. These cases need to be differentiated from those presumably occurring at the elbow as the treatment will be very different.
Anterior interosseous nerve syndrome leads to an inability to flex the interphalangeal joint of the thumb and the distal interphalangeal joint of the index finger (and occasionally the long finger) because of weakness of the flexor pollicis longus and the radial half of the flexor digitorum profundus. Although the pronator quadratus is also involved, the patient will not be aware of this involvement because of intact function of the pronator teres. During examination, the contribution of the pronator teres can be partially eliminated by testing the strength of resisted pronation with the elbow maximally flexed. The flexor pollicis longus and flexor digitorum profundus of the index finger are examined by asking the patient to make an "OK sign" by firmly opposing the tips of the thumb and index finger. A patient who has anterior interosseous nerve palsy will compensate with a key, or lateral, pinch using the thumb adductor and the first dorsal interosseous muscle (which are innervated by the ulnar nerve).
Pronator Syndrome
The symptoms and signs of pronator syndrome overlap those of carpal tunnel syndrome. Both conditions are associated with pain and paresthesias in the radial three and one-half digits, with the symptoms often being worse with certain activities. Patients with either of these conditions may have a tender nerve over both the wrist and the proximal part of the forearm, although the Tinel sign is most prominent at the main site of compression, which can include a ligament of Struthers originating from a supracondylar process, the pronator teres, the lacertus fibrosus, and the fibrous arcade of the flexor digitorum superficialis. There are, however, some features that may help to differentiate the two diagnoses. Painful symptoms that awaken the patient from sleep are much more common in association with carpal tunnel syndrome. A careful examination of a patient who has pronator syndrome may reveal decreased sensibility in the palm over the thenar eminence (innervated by the palmar cutaneous branch, which branches from the median nerve proximal to the carpal tunnel). The Phalen test will be negative. Provocative testing for pronator syndrome stresses structures that potentially can compress the median nerve. If symptoms are reproduced during resisted forearm pronation or resisted elbow flexion/supination, one should suspect the pronator teres or the lacertus fibrosus, respectively. Resisted grasp (the "middle finger flexion test") is less helpful because the result could be positive in association with either condition: contraction of the flexor digitorum superficialis muscle belly may compress the nerve proximally, and finger flexion may draw prominent lumbrical muscle bellies into the carpal canal and cause distal nerve compression8.
Evaluation of the Median Nerve
Standard radiographs of the elbow will aid in the diagnosis of structural anomalies such as a supracondylar process5. Magnetic resonance images are rarely made unless a mass is suspected. Electrodiagnostic studies can be very helpful for the evaluation of anterior interosseous nerve palsy but usually reveal normal findings in cases of pronator syndrome.
Treatment and Outcomes
Patients with pronator syndrome or anterior interosseous nerve syndrome should be managed with a trial of nonoperative treatment, which should include the avoidance of provocative activities that involve repetitive elbow flexion, forearm pronation, or forceful gripping. A posterior elbow splint may reinforce the importance of resting the extremity. Oral nonsteroidal anti-inflammatory medications may be helpful, particularly in cases of pronator syndrome.
It has been reported that 50% to 70% of patients with pronator syndrome may respond to conservative therapy9,10. Patients who do not respond to such therapy are candidates for operative decompression, which has been associated with a success rate of approximately 90%11,12. The indications for surgical decompression in cases of anterior interosseous nerve syndrome are more controversial. Recommendations in the literature have been empirical or anecdotal, based mainly on small retrospective series. The authors of those studies have recommended surgical decompression if no motor recovery is detected after eight to twelve weeks. Improvement has been documented anywhere from four weeks to two years after surgical release13,14. To further muddy the waters, a few reports have described series of patients with anterior interosseous nerve syndrome who recovered function without surgical intervention; spontaneous resolution occurred between three and twenty-four months after the onset of symptoms15,16.
During surgery, the nerve is decompressed from proximal to distal. The surgeon should begin the decompression proximal to the antecubital crease in order to easily identify the median nerve and to explore this region for a ligament of Struthers. If present, the ligament and the supracondylar process should be removed. Decompression in the forearm includes release of the lacertus fibrosus and the proximal fibrous edge of the flexor digitorum superficialis arch (Figs. 2-A and 2-B). The superficial head of the pronator should be mobilized; if still causing compression, it may be released with a step-cut at the distal tendon and then be repaired in a lengthened position17. The surgeon should also explore the region for the presence of anomalies such as Gantzer's muscle, an accessory muscle of the flexor pollicis longus arising from the medial epicondyle and occasionally the coronoid18. The arm is immobilized in a well-padded posterior splint with the elbow at 90° and the forearm in neutral rotation. The splint is discontinued and motion is allowed approximately one week after surgery.
Anatomy
The radial nerve is the terminal branch of the posterior cord and is actually the largest terminal branch of the brachial plexus. It is composed of C5 to C8 nerve roots and lies posterior to the axillary artery. As the nerve travels distally, it falls further posteriorly and passes through the triangular space to enter the posterior aspect of the brachium. It then runs along the anterior aspect of the long head of the triceps, leaving this area and coursing along the posterior aspect of the humerus in the spiral groove to lie on the lateral aspect of the humerus under the brachioradialis and on top of the brachialis. As the nerve courses distally, it is under the cover of the extensor carpi radialis longus and brevis, with the joint capsule and capitellum of the humerus lying posterior to the nerve at this point. This region, from the joint capsule to the proximal supinator, was originally referred to as the "radial tunnel" by Roles and Maudsley, who began decompressing the radial nerve as a part of the treatment for recalcitrant lateral epicondylitis19. The radial nerve divides into its two terminal branches, the sensory branch and the posterior interosseous nerve. The superficial sensory branch takes off from the radial nerve proximal to the radial tunnel and travels distally on the undersurface of the brachioradialis. The posterior interosseous nerve crosses the elbow joint and lies in fatty tissue anterior to the radiocapitellar joint capsule. The first potential structures causing nerve compression in the radial tunnel, tethering fibrous bands that arise between the brachioradialis and the joint capsule, can be found at the level of the radial head (Fig. 3). The next anatomical area of possible compression includes the branches of the radial recurrent artery that are believed to cause increased pressure on the posterior interosseous nerve. As the nerve courses distally, it lies underneath the extensor carpi radialis brevis, which can itself cause compression of the nerve under its tendinous edge. The nerve travels between the superficial and deep heads of the supinator after passing under the arcade of Frohse. Interestingly, this proximal fibrous arcade of the superficial head of the supinator does not appear to be present in the human fetus but develops at some point later in life in approximately 30% of individuals (with the rest having a membranous leading edge of the supinator)20. In a small minority of patients, the nerve also may be compressed under the distal edge of the supinator.
Before the nerve enters the supinator, it sends branches to the extensor carpi radialis brevis and supinator muscles. As it runs under the cover of the distal supinator, it branches into two parts; the superficial portion innervates the extensor carpi ulnaris, extensor digitorum communis, and extensor digiti minimi, and the deep branch serves the abductor pollicis longus, extensor pollicis brevis, extensor pollicis longus, and extensor indicis proprius. The bifurcated nerve then continues into the posterior aspect of the forearm, deep to the extensor digitorum communis and superficial to the outcropper muscles, to terminate as an articular branch to the radiocarpal joint.
Posterior Interosseous Nerve Syndrome and Radial Tunnel Syndrome
Compression of the radial nerve at the elbow can present in one of two forms: posterior interosseous nerve syndrome and radial tunnel syndrome. Patients with posterior interosseous nerve syndrome present with a motor palsy, whereas patients with radial tunnel syndrome complain of lateral elbow and forearm pain that can easily be confused with lateral epicondylitis. Both posterior interosseous nerve syndrome and radial tunnel syndrome have the same sites of compression; these sites can be recalled by the mnemonic FREAS, which stands for fibrous bands about the radiocapitellar joint, radial recurrent branches, extensor carpi radialis brevis, arcade of Frohse, and, finally, the distal edge of the supinator.
Posterior Interosseous Nerve Syndrome
A patient who has posterior interosseous nerve syndrome may present with a transient episode of forearm pain followed by progressive weakness of the digital extensors as well as the extensor carpi ulnaris. Posterior interosseous nerve palsy, while usually complete, occasionally involves only some of the extensor muscles (Fig. 4). Muscles of the mobile wad are spared as they are innervated more proximally. If the brachioradialis or the radial wrist extensors (the extensor carpi radialis brevis and extensor carpi radialis longus) are also weak, the examiner should suspect more proximal compression by, for example, the lateral head of the triceps or a humeral exostosis. Posterior interosseous nerve syndrome may occur idiopathically or may be caused by compression of the posterior interosseous nerve by masses such as ganglia. In the rheumatoid population, the examiner should palpate for rheumatoid nodules, radiocapitellar synovitis, or a subluxated radial head.
The examiner should not assume that a patient with rheumatoid arthritis who presents with a loss of digital extension necessarily has posterior interosseous nerve palsy. Other diagnoses to be considered in the differential include extensor tendon rupture (usually at the level of the distal part of the ulna), sagittal band rupture, or inflammatory involvement of the metacarpophalangeal joints leading to palmar-ulnar subluxation. Extensor tendon rupture is characterized by loss of the tenodesis effect (compensatory digital extension during passive wrist flexion). When the sagittal band ruptures, the extensor tendon usually subluxes ulnarly between the metacarpal heads. Passive extension of the finger allows the extensor mechanism to centralize over the metacarpophalangeal joint, and the patient is then able to actively maintain extension (but not to initiate digital extension).
Radial Tunnel Syndrome
Radial tunnel syndrome is also caused by compression of the posterior interosseous nerve in the proximal part of the forearm. The factors that cause radial tunnel syndrome to develop in some patients and posterior interosseous nerve syndrome to develop in others are not well understood. Radial tunnel syndrome is characterized by proximal dorsal forearm pain. Some patients may have weakness secondary to the pain, although actual motor involvement is rare. Symptoms may occur at night or may be aggravated with repetitive activities such as forearm rotation, elbow extension, and maximum wrist flexion-extension.
The symptoms of radial tunnel syndrome overlap those of the much more prevalent lateral epicondylitis (tennis elbow); a careful history and examination are required to differentiate the two. In fact, 5% of patients with tennis elbow also have radial tunnel syndrome21. Conversely, 50% of patients with radial tunnel syndrome may have concomitant lateral epicondylitis22. This may be due to the involvement of the extensor carpi radialis brevis in both conditions or to the fact that the superficial belly of supinator produces some tension of the common extensor23. The possibility of radial tunnel syndrome should always be considered in cases of "resistant" lateral epicondylitis19. The most common finding is tenderness over the mobile wad rather than over the lateral epicondyle. The diagnosis is strengthened with positive provocative maneuvers, including resisted supination with the elbow extended. The "middle finger extension test," heralded in some texts as pathognomonic for radial tunnel syndrome, has not been found to be useful in our experience as it is often positive in patients with lateral epicondylitis (because of stressful contraction of the extensor carpi radialis brevis). A small group of patients may present with point tenderness over the dorsal aspect of the middle part of the forearm, distal and ulnar to the mobile wad; these patients represent the 5% of cases in which the posterior interosseous nerve is compressed under the distal edge of the supinator. Because of the absence of objective findings associated with radial tunnel syndrome, the physical examination should be repeated on more than one occasion and the findings should be compared with those for the contralateral extremity.
Another rare cause of lateral elbow/proximal forearm pain that may be confused with radial tunnel syndrome is compression of the lateral antebrachial cutaneous nerve, first described in 1982 by Bassett and Nunley24. This condition is caused by compression of the lateral edge of the biceps tendon and may be aggravated by active elbow flexion and supination or maximum forearm pronation. The diagnostic dilemma exists because of the close proximity of the lateral antebrachial cutaneous nerve and the radial nerve in the antecubital region (Fig. 5).
Evaluation of the Radial Nerve
Differential lidocaine injections may be helpful for distinguishing among lateral epicondylitis, radial tunnel syndrome, and lateral antebrachial cutaneous nerve compression. Electrodiagnostic studies may be positive for patients with posterior interosseous nerve syndrome and those with lateral antebrachial cutaneous nerve pathology but are usually normal for patients with radial tunnel syndrome. Imaging studies may be helpful if a mass around the elbow or the proximal part of the forearm is suspected.
Treatment and Outcomes
Most cases of posterior interosseous nerve syndrome and radial tunnel syndrome initially should be treated nonoperatively with the avoidance of provocative activities, immobilization in a wrist or elbow splint, and use of oral nonsteroidal anti-inflammatory medications.
Surgical decompression should be considered for patients with a clear diagnosis of radial tunnel syndrome after three months of failed conservative treatment. For patients with posterior interosseous nerve syndrome, surgical indications include progressive weakness, late presentation with severe weakness and atrophy, or the absence of improvement after four to twelve weeks of conservative treatment.
The surgical approach to the radial nerve in the forearm may vary depending on the comfort of the surgeon with the procedure and with his or her knowledge of the offending anatomical structures. At least four approaches have been described; however, not all approaches allow equal access to all potential points of compression. The brachioradialis-splitting approach is most direct, but it involves blunt dissection through muscle and requires careful placement of the incision25. The relatively bloodless interval between the brachioradialis and the extensor carpi radialis longus requires a more extensive incision, but it can be combined with the surgical approach used for the treatment of lateral epicondylitis26. A dorsal approach, commonly used for the treatment of proximal radial diaphyseal fractures, provides better access to the distal edge of the supinator but provides poorer visualization of the posterior interosseous nerve more proximally and associated potential compressive structures27. An anterior approach may be useful for exposure of the anterior aspect of the joint and masses, but it requires a deeper dissection and provides poor visualization of the distal edge of the supinator28,29.
Regardless of approach, one key to successful decompression of the radial nerve is the identification of both the extensor carpi radialis brevis and the proximal edge of the supinator. If the surgeon believes that the decompression is finished after releasing the first muscle encountered, the arcade of Frohse may have been missed. One should realize that the extensor carpi radialis brevis passes over the posterior interosseous nerve obliquely, whereas fibers of the supinator run at right angles to the nerve (Figs. 6-A and 6-B).
After satisfactory release and wound closure, the arm is immobilized in a well-padded posterior splint with the elbow at 90° and the forearm in neutral rotation. The splint is discontinued and motion is allowed approximately one week after surgery.
Good to excellent recovery of function after the surgical treatment of posterior interosseous nerve syndrome may be expected in as many as 90% of cases30. The results of radial tunnel surgery are much more variable. Good to excellent relief of symptoms have been reported in 51% to 92% of cases31-33. This wide range of satisfactory results may be due to numerous factors, including the more subjective nature of the diagnosis, with poor localization of pain, a more chronic presentation leading to permanent changes within the nerve, or a more frequent association with Workers' Compensation claims.
The less common nerve compression syndromes about the elbow require a high level of suspicion and, in some cases, may be confidently diagnosed only after serial examinations. Each syndrome has a series of overlapping conditions that must be considered in the differential diagnosis. The majority of patients with pronator syndrome will respond to nonoperative management. The role of surgery for the treatment of anterior interosseous nerve syndrome is controversial, particularly in light of the fact that many cases may actually be caused by more proximal nerve pathology. Most patients with posterior interosseous nerve syndrome will recover good to excellent function after surgical release. The results of decompression in patients with radial tunnel syndrome can be equally gratifying if one adheres to very careful patient-selection criteria.