Clinical manifestations

Presentation and course

Radial neuropathy most commonly presents with sudden painless wrist and finger drop. There may be associated paresthesias or numbness over the hand dorsum. The clinical picture sets radial neuropathy apart from confounding conditions and generally allows for lesion localization.

Radial neuropathy is most commonly due to compression at or around the spiral groove. Due to its more proximal innervation, the triceps muscle is spared. Elbow flexion may seem to be intact due to the biceps muscle compensating for the weakened brachioradialis muscle. More commonly, brachioradialis muscle palpation during elbow flexion reveals a flaccid tone. Motor deficit involves wrist and finger extensor muscles. Sensation is impaired over the dorsum of the hand, mainly the first and second metacarpals. In one series of 29 patients with Saturday night palsy, the brachioradialis muscle was reportedly intact in 18%, and sensation was preserved in 40% (43). This may be due to anomalous nerve origin or fascicular arrangement of sensory fibers.

Radial nerve injury proximal to the spiral groove is less common. In addition to spiral groove lesion manifestations, the triceps muscle is weak, and the triceps deep tendon reflex is reduced or absent. When present, sensory loss may extend beyond the hand to the posterior forearm.

Lesions a few centimeters distal to the elbow result in posterior interosseous neuropathy. The acute to subacute progressive weakness involves one or more adjacent metacarpophalangeal joint extensors. Interphalangeal joint extension is innervated by the ulnar and median nerves and is, therefore, spared. Elbow pain occurs is 50% of these patients and usually clears in a few days. Tenderness can occur 5 centimeters distal to the elbow. Useful features are pain reproduction or exacerbation by pressure, resisted supination with 90-degree elbow flexion or middle finger extension, and relief by local anesthetic injection. There is typically no sensory loss. The spared extensor carpi radialis brevis and longus muscles result in radial deviation on wrist extension.

Superficial radial neuropathy induces paresthesias over the hand dorsum. At times, this may be painful. A handcuff or watchband may additionally compress the median and ulnar nerves. Superficial radial neuritis, or Wartenberg syndrome, is associated with pain at the radial styloid radiating to the hand and is exacerbated by ulnar deviation of the wrist. Tinel sign is usually positive over the junction of the middle-to-distal third of the radius.

Prognosis and complications

Many patients with Saturday night palsy will fully recover in 6 to 8 weeks (47). Denervation by electromyography indicates further delay in recovery. Incomplete recovery occurs in only 13% of cases (30). In a retrospective case series, all patients with compressive radial neuropathy in whom clinical follow-up was available experienced complete recovery with a mean duration from onset to resolution of 3.4 months (01). In 2007, Malikowski and colleagues underscored the value of electrodiagnostic studies in predicting prognosis for recovery after traumatic radial neuropathy. Good outcome is present in 65% of patients despite an absent radial motor response (26). The overwhelming majority of patients with tourniquet paralysis fully recover within 1 to 6 months (20). Recovery starts as early as a day to a few weeks. Radial neuropathy after closed fracture of the humerus recovers fully in the vast majority of patients (05). Idiopathic posterior interosseous neuropathy sometimes resolves within 3 to 6 months. Posterior interosseous neuropathy due to palpable lipoma responds well to surgical resection. Prognosis for full recovery is also good in newborn radial palsy. Very few patients with tennis elbow are refractory to conservative management (45).

Clinical vignette

A 56-year-old man reported to the clinician that he awoke 3 days earlier with left arm weakness. He had slept on his left side with the arm flexed at the elbow. He reported prior exposure to lead vapors but denied alcohol consumption. Urine heavy metal screen was normal. On physical examination, his muscles were well developed. Weakness was mild in the left brachioradialis but was severe at the wrist and long finger extensor muscles. When the wrist and metacarpophalangeal joints were passively extended, he could actively extend his interphalangeal joints. Similarly, wrist extension helped normalize his interosseus muscle’s strength. Pinprick was reduced over the left dorsal first web space, and deep tendon reflexes were symmetrically reduced at the arms. Routine left arm nerve conduction studies were normal. Radial motor conduction recording from the extensor digitorum communis muscle revealed 90% conduction block at the spiral groove.

Stimulation 2 centimeters distally abolished the conduction block. Radial motor conduction velocities were moderately reduced between the spiral groove and axilla. Electromyography of the left extensor digitorum communis muscle revealed only a few rapidly firing motor-unit potentials. The interference pattern was moderately reduced in the left brachioradialis muscle. These findings are diagnostic of severe demyelinating left radial neuropathy at the spiral groove due to compression during sleep.

Biological basis

Etiology and pathogenesis

Compression, trauma, and entrapment are the main mechanisms of radial nerve injury. Tumor, systemic disease, and birth trauma constitute less common causes.

Radial nerve compression is most frequently at or within close vicinity of the spiral groove (42). The radial nerve in Saturday night palsy may be compressed during sedation, deep sleep, inebriation, or by the head of a sleeping partner. Triceps muscle fibrosis induced by chronic analgesic intramuscular injection may cause severe progressive painless radial neuropathy (27). Radial nerve compression may occur more proximal due to inappropriate crutch pressure at the axilla, prolonged tourniquet application, frequent automated blood pressure cuff inflation, or poor arm position during anesthesia (40). Radial nerve compression at the axilla or arm may be associated with ulnar or median nerve dysfunction. Windmill pitchers may develop radial neuropathy at variable sites proximal to the spiral groove (37).

Due its course around the spiral groove, the radial nerve is injured in 15% of humeral fractures (05). Mechanisms include blunt trauma, stretch due to bone separation or reduction effort, laceration, and strangulation between sharp bone edges. There may be concomitant median and ulnar nerve damage. Tardy radial neuropathy is usually due to callus formation. Gunshot or stab wounds, elbow arthroscopy, or inadvertent nerve injection may also injure the radial nerve.

Posterior interosseous nerve entrapment occurring without clear trauma or compression has been referred to as radial tunnel syndrome, supinator syndrome, or supinator channel syndrome. These vague and confusing terms have also been used to describe patients with predominant elbow pain (35). When anatomically identifiable, compression most commonly is at the fibrous edge of the arcade of Frohse or more distally within the substance of the supinator muscle. Sustained forceful pronation or repetitive pronation-supination is hypothesized to predispose violinists, flutists, and mostly manual workers to this type of entrapment. Additionally, posterior interosseous neuropathy can follow radial fracture, dislocation, or upper radius orthopedic procedure. On the other hand, tennis elbow is related to repetitive trauma and extensor origin angiofibroblastic hyperplasia (31).

An enlarging lipoma in the vicinity of the posterior interosseous nerve causes a painless, compressible, freely movable mass. Ganglia or cysts around a degenerated elbow joint produce painful compression of the posterior interosseous nerve. Painless weakness without palpable mass may suggest a benign peripheral nerve tumor or hereditary neuropathy with liability to pressure palsy. The latter entraps the radial nerve at sites that vary from one attack to the next. Progressive painful weakness without a palpable mass raises concern about a malignant tumor.

Diabetes mellitus and multifocal motor neuropathy with conduction block may involve the radial nerve. Mononeuritis multiplex, leprosy, and lead poisoning selectively involve the radial nerve. Lead neuropathy is typically motor and spares the brachioradialis.

Cheiralgia paresthetica or superficial radial nerve compression complicates a small proportion of radial styloid fractures. It also is the result of tight casts, handcuffs, or watch bands (11). Iatrogenic superficial radial neuropathy can complicate intravenous catheter placement and de Quervain tenosynovectomy.

The radial nerve originates from the C5 through T1 spinal roots. It is the major continuation of the posterior cord of the brachial plexus (11). The radial nerve travels at the angle formed by the posterior wall of the axilla and the medial arm. The ulnar collateral branch of the radial nerve arises at the axilla and innervates the triceps (medial head), anconeus, and the elbow joint. The posterior cutaneous nerves of the arm arise from the radial nerve at the axilla, and the posterior cutaneous nerves of the forearm arise from between the axilla and spiral groove. The radial nerve delves between the medial and lateral heads of the triceps before looping posterolaterally around the spiral groove. It then traverses the lateral intermuscular septum below the insertion of the deltoid muscle. The radial nerve branches that arise at the lateral epicondyle usually innervate the brachioradialis, extensor carpi radialis muscles, and elbow joint. The nerve courses anteriorly and laterally at the elbow, between the brachioradialis and extensor carpi radialis and the brachialis muscles. That space forms the radial tunnel limited posteriorly by the humeral capitulum. Within the tunnel and few centimeters distal to the elbow, the radial nerve divides into the superficial radial nerve and the posterior interosseous nerve. The superficial radial nerve gives off variable sensory innervation to the dorsum of the hand and the first 2.5 to 3.5 digits. The supinator forms a fibrotendinous arch called the arcade of Frohse. This arcade is identified in 66% of adults and measures on average 2.3 cm long, 1.1 cm wide, and 0.07 cm thick (04). The posterior interosseous nerve pierces the arcade and then courses within the supinator muscle. This deep motor branch innervates the ulnar extensors of the wrist and finger extensors at the metacarpophalangeal joints.

Epidemiology

Latinovic studied the occurrence of entrapment neuropathies between 1992 and 2000 in 253 UK general practices participating in a research database (21). The annual incidence of radial neuropathy was 1.56/100,000 for women and 2.83/100,000 for men.

Fifty percent of radial neuropathies are due to compression at the spiral groove. Twenty-five percent are due to humeral fracture. A systematic review of the literature (36) indicates an 11.8% prevalence of radial neuropathy following humeral fracture.

Prevention

Adequate positioning during anesthesia and avoidance of poor sleep habits help minimize the risk of radial nerve compression. Care in the use of pneumatic tourniquets dramatically reduces the risk of compression palsy. Posterior interosseous neuropathy frequently responds to avoidance of precipitating activity.

Differential diagnosis

The differential diagnosis of radial neuropathy includes C7 radiculopathy, posterior cord brachial plexopathy, and extensor tendon rupture. Radial neuropathy can be confused with stroke due to apparent intrinsic hand muscle weakness. When the wrist and metacarpophalangeal joints are passively extended, patients with radial neuropathy can better activate median and ulnar-innervated lumbricals, thereby extending their interphalangeal joints. Similarly, wrist extension dramatically improves interosseus muscles strength. These maneuvers eliminate most mechanical disadvantage in radial neuropathy. They do not, however, improve stroke-related hand weakness.

Besides radiating neck pain, weakness from C7 radiculopathy also affects median (pronator teres and flexor carpi radialis) and ulnar (flexor carpi ulnaris) C7-innervated muscles. It spares the brachioradialis muscle. Sensation is lost over dorsal and palmar surfaces of the middle finger.

Posterior cord brachial plexopathy is characterized by additional weakness in the deltoid and latissimus dorsi muscles. Sensory loss also involves the axillary and posterior cutaneous distribution to the arm and forearm. Thumb and index finger extensor tendon rupture occurs in rheumatoid arthritis. Despite weakness, electromyography of the extensor pollicis and indicis proprius muscles is normal.

Posterior interosseous neuropathy must be distinguished from tennis elbow. Pain in the latter is maximal proximal to the elbow, at the extensor muscle origin from the lateral epicondyle. Passively stretching the extensor muscle tendons intensifies the pain. Tennis elbow is the result of chronic traumatic tendinous degeneration, most commonly without posterior interosseous nerve compromise. Occasionally, minor neurogenic involvement has been purported (49).

Diagnostic workup

Nerve conduction studies and electromyography are essential to precisely confirm, locate, and prognosticate radial nerve injuries. In addition to routine nerve conduction studies, bilateral recordings from the extensor digitorum communis and extensor indicis proprius muscles are indicated. In general, acute radial neurapraxia causes conduction block and fast-firing motor unit potentials as in our vignette. Axonotmesis reduces radial motor amplitudes within a week and eventually results in active denervation.

Radial neuropathy at the spiral groove is typically associated with motor conduction block. Additionally, 60% to 90% of cases display active denervation mostly due to secondary axonal loss (43). Distal muscle denervation may persist in 60% of cases for at least 2 to 4 years after full clinical recovery (29). In most instances, denervation involves the brachioradialis muscle as well as the extensor digitorum communis, extensor indicis proprius, and the extensor carpi radialis.

Posterior cutaneous sensory conduction studies may confirm a postganglionic C7 lesion localization or exclude a posterior interosseous neuropathy; when abnormal, they may predict a poor outcome (24).

The “window test” is an easy-to-perform, simple bedside maneuver that can support the diagnosis of posterior interosseous neuropathy (12). It is administered by an examiner facing the patient with the hand placed on the ulnar aspect of the patient's pronated forearm. The patient is asked to extend the wrist. A positive “window test” is detected when a gap or window appears between the examiner's forearm and the patient's hand ulnar aspect.

Nerve conduction studies in posterior interosseous neuropathy are of limited use (11). Stimulation at the spiral groove with the arm supinated while recording from the extensor indicis proprius muscle may occasionally show unilateral motor latency prolongation. It is unclear whether radial nerve latency change with pronation and supination is a reliable technique. Chronic polyphasic motor unit potentials or active denervation may be observed in the extensor digitorum communis and extensor indicis proprius muscles. Such electrodiagnostic localization may necessitate radiologic imaging of the elbow (40). When present, noncalcified lipoma is usually a radiolucent mass on routine roentgenogram. Computerized axial tomography or MRI may visualize a tumor, cyst, ganglion, hematoma, or lipoma around the radial nerve. To assess posterior interosseous nerve syndrome at the arcade of Frohse, ultrasound images and clinical data of 13 patients confirmed by neurologic examination and electrophysiological testing were evaluated retrospectively (09). In comparison to healthy volunteers, enlargement of the posterior interosseous nerve was seen in all patients with posterior interosseous nerve syndrome, but not in volunteers. MRI of tennis elbow may reveal thickening and, in more severe cases, tears in the lateral ulnar collateral ligament (06).

Management

With few exceptions, expectant management with clinical and electrodiagnostic follow-up is the rule for radial neuropathies (40). Most patients with radial palsies induced by compression recover fully within 2 to 12 months. Faster and more complete recovery is anticipated with neurapraxia. All patients with radial neuropathy should wear a splint with passive dorsiflexion at the wrist and fingers. In the absence of clinical and electromyographic recovery of severe radial neuropathy, surgical exploration after 3 to 5 months is indicated, and intraoperative radial nerve conduction studies are helpful.

Indications for early exploration of posttraumatic radial neuropathy include complex humeral fractures or weakness developing after closed reduction of a simple fracture. Spontaneous recovery occurs in more than 90% of radial neuropathies associated with closed humeral shaft fractures, and there was no benefit of early exploration on nerve recovery (14). Furthermore, closed humeral shaft fracture cases managed operatively have higher risk for developing secondary radial neuropathies. Therefore, exploration of a simple humeral fracture may be delayed for 2 months to allow for spontaneous radial nerve recovery. Successful nerve graft induces return of function in 12 to 18 months. Delayed tendon transfers may restore wrist extension.

In a retrospective chart review of 30 patients with isolated radial nerve injuries treated with tendon transfers and 16 patients managed with nerve transfers, both approaches yielded significant improvements in grip and pinch strength after surgery (32). Although the nerve transfer group had demonstrably better grip strength, pinch strength did not differ. Both groups had improved pain, function, and satisfaction following surgery.

Patients with clinically and electromyographically confirmed spontaneous posterior interosseous neuropathy should be evaluated clinically and radiographically for elbow deformity and soft tissue mass. Such patients usually require early surgical exploration. On the other hand, patients without elbow mass or deformity may respond to avoidance of provocative activities, splinting, and local anesthetic or steroid injection. If improvement is not forthcoming after 2 to 3 months, surgical exploration is necessary. Exploration of posterior interosseous neuropathies associated with simple fractures, dislocation, or surgical procedure of the elbow is also delayed 2 to 3 months to allow for spontaneous recovery (38).

Conservative management of spontaneous superficial radial neuropathy includes prevention of external compression at the wrist, avoidance of repetitive wrist movements, nocturnal wrist splints, and local anesthetic or steroid injection. Otherwise, surgical decompression is effective as it is for traumatic radial nerve laceration.

Most patients with tennis elbow respond well to 6 to 12 months of conservative therapy including rest, activity modification, physical therapy, nonsteroid anti-inflammatory drugs, and local steroid injection (45). In a meta-analysis, Kinesio tape was found to be effective for pain relief, grip strength, and functional improvement in patients with lateral epicondylitis undergoing rehabilitation (52). Very few cases require decompression of the posterior interosseous nerve to reduce tension at the lateral epicondyle. Ma and Wang provide a detailed review of therapeutic options for acute and chronic lateral epicondylitis (25).

A large randomized, controlled trial indicated that graded eccentric exercise is of more benefit than concentric therapy throughout the 1-year follow-up period in subjects recruited from primary care practice for tennis elbow lasting more than 3 months (33). A cost-effectiveness study compared corticosteroid injection, physical therapy, and a combination of these interventions to patients receiving a blinded placebo injection (07). The authors found that physiotherapy, which was more costly but more cost-effective than corticosteroids, should be considered first-line intervention. Corticosteroid injection alone had greater response variability, and was less cost-effective, whereas the combination was neither clinically more effective nor cost-effective. Intermittent corticosteroid injection is not a good long-term management option for lateral epicondylitis given their short-lasting benefit and the risk of tendon rupture with repeated use.

A prospective, randomized, double-blinded, placebo-controlled clinical trial compared counterforce bracing with placebo bracing in the management of acute tennis elbow (19). As compared to placebo bracing, the counterforce brace provided better reduction in the frequency and severity of pain in the first 3 months and in overall elbow function at 6 months. A randomized controlled trial of extracorporeal shock wave therapy found it to be ineffective in tennis elbow (39). An earlier meta-analysis of the literature suggested a role for extracorporeal shock wave therapy in chronic tennis elbow (50), but a 2021 meta-analysis refuted that conclusion based on low to moderate certainty of evidence (17). A literature meta-analysis provides support for use of botulinum toxin A injections into the forearm extensor muscles (60 units) for treatment of chronic, treatment-resistant lateral epicondylitis (16). A network meta-analysis suggested effectiveness of low-dose onabotulinumtoxinA (10 to 20 units) in reducing pain in lateral epicondylitis (41).

In a prospective randomized trial comparing leukocyte-rich platelet-rich plasma injection with surgical intervention for the management of refractory tennis elbow, both approaches produced equivalent functional outcomes. Although surgery may have led to lower pain scores at 1 year, most cases treated with platelet-rich plasma avoided surgery (48). In a systematic review and meta-analysis comparing platelet-rich plasma to local corticosteroids injections for lateral elbow epicondylitis, the results were mixed. Although recipients of local steroids fared better for up to 8 weeks, those receiving platelet-rich plasma did better at 24 weeks (22). A Cochrane database review evaluated autologous blood and platelet-rich plasma injection therapy for lateral elbow pain (18). It concluded that there was no support for the use of either of these injection modalities for lateral elbow pain. In a 2023 updated systematic review and meta-analysis of level 1 and 2 studies, the pooled estimate for pain scores at 1 month favored corticosteroids but favored platelet-rich plasma at 3 and 6 months (15). However, caution should be exercised in formulating a treatment recommendation as there was high risk of bias, and available studies were of moderate to low quality.

In lateral epicondylitis of at least 1-month duration, a single course of leech therapy was reportedly effective in relieving pain in the short-term and improved disability in the intermediate-term (02). In a pilot study, ultrasonographically guided percutaneous radiofrequency thermal lesioning for recalcitrant lateral epicondylitis resulted in 78% pain reduction (23). In a small prospective study, percutaneous ultrasonic tenotomy performed under local anesthesia appeared to improve pain and performance in chronic, refractory lateral, or medial elbow tendinopathy up to 1 year after the procedure (03). Low-quality published evidence suggests that acupuncture may be superior to drug or blocking therapy or sham acupuncture therapy for lateral epicondylitis (51). In a prospective randomized controlled study, dry needling was more effective than corticosteroid injections for up to 6 months in lateral epicondylitis refractory to 3 weeks of first-line therapy (44). In a systematic review and meta-analysis, high-intensity laser therapy reduced pain and improved quality of life in lateral epicondylitis, with low quality of evidence (10).

Special considerations

Pregnancy

Unilateral or bilateral newborn radial neuropathies are related to uterine constriction rings, umbilical cord pressure to the arm, or an amniotic band. These are usually suspected when newborns have arm bruising or fat necrosis (08). Radial nerve injury can complicate forceps delivery and humeral fracture.

Anesthesia

Adequate positioning and padding during general anesthesia dramatically reduces the risk of radial nerve palsy.