The radial nerve is commonly injured in both traumatic and iatrogenic settings. Although management principles of peripheral nerve injuries apply, continuing advances in the literature warrant periodic reexamination of this injury. Presented here is an overview of the anatomy, epidemiology, and management of radial nerve injuries with a focus on recent literature.
The posterior cord of the brachial plexus terminally divides into the radial and axillary nerves behind the pectoralis minor, atop the subscapularis muscle. A common misconception is that the radial nerve descends along the spiral groove after traversing the triangular interval. The spiral groove contains fibers of the brachialis origin that separate the nerve from the bone. Direct contact instead occurs about the deltoid tuberosity proximally and along the lateral metaphyseal flare distally. The radial nerve is tethered where it pierces the lateral intermuscular septum, more proximally than previously thought, 16 cm from the distal humerus or 47% of the shaft length.1
Interposed between the brachialis and brachioradialis, the radial nerve becomes increasingly anteriorly positioned as it travels distally and bifurcates into the posterior interosseus nerve (PIN) and superficial radial nerve (SRN) just distal and anterior to the radiocapitellar joint. The extensor carpi radialis longus (ECRL) is innervated proximal to the elbow, whereas the extensor carpi radialis brevis (ECRB) is innervated distal to the elbow, more often from the SRN than the PIN.
The SRN provides sensation to the anatomic snuffbox, first web space, and dorsal aspect of the thumb, index, and long fingers. Because of the overlap between the SRN and lateral antebrachial cutaneous nerve, loss of SRN function is well tolerated, but only symptomatic with neuroma formation.
Radial nerve injuries are often simply labeled high or low in relation to nerve bifurcation. Injuries can also be divided into four levels with functional implications,2 with all levels involving loss of thumb and finger extension with variable levels of more proximal paralysis. Level I is infraclavicular, proximal to the brachioaxillary inlet at the beginning of the spiral groove. Injuries at this level affect elbow extension. Level II injuries occur at the level of the spiral groove, proximal to the brachioaxillary outlet where the nerve pierces the lateral intermuscular septum. Elbow extension is spared; however, wrist extension is not. Level III injuries are between the septum and elbow joint, and wrist extension is variable. Physical examination cannot reliably differentiate loss of the ECRB if the ECRL is intact, although isolated ECRL function could put the wrist into exaggerated radial deviation given its insertion. The loss of one wrist extensor is enough to weaken grip strength. Level IV injuries are distal to the ECRB motor branch.
The radial nerve is the most commonly injured motor nerve. The incidence with humerus fractures is 12%. Small studies have shown trends toward higher incidence in the presence of skin wounds and major fracture displacement.3 Most of the palsies, 50% to 68%, present as complete motor loss.4 Other injuries, such as shoulder dislocations, humeral neck injuries, and Monteggia fractures, can also be associated with radial nerve injuries.
Iatrogenic injuries are not uncommon. The incidence of injury for each humeral shaft approach is as follows: lateral 1/5, posterior 1/9, and anterolateral 1/25.5 External fixation puts the nerve at risk during lateral pin placement, and despite anatomic studies detailing safe zones, open pin placement is still recommended.6 In addition, the radial nerve is the most commonly injured nerve during medial epicondyle fracture fixation, put at risk with bicortical penetration.
In addition to traumatic iatrogenic causes, intramuscular and intravenous injections have been shown to cause nerve injuries. In one study, radial nerve injection injuries, only 7 of 24, at the level of the arm recovered spontaneously.7 Symptom onset is delayed in 10% because the agent injected can take time to penetrate the nerve's protective layers. Last, the differential should include compressive causes because of inflammation and neoplastic processes (Figure 1).
A thorough history should elucidate both the cause and probable injury location. One should assess motor and sensory deficits, strength of potential donor muscles, and passive joint motion. A migrating Tinel's phenomenon is a good prognostic indicator.
Objective measures are limited because MRI and ultrasound have been of limited use. Electromyography and nerve conduction velocity studies are rarely helpful acutely except when continuity is unknown because slowed but intact conduction indicates some continuity. Study is recommended around 3 to 4 months after injury. Larger polyphasic motor action potentials of longer duration may be seen before clinical recovery, although the degree of recovery is unclear. If the amplitude of motor nerve conduction velocity is low, less than 0.3 mV, exploration can be undertaken, with repair remaining a viable option even 5 to 6 months after injury.2
Management and Outcomes
With few exceptions, an initial observation period is warranted, during which extension bracing of the fingers and wrist and motion to keep joints supple should be instituted. In the setting of humeral shaft fractures, 70% of radial nerve palsies demonstrate spontaneous recovery around 7 weeks, ranging from 2 weeks to 6.5 months.8 Early exploration has been shown not to improve the outcome except in the case of open fracture or concomitant forearm injuries.9 Humerus fracture management is generally unchanged by the presence of the radial nerve injury. A study of unreamed intramedullary nail fixation with a 40% rate of primary radial nerve injury demonstrated that the injury resolved spontaneously 93%,10 indicating that nerve exploration was not warranted even with operative cases, unless nerve visualization was required for an open approach. Secondary palsy during functional bracing was an indication for early exploration in the past; however, these have also been shown to resolve spontaneously.11
Past indications for early exploration have included secondary palsy after closed reduction, open fractures, distal third fractures including Holstein–Lewis fractures, penetrating injury (Figure 2), associated vascular injury, high-velocity gun-shot wounds, and severe soft-tissue injuries. Both secondary palsies and distal third fractures are no longer considered definite indications for early exploration, with larger studies showing higher rates of spontaneous recovery than indicated by initial reports. The arguments for early exploration are high rates of nerve entrapment, seen in 6% to 25% of cases, and nerve lacerations reported in 20% to 42% of humerus fractures4,8 that can be repaired before scarring when the nerve is maximally mobile. If there is an extensive zone of injury, a reliable minimally invasive approach to finding the nerve without damage to the triceps is with an incision along the lateral bicipital groove, where the posterior antebrachial cutaneous nerve is found posterior to the septum and traced proximally to where it joins the radial nerve and penetrates the intermuscular septum.12 A decision analysis model integrating 37 years of studies demonstrated early surgery to provide an 85% chance of nerve recovery in the setting of humeral shaft fracture.9 The downside of early exploration is potentially unnecessary surgery if the nerve is found to be in continuity.
Late exploration is the most common strategy for persistent palsies. Late surgery gives a 69% chance of recovery in the setting of humeral shaft fracture, with a 31% risk of no recovery at the end of treatment.9 As the radial nerve innervates extrinsic muscles, the distance from the level of injury to the motor end plates is shorter than that for the median and ulnar nerves. Nerve repair can, therefore, be successfully performed later, although timing may affect recovery. One study of 244 traumatic radial nerve injuries demonstrated good outcomes with repair, neurolysis, or grafting an average of 3 months after injury compared with unsuccessful outcomes an average of 5 to 6 months after injury, although repairs performed within 5 months were still superior to muscle or tendon transfers.2
All nerve repairs/reconstructions should be performed tension free. Adjacent joints should be stretched for the repair to allow free limb excursion without traction on the coaptation. Bone shortening, nerve mobilization, and transposition all reduce tension on the repair. A series of 27 penetrating injuries between the brachial inlet and elbow joint primarily repaired by 6 months demonstrated recovery of extension at the wrist in 93%, fingers in 74%, and thumb in 52%.13 Nerve grafting can be used for residual gap management. In Pan's study, all injuries above the brachial outlet required grafting, with longer grafts at the spiral groove level averaging 10.3 cm, compared to more distal grafts averaging 6.4 cm. Shorter grafts were associated with better outcomes. Overall motor recovery was 95% at the elbow, 80% at the wrist, and 30% in the digits.2
The sural nerve is the most commonly chosen autograft. Harvest sequelae include sensory loss, neuroma, deep venous thrombosis, and hematoma. There may be hesitation to incur this morbidity in the setting of guarded prognosis, in which case allografts can be considered. A report of 71 upper extremity nerve repairs using processed nerve allograft, including 2 radial nerve repairs, demonstrated functional results comparable to those of repairs using autografts for gaps measuring 5 to 50 mm.14 Synthetic conduits are alternatives to grafts for small gaps or can supplement direct repairs and grafts. They contain fluid leaking from nerve ends to help form the fibrin matrix and support cell migration.
Nerve transfers from the median to the radial nerve are among the most technically difficult, but carry the best quality of results. Ray and Mackinnon15 reported on 19 patients treated an average of 5.7 months after injury with transfer of redundant median nerve branches to the flexor carpi radialis (FCR) and flexor digitorum superficialis (FDS) to the PIN and ECRB branches, respectively, with a few variations. Medical research council muscle strength at 12-month follow-up was M4 or higher for wrist extension in 18 patients and finger/thumb extension in 12 patients. Good results have been reproduced in a similar study of the pronator teres (PT) branch to ECRL, FCR branch to PIN, and FDS branch to ECRB in six patients with complete radial nerve palsy. All achieved independent finger function with 93% grip strength.16
It should be noted that nine patients in Mackinnon's series had concomitant PT to ECRB tendon transfers as an internal splint. A supercharged end-to-side transfer is another option to “babysit” motor end plates while awaiting spontaneous nerve recovery or regeneration of a nerve transfer.17
Mackinnon18 notes that despite her success with nerve transfers, she still performs five tendon transfers for every nerve transfer when treating radial nerve palsies. Nerve transfers offer high-demand patients increased finger dexterity without nerve grafting and reduce the risk of sensory/motor bundle malalignment; however, the recovery time frame is significantly longer at 1 year compared to that for tendon transfers which offer immediate motor function. Technical pearls include tension-free repair with the donor branch taken distal and connected at a proximal end of the recipient branch, releasing recipient nerve sites of entrapment, motor reeducation when M1 recovery is present, and concomitant tendon transfers when appropriate.
Tendon transfers are a tried and true treatment option. Functional improvement is such that many choose this option without even exploring the nerve for potential repair. Principles include ensuring the presence of supple joints, expendable donors with adequate strength and excursion, straight line of pull, synergism, soft-tissue coverage with minimal scar tissue, and one tendon per function. The most common transfer for wrist extension is PT to ECRB. Finger extension is with FCR to extensor digitorum communis, although the flexor carpi ulnaris or FDS to the ring or long fingers is an option. Thumb extension is with palmaris longus or an FDS tendon transferred to the extensor pollicis longus. Repairs can be performed end to end for a straighter line of pull, or end to side when done early or in combination with nerve repair and some degree of nerve recovery is expected.
High levels of patient satisfaction and good range of motion are generally reported. Neuroplasticity is required and, therefore, younger and motivated patients tend to do better. Sequelae include under- or over-tensioning, loss of power grip, unnatural movement of the wrist and fingers, limited wrist flexion, and lack of individual finger movement.2
Stiff or painful joints will not function well with either tendon transfers or late recovery of muscles. In these cases, wrist fusion provides a durable solution that places the wrist in a position of power so that remaining finger and thumb function can be optimized.
Although lack of radial nerve sensation tends not to be functionally disabling, painful neuromas can cause significant allodynia. Options include desensitization with occupational therapy, anesthetic injection, sympathectomy, or proximal transection, or cauterizing and burying the end within muscle or other soft tissue.19
Knowledge of radial nerve anatomy, physiology, injury mechanisms, and potential for recovery after insult continues to be updated. As larger studies demonstrate higher spontaneous recovery rates and longer windows for successful late repair, early nonsurgical management for up to 6 months in adults and 9 months in children has expanded from closed humeral shaft fractures to include operative fractures that do not require nerve exposure, secondary palsies, and distal third humerus fractures. Early exploration continues to be the recommendation when the probability of recovery after observation is less than 40%, such as with penetrating injuries and high-energy open fractures. Treatment options include tendon transfers, direct repair with or without grafting, nerve transfers, or combination procedures. Further study is needed to determine the long-term outcomes of nerve repairs with modern techniques that emphasize tension-free repair and compare the use of allografts with that of autografts.
Evidence-based Medicine: Levels of evidence are described in the table of contents. In this article, references 2, 5, 8, 10, and 11 are level III studies. References 3, 7, and 13-17 are level IV studies. References 1, 4, 6, 9, 12, and 18 are level V studies.
References printed in bold type are those published within the past 5 years.
1. Bono CM, Grossman MG, Hochwald N, Tornetta P: Radial and axillary nerves: Anatomic considerations for humeral fixation. Clin Orthop Relat Res 2000:259–264.
2. Pan C-H, Chuang DC-C, Rodríguez-Lorenzo A: Outcomes of nerve reconstruction for radial nerve injuries based on the level of injury in 244 operative cases. J Hand Surg Eur Vol 2010;35:385–391.
3. Nachef N, Bariatinsky V, Sulimovic S, Fontaine C, Chantelot C: Predictors of radial nerve palsy recovery in humeral shaft fractures: A retrospective review of 17 patients. Orthop Traumatol Surg Res 2017;103:177–182.
4. DeFranco MJ, Lawton JN: Radial nerve injuries associated with humeral fractures. J Hand Surg Am 2006;31:655–663.
5. Claessen FM, Peters RM, Verbeek DO, Helfet DL, Ring D: Factors associated with radial nerve palsy after operative treatment of diaphyseal humeral shaft fractures. J Shoulder Elbow Surg 2015;24:e307–e311.
6. Wegmann K, Lappen S, Pfau DB, Neiss WF, Müller LP, Burkhart KJ: Course of the radial nerve in relation to the Center of rotation of the elbow: the need for a rational safe zone for lateral pin placement. J Hand Surg Am 2014;39:1136–1140.
7. Esquenazi Y, Park SH, Kline DG, Kim DH: Surgical management and outcome of iatrogenic radial nerve injection injuries. Clin Neurol Neurosurg 2016;142:98–103.
8. Shao YC, Harwood P, Grotz MRW, Limb D, Giannoudis PV: Radial nerve palsy associated with fractures of the shaft of the humerus: A systematic review. J Bone Joint Surg Br 2005;87:1647–1652.
9. Bishop J, Ring D: Management of radial nerve palsy associated with humeral shaft fracture: A decision analysis model. J Hand Surg Am 2016;34:991–996.e1.
10. Grass G, Kabir K, Ohse J, Rangger C, Besch L, Mathiak G: Primary exploration of radial nerve is not required for radial nerve palsy while treating humerus shaft fractures with unreamed humerus nails (UHN). Open Orthop J 2011;5:319–323.
11. Koch PP, Gross DFL, Gerber C: The results of functional (Sarmiento) bracing of humeral shaft fractures. J Shoulder Elbow Surg 2002;11:143–150.
12. Matzi V, Hörlesberger N, Hohenberger GM, et al: Minimally invasive approach to the radial nerve: A new technique. Injury 2015;46:2374–2378.
13. Laubscher M, Held M, Maree M, Solomons M: Radial nerve lacerations: The outcome of end-to-end repairs in penetrating trauma. Hand Surg 2015;20:67–72.
14. Cho MS, Rinker BD, Weber RV, et al: Functional outcome following nerve repair in the upper extremity using processed nerve allograft. J Hand Surg Am 2012;37:2340–2349.
15. Ray WZ, Mackinnon SE: Clinical outcomes following median to radial nerve transfers. J Hand Surg Am 2011;36:201–208.
16. García-López A, Navarro R, Martinez F, Rojas A: Nerve transfers from branches to the flexor carpi radialis and pronator teres to reconstruct the radial nerve. J Hand Surg Am 2014;39:50–56.
17. Barbour J, Yee A, Kahn LC, Mackinnon SE: Supercharged end-to-side anterior interosseous to ulnar motor nerve transfer for intrinsic musculature reinnervation. J Hand Surg Am 2012;37:2150–2159.
18. Mackinnon SE: Donor distal, recipient proximal and other personal perspectives on nerve transfers. Hand Clin 2016;32:141–151.
19. Wolfe SW, Pederson WC, Hotchkiss RN, Kozin SH, Cohen MS: Green's Operative Hand Surgery. Philadelphia, PA, Elsevier Health Sciences, 2016, pp 1.