Superficial Radial Nerve Anatomy

Origin. The nerve separates from the radial nerve at the level of the radiocapitellar joint, in the interval between brachialis and brachioradialis, at the same moment the posterior interosseous nerve is given off. The two separate here and never meet again. The deep branch dives into the supinator as the posterior interosseous nerve, which has no cutaneous territory, and the superficial branch runs distally on the deep surface of brachioradialis, hugging the radial side of the forearm.

Forearm course. Beneath the brachioradialis muscle belly it lies on the supinator, pronator teres and flexor digitorum superficialis in sequence. In the proximal third it lies lateral to the radial artery; in the middle third the two run closely applied, the nerve lateral to the artery, which is the relationship divided in a radial forearm flap harvest and encountered in a volar Henry approach. In the distal third the nerve diverges laterally and turns dorsally, while the artery continues volarly towards the wrist.

The fascial exit. The nerve pierces the deep fascia by passing between the tendons of brachioradialis and extensor carpi radialis longus, the pattern Leroux and colleagues found in 9 of 10 cadaveric specimens. This is the site of Wartenberg compression, and the reason a Tinel sign here is diagnostic.

Terminal divisions. The nerve divides into a dorsolateral (radial) branch to the thumb and a dorsomedial branch to the index, middle and radial ring digits, though the branching pattern is genuinely variable. The same cadaveric series gives the distances, referenced to the radial styloid, that govern every radial wrist incision:

• 7.8 cm proximal: mean point where the nerve pierces the deep fascia
• 4.5 cm proximal: mean point of the first division
• 3.0 cm proximal: mean point of the second division
• At the styloid: a mean of 5.1 branches crossing the operative field

Reading them in order. The further distal you operate, the more branches there are and the smaller each one is. Proximal to 8 cm you are looking for one nerve under fascia; distal to 3 cm you are looking for five nerves in fat. There is no safe zone in the distal 5 cm and no interval over the first compartment you can dissect through blindly, because the nerve is a fan of branches to be found rather than one cord to be avoided.

Relations at the wrist. By the level of the styloid the branches lie in the subcutaneous fat directly over the first dorsal compartment, crossing superficial to abductor pollicis longus and extensor pollicis brevis and then extensor pollicis longus. In the anatomical snuffbox the radial artery lies deep, on the floor against the scaphoid and trapezium, and the nerve branches lie superficial, in the roof. A wire aimed at the scaphoid from the snuffbox passes through nerve and vein before it reaches artery and bone.

The cephalic vein. The branches run with and around the cephalic vein and its tributaries. A vein encountered in a dorsoradial wrist incision is a reliable warning that a nerve branch is within millimetres, and bleeding from it is a signal to stop and look, not to reach for diathermy.

The lateral antebrachial cutaneous nerve lies more volar and radial in the distal forearm and is injured in the same incisions. A patient with persistent radial wrist dysaesthesia after surgery may have a lesion of either.

There is none, and that is the point. The superficial branch of the radial nerve supplies no muscle; its entire content is cutaneous sensory and vasomotor. Knowing that it is purely sensory is what makes the three-way discrimination below (sensory only, motor only, or both) the single most efficient way to localise a radial nerve problem at the bedside:

• Numbness in the first dorsal web with normal power: a lesion of the superficial branch alone, such as a Wartenberg lesion, a laceration or a compressive dressing
• The same numbness plus finger drop: a radial nerve lesion proximal to the division, above the radiocapitellar joint, with the finger drop coming from the deep branch
• Finger drop with normal sensation: an isolated posterior interosseous nerve lesion

The muscles it lies against. Although it drives no muscle, the muscles around it decide where it is compressed and where it is cut.

The territory. The nerve supplies the skin of the dorsoradial hand: the dorsum of the thumb, index and middle fingers and the radial half of the ring finger, and proximally the dorsoradial aspect of the distal forearm and wrist. On the digits it reaches only as far as the proximal interphalangeal joints. The dorsal skin distal to the PIP joints of the radial three and a half digits is supplied by the dorsal branches of the digital nerves from the median nerve.

The autonomous zone. The most reliable area to test is the dorsal aspect of the first web space, over the first dorsal interosseous, where overlap is least. Elsewhere the territory overlaps substantially with the lateral antebrachial cutaneous nerve (from the musculocutaneous nerve) proximally and radially, which is why a completely divided superficial radial nerve can leave a surprisingly small anaesthetic patch.

How far the overlap goes. In 20 mapped arms the two nerves actually intersected in 95 per cent, on average 2.25 times each. Dual innervation with the dorsal branch of the ulnar nerve is likewise frequent, and the lateral antebrachial cutaneous nerve is a common contributor to the thumb, so thumb-dorsum numbness does not by itself name the injured nerve.

Examining it. Test light touch and two-point discrimination in the first dorsal web, and compare with the contralateral side and with the adjacent ulnar and median territories. Map the border of the abnormal area with a pen: a sharply demarcated dorsoradial patch is a peripheral lesion, while a glove-like or non-anatomical distribution points to a functional or a central cause.

Always test the dorsoulnar hand as well. Preserved dorsoulnar sensation with abnormal dorsoradial sensation confirms an isolated superficial radial lesion. Abnormality in both suggests a more proximal or a systemic cause.

Cheiralgia paraesthetica. Wartenberg syndrome is compression of the superficial branch of the radial nerve at the point where it emerges through the deep fascia between the brachioradialis and extensor carpi radialis longus tendons.

The mechanism is a scissor. In forearm pronation the extensor carpi radialis longus tendon rotates beneath the brachioradialis tendon. The two behave like the blades of a scissor with the nerve between them, which is why hyperpronation is provocative and why symptoms are worse with repetitive rotation. Adding wrist flexion and ulnar deviation places the nerve under longitudinal tension on top of the transverse compression, which is why a Finkelstein manoeuvre reproduces the pain.

External compression. Tight watch straps, wrist bands, bracelets and identification bands give the classic history, and are the reason the syndrome has been called handcuff neuropathy. Casts and tight dressings over the distal radial forearm, direct blows in contact sport, and pressure from crutch or walking aid handles act in the same way.

Internal and pathological causes. Prior surgery or trauma with perineural scarring is by far the commonest cause seen in a hand clinic. The others:

• De Quervain tenosynovitis, with a thickened, inflamed first dorsal compartment directly under the nerve branches
• A ganglion arising from the first extensor compartment or the scaphotrapezial joint
• Distal radius fracture callus or a prominent implant on the radial column
• Anomalous muscle bellies, including an accessory abductor pollicis longus slip

Presentation. Pain and paraesthesia over the dorsoradial wrist, thumb and first web space, worse with repetitive pronation and supination, with wrist flexion and ulnar deviation, and at night. There is often a clear history of external compression: a tight watch strap, a wrist band, handcuffs, a cast, or a new work tool. There is no weakness; if there is weakness, reconsider the diagnosis.

Examination. A Tinel sign 4 to 8 cm proximal to the radial styloid, over the fascial exit, is the most useful single sign. The other findings are altered sensation in the dorsoradial territory, a positive hyperpronation test and tenderness over the nerve rather than over the compartment. The Finkelstein manoeuvre is positive but does not discriminate.

Non-operative first, and it usually works. Remove the offending compression, splint the wrist in slight extension with the forearm out of pronation, use a short course of anti-inflammatory medication and modify activity. In one series of 22 patients with Wartenberg neuritis, 19 responded to conservative treatment. A corticosteroid injection around the nerve at the fascial exit can be both diagnostic and therapeutic: inject small and inject accurately.

Surgical decompression is for refractory cases: a longitudinal incision over the fascial exit, release of the fascia between the two tendons over several centimetres, and neurolysis. Do not skeletonise the nerve; leave a bed of soft tissue under it. Warn the patient that operating on an already irritable subcutaneous sensory nerve can produce a scar-related neuroma, which is worse than the original complaint.

The single most examined differential on the radial wrist. Both conditions hurt on the radial side of the wrist and both are positive on Finkelstein testing. Releasing the first dorsal compartment in a patient with Wartenberg syndrome does not help, and places the already irritated nerve directly in the wound.

They coexist. A meaningful proportion of patients have both conditions, because the swollen first compartment irritates the nerve branches lying directly over it. In a French series of 44 patients with radial sensory nerve disorders, the authors specifically warned that a medicolegal problem arises when a coexisting Wartenberg neuritis has not been recognised before first compartment release, because the patient then attributes the persistent nerve symptoms to the operation.

Practical rule. Before booking any first dorsal compartment release, document the sensory examination and the presence or absence of a proximal Tinel sign in the notes. If both conditions are present, treat the tenosynovitis first and non-operatively, and re-examine.

Why it happens here. A divided or tethered cutaneous nerve in subcutaneous fat directly over a bony prominence is the worst possible location for a neuroma. That is precisely the situation a careless incision over the radial styloid creates.

Diagnosis. The sign is a pinpoint, exquisitely tender Tinel-positive spot at or just proximal to a previous incision or injury, with pain radiating into the thumb and index on percussion and a hypersensitive skin territory distally. The nerve carries sympathetic vasomotor fibres, and patients with a neuroma frequently describe cold intolerance and colour change in the dorsoradial hand alongside the pain, which overlaps with the presentation of complex regional pain syndrome.

The block decides. A diagnostic local anaesthetic block proximal to the point that abolishes the pain confirms the neuroma. It is the single most important step before offering surgery, because it separates a neuroma from a regional pain syndrome or a central sensitisation problem.

Non-operative management comes first.

• Desensitisation therapy with graded textures, mirror therapy and scar massage
• Neuropathic pain medication: gabapentinoids or tricyclics
• Protective padding and removal of the offending pressure

Treat any complex regional pain syndrome before considering operation. Operating in the middle of an untreated regional pain syndrome makes the hand worse.

Operative options. Neuroma excision alone gives poor results, because a new neuroma forms in the same subcutaneous plane; the proximal stump must be dealt with.

• Excision and proximal transposition into muscle is the workhorse: bury the stump in brachioradialis or in the pronator mass, well proximal to the wrist, away from any pressure point and without tension
• Transposition into bone or beneath the interosseous membrane is described as an alternative
• Targeted muscle reinnervation and regenerative peripheral nerve interface techniques give the regenerating axons an end organ instead of a scar, and are increasingly used for symptomatic cutaneous neuromas
• Nerve capping or conduit techniques, and direct repair or grafting where a distal stump exists and the territory is worth restoring

Reported outcomes in older series were modest, roughly two-thirds excellent by strict criteria, which reinforces that prevention is the real treatment.

Universal principles for any dorsoradial incision.

1. Mark the branches before you scrub, using the surface anatomy technique above.
2. Cut skin only. The dermis is the last thing you divide with a blade in this region.
3. Spread longitudinally with fine scissors through the subcutaneous fat, parallel to the expected course of the branches.
4. Treat every vein as a warning.
5. Hold branches with moist loops or a skin hook, never with a self-retaining retractor placed in fat.
6. Cross the branches at right angles, with the branch mobilised and visible, rather than dissecting along them.
7. Release the tourniquet and inspect before closure. An intact-looking branch under tension from a stitch will still form a neuroma.
8. Document the nerve findings in the operation note, both for the patient and for whoever sees them next.

Anaesthetic technique is protective too. A wrist block or regional block placed proximally avoids local infiltration of the subcutaneous fat around the branches, which distorts the tissue planes exactly where you need to see them. If local infiltration is used, place it intradermally along the incision line only, not into the subcutaneous plane.

First dorsal compartment release. The nerve branches lie in the subcutaneous fat directly over the compartment. They are the structure at risk, and superficial radial nerve injury is the recognised complication of the operation. On top of the universal principles:

• A transverse incision at the level of the styloid crosses the branches at right angles and makes them easier to identify; a longitudinal incision runs parallel to them and is more extensile but demands more care
• Spread with tenotomy scissors down to the retinaculum, and protect every branch with a moist loop before opening the sheath
• Release the compartment on its dorsal side to reduce the risk of volar tendon subluxation
• Look specifically for a separate extensor pollicis brevis subcompartment, present in 55 per cent of patients in one surgical series and a common cause of persistent symptoms if missed
• Never perform a blind percutaneous release of this compartment

Snuffbox and radial styloid wires. Cadaveric work in which K-wires were placed into the anatomical snuffbox found a branch of the superficial radial nerve injured in 22 per cent of specimens and the cephalic vein in 33 per cent. The radial artery was never transfixed, but a wire lay within 1 mm of it in 22 per cent. The authors' conclusion is the one to remember: an open approach is preferred to percutaneous wiring in this region. The same applies to percutaneous scaphoid fixation from a dorsal or snuffbox start point and to radial styloid pinning of a distal radius fracture.

Distal radius external fixator pins. Superficial radial nerve injury is the commonest neurological complication of bridging external fixation of the distal radius. The pin technique:

• Make a formal open incision of adequate length, not a stab
• Spread bluntly to bone with scissors, protecting the branches and the cephalic vein
• Use a soft tissue protection sleeve and drill through it; never drill freehand through skin
• Place radial shaft pins in the interval between extensor carpi radialis longus and extensor carpi radialis brevis on the dorsoradial shaft, where the nerve has already emerged and can be seen, rather than on the radial border where it lies unseen

Thumb base surgery. Trapeziectomy, first carpometacarpal arthroplasty and arthrodesis, and thumb metacarpophalangeal ulnar collateral ligament repair all involve incisions that the nerve branches cross transversely. In a dorsoradial approach to the trapeziometacarpal joint the branches lie in the roof of the wound before the capsule is reached, so elevate skin flaps carefully and retract the branches with the flap, not against it. In a Wagner volar-radial approach the branches are encountered at the radial extent of the incision, and a prominent Kirschner wire or screw head on the radial column can tether or abrade a branch late.

Other mechanisms. The nerve is also injured by radial artery cannulation and intravenous cannulation in the snuffbox, and by glass and knife lacerations of the dorsoradial forearm, a very common mechanism in trauma practice. In radial forearm flap harvest it is at risk along the length of the donor site.

Recognised at operation. Direct epineurial repair under magnification is the first choice when the ends are clean and can be brought together without tension. Sensory recovery in the territory is usually acceptable and, more importantly, an intact repair is far less likely to produce a painful neuroma than a free stump.

Gaps and irreparable ends. A short gap takes a conduit or a short interpositional graft; donors include the terminal posterior interosseous nerve from the fourth dorsal compartment, the medial antebrachial cutaneous nerve or a short segment of the same nerve distally. If repair is impossible, do not simply ligate and leave the stump in subcutaneous fat. Resect the stump proximally and bury it in muscle, away from the wrist and away from any pressure point.

Recognised afterwards. A new dysaesthetic territory after surgery with a Tinel sign at the incision is a divided or tethered branch. Early on, reassure, desensitise and observe, because a traction injury or a branch caught in a suture may settle. Persistent, well-localised pain with a positive diagnostic block at 3 to 6 months warrants exploration. The findings are usually a branch tethered in scar, treated by neurolysis, or a bulb neuroma, treated by excision with proximal transposition.

As a graft donor. The superficial radial nerve is a recognised donor for sensory nerve grafting when the sural nerve is unavailable, or when a small-calibre, well-matched graft is needed for digital nerve reconstruction. It can be harvested as a vascularised nerve graft on the radial artery, which historically made it attractive for scarred beds.

Donor morbidity is the objection. A numb dorsoradial hand is well tolerated, but the donor site is exactly where neuromas hurt most. For that reason it sits behind the sural and medial antebrachial cutaneous nerves in most surgeons' order of preference.

As a diagnostic resource. Superficial radial sensory nerve biopsy is a standard alternative to sural nerve biopsy when a distal sensory nerve specimen is required, for example in suspected vasculitic neuropathy or amyloid.

As a recipient. In selected reconstructions the nerve can be used to reinnervate a sensory flap on the dorsum of the hand or thumb. In a first dorsal metacarpal artery (kite) flap, the nerve branch travelling with the pedicle can be coapted to a digital nerve to give the flap protective sensation.