Palmaris Longus Anatomy

Origin. The medial epicondyle of the humerus via the common flexor origin, with additional fibres from the adjacent intermuscular septa and the deep antebrachial fascia.

Belly and tendon. A short, fusiform belly in the proximal third of the forearm gives way to a long, thin, round tendon that occupies the distal two-thirds. The tendon is typically 12 to 15 cm in usable length and 3 to 5 mm in diameter. Both dimensions make it ideal as a free graft: the length is adequate for most upper limb reconstructions, and the diameter is ideal for collateral ligament and pulley reconstruction.

View larger

Insertion. The flexor retinaculum (transverse carpal ligament) and, in direct continuity, the apex of the palmar aponeurosis. Some fibres continue into the thenar fascia and the abductor pollicis brevis region, the anatomical continuity exploited in the Camitz opponensplasty, where a strip of palmar aponeurosis is harvested in continuity with the tendon to gain length.

The defining fact. Palmaris longus passes superficial to the flexor retinaculum. It is the only long flexor tendon of the forearm that is not a carpal tunnel structure, so it is not decompressed by carpal tunnel release, and a structure identified at wrist harvest that dives deep to the retinaculum is not palmaris longus.

Nerve supply. The median nerve, root values C7 and C8 (a C6 contribution is described in some texts). The branch arises in the proximal forearm and enters the deep surface of the short proximal belly, in the proximal third of the forearm. It comes third in the order of median innervation in the forearm:

1. Pronator teres
2. Flexor carpi radialis
3. Palmaris longus
4. Flexor digitorum superficialis

The anterior interosseous nerve then supplies flexor pollicis longus, the radial half of flexor digitorum profundus and pronator quadratus.

Clinical significance. Palmaris longus weakness is essentially never diagnosed clinically: it is too weak a muscle to assess reliably and too often absent for a negative finding to mean anything. Its relevance to nerve injury is entirely surgical. It is a median-innervated donor, so in a high median nerve palsy it is denervated and unavailable as a transfer, exactly as flexor digitorum superficialis is.

Arterial supply. Muscular branches of the ulnar artery are the dominant supply to the belly, with the posterior ulnar recurrent artery proximally at the common flexor origin; a persistent median artery may contribute. The tendon itself is relatively avascular, nourished by paratenon vessels and by diffusion.

Why it matters for grafting. Because the tendon is thin, round and relatively avascular, it survives as a free graft by revascularisation from the recipient bed, in the same way as any other free tendon graft. The graft therefore must be laid in a healthy, vascularised bed. A scarred, irradiated or infected bed is the reason to stage the reconstruction with a silicone rod rather than to place a graft primarily.

Across the volar wrist, radial to ulnar. Radial artery, flexor carpi radialis, the palmar cutaneous branch of the median nerve, palmaris longus, the median nerve (deep), flexor digitorum superficialis, flexor digitorum profundus, the ulnar artery and nerve on the radial side of the flexor carpi ulnaris tendon, and flexor carpi ulnaris itself. Superficial to palmaris longus lie only skin, subcutaneous fat and the antebrachial fascia; it is the most superficial tendon at the volar wrist.

The median nerve. At the wrist the nerve lies deep and slightly radial to palmaris longus, the single most important surgical relation of this muscle. In some individuals it lies directly beneath the tendon instead, and that is the anatomy behind the reported catastrophe of a surgeon harvesting the median nerve believing it to be palmaris longus.

The palmar cutaneous branch. It arises 5 to 8 cm proximal to the wrist crease, runs in the interval between palmaris longus and flexor carpi radialis, radial to the palmaris axis, and pierces the fascia near the crease.

In the forearm. Palmaris longus lies between flexor carpi radialis and flexor carpi ulnaris in the superficial flexor layer, superficial to flexor digitorum superficialis. The median nerve is deep to flexor digitorum superficialis in the forearm and only becomes superficial, and adjacent to palmaris longus, in the distal few centimetres.

The figures. In a population study of 300 Caucasian subjects aged 18 to 40 (150 men and 150 women) assessed by clinical inspection, palmaris longus was absent unilaterally in 49 subjects (16 per cent, one in six people lacking it on one side) and bilaterally in 26 (9 per cent, roughly one in eleven with no palmaris longus at all). Both were more common in males, and in unilateral absence the right side was more commonly affected, but neither difference reached statistical significance: state the right-sided excess as a trend.

Two derived figures. Both follow from the counts:

• Any absence - 49 unilateral + 26 bilateral = 75 of the 300 subjects, 25 per cent. One patient in four is missing a palmaris longus somewhere; this is the figure that should shape your expectation in clinic, and the study does not print it.
• Absence per limb - (49 x 1) + (26 x 2) = 101 absent tendons in 600 limbs, about 17 per cent, or roughly one limb in six. This is the rate that applies when you need a graft from one nominated arm. It is close to the unilateral figure by coincidence, not because they measure the same thing.

Why the number matters. A patient consented for a thumb ulnar collateral ligament reconstruction, an elbow medial collateral ligament reconstruction, a flexor tendon graft or a pulley reconstruction on the assumption of an available palmaris longus has roughly a one in six chance of the tendon being absent on the operative side if it is never checked. The examination takes ten seconds. Not doing it is indefensible.

Population variation. Reported agenesis rates vary widely between populations. Rates in European-descended populations are generally the highest, in the region reported above, whereas published series from East Asian and African populations report substantially lower rates of absence. The principle outweighs any one group's figure: a surgeon should quote local or study-specific data rather than assume a universal figure, and should always examine the individual patient rather than rely on a population rate.

Inheritance. Absence appears to be inherited, with reported patterns most consistent with autosomal dominant transmission with variable penetrance. It is not associated with any specific syndrome in the general population.

Reversed palmaris longus. The muscle belly lies distally and the tendon proximally, the opposite of normal. The distal belly can present as a volar wrist mass and, by occupying space at the wrist, is a recognised cause of carpal tunnel syndrome and, less commonly, of ulnar nerve compression in Guyon canal. It is also useless as a conventional graft.

Palmaris profundus. An anomalous muscle running deep to the retinaculum, within the carpal tunnel, inserting into the palmar fascia or the thenar muscles. Because it lies inside the tunnel it is a genuine cause of median nerve compression, and unlike the normal palmaris longus it is not superficial to the retinaculum, which is the reason the retinaculum test is the definitive identification step at harvest.

Other variants include a digastric or bifid palmaris longus, with a central tendon between two bellies; a duplicated palmaris longus, occasionally providing two grafts; and accessory insertions into the thenar or hypothenar fascia.

Volar wrist mass. A volar wrist mass superficial to the retinaculum may be an anomalous palmaris belly, and an anomalous or reversed belly belongs in the differential of a volar wrist swelling alongside ganglion, lipoma, a persistent median artery and giant cell tumour of the tendon sheath. Dynamic ultrasound distinguishes them: a muscle belly contracts and moves with wrist flexion.

Dupuytren disease. The longitudinal pretendinous bands of the palmar aponeurosis, into which palmaris longus inserts, become the cords of Dupuytren disease. The continuity is direct, and a tense, prominent palmaris longus can make a pretendinous cord appear more obvious on wrist flexion. There is, however, no established causal association between palmaris longus presence or absence and the development of Dupuytren disease; be careful not to overstate this in an examination. At fasciectomy the tendon is encountered proximally at the wrist and is not routinely divided.

Injury. Palmaris longus is divided in almost any volar wrist laceration. Repair is not required, the tendon being functionally expendable, but its division is a useful marker that the laceration reached the deep fascia and that the median nerve must be formally examined. Spontaneous rupture is essentially unheard of and of no clinical significance.

The landmark, not the pathology. In carpal tunnel surgery the tendon lies superficial to the retinaculum and takes no part in the compression, but it defines where the incision goes.

The incision. Longitudinal, in line with the radial border of the ring finger, which places it ulnar to the palmaris longus axis, running from the distal wrist crease (or just distal to it) toward Kaplan cardinal line. Going ulnar keeps the incision away from the median nerve, deep and radial to palmaris longus, and from the palmar cutaneous branch, radial to it.

The crease. Do not cross the wrist crease transversely; cross obliquely if at all. A transverse scar over the crease is a classic route to a palmar cutaneous neuroma and a tender scar.

The ligament. Divide the transverse carpal ligament at its ulnar margin, immediately radial to the hook of hamate, keeping away from the median nerve and the recurrent motor branch. Depending on the Lanz variant, that branch may be extraligamentous, subligamentous or transligamentous.

Whose figures they are. Lanz's original operative series of 246 explored carpal tunnels described variations in the thenar branch course, accessory distal branches, high division and accessory proximal branches. The percentages everyone quotes alongside them are not Lanz's: the thenar branch was extraligamentous in 46 per cent, subligamentous in 31 per cent and transligamentous in 23 per cent in Poisel's 100 cadaver hands, which Lanz cites. Say whose data it is; an examiner who knows the paper is listening for it. The stated conclusion, to approach the median nerve from the ulnar side, is exactly the palmaris longus rule restated.

Beyond the carpal tunnel. The same relations guide other volar wrist procedures:

Exposure. A short transverse incision at the distal wrist crease, directly over the tendon. Incise skin only and spread bluntly; the tendon is immediately subcutaneous.

Confirm identity before dividing. One test is definitive:

Two further checks support it:

• Traction on it tensions the palmar aponeurosis and does not move a finger
• It is a tendon, firm and glistening white, without fascicular striations or longitudinal vasa nervorum. A nerve is soft, shows both, and does not glide with wrist flexion the way a tendon does, and in an awake patient traction on it produces paraesthesiae

Delivery. Divide distally, place a holding suture, and deliver the tendon proximally, either through one or two additional small transverse counter-incisions at intervals up the forearm or with a closed tendon stripper advanced proximally along the tendon to the musculotendinous junction. Preserve the paratenon on the graft where possible: it aids incorporation, and stripping it aggressively delays incorporation and increases adhesion. Confirm the graft length against the requirement before closing.

Hazards, and how to avoid them.

• Harvesting the median nerve - prevented absolutely by the retinaculum test, by never using a blind stripper before positive identification, and by operating with the anatomy explicitly in mind rather than by feel
• Blind stripper injury to the median nerve or to muscle if the stripper deviates from the tendon - advance it gently, in line, without force
• Palmar cutaneous branch injury from a transverse incision crossing the crease radially - keep the incision small and over the tendon, and cross the crease obliquely if extension is needed
• Injury to a superficial vein producing a haematoma - ligate rather than avulse
• Incomplete harvest where the tendon avulses at the musculotendinous junction - use a counter-incision if resistance is felt

What the graft is used for.

• Flexor tendon reconstruction: single-stage grafting and the second stage of a two-stage silicone rod reconstruction
• Thumb ulnar collateral ligament reconstruction for chronic gamekeeper thumb
• Elbow medial collateral ligament reconstruction, where palmaris longus is the classic donor for the docking and figure-of-eight techniques
• Pulley reconstruction (Kleinert-Weilby, Bunnell loop, Karev belt-loop) for A2 or A4 loss
• Scapholunate and other carpal ligament reconstructions, interposition arthroplasty and static suspension procedures
• Outside orthopaedics, eyelid slings for facial palsy, lip augmentation, tympanoplasty grafts and temporomandibular joint reconstruction, all exploiting the same expendability

If it is absent. The alternatives, set against palmaris longus:

Camitz opponensplasty. Indicated in severe carpal tunnel syndrome with marked thenar wasting and loss of palmar abduction, at the time of carpal tunnel release. Palmaris longus is harvested with a strip of palmar aponeurosis in continuity, which provides the extra length needed to reach the thumb without a separate graft, then tunnelled subcutaneously across the palm to the abductor pollicis brevis insertion and tensioned with the thumb in palmar abduction.

What it restores. Palmar abduction, not true opposition; it provides no meaningful pronation of the thumb. It is chosen for its simplicity and because it can be done through the same operation, not for the quality of the vector. That distinction must be explained preoperatively, so that the patient's expectation matches the operation.

Modifications. Adding a pulley at the ulnar remnant of the flexor retinaculum improves thumb flexion and pronation as well as abduction. A minimally invasive modification passes the transferred tendon beneath the abductor pollicis brevis fascia through palm and thumb incisions only, avoiding an incision across the wrist crease, and has been reported to reduce painful wrist scars and bowstringing of the transferred tendon compared with the conventional technique.

Its own complications. Bowstringing of the transferred tendon across the wrist is reduced by passing the transfer beneath the abductor pollicis brevis fascia or by using a formal pulley, and the painful scar of the conventional long incision straddling the wrist crease is the specific problem the minimally invasive modification addresses. Taking palmaris longus for a Camitz also removes it as a future graft donor in that limb.

Availability. Palmaris longus is median-innervated, so the Camitz is an operation for low median nerve palsy only. In a high median palsy, use extensor indicis proprius or an abductor digiti minimi (Huber) transfer.

Palmaris longus to extensor pollicis longus (radial nerve palsy). The standard third limb of the classical transfer set, alongside pronator teres to extensor carpi radialis brevis and flexor carpi radialis to extensor digitorum communis. Extensor pollicis longus is rerouted out of the third compartment to give a straight line of pull and to combine extension with palmar abduction of the thumb. Where palmaris longus is absent, the ring finger superficialis substitutes as the thumb extensor motor.