Extensor Pollicis Longus Rupture

Attritional rupture of the extensor pollicis longus is the commonest closed extensor tendon rupture of the hand and the prototype of attritional rupture at a fibro-osseous pulley. The classic story - sudden, painless loss of thumb extension weeks after an apparently trivial distal radius fracture - is an exam favourite because it tests tendon biomechanics, third-compartment anatomy and the principles of tendon transfer in a single question.

The numbers. Rupture follows about 0.3% of all distal radius fractures but 5% of strictly non-displaced ones, so the benign-looking fracture is the risk group. It declares itself 4-12 weeks after the injury as a rule, with reported cases from one month to a year, and it happens even in a cast.

What the patient loses. The EPL is the only tendon that retropulses the thumb, so its rupture takes thumb IP and MCP extension with it and leaves the thumb lying flat against whatever the hand rests on. Tip pinch, key pinch and grip strength all fall, and patients describe being unable to rest the hand on a table, climb, or pick a coin or other flat object off a surface. The hand also loses its palm-to-fist contour during grip as the thumb collapses into adduction.

The tendon and its tunnel. The EPL is the only tendon in the third dorsal compartment, and within it the tendon makes a sharp angular turn of 30-45 degrees around the Lister tubercle, a bony ridge of the distal radius that works as a fixed pulley. It inserts on the dorsal base of the distal phalanx and extends the IP and MCP joints while retropulsing the CMC joint, all in one line of pull. The extensor pollicis brevis, which travels in the first compartment with the abductor pollicis longus, extends the MCP joint only.

View larger

The two hits. The first is ischaemic. The intrinsic blood supply of the tendon is poorest at the level of the tubercle, a watershed zone, and the third compartment is a tight fibro-osseous tunnel in which haematoma, oedema, tenosynovitis, rheumatoid pannus or cast compression raise the pressure further. The second is mechanical: the weakened tendon grinds over an unforgiving bony ridge with every thumb movement. The two summate, and the tendon gives way at or just proximal to the Lister tubercle, often inside a retinaculum that looks intact - a combination unique to the EPL.

Why the undisplaced fracture is the dangerous one. A non-displaced fracture bleeds into an intact third compartment and raises the pressure inside it, and the cast then holds a retinaculum-bound tendon under sustained compression against bone. A radius that heals short sharpens the angle the tendon has to turn through, and the tendon frays and gives way with no fresh injury to explain it.

The volar-plate era version. Volar locking plating changed the mechanism. The plate is fixed from the front and its screws are measured from the front, so every millimetre a locking screw overshoots the dorsal cortex is a millimetre protruding into an extensor compartment - and the third compartment holds the EPL. Perry's cadaveric study passed a wire through every hole of three common volar plates and found that every plate has specific high-risk holes, the central screws most often, whose trajectory points directly at the EPL groove. This is not random bad luck; it is predictable from the plate design and the hole chosen.

Two rules follow. A screw is either long or it is not - there is no "slightly proud" - so check with a dorsal tangential view rather than trusting a lateral; Kiyak found the view detected screw penetration with no false positives in his series. And when an EPL ruptures after plating rather than after a simple fracture, think hardware first and check the plate position and the screw lengths before putting it down to attrition, because the management then includes dealing with the offending screw.

The rheumatoid wrist: caput ulnae. Dorsal synovitis attenuates the ECU subsheath and the ECU subluxates volarly; the DRUJ is destroyed and the ulnar head erodes dorsally into the caput ulnae. The extensors lying over that prominence fail first and the EPL, on the radial side, ruptures later as synovitis reaches the third compartment. The EIP may itself have ruptured, and the FPL may be failing at the scaphoid in the same hand - the Mannerfelt overlap. The EPL is easy to miss in a rheumatoid hand unless it is tested specifically.

Vaughan-Jackson syndrome. The extensor ruptures follow a named, examinable pattern: sequential, progressive rupture of the finger extensors beginning on the ulnar side and marching radially - extensor digiti minimi first, then the EDC to the small finger, then the ring, then the middle. The cause is attrition over the dorsally prominent, eroded distal ulna combined with dorsal tenosynovitis, since the most ulnar tendons sit directly over the offending bone. The clinical clue is a rheumatoid patient who cannot extend the little finger MCP joint, then weeks later the ring: the droop is at the MCP joints and passive tenodesis is lost.

Its flexor mirror. Mannerfelt syndrome is attritional rupture of the FPL over a scaphoid or trapezial osteophyte inside the carpal tunnel - the same disease on the opposite side of the wrist. Finding an EPL rupture in a rheumatoid hand should therefore prompt a search for ulnar extensors that have already gone, and any transfer here is combined with dorsal tenosynovectomy and treatment of the distal ulna (Darrach or Sauvé-Kapandji) to stop further ruptures.

Two axes matter. The aetiology dictates the reconstruction - transfer for attrition, repair for laceration - and the onset determines whether the bed of the tendon is healthy enough for a repair at all.

Metal belongs on the list too - a scaphoid screw, an external fixator pin, a prominent dorsal plate - and a knife, glass or a saw divides the tendon outright.

History. The rupture is usually described as spontaneous, so the questions that make the diagnosis are about the wrist rather than the thumb: a distal radius fracture, even one three to six months ago, rheumatoid disease, Kienböck disease, previous wrist surgery, scaphoid screws. Sudden, painless loss of thumb extension is the classic account, and the absence of pain at the moment of rupture is itself an important feature.

The retropulsion (drum) test. Lay the palm flat on a hard table with the fingers adducted and the thumb alongside the index, then ask the patient to lift the thumb straight up off the table. An intact EPL lifts it cleanly, retropulsing the CMC joint while extending the MCP and IP joints. After rupture the thumb stays down: the EPB may still produce a small amount of MCP extension, but it cannot retropulse the thumb or extend the IP joint. That is the value of the test - it isolates EPL from EPB, and a patient with a strong EPB and a positive sign has an EPL rupture until proven otherwise.

View larger

The rest of the examination. Four more checks, every time:

• Tenodesis test - wrist flexion no longer extends the thumb IP joint, because the EPL is discontinuous
• Palpation over the Lister tubercle - the tendon is no longer there, and a gap may be felt
• Every other extensor, and the FPL - the caput ulnae and Mannerfelt spectrum takes more than one tendon
• Finger and wrist extension - intact, which rules out a posterior interosseous nerve palsy

The diagnosis is clinical. Imaging is for the cause, and it identifies the cause far more often than it confirms the rupture - so in the post-fracture case, do not delay an EIP-to-EPL transfer waiting for an MRI.

Radiographs are mandatory and come first: PA, lateral and oblique views of the wrist and distal radius. Look for the healed fracture and its callus, rheumatoid change (ulnar styloid erosions, caput ulnae, carpal collapse), a sclerotic lunate, a scaphoid nonunion, and any screw or plate lying over the third compartment.

Ultrasound answers the equivocal examination - a palpable gap, a suspected partial rupture, or in-continuity degeneration - by showing discontinuity of the EPL fibres, the retracted stump, fluid in the third compartment and tenosynovial thickening. It is dynamic and cheap, needs no contrast, and can be compared with the other side.

MRI is a planning study rather than a diagnostic one: chronic ruptures with several tendons involved, and assessment of the third-compartment bed. It shows the gap, the scar tissue, marrow oedema in the lunate and active tenosynovitis with enhancement, which is what makes it valuable in a rheumatoid hand, where every ruptured tendon must be mapped before reconstruction.

The operation follows the tendon bed. An attritional rupture leaves frayed, retracted ends in a diseased third compartment and is reconstructed by transfer; a clean laceration in a healthy bed can be repaired directly.

Who does well. A young patient after a distal radius fracture, with a healthy contralateral hand, reconstructed within 3 months and rehabilitated under supervision.

Who does less well. The elderly rheumatoid hand with several ruptured tendons, a severe caput ulnae, and a late presentation with an established thumb adduction contracture; here the result depends on disease control.

The threshold. Somewhere between 3 and 6 months the outcomes begin to decline steadily, though delayed reconstruction still works. The transfer itself is reliable; the limiting factors are the bed, the disease and the rehabilitation.