The Intrinsic Hand: Intrinsic-Minus (Claw) and Intrinsic-Plus Deformity

INTRINSIC-PLUS is the opposite of claw: the intrinsics are too tight (contracture) or spastic, so they overpull, FLEXING the MCP joints and EXTENDING the IP joints, often with the thumb adducted. Causes include spasticity (stroke, cerebral palsy, traumatic brain injury), ischaemic (Volkmann) contracture, trauma/scarring of the intrinsic compartments, rheumatoid disease, and prolonged immobilisation in the wrong position (which is exactly why the protective 'intrinsic-plus' position - MCP flexed, IP extended - is used to splint a hand, since it keeps the collateral ligaments at length). The BUNNELL intrinsic tightness test detects intrinsic tightness: hold the MCP in EXTENSION (which stretches the intrinsics) and try to flex the PIP - if PIP flexion is limited with the MCP extended but improves when the MCP is flexed (relaxing the intrinsics), the test is positive for intrinsic tightness. (If PIP flexion is instead worse with the wrist/MCP flexed, suspect extrinsic/extensor tightness.)

• First, exclude an acute cause. An intrinsic-plus posture that has appeared acutely in a swollen, tender hand is an acute compartment syndrome needing decompression, not any of the measures below.
• Non-fixed/spastic: stretching and splinting, and for spasticity, botulinum toxin (or systemic/ neurosurgical spasticity management) to reduce intrinsic overpull.
• Fixed contracture: a DISTAL INTRINSIC RELEASE - excising the oblique fibres (the medial portion of the lateral bands/extensor expansion) distal to the MCP - relieves the deforming intrinsic pull while preserving MCP flexion; more proximal releases or muscle slide are options in severe ischaemic contracture.
• Treat the underlying cause (e.g. established Volkmann's) and rehabilitate.

• Force re-routing at the MCP. With the intrinsics gone, the long extensor's force is no longer shared with the MCP flexors, so it concentrates on hyperextending the MCP. Once the MCP sits hyperextended, the long extensor is "used up" proximal to the IP joints and can no longer transmit effective extension distally, so the IP joints stay flexed under the unopposed long flexors.
• Secondary soft-tissue changes. A persistently hyperextended MCP attenuates the volar plate and contracts the dorsal structures, while a persistently flexed PIP contracts the volar plate, checkrein ligaments and skin and lets the lateral bands drift volar - so a deformity that began as a purely dynamic imbalance becomes a fixed contracture that no transfer alone can overcome.
• Why this dictates the sequence. This is why a fixed PIP flexion contracture must be released/splinted to supple before any claw-correcting transfer, and why early anti-claw (lumbrical-bar) splinting - which blocks MCP hyperextension and forces the long extensor to reach the IP joints - both relieves the posture and prevents the cycle from fixing.
• The over-correction trap. Once MCP hyperextension is blocked or a transfer restores intrinsic pull, the same lateral-band system can tip the other way into a swan-neck (PIP hyperextension) deformity, which is why over-tensioning is avoided and PIP hyperextension is guarded against - the mechanism is set out in swan-neck deformity.

• What is lost. A low ulnar lesion paralyses the adductor pollicis, the deep head of flexor pollicis brevis and the first dorsal interosseous, so key pinch is weak and the index finger cannot be held abducted to stabilise pinch.
• Froment's sign. Asked to pinch a card, the patient flexes the thumb IP joint using FPL (median) to substitute for the weak adductor pollicis - a flexed thumb IP on pinch is Froment-positive.
• Jeanne's sign. The compensating long flexor pull, plus the lost adductor/FPB stabiliser, drives the thumb MCP into hyperextension during the same pinch - the MCP counterpart of Froment's.
• Wartenberg's sign. The little finger rests abducted because the ulnar-innervated third palmar interosseous (which adducts it) is out while the radially-innervated extensor digiti minimi is unopposed - the patient cannot adduct the little finger to the others.
• The intrinsic-plus thumb. In the tightness/spasticity (plus) hand the same adductor/intrinsic overpull drives a thumb-in-palm adduction posture that obstructs grasp.
• Reconstruction in principle. Beyond correcting the finger claw, restoring pinch may need an adductorplasty (a tendon transfer to recreate thumb adduction) and restoration of index abduction; the spastic thumb-in-palm needs tone management and selective release. The specific transfers are set out in tendon transfers for nerve palsy, and the broader ulnar work-up in ulnar nerve palsy and cubital tunnel syndrome.

The lumbrical origin/insertion, weak-but-proprioceptive role and the functional importance of the radial two lumbricals come from the cited Wang study, and the comparison of FDS four-tail versus Zancolli lasso for ulnar claw (and the prognostic effect of PIP extensor lag/paralysis duration) from the cited Ozkan study - a comparison of 25 patients, so the finding that the four-tail corrects the deformity better while the lasso restores more grip should be quoted as a small single-series signal rather than a settled result. The intrinsic action (flex MCP/extend IP), the ulnar paradox and the Bouvier/Bunnell tests are standard, well-established hand teaching. Leprosy is widely taught as the commonest cause of claw hand worldwide and is stated here on that basis; a PubMed search did not return a study giving a proportion or denominator for that claim, so no figure is quoted. The tenodesis (cascade) discriminator, the Bunnell-Littler naming and the Tsuge grading are taken in substance from our Volkmann's ischaemic contracture topic, and the hand-compartment decompression detail from region-specific compartment syndrome. See also ulnar nerve palsy, cubital tunnel syndrome, tendon transfers for nerve palsy and swan-neck deformity.