Hamate Fractures

Hamate fractures make up 2-4% of carpal fractures, and 80-90% of them are fractures of the hook. The hamate is the most ulnar bone of the distal carpal row, and the hook is what makes it interesting: it sits against the ulnar nerve, the flexor tendons to the ring and small fingers, and the handle of whatever the patient was gripping. A hook fracture is easy to miss, slow to declare itself and often fails to unite on its own, so the injury is small and the cost of ignoring it is not.

Mechanism. The classic hook fracture is a direct blow from the handle of sporting equipment striking the palm: a golf club (the most common cause, at the grip pressure point), a racquet in tennis, squash or badminton, the handle end of a bat in baseball, or a weightlifting bar or dumbbell handle in the gym. The other route is avulsion, through the pisohamate or pisometacarpal ligament, both of which attach to the hook. Stress fractures are rare and follow repetitive grip activity.

Body fractures are a different injury in a different patient. They follow a high-energy axial load, such as a punch, and are often part of a dorsal dislocation of the 4th and 5th CMC joints.

Who. Peak incidence is at 20-40 years, the majority are sports-related, and the diagnosis is often delayed by weeks to months.

Why the diagnosis is missed, and what it costs. Hook fractures are missed in up to 50% of initial presentations: standard radiographs often fail to show the hook, the symptoms are put down to a "wrist sprain" or "tendinitis", and associated injuries distract from the hook. The commonest consequence by far is nonunion, which follows almost as a rule rather than as a risk: in Scheufler's series five of six conservatively treated fractures went on to symptomatic nonunion. The others are ulnar nerve motor branch injury, chronic pain with grip weakness, and the feared late sequel, attritional rupture of the flexor tendons to the ring and small fingers, which is taken up under Complications.

The body. The hamate body articulates with the capitate radially, the triquetrum proximally and the bases of the 4th and 5th metacarpals distally. The joint with the triquetrum is helicoid, allows the dart-throwing motion, contributes to the ulnar column of the wrist and is important for grip strength. Distally the 4th metacarpal base is relatively immobile while the 5th allows 20-30 degrees of flexion, and that CMC motion is important for power grip.

The hook (hamulus). The hook projects 10-12mm from the volar surface of the body and is the attachment for the flexor retinaculum, the pisohamate and pisometacarpal ligaments and the hypothenar muscles. It forms the ulnar pillar of the carpal tunnel and the radial wall of Guyon's canal, which is why a hook fracture can produce both median nerve symptoms (carpal tunnel compression) and ulnar nerve symptoms (motor branch palsy).

The ulnar nerve. The motor branch curves around the radial aspect of the hook. Injury weakens the intrinsic muscles: Froment's sign becomes positive as FPL substitutes for thumb adduction, and first dorsal interosseous weakness impairs key pinch.

The ulnar artery passes superficially in Guyon's canal. A hook fracture can cause thrombosis or a pseudoaneurysm, and repetitive trauma to the same spot produces hypothenar hammer syndrome.

The flexor tendons. FDP to the ring and small fingers pass directly over the volar aspect of the hook, so a sharp fracture edge can wear through them. That attrition rupture is uncommon and has no established population rate, the literature being case reports and small series, but it is the reason a symptomatic nonunion is not left alone.

Blood supply. The hamate as a whole is supplied from several sources: dorsal branches from the dorsal carpal arch, volar branches from the palmar carpal arch and direct branches from the ulnar artery. The hook is not so fortunate. It receives single or few nutrient vessels from the ulnar artery, a tenuous supply that explains the high nonunion rate with conservative treatment.

Why the hook does not unite. Several things work against a hook fracture left in a cast: the tenuous blood supply, tensile force from the pisohamate ligament pulling on the fragment, motion transmitted by the adjacent flexor tendons, and the fact that a delayed presentation is already a nonunion by the time it is seen. That reasoning underlies the whole of management.

History. The hook fracture patient plays golf (the most common), tennis, baseball or squash, felt a "pop" or sharp pain with a swing, and has pain that is worse with gripping. Presentation may be delayed by weeks, and there may be paraesthesias in the ring and small fingers from the ulnar nerve. The body fracture patient has had a high-energy injury, a fall, a punch or a road traffic collision, or a direct blow to the dorsum of the hand, with immediate swelling and deformity, often in a polytrauma setting and with associated metacarpal injuries.

Examination. The hook lies 2cm distal and radial to the pisiform, and point tenderness there is the finding to seek. Grip is painful and weak compared with the other hand, and the pull test, resisted DIP flexion of the ring and small fingers, reproduces pain at the hook.

The ulnar nerve. Assess the motor branch in every suspected hook fracture. Test the first dorsal interosseous by index finger abduction against resistance, the palmar interossei by finger adduction (the paper grip test), and thumb adduction for Froment's sign, where FPL substitution is positive. Sensation is tested by two-point discrimination of the small finger and light touch over the hypothenar eminence (palmar cutaneous branch), and small finger abduction at rest, Wartenberg's sign, suggests an ulnar nerve palsy.

The flexor tendons. Test FDP to the ring and small fingers at every visit, because rupture, especially of the small finger tendon, is a late complication of the missed fracture. Stabilise the PIP joint to isolate the DIP, ask for active DIP flexion, compare strength with the other side and palpate for tendon continuity at the wrist.

Differential diagnosis. Ulnar-sided wrist and hypothenar pain has a short list of alternatives, and each has a finding that separates it from the hook.

Radiographs. On the standard PA view the body is visible but the hook is often obscured, and the lateral shows it poorly; standard views miss up to 50% of hook fractures. Obtain them if a fracture is suspected, and add the projections that profile the hook:

• Carpal tunnel view: wrist dorsiflexed, beam parallel to the palm
• 45-degree supinated oblique: profiles the hook
• Lateral with supination: improves hook visibility

These are more sensitive but still miss fractures. Andresen's cadaver study puts a number on it: even a complete radiographic series including the carpal tunnel view reached a sensitivity of only 72.2% against 100% for CT, so roughly one fracture in four escapes a full series and negative films do not exclude the diagnosis.

View larger

View larger

CT is the gold standard, with a sensitivity that approaches 100%. Get it when clinical suspicion persists despite negative radiographs, for preoperative planning before ORIF, and to evaluate union after conservative treatment. The protocol is 1mm axial slices through the carpus with sagittal and coronal reconstructions, comparing with the contralateral side if needed. It shows the fracture line through the hook at any level, displacement and fragment size, the signs of nonunion (sclerosis, gap, cyst formation) and any associated body fracture.

View larger

View larger

The decision. Timing, the level of the fracture and what the patient needs from the hand decide it. A hook fracture seen acutely, within 2 weeks, with a large basal fragment can be fixed, or excised early in an athlete; a symptomatic nonunion or a presentation beyond 6 weeks is excised; a displaced body fracture is fixed to restore the CMC joint. Conservative treatment of the hook fails often, with up to 50% going on to nonunion, for the reasons set out under Anatomy.

Excision or fixation. The evidence needs stating carefully. Excision is what almost all published patients received, it returns athletes slightly sooner than ORIF (6 versus 7.8 weeks in Donohue's review) and it removes any possibility of nonunion; most sports medicine surgeons recommend early excision for athletes who need to return to grip sports, citing predictable outcomes and faster recovery. The same review, though, could not determine a consistent difference between excision and ORIF, and Scheufler's series found the two comparable, so the optimal operation is not established and the choice is a judgement about this fracture and this patient.

Early and late. The early complications are nonunion after conservative treatment, ulnar nerve motor branch palsy, ulnar artery thrombosis, infection if the injury was open, and complex regional pain syndrome. The late ones are the FDP rupture above, chronic pain and grip weakness, chronic ulnar neuropathy from compression, CMC arthrosis after body fractures, and hypothenar atrophy.

The ulnar nerve. Where the nerve is injured along Guyon's canal determines what is lost, and hamate fractures injure it in one place above all.

A Zone II injury, to the deep motor branch as it passes around the hook, is the most common pattern with hamate fractures: weakness of the interossei and hypothenar muscles, a positive Froment's sign from adductor pollicis weakness, and visible first dorsal interosseous atrophy, with no sensory loss because the sensory branch takes a separate course. It usually improves after hook excision and decompression.

Grip after excision. The concern that removing the hook will weaken grip is reasonable and the evidence answers it. Grip strength averages 80-90% of the contralateral side after excision, there is no significant difference between excision and ORIF in athletes, a larger handle diameter compensates for any weakness, return to elite golf and racquet sports is documented, and patient satisfaction is high despite the measurable decrease. Set against ongoing pain or the tendon rupture risk of a nonunion, the reduction is minimal and well tolerated.

The athlete. For an athlete who needs a reliable return to grip sports, excision is preferred over conservative treatment or ORIF: 85-95% return to full sport, the timeline is predictable, there is no risk of nonunion because the hook is gone, grip is acceptably preserved, and return to elite-level golf, tennis and baseball is documented. ORIF may offer slightly better grip strength if union is achieved, but 15-25% require excision anyway because of nonunion, the rehabilitation is longer, and it is reserved for select acute cases with large fragments.

Body fractures. The prognosis is set by what was achieved at surgery and by the injury itself.

• Good: anatomic reduction, stable fixation, early presentation, an isolated injury without CMC dislocation, good compliance
• Poor: articular comminution, associated CMC dislocation, delay to treatment, polytrauma, tobacco use

In the long term. After hook excision most patients have no lasting problems; grip strength stabilises by 3-6 months, hypothenar weakness is rare and follows motor branch injury, and recurrent symptoms are very rare and usually mean the excision was incomplete. After body ORIF, CMC arthrosis may develop over 10-20 years, more likely with comminution or residual incongruity; it is usually manageable with activity modification, and CMC arthrodesis is rarely needed.

Global Epidemiology

Guidance and Society Positions

No major orthopaedic body (AAOS, BOA/BSSH, NICE, AO Foundation, EFORT) publishes a disease-specific clinical practice guideline dedicated to hamate fractures; management is guided by hand-surgery consensus and the systematic-review evidence below. There is no carpal-fracture-specific national joint registry (the AOANJRR/NJR/AJRR cover arthroplasty, not carpal trauma), so registry-level outcome data are not available for this injury.

Practice Variation

Sample MCQ Concepts