Kienböck's Disease

Kienböck's disease is avascular necrosis (AVN) of the lunate, leading to progressive collapse, carpal instability and wrist arthritis. It is a rare but disabling condition, and it falls on young, active patients.

Who. The peak age is 20-40 years, with a male predominance of 2:1, in manual labourers and athletes. It typically affects the dominant hand of a young worker. Bilateral disease is rare, less than 5%.

The aetiology is multifactorial.

Negative ulnar variance. It is present in 78% of patients against only 23% of the normal population, the ulna sitting 2-4mm shorter than the radius. A short ulna transmits more force through the lunate. In a neutral wrist 60% of axial load passes through the radius and 40% through the ulna; with 2mm of negative variance the lunate bears 95% of the radiocarpal load. Radial shortening osteotomy redistributes that load to the ulna.

Repetitive load and trauma. Cumulative stress on the lunate comes from repetitive wrist use: manual labour and jackhammer use, gymnastics and racquet sports. A prior wrist fracture or injury is also a risk factor.

Vascular anatomy. The extraosseous supply is profuse, with 2-3 dorsal and 3-4 volar vessels, but only 1-2 nutrient vessels enter the bone at each of the dorsal and volar poles (Gelberman). Capsular stripping, or a fracture crossing the centre, separates the bone from its entry points; it is stripping that devascularises the lunate.

Systemic associations.

• Corticosteroid use
• Systemic lupus erythematosus and vasculitis
• Gout and sickle cell disease
• Hypothyroidism and diabetes

The intraosseous patterns (Stahl classification). These are described from cadaveric injection studies rather than clinical angiography. Gelberman's study (PMID 7400565, 35 fresh cadaver limbs) describes three consistent intraosseous patterns, conventionally lettered Y, I and X, and found a dorsal-volar anastomosis in every specimen. The I pattern is conventionally taught as the least redundant, which is a reasonable reading of a single dominant intraosseous channel, but with an anastomosis in every lunate, "single vessel with no anastomosis" overstates it.

The vulnerable minority. The better modern basis is micro-CT work from van Alphen (PMID 27782999). Of 14 specimens, 12 had both volar and dorsal nutrient vessels and 2 had none dorsally; volar vessels were the more numerous (2.3 versus 1.4 on average). The authors therefore warn that a volar approach risks localised ischaemia in the no-dorsal-vessel subset.

Gelberman's conclusion. Gelberman's own conclusion favoured repeated compression fracture over a primary vascular deficiency as the cause of Kienböck's disease.

The cascade. From the first insult to end-stage wrist, the sequence runs:

1. Vascular insult or repetitive trauma
2. Ischaemia and bone marrow oedema
3. Osteocyte death and sclerosis
4. Structural collapse of the lunate
5. Scaphoid rotation and carpal instability
6. Progressive radiocarpal and midcarpal arthritis

Collapse. AVN weakens the subchondral bone. Repeated loading causes microfractures, progressive loss of height and eventually fragmentation. Once collapse begins (Stage III), progression is difficult to arrest, and early intervention aims to prevent it.

Lichtman. The modified Lichtman classification (2010) is the gold standard staging system, based on radiographic appearance and tied to treatment.

Stage IIIB is the critical prognostic divider. Scaphoid rotation means lunate collapse severe enough to destabilise the carpus, a fixed carpal collapse beyond salvage by joint-preserving techniques. Stages I-IIIA may benefit from joint levelling; IIIB-IV require salvage. That makes IIIA the last chance for joint-preserving surgery.

View larger

View larger

View larger

MRI staging. A five-stage MRI scheme is more sensitive than Lichtman for detecting early disease (its Stages 2-3) before X-ray changes, and it guides early intervention decisions.

Bain and Begg. Their classification is arthroscopic rather than MRI-based: it grades how many articular surfaces of the lunate and its neighbours are non-functional, and matches a procedure to each grade.

The zones. The 2017 integrated algorithm (Lichtman, Pientka and Bain) and several European centres pair the osseous Lichtman stage with the Schmitt perfusion/viability classification, which plain radiographs cannot show. Schmitt and Kalb described that a necrotic lunate remodels in band-shaped zones, demonstrable only on gadolinium-enhanced MRI:

• Proximal necrotic zone: dead marrow, no enhancement
• Intermediate reparative zone: hypervascular granulation or repair tissue that enhances avidly
• Distal viable zone: preserved marrow with normal perfusion

Intravenous gadolinium is mandatory, because only a perfusion study reliably separates the non-enhancing necrotic zone from the enhancing repair zone.

What it changes. The lunate with a substantial viable or reparative component is a candidate for radial shortening, a pedicled graft or a medial femoral condyle flap. The globally necrotic, non-enhancing lunate is effectively the "compromised lunate" of the 2017 "intact / compromised lunate / compromised wrist" framework, which this perfusion axis underlies. Reconstruction is unlikely to succeed, so PRC or fusion is the more honest choice.

Practical protocol. High-resolution CT best shows the osseous detail that stages the disease, the proximally located (stage IIIA) fractures and pancarpal arthritis (Stage IV), while contrast-enhanced MRI grades marrow viability. Used together, they classify Kienböck's disease by both morphology and perfusion, which plain films cannot do.

View larger

View larger

History. Onset is typically insidious over months. The patient describes dorsal wrist pain worse with activity, progressive stiffness and a weak grip, usually in the dominant hand of a manual labourer or someone whose work involves repetitive wrist use. 30% give a history of wrist injury.

Examination. Look for dorsal swelling over the lunate, tenderness over the lunate fossa and pain on axial loading of the third metacarpal. The Watson test may be positive (scaphoid instability). Flexion and extension fall to 50% of normal and grip by 30-50%, figures that depend on stage, as the table shows. The physical findings are non-specific, and imaging confirms the diagnosis.

Differential diagnosis. Dorsal or central wrist pain in a young adult has several mimics. MRI is the single best discriminator: AVN of the lunate shows diffuse low T1 marrow signal, whereas the alternatives have their own characteristic patterns.

Plain radiographs. First line: PA, lateral and clenched-fist views, read for the Lichtman findings above; scaphoid rotation (IIIB) is seen on the lateral. Measure ulnar variance on a neutral-rotation PA view.

MRI. The gold standard, and the only way to see Stage I disease. T1 shows low signal from marrow replacement, and T2 signal varies with stage. Bone marrow oedema and low T1 signal precede radiographic sclerosis by months, and early detection may allow non-operative treatment. MRI also assesses lunate viability and cartilage integrity.

CT. For surgical planning. It defines the extent of collapse, the fracture lines and articular involvement, and before salvage it assesses the capitate head (for PRC) or plans the fusion surfaces.

Bone scan. Rarely used, and of historical interest. It shows increased uptake in the lunate but is less specific than MRI, which has replaced it.

View larger

View larger

Conservative. Indicated for Stage I disease, for the patient who refuses surgery and for medical comorbidity. A short-arm cast or splint for 3-6 months aims to unload the lunate and allow revascularisation, with serial MRI every 3-4 months. The success rate in Stage I is 30-50%, although no series cited on this page quantifies arrest. Conservative treatment is rarely effective beyond Stage I and is temporising in most cases.

Adjuncts. Avoid heavy manual work and impact loading, with an ergonomic assessment and job modification or retraining where needed. NSAIDs control pain. Bisphosphonates are experimental with limited evidence, and iloprost (a prostacyclin analogue) may improve vascularity.

Joint levelling. Indicated for Stage II-IIIA with negative ulnar variance and a viable lunate. Radial shortening osteotomy is the most common: shortening the radius 2-4mm decreases lunate loading by 45% and redistributes load to the ulnocarpal joint. Pain relief is 60-80%, and it works best in Stage II. Ulnar lengthening is the alternative; it preserves radial length at the cost of a higher ulnar nonunion risk and DRUJ stiffness.

What levelling does not do. Progression is common without surgery, and radial shortening may slow it. The pooled long-term evidence (Shin et al) found Lichtman-stage progression no different after radial osteotomy than after non-operative care, although pain and motion were better.

Revascularisation. Indicated for Stage II-IIIA with a viable lunate on MRI, particularly in young patients. The most common graft is the 4+5 ECA pedicle graft: vascularised bone harvested from the dorsal distal radius on the 4th and 5th extensor compartment artery and inset into a drilled lunate core. It may restore lunate vascularity and does best in Stage II. Pain relief is 70% and success rates vary from 40-80% on moderate-quality evidence, though no series cited on this page quantifies it. The alternatives are a pisiform vascularised graft and the medial femoral condyle free flap (microsurgery). All are technically demanding with variable results.

View larger

Salvage. Indicated for Stage IIIB-IV, failed joint-preserving surgery and the painful wrist. Salvage surgery trades motion for pain relief and stability.

Proximal row carpectomy. The motion-preserving salvage for Stage IIIB without arthritis. The scaphoid, lunate and triquetrum are excised so that the capitate head articulates with the lunate fossa of the radius. It gives 80% pain relief, preserves 50% of wrist motion and leaves grip at 70% of normal, with good 10-year results. It needs intact cartilage on the capitate head and in the lunate fossa, and no midcarpal arthritis, assessed before surgery on CT or MRI; capitate head, radiocarpal or inflammatory arthritis contraindicates it.

View larger

Limited intercarpal fusion. Four-corner fusion joins the capitate, hamate, lunate and triquetrum and excises the scaphoid; it gives reliable pain relief and preserves 50% of motion. Radiolunate fusion joins the radius to the lunate and preserves some motion, and capitolunate fusion joins the capitate to the lunate. STT and scaphocapitate fusion are covered below.

Total wrist fusion. For Stage IV with generalised arthritis, or salvage of a failed PRC or fusion. The radius, carpus and metacarpals are fused with a plate. Pain relief is excellent and grip strong, but there is no motion and a significant limitation of daily activities.

Wrist arthroplasty. Reserved for low-demand patients and bilateral disease, to preserve motion in one wrist. It preserves motion at the risk of loosening or instability, and the young, high-demand Kienböck's patient is not an ideal candidate.

Arthroscopic scaphocapitate fusion. Scaphocapitate fusion is one limited intercarpal fusion used for stage III disease; the figures follow the arthroscopic technique from surface preparation to union.

View larger

View larger

View larger

View larger

View larger

Technique. The scaphoid, trapezium and trapezoid are fused with K-wires, staples, a circular plate or a memory-alloy concentrator, plus local bone graft, with the scaphoid fixed in a corrected, extended position. Unlike lunate excision alone, this preserves carpal height. In advanced disease it is frequently combined with debridement or excision of the necrotic lunate.

View larger

View larger

Results and the trade-off. Pain relief is reliable, grip recovers to roughly 60 to 65 percent of the opposite side, and most patients return to their original occupation. Wrist motion is reduced, radial and ulnar deviation especially, because the radial column is locked. Nonunion is the main early risk, and union typically takes about 7 to 11 weeks. Radioscaphoid arthrosis is the characteristic late complication, in around one quarter of cases at long follow-up.

View larger

Where it fits. STT fusion competes with radial shortening (for negative variance) and with the salvage options, PRC and four-corner fusion. It is most attractive when a joint-preserving load-shift is still wanted, but candidates must accept the loss of deviation and the risk of later radioscaphoid wear. The four-corner-fusion and proximal-row-carpectomy topics cover the fusions it sits alongside.

Arthritis after PRC. The table's 20% at 10 years sits beside Wagner's series of 144 carpectomies, which found radiocapitate arthrosis in 45% at a mean 13.4 years; it did not correlate with clinical outcome or revision.

CRPS. It follows 2-5% of wrist operations. Female sex and prolonged immobilisation are risk factors, early motion and vitamin C supplementation are the preventive measures, and management is by a multidisciplinary pain team.

Stiffness. Common after any wrist surgery, and worse with prolonged immobilisation. Prevent it with early therapy and by limiting postoperative casting to 6 weeks at most.

Return to work depends on the procedure:

• Desk work: 2-4 weeks
• Light manual work: 8-12 weeks
• Heavy manual labour: 12-16 weeks, or job modification

Surveillance. Annual radiographs for the first 3 years detect progression, with MRI if symptoms worsen or new pain appears. If a joint-preserving procedure fails, plan salvage.

Natural history. Untreated, the disease progresses over years to decades, and collapse and arthritis are the endpoint. Progression is inevitable without intervention in most cases: early disease (Stages I-II) may stabilise with treatment, progressive collapse is likely in IIIA, and by IIIB-IV the arthritis is irreversible. The timeline is 3-10 years from Stage I to IV, and a variable 3-15 years from onset to end-stage.

• Stage I: 50% progress to Stage II within 2 years
• Stage II: 80% progress to Stage III within 3-5 years
• Stage IIIB: 100% progress to Stage IV without treatment

Prognosis. Early diagnosis is critical to prevent collapse, and early diagnosis and intervention improve outcomes substantially.

Global epidemiology. Kienböck's disease is rare worldwide; most hand surgeons see only a handful of cases per year. It clusters in men aged 20-40, frequently in manual or vibration-exposed occupations (construction, drilling, mining). Negative ulnar variance is over-represented compared with the general population, though it is a predisposing rather than causal factor. There is no high-quality population incidence figure, and there is no implant registry for Kienböck's surgery because most procedures are osteotomies, fusions or carpectomies rather than arthroplasty.

No single global guideline — practice is built on cohort evidence and consensus. No society publishes a formal Kienböck's treatment guideline equivalent to a fracture protocol. Practice converges on staging-driven algorithms.

Registry and high- vs limited-resource practice. Where MRI is freely available, Stage I disease is detected early and unloading or revascularization is offered before collapse. Where MRI is scarce, diagnosis is later and total wrist fusion becomes the dependable salvage because it needs no specialist implant and reliably abolishes pain. Vascularized bone grafting and medial femoral condyle free flaps are concentrated in microsurgical referral units. Documenting the neutral-rotation ulnar variance on every wrist radiograph remains a universal, resource-independent standard because it determines whether radial shortening or capitate shortening is the appropriate unloading procedure.

Consent essentials (universal). For salvage surgery, counsel explicitly on the motion-versus-pain-relief trade-off (PRC preserves roughly half of wrist motion; total fusion removes motion but most reliably relieves pain), and on progression risk if early-stage disease is managed non-operatively.