Kienbock Disease

Updated: Sep 11, 2023
Author: Brian J Divelbiss, MD; Chief Editor: Harris Gellman, MD 


Practice Essentials

Kienböck disease is a condition of uncertain etiology that results in osteonecrosis of the carpal lunate.[1, 2]  In 1843, Peste presented the initial description of lunate collapse in the French literature. Nearly 70 years later, the Viennese radiologist Robert Kienböck introduced the term lunatomalacia to describe the condition that now bears his name. Kienböck believed that traumatic rupture of the ligaments and vessels around the lunate produced lunate fracture with subsequent collapse.[3]

In 1928, Hulten noted an association between Kienböck disease and the presence of negative ulnar variance. He advanced the progress of treatment by advocating shortening of the radius. Shortly thereafter, Persson presented the option of lengthening the ulna to restore normal ulnar variance.

The true natural history of this condition has not been fully elucidated, and this has hampered the determination of the ideal treatment. As with many conditions that affect the wrist, the clinical condition of the patient does not necessarily correlate well with the radiographic appearance.

Patients typically present with reports of activity-related dorsal wrist pain, decreased wrist motion in the flexion-extension arc, and poor grip strength, more often in the dominant than in the nondominant hand. Dorsal wrist swelling and tenderness may be present over the radiocarpal joint. There may be a history of trauma.

Staging and treatment are based on plain films. Tomograms may help determine the extent of disease, but in view of their limited availability, computed tomography (CT) may be best for evaluating the bony architecture of the lunate. Bone scanning may help exclude Kienböck disease but is not highly specific. Magnetic resonance imaging (MRI) is sensitive and specific for marrow changes consistent with osteonecrosis and may help demonstrate revascularization after operative treatment. Wrist arthroscopy may be used to assess the lunate and as a therapeutic tool for performing debridement, resection, or arthrodesis; it has also been used to develop an articular-based classification of Kienböck disease.

The primary methods of nonoperative treatment are immobilization and anti-inflammatory medications. Treatment is primarily directed by the level of symptoms. The primary indication for operative treatment in Kienböck disease is persistent pain that does not respond to conservative nonoperative treatment. Most patients with Kienböck disease who have moderate-to-severe symptoms are candidates for operative intervention. The choice of operative technique is based on patient age, disease stage, and the presence or absence of ulnar variance.


Force transmission studies have formed an important part of the understanding of the pathophysiology of Kienböck disease. In the normal wrist in neutral alignment, 80% of the axial load through the wrist is transmitted through the radiocarpal joint, while the remaining 20% goes through the ulnocarpal joint.

Two-dimensional theoretical models of force transmission in patients with Kienböck disease have demonstrated that in the early stages of the disease (II and IIIa), the normally positioned scaphoid prevents excessive forces on the lunate. However, as the scaphoid assumes its flexed position in stage IIIb, loads across the lunate are increased. These excessive loads may further accelerate the process of fracture and fragmentation leading to collapse.


The etiology of Kienböck disease has not been clearly determined. Most likely, it occurs as a result of repeated loads to a "lunate at risk" by virtue of its unique vascular or mechanical environment. Although the underlying etiology of this condition is not known, the final results of fragmentation and collapse are secondary to osteonecrosis. Intrinsic and extrinsic factors have been implicated.

Intrinsic factors

Vascular supply

The vascular supply of the lunate has been well studied by Gelberman.[4, 5] It consists of both extraosseous and intraosseous vessels running in the dorsal and volar radiocarpal ligaments. Three vessel patterns of intraosseous supply have been noted. In 70% of lunates, multiple vessels enter either volarly or dorsally (X or Y pattern). In the remaining 30% (I pattern), only a single vessel is present palmarly and dorsally, which theoretically places these lunates at increased risk of losing vascular supply.

Kienböck disease has not been reported following perilunate dislocations where the vascular supply has been damaged completely. Early signs of osteonecrosis (eg, increased radiodensity on plain radiographs) may be seen but have not been followed by progression to collapse. Increased intraosseous pressure has been shown to occur in lunates with Kienböck disease, but it is unclear whether this is a primary or secondary finding.

Lunate and distal radius geometry

Zapico classified lunate geometry into three types: Type I lunates occur in ulnar-negative wrists, whereas type II and III lunates are seen in ulnar-neutral or ulnar-positive wrists. His thesis was that the weakest trabecular pattern was seen in the type I lunate, which helped explain the relation between the ulnar-minus variant and the disease. Later work by Tsuge, however, failed to show an association between lunate geometry and Kienböck disease. Mirabello demonstrated that patients with Kienböck disease with decreased radial inclination developed the disease at an earlier age. The exact biomechanical effect of de-novo decreased radial inclination has not been determined.

Extrinsic factors

Relation between radii of curvature of lunate and capitate

Compressive axial forces are concentrated on the distal articular surface of the lunate because the radius of curvature of the capitate is less than that of its articulating surface on the lunate. As the capitate settles proximally later in the disease process, it can act as a wedge to split the lunate into dorsal and volar halves.

Repetitive trauma

Although no specific data support a causal relation, a history of repetitive microtrauma is often noted in patients with Kienböck disease.[6]

Ulnar variance

A statistical relation between negative ulnar variance and Kienböck disease appears evident, though it is not currently thought to be a causal one. In Hulten's original work, he noted that 23% of the general population has negative ulnar variance, whereas 74% of his patients with the disorder were ulnar-minus. The ulnar-minus variant has been shown experimentally to cause an abnormal increase in the force transmitted across the lunate. In addition, the triangular fibrocartilage complex (TFCC) is thicker in these patients, and the differential loading between it and the ulnar edge of the radius is increased.

De Smet, however, countered that a true correlation between the ulnar-minus variant and Kienböck disease has not been proved when appropriate sex- and age-matched controls and radiographs have been used.[7]


Kienböck disease usually affects the dominant wrist of men aged 20-40 years.


Given that relatively few reported series have focused exclusively on nonoperative treatment of Kienböck disease, there remains room for better definition of the natural history of this condition. Kristensen monitored 49 patients nonoperatively for an average of 20.5 years and found that 80% of the patients had no pain or had pain only with heavy labor.[8] Most patients reported a gradual lessening of symptoms over time. This benign clinical picture was not mirrored by radiographic findings; degenerative changes in the wrist were common and every lunate was deformed.

In a retrospective study comparing surgical treatment with nonsurgical treatment for patients with Kienböck disease, Delaere noted that surgical management did not appear to show superiority over nonsurgical treatment at 5-year follow-up.[9] Care should be taken in interpreting these results, however, because scaphotrapeziotrapezoid (STT; triscaphe) fusions were performed in the majority of those treated surgically. This procedure has not been shown to produce long-lasting good results. In addition, patients who had more advanced disease were more likely to undergo surgery.

Mikkelsen noted that 15 of 25 patients treated conservatively had daily problems with the wrist.[10] He concluded that nonoperative treatment was not indicated for Kienböck disease. In the early stages of Kienböck disease, a short trial of casting may alleviate symptoms and obviate the need for surgery. These patients should be monitored closely so that surgery, if necessary, can be performed when radial shortening is still feasible.

A 2021 retrospective study by DeGeorge et al, aimed at describing the natural history of Kienböck disease among patients who chose nonoperative treatment, found that over time, the majority of these patients achieved significant improvements in pain, grip strength, and Mayo wrist score, despite radiographic progression of disease.[11]

Although initial success was seen with the use of a silicone spacer following lunate resection, this implant is no longer indicated. Alexander presented a 5-year follow-up of a group of 10 patients with silicone lunate replacements and noted 50% unsatisfactory results.[12] Sixty percent of patients who had radiographs at final follow-up demonstrated evidence of silicone particulate synovitis. Better results may be achievable with implants composed of other substances, such as pyrocarbon.[13]

Lunate excision is not commonly recommended, because of concerns of progression of carpal collapse. A fascial or palmaris anchovy replacement has had variable success in preventing subsequent collapse, though Carroll reported long-term (>10 y) success in a series of 10 patients treated with a fascial implant following lunate excision. He noted no evidence of carpal collapse, and all patients had unrestricted use of their hands.

Radial shortening remains a mainstay of treatment. It is a reliable, reproducible procedure with good results. Weiss and Quenzer presented large series of radial shortenings,[14, 15, 16] noting decreased pain in about 90% of patients, as well as improved motion and grip strength at 4-year follow-up. Weiss also demonstrated that good results can be obtained in patients with stage III disease who have evidence of lunate collapse.

Despite good clinical outcomes, radiographic signs of continued collapse and degeneration are common. Although ulnar lengthening has demonstrated similar clinical outcomes, it is also associated with a higher complication rate. For this reason, radial shortening is the preferred joint-leveling method.

Watson reported on the use of STT (triscaphe) fusion for stage III disease and noted nearly 80% of his patients had good or excellent pain relief at 51-month follow-up.[17] Caution should be used in evaluating these results, in that nearly 40% of his patients required additional procedures to achieve this outcome. Compared with STT fusion, scaphocapitate (SC) fusion has demonstrated similar pain relief rates but slightly decreased motion. Currently, no long-term results are available on the use of capitate shortening with or without capitohamate arthrodesis.[18]

In a series of 51 patients who underwent vascular bundle implantation, 98% showed a reduction or resolution of pain at long-term follow-up. Again, these encouraging clinical results were not mirrored by radiographic improvement: 20% had further degeneration and 10% had frank fragmentation. Results of the newer technique of distal radial vascular bone pedicle were noted previously.

The expected outcome following proximal-row carpectomy (PRC) is 75° in the flexion-extension arc and grip strength of 75% of the opposite side. Pain relief has been demonstrated in 80% of patients who underwent PRC for Kienböck disease. As noted, a PRC can be converted successfully to wrist arthrodesis.

Several other options for treatment have been reported. Ruby performed cancellous bone grafting supplemented by postoperative external fixation and found good pain relief in 80% of patients.[19] In patients with stage III disease and mechanical symptoms, arthroscopic debridement was found to decrease pain and increase motion. Wrist denervation is a simple, safe option that can also be combined with other procedures to aid in postoperative pain reduction. In fact, denervation itself may provide much of the perceived pain reduction in cases where the dorsal capsule is incised to perform the index procedure.

Gay et al evaluated the use of a simple capitate osteotomy without arthrodesis, through a dorsal medial approach and fixed with staples, in 11 patients with mild forms (stage I to IIIA, Lichtman classification) of Kienböck disease.[18] At final follow-up (mean, 67.4 mo), the mean visual analogue scale (VAS) score was 1.7 (range, 0-7). There were six good results, two fair results, and three poor results. Compared with the healthy side, mean strength improvement was 25%. In two cases, revision surgery was necessary.

Takahara et al reported on 13 patients (6 with stage II, 4 with stage IIIA, and 3 with stage IIIB) who responded to a DASH (Disabilities of the Arm, Shoulder and Hand) questionnaire regarding previous radial osteotomy for Kienböck disease (mean follow-up, 21 y).[20] Mean DASH score was 8 points (range, 0-23), and patient satisfaction was high. The DASH scores tended to be worse in patients with stage IIIB disease.

Follow-up radiographs revealed that the Lichtman stage had progressed in six of the 12 patients.[20] Compared with the contralateral wrist, mean range of motion was 81% in flexion and 82% in extension; mean grip strength was 88%. Clinical results were considered excellent in six patients, good in five, and moderate in one.



History and Physical Examination

The most common patient with Kienböck disease is a man aged between 20 and 40 years who either works as a manual laborer or participates in recreational activities that repetitively load the wrist. Patients present with reports of activity-related dorsal wrist pain, decreased wrist motion in the flexion-extension arc, and poor grip strength. The symptoms tend to occur more often in the dominant hand. Dorsal wrist swelling and tenderness are frequently present over the radiocarpal joint. A history of trauma is variable and may be in the distant past.

A review that focused on Kienböck disease in women revealed that men and women have different presentations of Kienböck disease. Women had roughly equivalent involvement of the dominant and nondominant sides, and they tended to present at a much older age than men did (46 vs 31 y).



Imaging Studies

Plain radiography

Plain films form the basis for staging and treatment of Kienböck disease.[21, 22] Lichtman's modification of Stahl's classification, the system that has been most widely used, divided the disease into five stages, as follows:

  • Stage I - Normal radiograph
  • Stage II - Increased radiodensity of lunate with possible decrease of lunate height on radial side only
  • Stage IIIa - Lunate collapse, no scaphoid rotation
  • Stage IIIb - Lunate collapse, fixed scaphoid rotation
  • Stage IV - Degenerative changes around the lunate

Plain films must also be examined to determine the amount of ulnar variance present. This will directly impact the choice of operative technique. A true posteroanterior view of the wrist is necessary for an adequate determination of ulnar variance.


Tomograms may be useful in determining the true extent of disease. Tomograms have been found to result in the upgrading of many patients with stage II disease to stage III by more clearly demonstrating collapse. In addition, coronal fractures that split the lunate into volar and dorsal halves are more evident with tomograms.

In view of the limited availability of tomograms at this time, a computed tomography (CT) scan would be the best imaging modality for evaluating the bony architecture of the lunate.

Bone scanning

Bone scanning may help exclude the presence of Kienböck disease, but it is not specific enough to exclude the many other causes of increased uptake in the area of the lunate. It may be of some help in the patient with known Kienböck disease who develops wrist pain in the contralateral side.

Magnetic resonance imaging

Magnetic resonance imaging (MRI) is most helpful early in the course of the disease when plain films are not diagnostic.

T1- and T2-weighted images reveal decreased signal intensity. Patterns of signal loss can be focal or generalized; however, primary involvement of the ulnar proximal portion of the lunate indicates potential ulnar abutment syndrome. T1-weighted images showing focal loss on the radial half of the lunate suggest early involvement, particularly if the corresponding T2-weighted images show normal or increased intensity.

MRI is an extremely sensitive and specific test for detecting the presence of marrow changes consistent with osteonecrosis. MRI has also been used for indirectly demonstrating revascularization after operative treatment.

Bae et al proposed an MRI-based classification system for Kienböck disease, which they found to be more reliable than the modified Lichtman system and more appropriate for classification into stages IIIA and IIIB.[23]

Other Tests

MacLean et al evaluated the use of wrist arthroscopy to assess the lunate and concluded that the procedure can also be a therapeutic tool for performing debridement, resection, or arthrodesis procedures.[24]

Bain and Begg used arthroscopy to develop an articular-based classification of Kienböck disease that included the following grades[25] :

  • Grade 0 - All articular surfaces are functional
  • Grade 1 - The proximal lunate articular surface is nonfunctional
  • Grade 2A - The proximal articular surface of the lunate and the lunate fossa are both nonfunctional
  • Grade 2B - The proximal and distal articular surfaces of the lunate are nonfunctional
  • Grade 3 - Three articular surfaces are nonfunctional; usually, there will be only a functional capitate articular surface
  • Grade 4 - All four articular surfaces are nonfunctional


Approach Considerations

The primary indication for operative treatment of Kienböck disease is persistent pain that does not respond to conservative treatment such as nonsteroidal anti-inflammatory drugs (NSAIDs) and immobilization. Most patients with Kienböck disease who have moderate-to-severe symptoms are candidates for operative intervention. The choice of operative technique is based on patient age, disease stage, and the presence or absence of ulnar variance.

The primary contraindication to consider in the operative treatment of Kienböck disease is ulnar-positive or ulnar-neutral variance because in patients with such variance, joint-leveling procedures (radial shortening and ulnar lengthening) cannot be performed.

Medical Therapy

The primary methods of nonoperative treatment are immobilization and adminstration of anti-inflammatory medications. Because the natural history of Kienböck disease is not well determined, treatment is primarily directed by the level of symptoms. Certainly, a very young patient, though an unusual presentation of Kienböck disease, should be given an adequate trial of immobilization in hopes of allowing revascularization of the lunate and preventing disease progression.

Surgical Therapy

A number of options are available for surgical management of Kienböck disease. The two most important pieces of information are the stage of the disease and the presence or absence of ulnar variance.

Directly comparing the results of different techniques is difficult because most studies have a fairly small number of patients and short follow-up. However, review of the literature reveals that many of the techniques result in very similar rates of good outcomes.

Operative treatment can be classified broadly into six categories, as follows:

  • Lunate excision with or without replacement
  • Joint-leveling procedures
  • Intercarpal fusions
  • Revascularization
  • Salvage procedures
  • Other procedures

Lunate excision with or without replacement

Available approaches include the following:

  • Simple lunate excision
  • Excision with soft-tissue (fascial or palmaris longus tendon graft) replacement
  • Replacement arthroplasty

Joint-leveling procedures

Radial shortening[26, 27] and ulnar lengthening are two options for leveling the joint. The goal is to produce a wrist with neutral ulnar variance, though correction should probably not exceed 4 mm, because nearly all strain reduction occurs in the first 2 mm of correction. Strains at the lunate can be reduced by 70% with an appropriate radial shortening or ulnar lengthening.

Currently, radial shortening with a volar distal radius locking plate is preferred to ulnar lengthening because there is a lower complication rate with the volar-shortening procedure and because the two procedures have shown similarly good outcomes. In patients with neutral or positive ulnar variance, shortening the radius is contraindicated. In this clinical situation, radial wedge osteotomies designed to decrease the radial inclination have been proposed.[28, 29, 30]

Intercarpal fusions

Various intercarpal fusions for the treatment of Kienböck disease have been reported. The goal is to reduce lunate strain and, in procedures that involve the scaphoid, to correct and maintain proper scaphoid position.

Of the limited intercarpal fusions reported, the greatest experience has been with scaphotrapeziotrapezoid (STT; triscaphe) fusion. STT arthrodesis does decrease lunate strain but merely by shifting it to the radioscaphoid joint. STT fusion in a cadaver model was found to provide strain reduction similar to that of joint-leveling procedures but with greater loss of motion. The use of STT fusion has waned in recent years because of complications and longer-term follow-up that revealed decreased success rates.[31]

Several authors have reported scaphocapitate (SC) fusion.[32] Biomechanically, this fusion has been shown to reduce strain at the radiolunate joint by about 10%. Some authors prefer this fusion because it requires only one fusion site and is technically easier to perform.

Scapholunocapitate (SLC) fusion with proximal lunate articular surface preservation has been described in the management of grade IIIA Kienböck disease.[33]

Capitohamate fusion has been reported in a study with a short follow-up. However, this fusion alone has been shown to be biomechanically ineffective in reducing lunate strain. If this fusion is combined with capitate shortening, significant reductions of load across the radiolunate and SC joints have been noted. This load reduction is offset by large force increases at the ulnotriquetral, triquetrohamate, and scaphotrapezial joints. This method also does not address the scaphoid rotation that occurs with stage IIIb disease.

At present, intercarpal fusions are more likely to be reserved for patients with neutral or positive ulnar variance in whom a joint-leveling procedure is contraindicated.


In the late 1970s, Hori presented his initial work on the use of a vascular pedicle directly implanted into the lunate. Nearly all of his patients showed improvement in their pain at later follow-up.

Currently, most revascularizations use vascularized bone pedicles.[34, 35, 36] There are several sources for the pedicles, including the distal radius,[37] pisiform,[38] and pronator quadratus (Braun). Results with the use of pedicled distal radius grafts have shown improved grip strengths and progressive evidence of revascularization on magnetic resonance imaging (MRI) over an 18- to 36-month period.[39]

A retrospective study by Windhofer et al reported good short-term and midterm results from the use of a lateral femoral trochlea osteochondral graft for lunate reconstruction in 27 patients with stage III Kienböck disease.[40]

A systematic review and single-arm meta-analysis by Park et al compared the long-term outcomes of vascularized bone graft (VBG) and nonoperative treatment in patients with Kienböck disease.[41]  Greater improvements in radiographic stage and wrist pain were achieved with VBG than with nonoperative treatment, but meaningful differences in parameters were not observed.

Revascularization techniques may also be combined with other previously mentioned approaches. Revascularization may be especially attractive for the young patient with ulnar-neutral or ulnar-positive variance in whom a radial shortening is not an option and for the patient who wishes to avoid an intercarpal fusion and resultant loss of motion.

Salvage procedures

Salvage procedures are reserved for later stages of disease and for failures of other treatments. Proximal-row carpectomy (PRC) has been shown to provide relatively good results for Kienböck disease, as well as for other wrist problems.[42] Wrist arthrodesis is the final option for patients with global wrist degeneration. Arthrodesis can be achieved successfully after a failed PRC. SC arthrodesis has been suggested as an option for wrist salvage in cases of advanced Kienböck disease.[43]

Other procedures

Other possible procedures include the following:

  • Cancellous bone grafting plus external fixation
  • Arthroscopic debridement [44]
  • Wrist denervation
  • Metaphyseal decompression
  • Lunate core decompression [45, 46]

Some studies have found partial capitate shortening to be effective for stage II and III Kienböck disease.[47, 48, 49]

Treatment based on disease stage

A reasonable approach to determining the surgical treatment of Kienböck disease on the basis of the stage of the disease is as follows:

  • Stage 0, I, II, or IIIa with ulnar-negative variance – Radial shortening, revascularization, denervation
  • Stage 0, I, II, or IIIa with ulnar-neutral or ulnar-positive variance – Revascularization, capitohamate (CH) fusion with capitate shortening, distal radius wedge osteotomy, denervation
  • Stage IIIb – SC fusion, radial shortening, denervation
  • Stage IV – PRC, total wrist arthrodesis, denervation