Kienböck disease, or osteonecrosis of the lunate, most often affects patients between the ages of 20 and 40 years. There are 4 major stages of the disease, and treatment is based on the stage of disease. Advancements are still being made with regards to the cause, pathophysiology, and preferred method of treatment of each stage. Although the goals of pain relief, motion preservation, strength maintenance, and function outcomes are paramount to success, no 1 procedure consistently and reliably achieves these outcomes. Further advancements in treatment and results of long-term outcome studies should resolve some of these topics.
Key points
- •
Kienböck disease is defined by avascular necrosis of the lunate with a predictable pattern of lunate collapse, carpal change, and degeneration, resulting from an apparent combination of vascular, anatomic, and traumatic insults.
- •
Advancements are still being made with regards to the cause, pathophysiology, and preferred method of treatment of each of the various stages of Kienböck disease.
- •
Although the goals of pain relief, motion preservation, strength maintenance, and function outcomes are paramount to success, there is no 1 procedure that consistently and reliably achieves these outcomes.
- •
With further advancements in treatment and results of long-term outcome studies, answers should be obtained to some of these still unresolved topics.
Introduction
Kienböck disease, also known as avascular necrosis (AVN) or osteomalacia of the lunate, was first described in 1843 by Peste through his cadaveric work and observations of lunate collapse. However, the term Kienböck disease was coined after Dr Robert Kienböck, an Austrian radiologist, when he presented clinical and radiographic evidence of 16 patients with osteomalacia of the lunate in 1910. He described a stepwise progression of disease from isolated proximal lunate involvement, to fragmentation and collapse of the lunate, evolving to radiocarpal involvement with degenerative changes. He believed the osteomalacia to be a result of “a disturbance in the nutrition of the lunate caused by rupture of the ligaments and blood vessels during contusions, sprains, or subluxations. ”
Since the early descriptions by Kienböck, there has been research to explore the cause, pathophysiology, staging, treatments, and outcomes of Kienböck disease.
Introduction
Kienböck disease, also known as avascular necrosis (AVN) or osteomalacia of the lunate, was first described in 1843 by Peste through his cadaveric work and observations of lunate collapse. However, the term Kienböck disease was coined after Dr Robert Kienböck, an Austrian radiologist, when he presented clinical and radiographic evidence of 16 patients with osteomalacia of the lunate in 1910. He described a stepwise progression of disease from isolated proximal lunate involvement, to fragmentation and collapse of the lunate, evolving to radiocarpal involvement with degenerative changes. He believed the osteomalacia to be a result of “a disturbance in the nutrition of the lunate caused by rupture of the ligaments and blood vessels during contusions, sprains, or subluxations. ”
Since the early descriptions by Kienböck, there has been research to explore the cause, pathophysiology, staging, treatments, and outcomes of Kienböck disease.
Cause and pathophysiology
The cause of Kienböck disease is still debated, but it is likely multifactorial and seems to result from varying degrees of anatomic factors, interrupted vascularity, and traumatic insults to the lunate.
Various anatomic and mechanical factors have been associated with the development of Kienböck disease. Hulten was the first to describe the relationship between Kienböck disease and ulnar negative variance. In 1928, he described a series of patients with Kienböck disease, in whom 78% of affected patients were noted to have ulnar negative variance, and 23% of individuals with neutral or positive ulnar variance were affected. His work was followed by Gelberman and colleagues, who also found a significant relationship between ulnar negativity and Kienböck disease when they studied various anatomic and vascular differences among 35 cadaver specimens.
Antuna Zapico described 3 types of lunate morphologies and related them to ulnar variance and described their association with Kienböck disease. A type I lunate has a proximal apex, which is more likely to be seen in wrists with ulnar negative variance. In type II and III lunates, the shape is more rectangular and is associated with ulnar neutral or positive wrists. Type I lunates were associated with higher rates of stress and fragmentation under loading, and the investigators concluded that they were the weakest configuration. Despite this evidence, several subsequent studies have failed to show a definitive relationship between ulnar variance and Kienböck disease.
Other studies have described the relationship between a lower angle of radial inclination and Kienböck disease. Although there is no general consensus regarding radiocarpal anatomy and the development of Kienböck disease, there seems to be a relationship between unequal load distribution through the radiocarpal joint, whereby the lunate receives an abnormal distribution of the force, and the lunate subsequently becomes at risk for the development of Kienböck disease.
The vascular supply to the lunate is variable, but most have both a dorsal and volar nutritional supply. Three major patterns of vascularity have been described: Y, X, and I types, which refer to the geometric branching of the intraosseous arteries. In the I pattern, there is a single vessel supplying the lunate, which may increase the risk for the development of osteonecrosis. It has been reported that approximately 7% to 23% of patients have a single contributing vessel to the lunate, and up to 31% of individuals have minimal branching of the anastamotic vessels that provide nutrition to the lunate. In addition, AVN of the lunate has been linked to a vascular insult caused by fracture, ligament collapse, primary circulatory collapse, systemic disease, ligamentous collapse, and venous congestion. However, none of these occurs with significant regularity to allow for generalized screening. Schiltenwolf and colleagues concluded that necrotic lunates showed significantly higher levels of intraosseous vascular congestion compared with normal lunates, but it was unclear whether or not this was related to lunate collapse or was a cause of the disease process itself.
The relationship between Kienböck disease and traumatic insults has been studied extensively, and there seems to be a link between trauma and the development of osteonecrosis in the lunate. Repetitive microtrauma, such as that seen in heavy laborers, seems to play a role, but there is often no association between a single traumatic event and the development of Kienböck disease. Although early increased radiodensity of the lunate can be seen after a perilunate dislocation, the collapse and fragmentation that are seen in Kienböck disease rarely occur in this subset of patients, as confirmed by Takami and colleagues. The volar lunate vasculature typically remains attached to the volar capsule during this injury, thus preserving the blood supply of the lunate.
Although there is no single definitive cause of Kienböck disease, a complex interplay of vascular and anatomic variations, combined with varying degrees of microtrauma and insults, contribute to its development.
Clinical presentation
Kienböck disease most commonly affects men between the ages of 20 and 40 years. Many patients describe a history of trauma, but this is not always present. Although symptoms can vary depending on their stage at initial presentation, patients typically present with pain localized to the radiolunate facet, decreased motion, swelling, and weakness in the affected hand. Pain is classically insidious in onset, often related to activity, and can be present for extended periods before presentation. The disease is rarely bilateral.
Physical examination reveals tenderness over the dorsal lunate and radiolunate facet. An effusion or bogginess overlying the radiocarpal joint is not an uncommon finding. Motion in the flexion and extension arc is often decreased, and average grip strength may decrease up to 50% of the contralateral side.
Radiographic imaging
Standard neutral rotation posteroanterior (PA) and lateral radiographs of the affected hand should be ordered on any patient suspected to have Kienböck disease. Plain radiographs may be negative early in the disease process, but typically progress to show increased lunate density, which indicates osteonecrosis. With worsening disease, collapse of the lunate and fragmentation of the lunate are visualized. Advanced disease yields proximal migration of the capitate, dorsal intercalary segmental instability (DISI), showing scaphoid flexion coupled with lunate extension and degenerative disease involving the radiocarpal joint.
Advanced imaging can aid in the diagnosis and staging of Kienböck disease. Bone scintigraphy, which has largely been replaced by the ease and convenience of magnetic resonance imaging (MRI), can detect early stages of the disease with increased signal uptake. MRI typically shows decreased signal intensity on T1-weighted and T2-weighted images indicating impaired vascularity. In patients with other diseases, such as perilunate dislocation or ulnar impaction syndrome, changes within the lunate may appear similar to the osteonecrotic changes seen in Kienböck disease. However, these changes are often focal and nonprogressive, which is different from the uniform osteonecrotic changes seen in the lunate of patients with Kienböck disease. Hashizume and colleagues compared diagnostic modalities and concluded that although MRI is useful in detecting early stages of disease before collapse has occurred, other imaging modalities such as computed tomography (CT) or tomography best characterize lunate necrosis and trabecular destruction once collapse has occurred.
Trispiral tomography is another useful adjunct to the diagnosis of Kienböck disease. In a study conducted by Quezner and colleagues, 105 patients with radiographic evidence of stage I disease were evaluated with trispiral tomography, 87% of patients were upgraded to stage II after imaging. Likewise 71% of patients with stage II disease and 9% of patients with stage III disease were upgraded to stage III and IV, respectively, after trispiral tomography imaging. However, this imaging modality is not widely available.
Staging
Stahl originally described the classification of Kienböck disease in 1947. Lichtman and colleagues introduced a modified classification scheme that has remained the most frequently used. A summary of the Lichtman classification can be found in Table 1 .
| Stage I | Normal radiographs, ± linear fracture lines. MRI shows uniform signal decrease on T1-weighted images. Bone scan positive but nonspecific |
| Stage II | Plain radiographs show lunate sclerosis, ± fracture lines. No collapse of lunate |
| Stage IIIA | Lunate collapse, with maintenance of carpal height and alignment |
| Stage IIIIB | Lunate collapse plus any of the following: loss of carpal height, proximal capitate migration, flexed and rotated scaphoid |
| Stage IV | Stage IIIB + radiocarpal or midcarpal degenerative changes |
Stage I
Patients presenting with stage I disease typically complain of nonspecific, intermittent wrist pain and synovitis, which mimic a wrist sprain. Plain films are either normal or show small linear compression fractures through the lunate. There is no collapse, sclerosis, or increased radiodensity of the lunate.
MRI shows uniformly decreased signal uptake on both T1-weighted and T2-weighted images, indicating osteonecrosis of the lunate. If revascularization is occurring, such as after operative intervention, increased signal uptake on the T2-weighted images would be noted.
Bone scintigraphy at this stage typically shows decreased signal uptake and changes related to reactive synovitis.
Stage II
Stage II is characterized clinically by increased swelling, varying degrees of stiffness, and progressive pain. Radiographs show lunate sclerosis, with or without compression fracture lines. The lunate appears more radiodense, but there is no evidence of collapse and lunate height is maintained. The remainder of the carpus remains without degenerative change and the relationship between the lunate and the proximal carpal row is maintained.
Although other pathologic conditions may create increased sclerosis of the lunate, such as after a perilunate dislocation or with ulnar impaction syndrome, this process is typically focal and generally without significant progression. Patients with Kienböck disease present with diffuse lunate changes that are typically progressive. Ulnar variance (negative, neutral, or positive) should be noted.
Fig. 1 shows an individual with stage II disease.
Stage III
Stage III disease is defined by continued sclerosis and collapse of the lunate. This stage is divided into 2 separate subgroups, depending on the alignment and maintenance of carpal relationships. Attention must be paid to the scapholunate angle to evaluate for a DISI deformity as well as the carpal height. Carpal height is measured as the distance between the distal articular surface of the capitate to the lunate fossa of the distal radius. The carpal height ration is the carpal height divided by the length of the third metacarpal and is approximately 0.53. A decrease of the carpal height and carpal height ratio suggests collapse or osteoarthritic change within the carpus.
Stage IIIA is characterized by collapse of the lunate, with preservation of carpal height and intercarpal alignment. Thus, the capitate has not migrated proximally and the scaphoid has not assumed a flexed position. On a lateral radiograph, the lunate appears widened in the anteroposterior plane as a result of the coronal plane collapse. The scapholunate angle is also preserved at –10° to 10°.
Stage IIIB is characterized by both collapse of the lunate and characteristic changes of the surrounding capitate and scaphoid. The capitate migrates proximally, and carpal height becomes diminished. As the scaphoid flexes and subsequently rotates, a DISI pattern may become evident. Scaphoid flexion may be appreciated on a PA radiograph as a cortical ring sign. The triquetrium shifts ulnarly as a result of the surrounding carpal changes and widening at the scapholunate interval.
Symptoms at this stage have typically progressed from vague wrist pain and synovitis to symptoms of instability, clunking with radial and ulnar deviation, progressive pain, and decreased grip strength. An example of stage IIIB disease is shown in Fig. 2 .
Stage IV
Stage IV disease is characterized by progressive carpal collapse, leading to radiocarpal and midcarpal degenerative changes. Radiographs show classic joint space narrowing, subchondral sclerosis, degenerative cysts, and osteophyte formation. Symptoms have typically progressed to stiffness, constant pain, and swelling.
Treatment
There are several treatment options for Kienböck disease, largely based on the stage at presentation. Although options vary, they typically fall into several broad categories: unload the lunate, revascularize the lunate, or treat carpal instability and collapse with salvage procedures. A summary is given in Table 2 .
| Stage I | Cast immobilization for 3 mo. If disease progresses after 3 mo, consider stage II/IIIA treatment options |
| Stage II/IIIA with ulnar negativity | Joint leveling procedures; radial shortening osteotomy, ulnar lengthening osteotomy |
| Stage II/IIIA with ulnar positivity or neutrality | Revascularization procedures ± stabilization; capitate shortening; radial wedge or dome osteotomies; core decompression; combination of revascularization with joint leveling procedures |
| Stage IIIIB | PRC, radial shortening osteotomy, limited intercarpal arthrodesis (STT or SC fusion), lunate excision with interposition grafting |
| Stage IV | PRC, wrist arthrodesis, wrist arthroplasty, wrist denervation |
Stage I
Conservative treatment with 3 months of immobilization is typically recommended first for patients with stage I disease, although some have argued that this has minimal effect. Taniguchi and colleagues reported 35-year outcomes on a series of 20 patients, and found that although 70% of patients had radiographic progression of the disease, only 20% were symptomatic. Likewise, Kristensen and colleagues followed 46 patients treated nonoperatively for a minimum of 5 years, and reported that 66% of patients had arthritic changes on imaging but only 25% were symptomatic. Delaere and colleagues followed 21 surgically treated patients and 22 conservatively treated patients for 65 months and found no difference in outcomes, but did find change in social activities and loss of motion in nearly 25% of the operative group.
However, in contrast to these findings, several studies have reported poor outcomes with nonoperative management of stage I disease. Keith and colleagues reported decreased DASH (Disabilities of the Arm, Shoulder and Hand) scores, motion, and grip strength in a series of 33 patients treated nonoperatively. Likewise, Lichtman reported results of 22 patients treated nonoperatively, and results were poor: 17 patients had disease progression, and 5 patients had no improvement in symptoms.
Thus, although a 3-month trial of immobilization for stage I disease should be considered, the patients should continue to be monitored and if symptoms or radiographs progress, surgical discussion should commence.
Stage II or IIIA with Negative Ulnar Variance
Stage II and IIIA are considered together when considering treatment options. Although lunate collapse has occurred with stage IIIA, there remains a normal intercarpal relationship. Decreased vascularity has occurred with both stages.
The goal of treatment of stage II and IIIA with ulnar negative variance is generally centered toward unloading the lunate in an attempt to reduce intracarpal stress and allow for revascularization. Radial shortening osteotomies or ulnar lengthening procedures are joint leveling procedures that redistribute lunate load forces. The desired outcome is ulnar neutrality to 1 mm of ulnar positivity. Greater ulnar positive variance may result in ulnar impaction against the lunate or triquetrium, and cause ulnar-sided wrist pain or disease.
Horii and colleagues developed a two-dimensional model that looked at various joint leveling procedures in an attempt to quantify the amount of lunate unloading. Either ulnar lengthening or radial shortening of 4 mm led to a 45% decrease in force across the radiolunate joint. Likewise, Trumble and colleagues investigated treatment of Kienböck disease and reported that a 70% decrease in lunate strain was seen with ulnar lengthening, radial shortening, and scaphotrapeziotrapezoidal (STT) fusion. Specific to radial shortening and ulnar lengthening procedures, these investigators noted that 90% of the reduction in lunate strain occurred with the first 2 mm of length change. Joint leveling procedures were preferred for this stage of disease, because it maintained wrist motion, whereas STT fusion resulted in decreased radial deviation and wrist extension. Nakamura and colleagues concluded that radial shortening greater than 4 mm led to ulnar-sided wrist pain, and thus, 3 mm should be the maximum amount of radial shortening attempted.
Radial shortening
With regards to clinical outcomes of joint leveling procedures, Quenzer and colleagues reported results on 68 patients who underwent a radial shortening osteotomy; a 93% decrease in pain was reported at an average follow-up of 52 months. Grip strength increased in 74% of patients, whereas motion improved in 52% of patients and decreased in 19% of patients. In addition, one-third of patients showed signs of revascularization. Watanabe and colleagues reported long-term outcomes of radial shortening osteotomy procedures for stage II and IIIA disease, with an average 21-year follow-up, and found that motion and grip strength remained greater than 80% compared with the contralateral wrist, average DASH scores were 8, and patient satisfaction was high. Other studies looking at long-term outcomes of joint leveling procedures have concluded that radial shortening osteotomy is an effective, safe, and reliable procedure for stage II and IIIA disease. Radial shortening osteotomies should not be considered for patients who have progressed to stage IIIB disease, because the results are often less favorable.
Ulnar shortening
As an alternative to radial shortening, ulnar lengthening procedures achieve the same goal of unloading the lunate to allow for revascularization. Linscheid described an ulnar lengthening technique using an iliac crest bone graft. He reported results of 64 patients over a 14-year period and described high patient satisfaction, favorable range of motion, and improved grip strength. However, there was a 22% complication rate, which included 9 patients with a nonunion and 5 patients with ulnar impingement syndrome. Seven patients required an additional procedure. Although this surgery is an option for patients with stage II and IIIA disease, more favorable results with less morbidity and fewer complications are typically seen with radial shortening procedures.
Stage II and IIIA with Ulnar Neutral or Positive Variance
Direct revascularization allows the potential for salvage of the lunate and possible reversal of destruction of the lunate through neoangiogenesis. Many revascularization procedures have been described, including distal radius pedicle grafts, vascularized pisiform grafts, fourth or fifth extensor compartment artery transfers, or first, second, or third dorsal metacarpal artery transfers. Sheetz and colleagues reported the anatomic relationships of several vascularized pedicles from the distal radius and ulna that can be harvested and transferred to the lunate for revascularization.
Revascularization
Hori and colleagues described results of a vascularization procedure that transferred the second dorsal metacarpal vascular pedicle to the lunate. Long-term outcomes of this procedure included reduced pain in 50 of 51 patients and improved strength in all 51 patients. Despite these favorable outcomes, 10% showed progression of lunate fragmentation, and 20% showed increased degenerative changes. Bochud and colleagues reported 2-year outcomes of 32 patients treated with a vascularized pisiform transfer and found that 95% of patients showed restoration of lunate anatomy, but only 33% of patients maintained this correction. Fair to poor results were seen in nearly half of cases at final follow-up. Alternatively, Moran and colleagues described revascularization in 48 patients with an average 10-year follow-up. Significant pain relief was related by 98% of patients, and MRI evidence of lunate revascularization occurred in 60% of patients.
Osteotomies
Osteotomies provide an alternative to revascularization and include capitate shortening osteotomies with or without capitohamate fusion, radial closing-wedge osteotomies, and radial-dome osteotomies. The goal of these various procedures is to unload the lunate in an attempt to decrease stress across the radiolunate joint, to allow for revascularization and prevention of disease progression. Almquist described a capitate shortening procedure and its associated results, reporting 83% revascularization of the lunate. In addition to this, Viola and colleagues reported significant reduction in load across the radiolunate joint after capitate shortening osteotomy with capitohamate fusion. Radial closing-wedge osteotomies have been described for patients with ulnar positive wrists and act to reduce radial inclination and, subsequently, the stress across the radiolunate articulation. Various long-term studies have reported favorable outcomes in terms of pain, grip strength, and motion, despite evidence of disease progression. Takahara and colleagues recently described the surgical technique of radial closing-wedge shortening osteotomy for patients with ulnar positive wrists, and recommend 5° to 10° of a closing wedge with a total of 2 mm of shortening.
Core decompression
Core decompression of the radius and ulna creates a local vascular healing response and has been suggested for patients with stage I to IIIA disease. Illarramendi and colleagues described 10-year follow-up results of 22 patients treated with distal radial and ulna metaphyseal core decompression, using cortical windows and a small curette. Average grip strength was 75%, and motion was 77% compared with the contralateral wrist. Seventy-three percent of patients reported no pain, and 20 patients were able to return to their occupations.
Stage IIIB
By this stage, the surrounding carpus has been altered because of progressive lunate collapse and fragmentation, with resulting scaphoid flexion, proximal capitate migration, and decreased carpal height. Thus, procedures are needed to address the destabilized carpus, prevent further carpal collapse, and decrease load across the radiolunate joint. There are several procedures that can accomplish these goals, including proximal row carpectomy (PRC); STT and scaphocapitate (SC) arthrodesis; radial shortening osteotomy; fusion with or without lunate excision; and interposition grafting.
PRC
PRC is a procedure that excises the scaphoid, lunate, and triquetrium, transferring load from the capitate directly to the lunate facet of the distal radius. Wall and Stern recently reviewed PRC as a treatment option for advanced Kienböck disease and reported favorable results, few complications, painless motion, and adequate grip strength. They also concluded that the best candidates for PRC are patients older than 35 years with an intact lunate facet of the distal radius and an intact capitate head. Richou and colleagues reported long-term outcomes of PRC, with an average 9.6-year follow-up, in which 83% of patients were satisfied with their outcomes. Pain improved in all patients, grip strength was 76% of the contralateral side, total arc of flexion-extension was 76°, and average DASH scores were 31. Despite favorable clinical results, advancing radiographic staging was seen in 52% of patients.
Hohendorff and colleagues recently published 1-year outcomes from an ongoing study, which compares 8 patients with STT fusion versus 11 patients with a PRC for stage IIIB Kienböck disease. These investigators reported slightly improved outcomes for the PRC group in terms of DASH scores, Mayo wrist scores, grip strength, and motion.
Crog and Stern reported 10-year follow-up data on 21 patients treated with PRC, in whom 3 patients required radiocapitate arthrodesis. Even although 2 of those 3 patients had stage IV disease, they concluded that although it is a reliable procedure, PRC must be performed with caution in patients with advanced stage disease. Likewise, DiDonna and colleagues reported similar findings with PRC, cautioning against this procedure in patients younger than 35 years.
In an attempt to address capitoradial arthritis, Salomon and Eaton also recommended an interpositional arthroplasty of the dorsal wrist capsule as an adjunct to PRC for patients with evidence of capitate head degenerative changes. In their study of 12 patients, with an average of 55 months of follow-up, 4 patients were treated with the addition of a dorsal capsule interpositional arthroplasty. There was a significant increase in postoperative flexion and grip strength, and most patients reported no pain. An example of a patient with stage IIIB disease treated successfully with PRC can be seen in Fig. 3 .
