David M. Lichtman, MD, and Gregory G. Degnan, MD
Since its original description by Peste' in 1843 and subsequently by Kienbbck in 1910, Kienbock's disease has remained an enigma to orthopedic surgeons. Despite con-tinued efforts by various authors, neither the exact cause nor a universally accepted treat-ment for this condition has been established. The consistent histologic findings of de-creased lunate vascularity, fragmentation, and osseous necrosis have led to the accep-tance of avascular necrosis as being the essen-tial change in this disorder. The actual cause of the vascular impairment, however, contin-ues to elude us.
Because this disease has several potential causes, it is not surprising that a myriad of treatment options have evolved over the years. The following discussions of etiology and staging for Kienbock's disease make it easier for the surgeon to individualize treat-ment for the patient who presents with this relatively uncommon condition.
ETIOLOGY
Peste in 1843, first described collapse of the carpal lunate in anatomic specimens. He believed it was the result of a fracture with a traumatic cause. Kienbock, in 1910, subse-quently published his description and stated that he believed the lesion to be the result of trauma. Repeated sprains or contusions were thought to lead to ligamentous and vascular injury. Since these early descriptions, a multi-tude of authors have described the pathologic changes as avascular necrosis.
Numerous authors since Peste and Kien-bock have implicated trauma or acute fracture as a cause of the avascular necrosis, Beckenbaugh reported an 82% incidence of fracture lines on radiographs of his patients. The use of tomography has increased the number of reports on the presence of these fractures. It is still not clear, however, whether these fractures are the cause or the result of the avascular necrosis.
White and Omer in 1984, pointed out that classic avascular necrosis of the lunate is rare after fracture-dislocation or dislocation of the carpus. They presented a series of cases dem-onstrating a transient vascular compromise of the lunate after these injuries. These cases were characterized by an abnormal radio-graphic density consistent with Kienbock's disease. They note, however, that the clinical course is one of resolution rather than one of progression.
Stahle postulated that in a lunate with an already tenuous blood supply, traumatic compression fracture leads to avascular ne-crosis. Lee, in his cadaver experiments, found that the lunate has three predictable vascular patterns: (1) a single volar or dorsal vessel supplying the entire bone; (2) several vessels, volar and dorsal, without central anastomosis; and (3) several vessels, volar and dorsal, with central anastomosis. He theorized that patients with either of the first two patterns are at greater risk for developing Kienbbck's disease.
Gelberman et all' studied the extra and in-traosseous blood supply in fresh specimens. They found an extensive extraosseous blood supply volarly and dorsally in the vast major-ity of specimens. In only 7% was the extraos-seous contribution from the volar side only. When reviewing the intraosseous supply, they found three patterns: Y in 59%, 1 in 31%, and X in 10%. The central anastomosis was just distal to the center of the lunate in all patterns. This study revealed that the least vascular zone was in the proximal subchon-dral area adjacent to the radial articular sur-face. Based on this work, Gelberman sug-gested that it is intraosseous disruption, caused by repeated trauma, that causes the avascular necrosis.
Hulten.20 in 1928, described the ulna-minus variant. He noted the presence of a short ulna in 78% of his patients with Kienb6ck's disease as compared with 23% of the normal popula-tion. Since this description, numerous authors have confirmed the increased incidence of ulnar-minus variance in Kienbock's disease. It is theorized that a short ulna causes increased shear forces on the lunate. These forces are thought by most authors to be a contributing factor in the development of this disorder. In summary, it is currently the consensus that certain "lunates, [are] at risk." These lu-nates include those associated with a short ulna and those that have a single nutrient ves-sel or a limited intraosseous blood supply. When these lunates are exposed to repeated minor trauma, interruption of the blood sup-ply results in avascular necrosis.
DIAGNOSIS AND STAGING
Kienbock's disease is diagnosed radio-graphically. The characteristic changes of the lunate include increased density, fracture lines, fragmentation, and progressive col-lapse. The early finding of increased density must be distinguished from posttraurnatic transient vascular compromise as described by White and Omer. The key feature of this disorder is that no radiographic progression occurs. In true Kienbock's changes follow a predictable sequence.
Initially, in the early phases of the disease, the changes are localized to the lunate. In these phases, radiographs reveal increased density, which progresses to lunate collapse. In the later phases, the pathology involves the mechanical structure and kinematics of the wrist, not just the lunate. In the more severe disease, the capitate migrates proximally, the proximal row widens, and rotatory instability of the scaphoid and diffuse degenerative changes occur.
The degree of structural and kinematic al-teration has significant ramifications when considering a treatment approach, which was the impetus for the development of a classifi-cation scheme that would help the surgeon select an appropriate treatment from the myr-iad of choices available.
In 1977, Lichtman and coworkers modi-fied Stahle's original radiographic classifica-tion in an attempt to help guide the choice of treatment options. This classification consists of four stages and differentiates those phases with isolated involvement of the lunate from those phases in which the structural and kine-matic architecture of the wrist is altered.
Stage I
In this stage, radiographs are normal except for the possibility of a linear or compression fracture that may be shown on tomog-raphy. In this phase, bone scan usually is ab-normal, and magnetic resonance imaging (MRI) is currently diagnostic.
It should be noted that MRI was not available for the diagnosis of this disorder when this classification was devised. Many have suggested that stage 0 should be added for those cases in which MRI changes are the only diagnostic clues. I should be clearly understood that these cases fit into stage I.
Stage I was a stage awaiting a diagnostic tool. In fact, Lichtman states in his paper that "ra-dionucleide scanning in stage I may be abnor-mal, and refinements in current techniques may soon permit reliable early diagnosis.
Stage II
Disease in this stage is still localized to the lunate. The size, shape, anatomic relationship, and kinematics of the carpal bones are not significantly altered. The lunate has a definite increased density relative to the other carpal bones. Late in this stage, some-height may be lost on the radial side of the lunate fracture that may be shown on tomog-raphy. In this phase, bone scan usually is ab-normal, and magnetic resonance imaging (MRI) is currently diagnostic.
Stage III
This stage is the transitional one in which the disease begins to affect the carpal structure and kinematics. At this point, the lunate has collapsed in the frontal plane and elon-gated in the sagittal plane. The capi-tate begins to migrate proximally. Foreshort-ening of the scaphoid (the ring sign), scapholunate dissociation, and uInar devia-tion of the triquetrum may or may not be present radiographically. Measurement of the carpal height ratios, defined by Yourn et al , or the lunar-perimeter and lunar-area indices, described by Weiss,' help determine the de-gree of carpal collapse.
This transitional stage is divided into Stages IIIA and IIIB- In stage IIIA, lunate collapse occurs without fixed scaphoid rotation; therefore, the kinematics of the prox-imal row are altered minimally. This charac-teristic has significant implications when considering treatment options. Stage IIIB has lunate collapse with fixed scaphoid rotation and other secondary derangements.
Stage IV
All the findings of stage III are present as well as generalized carpal degeneration.
TREATMENT
Since Kienbock's description in 1910, numerous nonoperative and operative treat-ments have been advocated for this disorder. Simple immobilization," wrist denervation,' simple excision of the lunate, and dor-sal flap arthroplast were among the early proposed treatments. Later, equalization pro-cedures, uInar lengthening, and radial short-ening, were performed for u1nar minus variants. Silicone replacement ar-throplasty, soft-tissue replacement arthroplasty, and various limited intercarpal fusions, have all been re-ported with reasonable results. Finally, wrist fusion and proximal row carpectomy have been advocated as salvage procedures.
The problem with reviewing the literature on these procedures is that with most series, and with virtually all series prior to 1977, either the stage of the various patients is not defined or the results are not well correlated with the stage. This makes analysis of the data difficult. It was this problem that stimu-lated this senior author to propose a classifi-cation scheme that would be helpful clinically in guiding a treatment approach.
Two major radiographic features influence our treatment choice. These factors are the presence or absence of an u1nar-minus, vari-ance and the stage of the disease. The pres-ence of a short ulna allows us the option of an equalization procedure in the early stages to attempt to decrease the shear stresses and prevent carpal collapse. The staging of the disease tells us whether the procedure should attempt to salvage the lunate or whether it should address the already altered carpal structure and kinematics.
Immobilization
The theory behind the use of immobiliza-tion in Kienbock's disease is that by decreas-ing the forces and shear stresses across the carpus, the lunate may be able to revascular-ize. Stahle initially advocated immobiliza-tion, and it has been tried in all stages of the disease by various authors. In most series, however, the results have been less than sat-isfactory and progressive collapse is common. In stage I, however, immobilization may still be a reasonable approach to minimize the vascular insult and allow the lunate to heal.
Revascularization
In stages 1, 11, or IIIA, it is theoretically fea-sible for the lunate to regain blood supply and reconstitute. Braun has described a ped-icle graft using a piece of volar radial bone attached to the pronator quadratus. This pro-cedure has been reported to be successful with follow-up of up to 7 years.
Our experience with revascularization has been with direct transplantation of a vascular bundle into the lunate as described by Hori. Results in our limited series have been encouraging. This procedure is one we would consider for stages 1, 11, or IIIA with uInar-neutral or u1nar-positive variance.
Equalization
Those patients who have u1nar-minus vari-ance may benefit from an equalization of the distal articular surfaces to reduce shear stresses across the limate. txcellent results have been reported with both radial shorten-ing and u1nar lengthening. These procedures cannot, however, restore an already collapsed lunate with carpal instabil-ity, and they should not, therefore, be used in advanced stages IIIB or IV disease.
Both of these procedures require osteotomy and fixation usually with a compression plate. The ulnar variance should be changed to 1 to 2 mm u1nar positive. Preoperatively, the degree of uInar variance should be meas-ured using the technique of Gelberman and coworkers." This variance should be measured from the standard radiographic view described by Palmer et al.
Limited Intercarpal Fusion
In stage IIIB, the rotatory subluxation of the scaphoid must be addressed.
Triscaphe Fusion
Triscaphe fusion and scaphocapitate fusion are good options in stage IIIB disease because they address the flexion deformity of the scaphoid and prevent proximal migration of the distal carpal row. In 1985, Watson et all' reported their results with triscaphe arthro-desis performed with and without silicone re-placement arthroplasty. They reported good results in 10 of 16 patients. This procedure is an attractive option in stage IIIB disease because it addresses the rotatory subluxation of the scaphoid and prevents proximal migra-tion of the distal carpal row.
Scaphocapitate fusion may accomplish the same goals with a much more limited expo-sure. Capitohamate fusion has been advo-cated by some , but we do not currently reccommend this technique because migration of the capitate is accompanied by migration of the entire distal row, including the hamate.
Salvage Procedures
In stage IV Kienbock's disease, good results have been reported with proximal row car-pectomy, and wrist arthrodesis . The choice between these two procedures is deter-mined by the needs and desires of the patient and the integrity of the articular surfaces of the lunate fossa and the capitate. Proximal row carpectomy should be reserved for those patients who desire motion over strength, whereas the wrist fusion is preferable in those patients who need strength. Total wrist ar-throplasty is generally contraindicated in these young, active patients.
SUMMARY
For more than 80 years, surgeons have staged an unsuccessful search for a univer-sally acceptable treatment for Kienbock's dis-ease. It is our contention that no single treatment will be universally successful. Treatment choice must be based on a number of variables, including the experience of the surgeon, the desires and activity level of the patient, the anatomic variation of the ulna, and most importantly, on the stage of the dis-ease. In the early stages, efforts should be made to salvage the lunate and prevent loss of normal architecture. In the later stages, ef-forts should be made to restore that architec-ture. In end stage, normal architecture must be sacrificed to restore function.
Currently we recommend immobilization with possible equalization procedures for pa-tients with u1nar-minus variance and stage I disease. In a patient with stage I disease and ulnar-positive variance, we recommend im-mobilization with consideration for a revas-cularization procedure. For stages 11 or IIIA disease with u1nar-minus variance, we at- Our experience with revascularization has been with direct transplantation of a vascular bundle into the lunate as described by Hori. Results in our limited series have been en-couraging. This procedure is one we would consider for stages 1, 11, or IIIA with uInar-neutral or u1nar-positive variance.
Equalization
Those patients who have u1nar-minus vari-ance may benefit from an equalization of the distal articular surfaces to reduce shear stresses across the limate. txcellent results have been reported with both radial shorten-ing and u1nar lengthening. These procedures cannot, however, restore an already collapsed lunate with carpal instabil-ity, and they should not, therefore, be used in advanced stages IIIB or IV disease.
Both of these procedures require osteotomy and fixation usually with a compression plate. The ulnar variance should be changed to 1 to 2 mm u1nar positive. Preoperatively, the degree of uInar variance should be meas-ured using the technique of Gelberman and coworkers." This variance should be meas-ured from the standard radiographic view described by Palmer et al.
Limited Intercarpal Fusion
In stage IIIB, the rotatory subluxation of the scaphoid must be addressed.
Triscaphe Fusion
Triscaphe fusion and scaphocapitate fusion are good options in stage IIIB disease because they address the flexion deformity of the scaphoid and prevent proximal migration of the distal carpal row. In 1985, Watson et all' reported their results with triscaphe arthro-desis performed with and without silicone replacement arthroplasty. They reported good results in 10 of 16 patients. This procedure is an attractive option in stage IIIB disease because it addresses the rotatory subluxation of the scaphoid and prevents proximal migra-tion of the distal carpal row.
Scaphocapitate fusion may accomplish the same goals with a much more limited expo-sure. Capitohamate fusion has been advo-cated by some 8 but we do not currently rec-ommend this technique because migration of the capitate is accompanied by migration of the entire distal row, including the hamate.
Salvage Procedures
In stage IV Kienb6ck's disease, good results have been reported with proximal row car-pectomy, and wrist arthrodesis . The choice between these two procedures is deter-mined by the needs and desires of the patient and the integrity of the articular surfaces of the lunate fossa and the capitate. Proximal row carpectomy should be reserved for those patients who desire motion over strength, whereas the wrist fusion is preferable in those patients who need strength. Total wrist ar-throplasty is generally contraindicated in these young, active patients.
SUMMARY
For more than 80 years, surgeons have staged an unsuccessful search for a univer-sally acceptable treatment for Kienbock's dis-ease. It is our contention that no single treatment will be universally successful. Treatment choice must be based on a number of variables, including the experience of the surgeon, the desires and activity level of the patient, the anatomic variation of the ulna, and most importantly, on the stage of the dis-ease. In the early stages, efforts should be made to salvage the lunate and prevent loss of normal architecture. In the later stages, efforts should be made to restore that architec-ture. In end stage, normal architecture must be sacrificed to restore function.
Currently we recommend immobilization with possible equalization procedures for pa-tients with u1nar-minus variance and stage I disease. In a patient with stage I disease and ulnar-positive variance, we recommend im-mobilization with consideration for a revas-cularization procedure. For stages 11 or IIIA disease with u1nar-minus variance, we at- tempt an equalization procedure. For stages 11 or IIIA disease with u1nar-positive variance, we recommend revascularization as per-formed by Hori In stage IIIB disease, we prefer a triscaphe fusion to restore carpal sta-bility and prevent further degeneration. In stage IV disease, proximal row carpectomy or wrist arthodesis is indicated.