Jeffrey E. Budoff
DEFINITION
Kienböck disease is a disorder of undetermined etiology that results in avascular necrosis (AVN) of the lunate.5
ANATOMY
Lunate Vascularity
The extraosseous blood supply of the lunate is extensive: branches of the radial and anterior interosseous arteries form a dorsal lunate plexus and branches of the radial, ulnar, and anterior interosseous arteries as well as the recurrent deep palmar arch form a volar plexus.
The intraosseous blood supply is variable. Because the lunate is covered by cartilage proximally and distally, vessels can enter the bone only at its dorsal and volar poles.1,13
Three studies have identified “lunates at risk” from a vascular standpoint. The vulnerable lunate is one that has large areas of bone dependent on a single intraosseous vessel, which occurs in 7% to 20%. In addition, 31% of lunates have no internal arterial branching.6,7,16 These internal vascular arrangements may render the lunate more vulnerable to AVN, as injury to the single vessel could not be compensated for by collateral flow.
Ulnar Variance
The standard posteroanterior (PA) wrist radiograph is taken with the shoulder and elbow at 90 degrees and the forearm in neutral rotation.
In this view, the length of the distal ulna with respect to the distal radius is called ulnar variance (FIG 1).
When the ulna is the same length as the radius, it is said to have neutral ulnar variance. When the ulna is shorter than the radius, it is referred to as negative ulnar variance, and when the ulna is longer than the radius it is referred to as positive ulnar variance.
Theoretically, a negative ulna variance increases shear forces on the lunate.
The triangular fibrocartilage complex (TFCC) is thicker in these patients and the difference in compliance between it and the ulnar edge of the radius is accentuated, leading to greater shear force.
In addition, loads across the radiocarpal joint are borne disproportionately by the radius.5
In the North American population, Kienböck disease is associated with an ulnar negative variance.
This relationship does not hold true in the Japanese literature.1
Other authors have noted a tendency toward smaller lunates in patients with Kienböck disease.3
PATHOGENESIS
The cause of Kienböck disease is incompletely understood. Current thinking is that acute or repetitive trauma causes excessive shear forces on a lunate at risk, interrupting its intraosseous vascularity and leading to AVN.1,2
While a history of injury is elicited in over 50% of cases, the absence of a single traumatic event is still very common.
Fracture of the lunate has been reported in up to 82% of lunates with Kienböck disease.1 However, it remains unclear whether these fractures are the cause or the result of AVN.
Kienböck disease is not seen after lunate or perilunate dislocations.5,13
Although transient ischemia may be seen after carpal fracture-dislocations, this spontaneously resolves after 5 to 32 months and should be treated expectantly.1,2
The key feature of transient ischemia is that no progressive radiographic collapse occurs, as opposed to Kienböck disease, where radiographic changes and collapse are predictable.
It has been suggested that Kienböck disease may be due to venous outflow obstruction with intraosseous vascular congestion, rather than arterial insufficiency. Increased intraosseous pressure has been shown in lunates with Kienböck disease, as well as in femoral heads with AVN.
FIG 1 • Measurement of ulnar variance. Ulnar variance is determined by extending a line from the radius's articular surface ulnarward and measuring the distance between this line and the distal surface of the ulnar head. Neutral ulnar variance occurs when the carpal surface of the radius and ulna are equal in height. If the ulna is shorter than the radius, negative ulnar variance exists; if the ulna is longer than the radius, positive ulnar variance exists.
This is more consistent with venous stasis than arterial compromise.
This increased pressure could also be due to bony collapse.3,5
Once the lunate becomes avascular, stress fractures occur first in the proximal lunate adjacent to the radial articular surface, where the blood supply is poorest.1,6,13 Consequently, the proximal lunate is usually more involved and more flattened than the distal lunate. In addition, the radial lunate that articulates with the distal radius is usually more involved than the ulnar lunate that overlies the triangular fibrocartilage, probably because of the difference in compliance between the two supporting surfaces. This difference is accentuated in patients with ulnar-negative variants.13
Lunate collapse leads to loss of carpal height. If a coronal plane fracture is present, the compressive forces of the capitate displace these two fragments volarly and dorsally.13
NATURAL HISTORY
The natural history of Kienböck disease is one of progressive fragmentation and collapse of the lunate, loss of carpal height with scaphoid flexion, and proximal capitate migration leading to perilunate arthritis. However, these changes do not universally lead to a poor clinical outcome.5
A follow-up study of 49 patients compared 23 wrists treated with mean 8 weeks of immobilization and 26 without treatment.5,11
In both groups, the majority reported a gradual decrease in symptoms over time.
At mean 20.5 years of follow-up, 83% of the wrists in the immobilized group were pain-free or were painful only with heavy work.
In the nontreated group, this was true for 77%.
In all wrists, the lunate was deformed and 67% developed radiocarpal arthritis on radiographs.
The authors concluded that Kienböck disease has a naturally benign course.
There was no correlation between residual symptoms and the radiographic appearance, including the appearance of arthritis.
In this study, immobilization did not lead to any longterm benefit.
PATIENT HISTORY AND PHYSICAL FINDINGS
Most patients with Kienböck disease are young, active patients between 20 to 40 years of age.
This has led to significant concerns about the long-term effects of this disorder.
The male–female ratio is approximately 3:1 to 7:1. It is rarely bilateral.3,21
Regardless of gender, more than 95% of patients are engaged in heavy manual labor.21
The most common complaints are dorsal central wrist pain, stiffness, and significant weakness of grip, which is often reduced to 50% of the opposite hand.1,5,13
There may be a long history of symptoms before presentation.
The pain may vary in intensity from mild discomfort to constant, debilitating pain. It is often activity-related and improves with rest and immobilization.
A history of trauma is variable.1,5
The wrist is typically mildly swollen dorsally, consistent with synovitis, and is tender over the lunate.
Flexion and extension are predictably diminished.
Wrist flexion is more likely to be limited than extension because the volar pole of the lunate often extrudes so that it impinges against the volar rim of the distal radius.
Forearm rotation is not affected.13
While Kienböck disease has been reported in association with steroid use, septic emboli, sickle cell disease, gout, carpal coalition, and cerebral palsy, there is no well-defined correlation with any systemic or neuromuscular process that warrants screening when considering the diagnosis.3
IMAGING AND OTHER DIAGNOSTIC STUDIES
Radiographic Classification
Kienböck disease is diagnosed radiographically,12 and staging is based on plain radiographs.
In 1977, Lichtman and Degnan12 modified Stahle's original radiographic classification in an attempt to help guide treatment decisions (FIG 2).
Stage I
Radiographs are normal, although a linear fracture without sclerosis or lunate collapse is occasionally present.
MRI shows the characteristic changes of AVN (FIG 3A).5,21
Stage II
The lunate becomes sclerotic and radiodense, similar to the radiologic appearance of other bones with AVN (FIG 3B). A coronal fracture splitting the lunate into dorsal and volar fragments may be noted.
Late in stage II, some loss of lunate height on the radial side may be evident.
The lunate retains its overall shape, and its anatomic relationship to the other carpal bones is not significantly altered.12,21
Stage III
The lunate collapses in the coronal plane and elongates in the sagittal plane. The carpal architecture is altered and the capitate begins to migrate proximally.
Stage IIIA
Lunate collapse has occurred, but carpal height is relatively unchanged and carpal collapse has not yet led to proximal migration of the capitate or scaphoid flexion. Therefore, the carpal kinematics have not yet been significantly altered.
Stage IIIB
The carpal collapse with proximal capitate migration has led to fixed scaphoid flexion, which may be noted on the AP radiograph as the “cortical ring sign.”3,12,21
Stage IV
Arthritis of the radiocarpal or midcarpal joint has resulted from the collapse, fractures, and altered carpal kinematics, leading to joint space narrowing, osteophyte formation, subchondral sclerosis, and degenerative cysts.3,21
MRI and CT
MRI is extremely sensitive in detecting changes in marrow fat that are consistent with, but not diagnostic of, AVN.
Decreased signal on T1 sequences represents replacement of the normal fatty marrow by dead bone or fibrous tissue.21
FIG 2 • Kienböck disease stage classification based on radiographic appearance.
Because MRI detects only the loss of marrow fat and not AVN specifically, to consider an MRI diagnostic for Kienböck disease over 50% of the lunate should be hypointense on T1 because the changes of Kienböck disease are diffuse, as opposed to other conditions such as ulnocarpal impaction, fractures, and intraosseous tumors, which cause more focal MRI changes.4,20,22
It is possible that a large enchondroma, interosseous ganglion, or other marrow-replacing lesion could lead to MRI changes in over 50% of the lunate. Thus, there is currently no truly pathognomonic imaging sign for Kienböck disease.4
T2 images typically show low signal intensity, which represents replacement of the normal fatty marrow by fibrosis.21
An increased T2 signal may occur if intramedullary edema is present or if revascularization is occurring.3,4,20 Thus, when the T2 images show normal or increased signal intensity, an earlier stage of disease with a better prognosis can be inferred.20,21
Although it cannot diagnose AVN directly, MRI is still the optimal imaging modality and gold standard for diagnosing Kienböck disease, especially before trabecular bone has been destroyed.
Gadolinium-enhanced MRI may provide a more sensitive means of evaluating lunate vascularity.
CT may upstage the disease compared with radiographs in 89% of those originally considered to have stage I, 71% with apparent stage II, and 9% with apparent stage III disease on radiographs.3
Once lunate collapse has occurred, CT best reveals the extent of necrosis and trabecular destruction.3
FIG 3 • A. Magnetic resonance image of wrist with Kienböck disease demonstrates diminished signal intensity of the lunate. B. Radiograph showing density changes in the lunate in Kienböck disease. (From Bishop AT, Pelzer M. Avascular necrosis. In: Berger RA, Weiss A-PC, eds. Hand surgery. Vol 1. Philadelphia: Lippincott Williams & Wilkins, 2004:554.)
DIFFERENTIAL DIAGNOSIS
Ulnocarpal impaction
Rheumatoid arthritis
Radial-sided triangular fibrocartilage tears
Posttraumatic arthritis
Acute fracture
Carpal instability
Lunate fracture
Enchondroma
Osteoid osteoma
Bone island
Occult or intraosseous ganglion
Intraosseous cyst
Transient ischemia
“Bone bruise”
Paget disease
Gaucher disease4
NONOPERATIVE MANAGEMENT
A trial of 2 weeks to 3 months of immobilization may be attempted for patients with stage I Kienböck disease, especially young patients with hyperintense lunates on T2 MR images.
The theory behind the use of immobilization is that by decreasing the forces across the carpus, the lunate may be able to revascularize.12
Most series report poor results with immobilization, and progressive collapse is common.
There is no study of immobilization consisting of patients with only stage I Kienböck disease. Consequently, the efficacy of immobilization in patients with stage I disease is anecdotal.
Immobilization does not decrease compressive forces across the lunate, which are imparted by the capitate. The capitate may still force any fracture fragments apart, leading to collapse and displacement.
Immobilization leads to stiffness.
The earlier the lunate is unloaded, the less collapse is anticipated. For this reason, early surgical decompression may be considered rather than immobilization, and many clinicians treat stage I disease surgically.13
In Trumble and Irving's series of 22 patients with various stages of Kienböck disease treated with immobilization, 17 showed disease progression with continued collapse of the lunate and 5 showed no improvement.22
In Lichtman et al's series, 19 of 22 had unsatisfactory results.2,3
When immobilization fails to reverse the avascular changes, the process will almost always advance to stage II, where surgical management is strongly recommended.2,3
In a series of patients with stage II or more advanced disease treated with immobilization, 76% (19/25) had either undergone total wrist arthrodesis or experienced daily problems with their wrists at mean 8 (1–11) year follow-up.15
A study of 18 patients with stage II or III disease treated nonoperatively were compared with those treated by radius shortening.
Patients treated surgically had less pain and better grip strength.
In some patients with stage III disease treated nonoperatively there was rapid deterioration to carpal collapse.
Although radius shortening did not reverse or prevent carpal collapse, it slowed the process.18
SURGICAL MANAGEMENT
There is no agreement on the optimal way to treat Kienböck disease.5 Multiple options for surgical management exist and the results do not vary significantly between the different procedures.
The mainstays of treatment are radius-shortening osteotomy and proximal row carpectomy.5
Two major radiographic features influence treatment choice: the stage of the disease and ulnar variance.12
Radius-shortening osteotomy is currently the benchmark against which other treatments are judged.13
For stages I to IIIB, radius-shortening osteotomy is a very popular option in patients who are ulnar negative. While the use of radius shortening in stage IIIB is controversial, because lunate height and normal carpal kinematics will not be re-established, potentially leading to progressive degenerative changes, very good results have been demonstrated in these patients with this procedure.2,24,25 Radius shortening is contraindicated for stage IV disease unless symptoms are severe and salvage procedures are not desired.24
Radius shortening decreases joint compression forces at the radiolunate joint by redistributing them to the radioscaphoid and ulnolunate joints. In addition, it relatively lengthens the tendons crossing the wrist, diminishing overall joint compressive forces.17
As opposed to ulnar lengthening, no intercalary bone graft is required and only one interface needs to heal, instead of two.
In addition, radial shortening leads to a relative lengthening of the musculotendinous units crossing the wrist, resulting in less force transmission across the carpus. Ulnar lengthening does not provide this particular advantage.24
After radial shortening, the ulnar head and TFCC support more of the wrist's compressive load through the triquetrum and the ulnar aspect of the lunate. The TFC is thicker in patients with ulna-minus variance, which provides a compliant pad to support the ulnar carpus.
Because radial-shortening osteotomy is an extra-articular procedure, it does not alter normal carpal joints or interfere with intracarpal relationships. It “burns no bridges,” and intracarpal procedures can always be undertaken at a later date if the radial shortening is ineffective and disease progression occurs.24
In patients who are ulnar-neutral or ulnar-positive, a radial closing wedge osteotomy (FIG 4) or capitate shortening with or without capitohamate fusion can be performed.
While radius shortening in patients with neutral or positive ulnar variance is not advised, good results have been reported even in these patients.1,24
For stages I to IIIA, revascularization using a vascularized pedicle or bone graft may be performed and may be combined with radius shortening or another unloading procedure (see Chap. HA-24).
In patients with stage IIIB disease, proximal row carpectomy, scaphotrapeziotrapezoid fusion, or scaphocapitate fusion may be performed with or without lunate excision and soft tissue interposition.
For stage IV disease, proximal row carpectomy or total wrist fusion may be indicated. A study of arthroscopic débridement for stage III or IV disease showed some pain relief at 19 months of follow-up.14
FIG 4 • Lateral closing wedge osteotomy. (Adapted from Soejima O, Iida H, Komine S, et al. Lateral closing wedge osteotomy of the distal radius for advanced stages of Kienbock's disease. J Hand Surg Am 2002;27A:31–36.)
Based on the hypothesis that Kienböck disease is due to venous obstruction, “metaphyseal core decompression” of the distal radius has also been reported with good results.8
Wrist denervation may also be considered and can be used as an adjunct at any stage.3
Lateral closing wedge osteotomies increase lunate coverage (joint contact area) in proportion to the decrease in radial inclination.
This transfers the compressive forces of the capitate from the lunate to the scaphoid, decreasing pressure at the radiolunate joint.17,23
To keep the wrist straight in relation to the forearm, the patient is forced to ulnarly deviate the wrist, extending the scaphoid, which may further transfer forces from the capitate to the scaphoid and decrease forces on the lunate.19
Preoperative Planning
Good-quality, standard preoperative PA radiographs should be taken with the shoulder and elbow flexed 90 degrees and the forearm in neutral rotation.
While many authors have recommended removing sufficient bone during radial shortening to result in an ulnar-neutral to 1-mm-positive variance,3 90% of the strain reduction occurs within the first 2 mm of shortening.1,5
Good results with excellent relief of symptoms have been reported removing only 2 mm of bone, regardless of variance. This has the advantages of technical ease and decreases the risk of distal radioulnar joint (DRUJ) incongruity and ulnocarpal impaction, which may occur with excessive shortening.
In patients with significant obliquity of the sigmoid notch, radial shortening should be limited to 2 mm to avoid overcompressing the DRUJ.
Postoperative ulnocarpal impaction and DRUJ incongruity are especially likely with shortenings of 4 mm or more, leading to pain with forearm rotation or limitation of forearm rotation.24 Therefore, shortening of the radius by over 4 mm is not recommended.
Patients with more than 4 mm of shortening and age greater than 30 years were found to be more likely to have poor results.1
Positioning
The patient is positioned supine with the arm on a radiolucent armboard.
Approach
A volar approach to the radius is performed.
TECHNIQUES
VOLAR APPROACH
A longitudinal incision is made over the flexor carpi radialis (FCR) tendon, ending distally at the distal volar wrist crease (TECH FIG 1A).
The approach is continued through the FCR sheath (TECH FIG 1B), with the FCR tendon retracted ulnarly to protect the palmar cutaneous branch of the median nerve (TECH FIG 1C).
The plane between the FCR and deep muscles of the radius (pronator quadratus and FCR) is bluntly dissected (TECH FIG 1D).
The distal border of the pronator quadratus and the radial insertions of the pronator quadratus and flexor pollicis longus muscles are incised with Bovie electrocautery, with care taken to retract and protect the radial artery, which does not need to be formally identified.
The volar surface of the radius is then subperiosteally exposed in a radial to ulnar direction (TECH FIG 1E).
Circumferential subperiosteal dissection should be avoided to preserve maximal blood supply to the osteotomy.
TECH FIG 1 • A. Incision. B. Dissection proceeds through the flexor carpi radialis (FCR) sheath. C. The FCR is retracted ulnarly to protect the palmar cutaneous branch of the median nerve. D. The pronator quadratus is exposed. E. The volar distal radius is subperiosteally exposed.
RADIUS-SHORTENING OSTEOTOMY
Initial Plate Application
Traditionally, a seven-hole 3.5-mm dynamic compression plate is placed as far distally as possible without riding up the volar lip of the distal radius.24
However, the newer fixed-angle volar plates used for fixation of distal radius fractures work very well and allow the osteotomy to be placed in metaphyseal bone.
To decrease the risk of nonunion, the osteotomy should be performed as distal as possible to be through metaphyseal cancellous bone, staying proximal to the DRUJ.
The plate is placed over the distal radius so that its distal fixation will be within 2 to 3 mm of the subchondral bone, without intra-articular penetration. The plate is provisionally fixed with Kirschner wires (TECH FIG 2).
Following fluoroscopic confirmation of appropriate placement, four fixed-angle screws are placed distally.
Radius Osteotomy
The osteotomy is marked proximal to the distal fixation and proximal to the DRUJ (TECH FIG 3A).
The plate is removed and the osteotomy is made at a 45-degree angle, from distal volar to proximal dorsal (TECH FIG 3B).
An oblique osteotomy has less potential for nonunion than a transverse osteotomy1 and allows placement of an interfragmentary compression screw for additional fixation.
TECH FIG 2 • A,B. The volar locking plate is placed so that its distal fixation (represented radiographically by a Kirschner wire) travels just proximal to the subchondral surface. Distal locking screw fixation is placed but not fully tightened.
TECH FIG 3 • A. The osteotomy site is marked between the plate's proximal and distal fixation. When using a plate specifically designed for the fixation of distal radius fractures, this automatically places the osteotomy proximal to the distal radioulnar joint. B. The osteotomy is created with a saw at a 45-degree angle from distal volar to proximal dorsal. C. The saw is used to remove 2 to 3 mm of bone, proceeding from volar to dorsal. D.The oblique osteotomy is finished and 2 to 3 mm of dorsal cortex is removed.
A longitudinal line may be marked across the osteotomy site to allow rotational assessment. However, the flat surface of the volar cortex allows for easy assessment of rotation.
An elevator can be placed on the dorsal surface of the osteotomy to protect the extensor tendons from the saw.
Two to 3 mm of shortening may be appropriate regardless of the amount of negative ulnar variance present.
For the reasons noted above, I prefer to shorten the radius by only 2 to 3 mm.
Excellent results have been reported with osteotomies that do not fully correct the radius length to neutral variance.24,25
The 2 to 3 mm to be taken is measured out and marked and the full amount of bone to be taken is removed from volar to dorsal so that the dorsal cortex remains intact to stabilize the bone during bone removal (TECH FIG 3C).
The dorsal cortex is then removed last (TECH FIG 3D).
During the osteotomy, constant cool irrigant is used to avoid thermal osteonecrosis.
While a slight (1 mm) concave bend in the plate over the osteotomy site may occasionally be needed to achieve compression of the dorsal osteotomy surface, this is not usually necessary.
Final Plate Application and Osteotomy Fixation
The plate and its distal fixation are then replaced.
Approximation of the two bone ends may also be facilitated by radial deviation of the wrist24 and use of a Verbrugge clamp.
A bicortical screw is placed 1 cm proximal to (not through) the plate (TECH FIG 4A).
TECH FIG 4 • A. The plate and its distal fixation are replaced. A bicortical screw a few millimeters longer than the bone width is placed 2 to 3 cm proximal to the plate and left proud. A retractor may provide the necessary proximal exposure without lengthening the incision. B. A Verbrugge clamp is placed in the most proximal plate hole and around the proximal screw. Squeezing the clamp provides a tremendous mechanical advantage to facilitate osteotomy closure. C. With the clamp compressing the osteotomy, the first proximal screw is drilled eccentrically through the most proximal aspect of a plate hole to provide additional compression. D. Three proximal bicortical screws are placed. E. The most distal of the proximal screws may be placed in lag fashion, overdrilling the near cortex. F. A stably fixed and well-compressed osteotomy. The tip of a scalpel blade could not be forced into the osteotomy. G,H. Intraoperative PA and lateral radiographs. The significant obliquity of the distal radioulnar joint (DRUJ) led to some radial displacement of the distal fragment. This should be allowed, as it “decompresses” the DRUJ. In patients with significant DRUJ obliquity, only 2 mm of shortening should be performed. I.Osteoperiosteal shingling of the volar cortex, which may facilitate osteotomy healing. Synthetic bone substitute was then placed over the shingled volar cortex.
The hooked end of the Verbrugge is placed in the plate's most proximal screw hole and the bifid end is placed around the screw proximal to the plate (TECH FIG 4B).
The Verbrugge clamp is closed manually, imparting tremendous mechanical advantage to compress the osteotomy.
The first screw is placed in a compression mode eccentrically in the plate hole just proximal to the osteotomy (TECH FIG 4C).
Reduction of the osteotomy and fixation are evaluated fluoroscopically.
Adjustments are made as necessary and the remaining screws are placed (TECH FIG 4D).
A lag screw is placed obliquely across the osteotomy through the most distal of the proximal plate holes for additional fixation (TECH FIG 4E–H).
After irrigation, osteoperiosteal shingling may be performed with allograft or bone substitute placed over the shingled cortex to facilitate healing (TECH FIG 4I).
Forearm rotation should be checked to ensure that it is full.
If forearm rotation is limited after osteotomy, the radius should be translated radially or a lateral closing wedge component added.17
Radiographs often show some mild residual gap at the osteotomy site even with full compression under direct vision.24
Intraoperative radiographs may not demonstrate the eventual ulnar variance (amount of radial shortening) because of soft tissue restraints at the DRUJ. In these cases, postoperative radiographs will demonstrate the anticipated correction.17
RADIUS CLOSING WEDGE OSTEOTOMY
A radial closing wedge osteotomy may be performed through the same approach with the same fixation.
A 15-degree radial closing wedge osteotomy is performed 4 to 5 cm proximal to the tip of the radial styloid and proximal to the DRUJ.23
POSTOPERATIVE CARE
The extra-articular nature of this procedure combined with stable internal fixation allows for quick postoperative rehabilitation.
The wrist is splinted for 2 weeks, after which a removable splint may be used and gentle motion started.
The osteotomy usually heals in 2 to 3 months, although 4 or 5 months is occasionally required.
OUTCOMES
A review of the reported series by Weiss24 in 1993 included 121 patients treated with radius shortening, with about 85% good or excellent results at just over 4 years of follow-up.
One study reviewed 30 wrists after radial shortening osteotomy for stages I to IIIB Kienböck disease at mean 3.8 years of follow-up.25
Pain decreased in 87% and grip strength improved in 49%. However, the radiographic appearance of the lunate changed little if at all.
The authors noted that good results could be obtained by shortening less than that required to attain neutral ulnar variance.
The exact amount of radius shortening may not be as important as the relative unloading of the lunate resulting from the shortening of the radius. The amount of shortening needed to be effective may be only about 2 mm. Radial shortening may therefore be used in ulnar-neutral wrists.
In addition, excellent results were realized in patients with stage IIIA and IIIB disease. There was one nonunion. Only 10 of 30 wrists had evidence of possible lunate revascularization, as indicated by decreased sclerosis and a more normal trabecular pattern.
Clinical improvement after radius shortening or radial wedge osteotomy does not necessarily correlate with the radiographic results.1,5,19 It appears that the lunate “stands still in time” after radius shortening, with no significant further deterioration or improvement in the lunate architecture or height.24
Another study reviewed 68 radius-shortening osteotomies at a mean of 52 months of follow-up.17
Pain was diminished in 93%, grip strength was improved in 74%, and motion was improved in 52% and worsened in 19%.
Twenty-five patients had undergone one or more additional procedures concurrently, which did not lead to a significant difference in clinical outcomes.
Complications were uncommon; there were no nonunions, but ulnocarpal impaction developed in two patients.
Lunate density was improved in 40%, unchanged in 46%, and increased (worsened) in 14%.
Fifty-five percent of wrists that underwent concurrent vascularized bone grafting of the lunate had an improved radiographic appearance, compared to only 20% that underwent isolated radius shortening.
It has been suggested that prognosis is improved in younger patients due to increased remodeling potential.12
Teenage patients (aged 11 to 19 years) were treated by radius shortening or lateral closing wedge osteotomy.9 Two had neutral or positive ulnar variance. At a mean 50 months of follow-up, 10 of 11 were pain-free. Five of six with stage IIIB disease had excellent outcomes.
The other patient had moderate wrist pain during strenuous activity, leading to only a fair result after lateral closing wedge osteotomy for stage IIIB disease.
Radiographic improvement, indicating possible lunate revascularization, was seen in 8 of 11 patients.
There were no complications of radial overgrowth or growth abnormalities in these patients.
Twenty-five patients were followed for a minimum of 10 years (mean 14.5 years) after radial osteotomy.10
Ninety-six percent had good or excellent results.
Pain, motion, and grip strength were all significantly improved after surgery and the results were maintained.
Although radiologic improvement was not drastic and carpal height did not significantly improve, sclerosis and bone cysts improved and there was evidence of improved lunate revascularization over time.
Osteoarthritic changes were observed in 54% at 5 years and in 73% at the time of final follow-up, but the arthrosis was generally mild and did not affect the clinical results.
Severe osteoarthritis and proximal migration of the capitate were avoided.
Radius shortening was used for patients with ulnarnegative variants and closing wedge osteotomy for those with ulnar-positive variants. These procedures gave identical outcomes.
Iwasaki et al9 also noted that both radius shortening and lateral closing wedge osteotomies gave equally acceptable results in adult patients.
Good long-term results were reported in 100% of 13 patients at a mean of 14 years after radial closing wedge osteotomy.23
Pain relief was good, and improvements in grip strength and range of motion were seen.
Radiographic changes improved in one, did not change in four, and advanced in eight.
COMPLICATIONS
Nonunion has been reported in up to 6% of cases.5
If the fixation remains stable, treatment should consist of autogenous cancellous bone grafting if healing has not occurred by 5 or 6 months.
A second operation may occasionally be necessary for plate removal, but this is uncommon.
Care must be taken not to overshorten the radius, or DRUJ incongruity or ulnocarpal impaction may occur.24
REFERENCES
1. Alexander CE, Alexander AH, Lichtman DM. Kienbock's disease and idiopathic necrosis of carpal bones. In: Lichtman DM, ed. The Wrist and Its Disorders, 2nd ed. Philadelphia: WB Saunders, 1997:329–346.
2. Alexander AH, Lichtman DM. Kienbock's disease. Orthop Clin North Am 1986;17:461–472.
3. Allan CH, Joshi A, Lichtman DM. Kienbock's disease: diagnosis and treatment. J Am Acad Orthop Surg 2001;9:128–136.
4. Budoff JE, Lichtman DM. Spontaneous wrist fusion: an unusual complication of Kienbock's disease. J Hand Surg Am 2005;30A:59–64.
5. Divelbiss B, Baratz ME. Kienbock's disease. J Am Soc Surg Hand 2000;1:1–12.
6. Gelberman RH, Bauman TD, Menon J, et al. The vascularity of the lunate bone and Kienbock's disease. J Hand Surg Am 1980;5A:272–278.
7. Gelberman RH, Szabo RM. Kienbock's disease. Orthop Clin North Am 1984;15:355–367.
8. Illarramendi AA, Schulz C, De Carli P. The surgical treatment of Kienbock's disease by radius and ulna metaphyseal core decompression. J Hand Surg Am 2001;26A:252–260.
9. Iwasaki N, Minami A, Ishikawa J, et al. Radial osteotomies for teenage patients with Kienbock disease. Clin Orthop Relat Res 2005;439:116–122.
10. Koh S, Nakamura R, Horii E, Nakao E, et al. Surgical outcome of radial osteotomy for Kienbock's disease: minimum 10 years of follow-up. J Hand Surg Am 2003;28A:910–916.
11. Kristensen SS, Thomassen E, Christensen F. Kienbock's disease: late results by non-surgical treatment: a follow-up study. J Hand Surg Br 1986;11B:422–425.
12. Lichtman DM, Degnan GG. Staging and its use in the determination of treatment modalities for Kienbock's disease. Hand Clin 1993;9:409–416.
13. Linscheid RL. Kienbock's disease. AAOS Instr Course Lect 1992;41:45–53.
14. Menth-Chiari WA, Poehling GG, Wiesler ER, et al. Arthroscopic debridement for the treatment of Kienbock's disease. Arthroscopy 1999;15:12–19.
15. Mikkelsen SS, Gelineck J. Poor function after nonoperative treatment of Kienbock's disease. Acta Orthop Scand 1987;58:241–243.
16. Panagis JS, Gelberman RH, Taleisnik J, et al. The arterial anatomy of the human carpus. Part II: The intraosseous vascularity. J Hand Surg Am 1983;8A:375–382.
17. Quenzer DE, Dobyns JH, Linscheid RL, et al. Radial recession osteotomy for Kienbock's disease. J Hand Surg Am 1997;22A:386–395.
18. Salmon J, Stanley JK, Trail IA. Kienbock's disease: conservative management versus radial shortening. J Bone Joint Surg Br 2000;82B:820–823.
19. Soejima O, Iida H, Komine S, et al. Lateral closing wedge osteotomy of the distal radius for advanced stages of Kienbock's disease. J Hand Surg Am 2002;27A:31–36.
20. Sowa DT, Holder LE, Patt PG, et al. Application of magnetic resonance imaging to ischemic necrosis of the lunate. J Hand Surg 1989;14A:1008–1016.
21. Szabo RM, Greenspan A. Diagnosis and clinical findings of Kienbock's disease. Hand Clin 1993;9:399–408.
22. Trumble TE, Irving J. Histologic and magnetic resonance imaging correlations in Kienbock's disease. J Hand Surg Am 1990;15A:879–884.
23. Wada A, Miura H, Kubota H, et al. Radial closing wedge osteotomy for Kienbock's disease: an over-10-year clinical and radiographic follow-up. J Hand Surg Br 2002;27B:175–179.
24. Weiss AP. Radial shortening. Hand Clin 1993;9:475–482.
25. Weiss AP, Weiland AJ, Moore JR, et al. Radial shortening for Kienbock disease. J Bone Joint Surg Am 1991;73A:384–391.