Skeletal Radiol DOI 10.1007/s00256-016-2332-8
REVIEW ARTICLE
Treatments for Kienböck disease: what the radiologist needs to know Carissa White 1
&
Prosper Benhaim 2 & Benjamin Plotkin 3
Received: 28 September 2015 / Revised: 12 December 2015 / Accepted: 6 January 2016 # ISS 2016
Abstract The etiology of Kienböck disease, or avascular necrosis of the lunate, is controversial, and there are a myriad of treatments aimed at correcting the various hypothesized pathologies. Interventions to reduce mechanical stress on the lunate have been used for decades, including radial osteotomy with or without radial shortening, ulnar lengthening and metaphyseal core decompression procedures. However, these procedures require preservation of lunate architecture. Newer procedures to revascularize the lunate bone have emerged in the last 10 years, such as pedicled corticoperiosteal vascularized bone grafting. Once there is collapse of the radiocarpal joint or midcarpal arthrosis, the conventional treatments have included proximal row carpectomy and complete or partial wrist joint arthrodesis. Newer salvage procedures such as lunate excision with autologous or synthetic interposition grafts are now being used when possible. As this disease is relatively rare, radiologists may not be familiar with the expected post-operative radiologic findings and complications, especially of the newer treatments. The goals of this paper are to review the available treatment options and their expected appearance on postoperative imaging, with discussion of possible complications when appropriate.
* Carissa White [email protected]
Keywords Kienböck . Lunate . Avascular necrosis . Wrist
Introduction Kienböck disease, or avascular necrosis of the lunate, is a slowly progressive disease that can eventually lead to complete loss of wrist and hand function. Kienböck disease tends to occur in people who axially load their wrist in dorsiflexion, such as manual workers, with a 2:1 male to female predominance. The typical patient is 20–40 years of age [1], although recent studies have shown radiographic findings of the disease to be present in 0.10 % of asymptomatic patients, many of whom are older than the 5th decade [2]. The disease is usually unilateral, with bilateral disease being reported only 4 % of the time, often with one of the wrists being asymptomatic [3]. There are case reports of bilateral Kienböck disease is association with type 1 diabetes mellitus [4], systemic lupus erythematosus [5] and Legg-Calve-Perthes disease [6], but these diseases do not seem to increase the risk for bilateral as opposed to unilateral disease. While the overall disease prevalence (including symptomatic and asymptomatic patients) is only 0.27 % [2], the frequency is likely increased in the setting of a specialized orthopedic or tertiary care center. Given the high morbidity of this disease, it is imperative for both the general and musculoskeletal radiologist not only to be able to accurately diagnose the disease, but also to know how to effectively evaluate the patient after conservative or surgical treatment.
1
Department of Radiology, University of California, Los Angeles, 757 Westwood Blvd. Suite 1638, Los Angeles, CA 90095, USA
2
Department of Orthopaedic Surgery, University of California, Los Angeles, 10945 Le Conte Ave, Room 33-55 PVUB, Box 957326, Los Angeles, CA 90095, USA
Etiology
Department of Radiology, University of California, Los Angeles, 1250 Sixteenth Street, Box 957036, Santa Monica, CA 90404, USA
The definitive etiology for Kienböck disease is unknown. In the past, it was thought that negative ulnar variance, where the
3
Skeletal Radiol Fig. 1 a PA radiograph demonstrating positive ulnar variance, with the ulnar articular surface projecting distal to the radial articular surface. b PA radiograph demonstrating negative ulnar variance, with the ulnar articular surface projecting proximal to the radial articular surface
ulnar articular surface projects proximal to the radial articular surface on a posterior-anterior (PA) radiograph (Fig. 1), predisposed to lunate necrosis by altering the load transmitted through the radial column. It is important that the patient be properly positioned when assessing for ulnar variance on a radiograph. The wrist should be in neutral position with the elbow flexed to 90° and the shoulder abducted to 90°. Maximum forearm pronation increases the ulnar variance (makes it more positive) while maximum supination decreases the ulnar variance (can make the variance appear falsely negative)[7]. However, recent studies have demonstrated no evidence for causation between ulnar variance and Kienböck disease [8]; moreover, newer interventions that do not alter the mechanical load on the lunate have been shown to halt and even reverse lunate necrosis [9–11]. Ulnar positive variance, on the other hand, has been shown to predispose to ulnar impaction/abutment syndrome, where the distal ulna repeatedly knocks up against the triangular fibrocartilage (TFCC) and the proximal carpal bones, leading to tears of the TFCC, and marrow edema, subchondral sclerosis and cystic changes of the lunate (87 %) and or triquetrum (43 %) [12] (Fig. 2). Other theories have concentrated on the lunate shape such as the Antuna Zapico classification, determined by the angle made by the lateral scaphoid side with the proximal radial side of the lunate. In this analysis, it was found that a square or rectangular shape of the lunate bone may increase the risk for lunate necrosis compared to a trapezoidal shape, due to the Table 1 Lichtman staging as it pertains to appearance of the lunate bone on imaging Stage
Radiographs
MRI
I II IIIa IIIb IV
Normal Increased density w/o collapse Collapse w/o carpal instability Collapse with carpal instability Collapse with radiocarpal or midcarpal arthrosis
↓ T1 signal ↓ T1 signal ↓ T1 signal ↑ T2 signal ↓ T1 signal ↑ T2 signal ↓ T1 signal ↑ T2 signal
decreased number of articular surfaces with the radius and therefore different distribution of forces on the bone [1, 13]. However, most compelling are the theories that focus on the native carpal blood supply. Normally, the lunate receives dual blood supply on both the dorsal and volar sides, with a variable intraosseous branching pattern. It has been shown that up to 7 % of people may have only a single palmar vessel [14], and up to one-third of patients may have comparatively less intraosseous branching, both of which may predispose to ischemia, especially in the setting of acute or chronic repetitive microtrauma [14–16]. Other theories suggest there may be outflow obstruction in the draining venous plexus [14], supported by the finding of increased intraosseous pressure in necrotic lunate bones [17]. However, it is unknown whether the increased venous pressure is a cause or a result of the disease [15]. Theoretically any systemic disease that predisposes to thrombosis or vascular pathology such as sickle cell disease, septic emboli, systemic lupus erythematosus or exogenous steroids, would also increase the risk for Kienböck disease, although there does not seem to be a direct correlation [5, 14].
Staging The radiological-pathological staging system for Kienböck disease was developed by Stahl 1947 and modified by Lichtman in 1977 [18], based solely on radiographs. MRI is now routinely used and has added greatly to diagnostic sensitivity, especially in the early stages of the disease (Table 1). In stage I, the lunate appears normal on plain radiographs, but demonstrates mildly decreased signal on T1-weighted MR images. If there is corresponding increased lunate marrow signal on T2-weighted images, the changes likely represent marrow edema versus innate revascularization attempts [19]. If intravenous contrast is given, the lunate will often demonstrate intense enhancement at this stage [20]. Alternatively, corresponding decreased lunate signal on T2-weighted images represents completed necrosis [21], a finding which is more likely
Skeletal Radiol
Fig. 2 Ulnar abutment syndrome. PA radiograph (a) of the wrist demonstrating only mild radiocarpal joint space narrowing. Sequential coronal T2-weighted MR images (b–c) of the wrist, demonstrating cystic
changes of the lunate and bone marrow edema in the distal ulna. In this case the ulnar variance was neutral
to be seen in stage II. The radiologist should look for increased T2 signal in the distal ulna and/or triquetrum, which in the presence of ulnar positive variance, would suggest an alternative diagnosis of ulnar abutment syndrome (UAS). A bone scan is also abnormal in stage I, although this is not routinely incorporated in the work-up. In stage II, radiographs demonstrate increased lunate density and sclerosis, without change of lunate shape. When the disease progresses to stage III, there is collapse of the lunate bone with marked hyperintensity of the lunate marrow on T2-weighted imaging (Fig. 3). The degree of lunate collapse can be quantified using the Stahl index, defined as the height (superior-inferior dimension) of the lunate divided by its diameter (anterior-posterior dimension) as measured on a lateral radiograph, although in clinical practice most radiologists can simply visually assess this parameter. An alternative measurement for assessing carpal collapse is the carpal height ratio, defined as the carpal height divided by the length of the 3rd metacarpal as measured on the PA radiograph [10]. In stage IIIa disease, there is no evidence of carpal instability (radioscaphoid angle < 60°), while stage IIIb is defined as
having a radioscaphoid angle of > 60° [22]. The radioscaphoid angle is defined as the angle between a line drawn through the center of the radius and a line tangential to the volar aspect of the scaphoid on the lateral radiograph [23] (see Fig. 4a). While computed tomography (CT) is not routinely used in the workup of Kienböck disease, it can be used to better discriminate the degree of carpal collapse and associated arthritis, especially if the patient cannot undergo MRI [1]. It has been reported that up to 50 % of patients are upgraded to a higher stage of disease on CT compared initial radiographs [20]. Contrastenhanced MRI can also be helpful to discriminate between partial and complete lunate necrosis, as the dead portions of the bone will not enhance [19]. An additional stage IIIc was added by Lichtman in 2010 for cases with chronic coronal lunate fracture [19]. Stage IV disease includes the sequlae of chronic lunate collapse, with secondary radiocarpal and midcarpal arthrosis, manifested by joint space narrowing, subchondral sclerosis and osteophyte formation. An alternative three stage articular-based staging system based on arthroscopic findings was put forth by Bain in 2011 [24], but this is less widely used.
Fig. 3 PA radiograph (a) coronal T1-weighted (b) and coronal fluid sensitive (c) MR images of the wrist, demonstrating sclerotic and lytic changes of the lunate with lunate collapse and widening of the
scapholunate interval. There was no evidence of carpal instability. The lunate is hypointense on T1-weighted imaging with areas of T2 hyperintensity on the fluid sensitive sequence
Skeletal Radiol
Fig. 4 a Measurement of the radioscaphoid angleb Measurement of the scaphocapitate angle c Schematic of measurement of the carpal-ulnar distance, indicated by a solid yellow line
Mechanical unloading Treatment options for Kienböck disease depend on the stage of the disease, as well as the patient’s age, hand dominance and functional status. Stage I disease is treated conservatively, with noninvasive bracing and activity modification, aimed at reducing mechanical stress on the lunate bone [25, 26]. Surgical intervention is typically indicated for stage II–IIIb disease, primarily with the intent to reduce mechanical stress on the lunate, halt disease progression and ideally facilitate spontaneous lunate revascularization. Based on the older evidence linking negative ulnar variance with lunate necrosis, the
Fig. 5 a Pre-op PA and lateral radiographs of the right wrist, demonstrating lunate sclerosis and mild lunate collapse, without evidence of arthrosis. There is negative ulnar variance. b Post-op PA and lateral radiographs 2 months after radial shortening osteotomy, showing intact plate and
most common procedure for stage II–IIIa disease is still radial osteotomy, with or without radial shortening (Fig. 5) [10, 27]. Multiple studies have demonstrated success with radial osteotomy alone, again bringing into question the hypothesis that ulnar variance alone is responsible for lunate avascular necrosis [10, 11]. Some have theorized that the therapeutic effect of radial osteotomy results from the secondary increase in local blood flow related to healing of the osteotomy. Similarly, although less commonly performed, are various ulnar lengthening procedures. In either case, postoperative radiographs will show plate and screw fixation of the osteotomy site, with some degree of improvement of the negative ulnar variance if bone
screw fixation and healing callus at the osteotomy site. There is now neutral ulnar variance. In this case, there has been no increased lunate sclerosis
Skeletal Radiol
Fig. 8 Schematic demonstrating complete transfer of the pisiform into the defect in the lunate. The pisiform maintains its anatomic blood supply from a dorsal branch of the ulnar artery Fig. 6 Post-op PA (a) and lateral (b) radiographs after capitateshortening osteotomy alone. In this case there is no evidence of capitate migration or proximal carpal row translation. Images courtesy of Dr. Michael Richardson, Professor of Radiology at University of Washington
shortening or lengthening was performed. Bony union at the osteotomy site is usually seen within 3 months [10]. A similar procedure which is reported to have good clinical success in grades II–IIIa disease without altering the biomechanics of the wrist joint is metaphyseal core decompression. In this procedure, the surgeon creates a cortical window, then curettes and impacts the cancellous bone of the radial and sometimes also the ulnar metaphysis without actually removing any osseous tissue [9]. Although the biologic response to this procedure is not yet well understood, it again highlights that Kienböck disease cannot be purely mechanical in nature and that perhaps changes in regional vascular flow and pressure may be more important. A newer technique for unloading the lunate bone is capitateshortening osteotomy [28], in which a central bony wedge of the capitate is resected, followed by fusion of the two remaining
Fig. 7 Schematic demonstrating harvest of a DDR graft supplied by the 4th extensor compartment branch of the anterior interosseous artery (AIA), which is placed into the carved out defect in the lunate
capitate bone segments. The goal is to increase the scaphocapitate angle, defined as the angle between the longitudinal axes of the capitate and scaphoid on posterior-anterior radiograph (see Fig. 4b). Although there is no defined normal range for the scaphocapitate angle, an increased post-operative to pre-operative angle has been correlated with decreased lunate stress and subjective pain relief [28]. In contrast, if a decreased scaphocapitate angle is seen on post-operative
Fig. 9 a Pre-op PA and lateral radiographs in a patient with stage IIIa Kienböck disease, demonstrating lunate sclerosis and collapse. b Post-op PA and lateral radiographs 19 years after vascularized pisiform transfer demonstrating increased carpal height and no progression of disease. Images reproduced with permission from Elsevier Publishing from Daecke et al. [32]
Skeletal Radiol
imaging. It is also helpful to comment on the degree of bony fusion at the osteotomy site. Post-operative imaging in many patients shows improvement in cystic change, sclerosis and collapse of the lunate. However, while there is overall favorable clinical outcome reported for these joint-leveling and metaphyseal decompression procedures, many studies show some discrepancy in the radiological outcome, with a small portion of patients going on to show progressive lunate collapse even while their pain and functional status were improved [9]. As such, while the radiologist should comment on increased lunate sclerosis and collapse if it is seen post-operatively, it should be noted that this finding does not necessarily indicate procedure failure clinically. Fig. 10 Schematic illustrating bone and cartilage graft site on the medial femoral trochlea and orientation within the resection cavity along the proximal lunate
Revascularization
radiographs, it is indicative of further lunate collapse. As there is some risk of post-operative migration of the proximal carpal row, newer modified procedures may include capitate shortening in conjunction with capitohamate fusion and/or capitometacarpal fusion. In the latter case, the cartilage on the proximal 3rd metacarpal is stripped and a plate is placed across the capitometacarpal joint. In the case of capitohamate fusion, there has been some report of post-operative carpal collapse [28, 29], which is important to note on post-operative imaging if present (Fig. 6). Radiologist should look for proximal capitate migration, proximal carpal row translation and/or carpal collapse. It is also helpful to comment on the degree of bony fusion at the osteotomy site. After any mechanical unloading procedure, the lunate morphology should be preserved. The radiologist should look carefully for any evidence of proximal capitate migration, proximal carpal row translation or carpal collapse on post-operative
Historically, once lunate collapse occurred, the patient would be relegated to salvage procedures. However, understanding that the ultimate pathology results from vascular compromise to the lunate, there has been development of various novel treatment options intended to actively revascularize the ischemic bone. In general, these procedures involve resection of the dead portions of the lunate with subsequent insertion of a pedicled vascularized bone graft to reinstate the lunate blood supply. Cortical and cancellous bone graft from the radius (Fig. 7) [30] or scaphoid [31] can be used to partially replace the lunate, while retaining its anatomical blood supply. A similar procedure may be performed using the entire ipsilateral pedicled pisiform, which maintains its anatomical blood supply from a dorsal branch of the ulnar artery (Figs. 8 and 9) [32]. A graft from the 2nd metacarpal joint has so far only been performed in cadavers [33]. For all pedicled cases, the surgeon must be able to re-expand the intact lunate cartilage, meaning that it cannot be used in patients with stage IIIb or later disease.
Fig. 11 a Pre-op PA radiograph in a 30-year-old patient with stage IIIa Kienböck disease, demonstrating lunate sclerosis and fracture. b Post-op AP radiograph in the same patient 19.5 months after vascularized bone grafting from the MFT, demonstrating improved carpal height, lunate
substance and conversion to stage II disease. There is no carpal step-off in this patient. Images reproduced with permission of Thieme Publishing from Higgins and Burger [36]
Skeletal Radiol
Fig. 12 a Pre-op PA radiograph in a 23-year-old patient with stage IIIa Kienböck disease, with history of prior radial shortening osteotomy, who presented with recurrent pain and a new coronal fracture of the lunate. b Post-op AP radiograph in the same patient 13 months after vascularized
bone grafting from the MFT, demonstrating improved carpal height, lunate substance and conversion to stage II disease. Images reproduced with permission of Thieme Publishing from Higgins and Burger [36]
Lunate buttressing and replacement
plate, screw or temporary K-wires passed through scaphoid into the radial side of graft and through the triquetrum through ulnar side of graft. It should be noted that these grafts are dependent on the local blood supply, and may suffer nonunion. Because the cartilage of the medial femoral trochlea is usually much thicker than that of the native carpal cartilage, it is also normal to see an apparent step-off of the proximal carpal arc on post-operative radiographs [35] (Figs. 10, 11, and 12). The lunate may also be replaced with interposition of a silicone or pyrocarbon prosthesis [37, 38]. These procedures routinely require partial radial styloidectomy and partial resection of the scaphoid and head of the capitate, in addition to total lunectomy. The surgeon may also need to deepen the radial glenoid concavity or perform capsular plication. Postoperatively, it is important to check for lysis, densification or ossification of the interposition material, as well as to exclude subluxation, dislocation or impaction of the prosthesis. A unique procedure involving insertion of a palmaris longus tendon ball [39] has also been tried. In these cases, it is normal to see calcification or ossification in the defect filled by the tendon ball [40] (Fig. 13).
In the setting of a completely necrotic lunate bone, revascularization procedures are unlikely to be successful; however, there are other newer techniques that allow for replacement of the lunate. For example, in the pisiform transfer procedure described in the preceding, if there is not a sufficient amount of lunate cortex remaining after debridement, a modified procedure may be performed where the lunate is entirely replaced by the pedicled pisiform. However, in this case, some wrist function will likely be sacrificed. For cases where the lunate cartilage is inadequate, a free osteocartilaginous graft from a remote donor site such as the iliac bone [34] or medial femoral trochlea (MFT) [35, 36] can be used to restore a more natural lunate shape. In the latter case, the cartilage on the radial aspect of the lunate is replaced by cartilage from the knee, holding promise for patients with later-stage disease and radiocarpal arthrosis. Postoperatively, radiographs will demonstrate fixation of the graft with a mini-
Partial and complete fusion
Fig. 13 Schematic of a pyrocarbon prosthesis
For stage IIIb disease and beyond, the conventional treatment is proximal row carpectomy (PRC). PRC requires intact cartilage on the capitate head and in the lunate fossa, excluding it as an option for the most advanced cases [41]. Once there is collapse of the radiocarpal joint or midcarpal arthrosis, complete or partial wrist joint arthrodesis may be performed to relieve pain at the expense of some additional loss of wrist function. Modifications of the historical Graner fusion
Skeletal Radiol
Fig. 14 a Pre-operative PA radiograph in a patient with stage IIIb Kienböck disease, demonstrating lunate sclerosis and collapse. b–c Post-operative PA and lateral radiographs after scaphocapitate fusion,
demonstrating bony fusion between the capitate and scaphoid and preserved radioscaphoid alignment and carpal-ulnar distance
procedures [42] include scaphoid-trapezium-trapezoid fusion [43] and scaphoid-capitate fusion (Fig. 14) [44]. These are all designed to minimize mechanical stress and force transmission through the lunate bone. Scaphoid-lunate fusion is another option, although it carries a high risk of nonunion given the tenuous lunate blood supply at baseline. On follow-up imaging after wrist fusion, there should be correct radioscaphoid alignment (ideal radioscaphoid angle is between 30 and 57° [45]), without evidence of ulnar translation [46, 47]. Ulnar translation can be evaluated by measuring the distance from the center of the head of the capitate to a line drawn down the center of the ulna, also known as the carpalulnar distance (Fig. 4c). The ratio of the carpal-ulnar distance to
the length of the 3rd metacarpal is normally approximately 0.3, with a lower ratio indicating possible ulnar translation [48]. Comparison to pre-operative imaging should show no significant change in this measurement. It is also important to note union at the fusion site, and in the case of partial fusion, to evaluate for progression of radioscaphoid arthrosis, as these issues can relate to any persistent patient complaint of pain.
Palliative procedures Total wrist arthroplasty is now being explored as an alternative to total wrist arthrodesis, as it can preserve some wrist function. This option is typically reserved for the older, lower functional demand patient. As with implants throughout the rest of the body, the radiologist should familiarize themselves with the particular model and brand of implant, as well as evaluate for evidence of loosening, peri-implant fracture or Table 2 Checklist for evaluating Kienböck disease on initial diagnostic radiographs Pre-operative radiograph checklist ✓ Ulnar variance positive, negative or neutrala ✓ Lunate morphology (shape, height, Stahl index) ✓ Lunate sclerosis or cystic change ✓ Radioscaphoid angle (> or < 60°) ✓ Scaphocapitate angle ✓ Scapholunate interval (widened or normal) ✓ Radiocarpal or intercarpal arthrosis ✓ Evidence of carpal instability ✓ Lichtman stage ✓ Contralateral wrist radiograph (bilateral disease?)
Fig. 15 PA and lateral radiographs in a patient after total wrist arthroplasty. Images courtesy of Alice Ha, MD, Associate Professor at University of Washington
a Although the contribution of ulnar variance is controversial, it should still be mentioned as the surgeon may value this information
Skeletal Radiol Table 3 MRI
Checklist for evaluating Kienböck disease on initial diagnostic
Pre-operative MRI checklist
successfully evaluate pre-operative (Tables 2 and 3) and post-operative imaging (Table 4) in a way that adds the most possible value to clinical patient care.
✓ Lunate morphology (shape, height, Stahl index)
Compliance with ethical standards
✓ Lunate bone marrow signal ✓ Distal ulnar and triquietral bone marrow signal (distinguish from UAS) ✓ Scaphocapitate ✓ Scapholunate ligament (intact or torn) ✓ Enhancement of lunate (if IV contrast was administered) ✓ Evidence of carpal instability ✓ Lichtman stage
Funding This study did not require funding. Conflict of interest The authors declare that they have no conflict of interest. Research involving Human Participants and/or Animals This article does not contain any studies with human participants or animals performed by any of the authors. Informed consent This article does not contain patient data.
malalignment (Fig. 15). A wrist neurectomy may also be performed for symptom relief, either as a single procedure, or in conjunction with any of the aforementioned procedures [49]. The same post-operative imaging checklists apply, with the exception that after neurectomy alone, the patient is usually just followed clinically.
References 1.
2.
Conclusion 3.
Although the etiology of Kienböck disease is controversial, the pathogenesis is likely a combination of predisposing anatomy and exogenous factors such as trauma and/or medications, which ultimately compromise vascular supply to the lunate bone. Successful evaluation of post-operative imaging for Kienböck disease requires an understanding of the various surgical treatments, including the newer revascularization, buttressing and salvage techniques. While it is beyond the scope of this paper to compare effectiveness of the myriad of surgical treatments for Kienböck disease, the provided checklists and imaging examples can help the radiologist
4.
5. 6.
7.
8. Table 4 Checklist for evaluating radiographs after conservative or surgical treatment for Kienböck disease. Suggestions are provided for comments in comparison to pre-operative radiographs
9.
Post-operative radiograph checklist 10. ✓ Ulnar variance (negative or neutral)a ✓ Lunate morphology (shape, height, Stahl index) ✓ Lunate sclerosis or cystic change (increased, decreased or stable) ✓ Radioscaphoid angle (ideally 30–60°) ✓ Scaphocapitate agle (ideally increased compared to pre-op) ✓ Scapholunate interval (widened or normal)
11.
✓ Radiocarpal or intercarpal arthrosis (progressed or unchanged) ✓ Evidence of carpal row translation ✓ Fixation hardware (proper placement, any complication) ✓ Bony fusion at osteotomy/fusion site (usually seen by 3 months)
13.
a Although the contribution of ulnar variance is controversial, it should still be mentioned as the surgeon may value this information
12.
14. 15.
Arnaiz J, Piedra T, Cerezal L, Ward J, Thompson A, Vidal JA, et al. Imaging of Kienböck disease. AJR Am J Roentgenol. 2014;203(1): 131–9. van Leeuwen WF, Janssen SJ, Ter Meulen DP, Ring D. What is the radiographic prevalence of incidental Kienböck disease? Clin Orthop Relat Res. 2015. Yazaki N, Nakamura R, Nakao E, Iwata Y, Tatebe M, Hattori T. Bilateral Kienböck’s disease. J Hand Surg Br Eur Vol. 2005;30(2): 133–6. Seah KT, McEachan J, Davidson D. Bilateral Kienböck’s disease in association with type 1 diabetes. J Hand Surg Eur Vol. 2012;37(7): 697–8. Mok CC, Lau CS, Cheng PW, Ip WY. Bilateral Kienböck’s disease in SLE. Scand J Rheumatol. 1997;26(6):485–7. Wollstein A, Tantawi D, Wollstein R. Bilateral Kienbock’s disease concomitant with bilateral Legg-Calve-Perthes disease: a case report. Hand (New York, NY). 2013;8(1):120–2. Cerezal L, del Pinal F, Abascal F, Garcia-Valtuille R, Pereda T, Canga A. Imaging findings in ulnar-sided wrist impaction syndromes. Radiographics. 2002;22(1):105–21. Stahl S, Stahl AS, Meisner C, Hentschel PJ, Valina S, Luz O, et al. Critical analysis of causality between negative ulnar variance and Kienböck disease. Plast Reconstr Surg. 2013;132(4):899–909. Illarramendi AA, Schulz C, De Carli P. The surgical treatment of Kienböck’s disease by radius and ulna metaphyseal core decompression. J Hand Surg. 2001;26(2):252–60. Iwasaki N, Minami A, Oizumi N, Suenaga N, Kato H, Minami M. Radial osteotomy for late-stage Kienböck’s disease: wedge osteotomy versus radial shortening. J Bone Joint Surg Br Vol. 2002;84(5):673–7. Blanco RH, Blanco FR. Osteotomy of the radius without shortening for Kienböck disease: a 10-year follow-up. J Hand Surg. 2012;37(11):2221–5. Imaeda T, Nakamura R, Shionoya K, Makino N. Ulnar impaction syndrome: MR imaging findings. Radiology. 1996;201(2):495– 500. JM. AaZ. Malacia del semilunar. Valladolid, Spain: University of Valladolid; 1966. Beredjiklian PK. Kienböck’s disease. J Hand Surg. 2009;34(1): 167–75. Lutsky K, Beredjiklian PK. Kienböck disease. J Hand Surg. 2012;37(9):1942–52.
Skeletal Radiol 16.
Gelberman RH, Bauman TD, Menon J, Akeson WH. The vascularity of the lunate bone and Kienböck’s disease. J Hand Surg. 1980;5(3):272–8. 17. Schiltenwolf M, Martini AK, Mau HC, Eversheim S, Brocai DR, Jensen CH. Further investigations of the intraosseous pressure characteristics in necrotic lunates (Kienböck’s disease). J Hand Surg. 1996;21(5):754–8. 18. Lichtman DM, Mack GR, MacDonald RI, Gunther SF, Wilson JN. Kienböck’s disease: the role of silicone replacement arthroplasty. J Bone Joint Surg Am Vol. 1977;59(7):899–908. 19. Lichtman DM, Lesley NE, Simmons SP. The classification and treatment of Kienböck’s disease: the state of the art and a look at the future. J Hand Surg Eur Vol. 2010;35(7):549–54. 20. Schmitt R, Heinze A, Fellner F, Obletter N, Struhn R, Bautz W. Imaging and staging of avascular osteonecroses at the wrist and hand. Eur J Radiol. 1997;25(2):92–103. 21. Luo J, Diao E. Kienböck’s disease: an approach to treatment. Hand Clin. 2006;22(4):465–73. abstract vi. 22. Goldfarb CA, Hsu J, Gelberman RH, Boyer MI. The Lichtman classification for Kienböck’s disease: an assessment of reliability. J Hand Surg. 2003;28(1):74–80. 23. Larsen CF, Mathiesen FK, Lindequist S. Measurements of carpal bone angles on lateral wrist radiographs. J Hand Surg. 1991;16(5): 888–93. 24. Bain GI, Durrant A. An articular-based approach to Kienböck avascular necrosis of the lunate. Tech Hand Upper Extrem Surg. 2011;15(1):41–7. 25. Wollstein R, Wollstein A, Rodgers J, Ogden TJ. A hand therapy protocol for the treatment of lunate overload or early Kienböck’s disease. J Hand Ther. 2013;26(3):255–9. quiz 260. 26. Meena D, Saini N, Kundanani V, Chaudhary L, Meena D. Distraction histiogenesis for treatment of Kienböck’s disease: a 2to 8-year follow-up. Ind J Orthop. 2009;43(2):189–93. 27. Matsui Y, Funakoshi T, Motomiya M, Urita A, Minami M, Iwasaki N. Radial shortening osteotomy for Kienböck disease: minimum 10-year follow-up. J Hand Surg. 2014;39(4):679–85. 28. Fouly EH, Sadek AF, Amin MF. Distal capitate shortening with capitometacarpal fusion for management of the early stages of Kienböck disease with neutral ulnar variance: case series. J Orthop Surg Res. 2014;9(1):86. 29. Okamoto MAM, Shirai H, Ueda N, Kagawa Y. Capitate shortening for Kienböck’s disease. J Jpn Soc Surg Hand. 1999;15:674–6. 30. Kakar S, Shin AY. Vascularized bone grafting from the dorsal distal radius for Kienböck’s disease: technique, indications and review of the literature. Chir de la main. 2010;29 Suppl 1:S104–11. 31. Mir X, Barrera-Ochoa S, Lluch A, Llusa M, Haddad S, Vidal N, et al. New surgical approach to advanced Kienböck disease: lunate replacement with pedicled vascularized scaphoid graft and radioscaphoidal partial arthrodesis. Techn Hand Upper Extrem Surg. 2013;17(2):72–9. 32. Daecke W, Lorenz S, Wieloch P, Jung M, Martini AK. Vascularized os pisiform for reinforcement of the lunate in Kienböck’s disease: an average of 12 years of follow-up study. J Hand Surg. 2005;30(5): 915–22.
33.
Vilkki SK. A new concept and technique for custom-made microvascular lunate bone reconstruction. J Reconstr Microsurg. 2007;23(7):351–9. 34. Arora R, Lutz M, Deml C, Krappinger D, Zimmermann R, Gabl M. Long-term subjective and radiological outcome after reconstruction of Kienböck’s disease stage 3 treated by a free vascularized iliac bone graft. J Hand Surg. 2008;33(2):175–81. 35. Burger HK, Windhofer C, Gaggl AJ, Higgins JP. Vascularized medial femoral trochlea osteochondral flap reconstruction of advanced Kienböck disease. J Hand Surg. 2014;39(7):1313–22. 36. Higgins JP, Burger HK. Osteochondral flaps from the distal femur: expanding applications, harvest sites, and indications. J Reconstr Microsurg. 2014. 37. Bellemere P, Maes-Clavier C, Loubersac T, Gaisne E, Kerjean Y. Amandys® implant: novel pyrocarbon arthroplasty for the wrist. Chir de la main. 2012;31(4):176–87. 38. Bellemere P, Maes-Clavier C, Loubersac T, Gaisne E, Kerjean Y, Collon S. Pyrocarbon interposition wrist arthroplasty in the treatment of failed wrist procedures. J Wrist Surg. 2012;1(1):31–8. 39. Ueba Y, Nosaka K, Seto Y, Ikeda N, Nakamura T. An operative procedure for advanced Kienböck’s disease: excision of the lunate and subsequent replacement with a tendon-ball implant. J Orthop Sci. 1999;4(3):207–15. 40. Mariconda M, Soscia E, Sirignano C, Smeraglia F, Soldati A, Balato G. Long-term clinical results and MRI changes after tendon ball arthroplasty for advanced Kienböck’s disease. J Hand Surg Eur Vol. 2013;38(5):508–14. 41. Chim H, Moran SL. Long-term outcomes of proximal row carpectomy: a systematic review of the literature. J Wrist Surg. 2012;1(2):141–8. 42. Facca S, Gondrand I, Naito K, Lequint T, Nonnenmacher J, Liverneaux P. Graner’s procedure in Kienböck disease: a series of four cases with 25 years of follow-up. Chir de la main. 2013;32(5): 305–9. 43. Lee JS, Park MJ, Kang HJ. Scaphotrapeziotrapezoid arthrodesis and lunate excision for advanced Kienböck disease. J Hand Surg. 2012;37(11):2226–32. 44. Luegmair M, Saffar P. Scaphocapitate arthrodesis for treatment of late stage Kienböck disease. J Hand Surg Eur Vol. 2014;39(4):416– 22. 45. Minamikawa Y, Peimer CA, Yamaguchi T, Medige J, Sherwin FS. Ideal scaphoid angle for intercarpal arthrodesis. J Hand Surg. 1992;17(2):370–5. 46. Bain GI, McGuire DT. Decision making for partial carpal fusions. J Wrist Surg. 2012;1(2):103–14. 47. Budoff JE, Gable G. Ulnar translation of scaphocapitate arthrodeses in Kienböck’s disease: two case reports. J Hand Surg. 2005;30(1): 65–8. 48. Youm Y, McMurthy RY, Flatt AE, Gillespie TE. Kinematics of the wrist. I. An experimental study of radial-ulnar deviation and flexion-extension. J Bone Joint Surg Am Vol. 1978;60(4):423–31. 49. Buck-Gramcko D. Wrist denervation procedures in the treatment of Kienböck’s disease. Hand Clin. 1993;9(3):517–20.
REVIEW ARTICLE
Treatments for Kienböck disease: what the radiologist needs to know Carissa White 1
&
Prosper Benhaim 2 & Benjamin Plotkin 3
Received: 28 September 2015 / Revised: 12 December 2015 / Accepted: 6 January 2016 # ISS 2016
Abstract The etiology of Kienböck disease, or avascular necrosis of the lunate, is controversial, and there are a myriad of treatments aimed at correcting the various hypothesized pathologies. Interventions to reduce mechanical stress on the lunate have been used for decades, including radial osteotomy with or without radial shortening, ulnar lengthening and metaphyseal core decompression procedures. However, these procedures require preservation of lunate architecture. Newer procedures to revascularize the lunate bone have emerged in the last 10 years, such as pedicled corticoperiosteal vascularized bone grafting. Once there is collapse of the radiocarpal joint or midcarpal arthrosis, the conventional treatments have included proximal row carpectomy and complete or partial wrist joint arthrodesis. Newer salvage procedures such as lunate excision with autologous or synthetic interposition grafts are now being used when possible. As this disease is relatively rare, radiologists may not be familiar with the expected post-operative radiologic findings and complications, especially of the newer treatments. The goals of this paper are to review the available treatment options and their expected appearance on postoperative imaging, with discussion of possible complications when appropriate.
* Carissa White [email protected]
Keywords Kienböck . Lunate . Avascular necrosis . Wrist
Introduction Kienböck disease, or avascular necrosis of the lunate, is a slowly progressive disease that can eventually lead to complete loss of wrist and hand function. Kienböck disease tends to occur in people who axially load their wrist in dorsiflexion, such as manual workers, with a 2:1 male to female predominance. The typical patient is 20–40 years of age [1], although recent studies have shown radiographic findings of the disease to be present in 0.10 % of asymptomatic patients, many of whom are older than the 5th decade [2]. The disease is usually unilateral, with bilateral disease being reported only 4 % of the time, often with one of the wrists being asymptomatic [3]. There are case reports of bilateral Kienböck disease is association with type 1 diabetes mellitus [4], systemic lupus erythematosus [5] and Legg-Calve-Perthes disease [6], but these diseases do not seem to increase the risk for bilateral as opposed to unilateral disease. While the overall disease prevalence (including symptomatic and asymptomatic patients) is only 0.27 % [2], the frequency is likely increased in the setting of a specialized orthopedic or tertiary care center. Given the high morbidity of this disease, it is imperative for both the general and musculoskeletal radiologist not only to be able to accurately diagnose the disease, but also to know how to effectively evaluate the patient after conservative or surgical treatment.
1
Department of Radiology, University of California, Los Angeles, 757 Westwood Blvd. Suite 1638, Los Angeles, CA 90095, USA
2
Department of Orthopaedic Surgery, University of California, Los Angeles, 10945 Le Conte Ave, Room 33-55 PVUB, Box 957326, Los Angeles, CA 90095, USA
Etiology
Department of Radiology, University of California, Los Angeles, 1250 Sixteenth Street, Box 957036, Santa Monica, CA 90404, USA
The definitive etiology for Kienböck disease is unknown. In the past, it was thought that negative ulnar variance, where the
3
Skeletal Radiol Fig. 1 a PA radiograph demonstrating positive ulnar variance, with the ulnar articular surface projecting distal to the radial articular surface. b PA radiograph demonstrating negative ulnar variance, with the ulnar articular surface projecting proximal to the radial articular surface
ulnar articular surface projects proximal to the radial articular surface on a posterior-anterior (PA) radiograph (Fig. 1), predisposed to lunate necrosis by altering the load transmitted through the radial column. It is important that the patient be properly positioned when assessing for ulnar variance on a radiograph. The wrist should be in neutral position with the elbow flexed to 90° and the shoulder abducted to 90°. Maximum forearm pronation increases the ulnar variance (makes it more positive) while maximum supination decreases the ulnar variance (can make the variance appear falsely negative)[7]. However, recent studies have demonstrated no evidence for causation between ulnar variance and Kienböck disease [8]; moreover, newer interventions that do not alter the mechanical load on the lunate have been shown to halt and even reverse lunate necrosis [9–11]. Ulnar positive variance, on the other hand, has been shown to predispose to ulnar impaction/abutment syndrome, where the distal ulna repeatedly knocks up against the triangular fibrocartilage (TFCC) and the proximal carpal bones, leading to tears of the TFCC, and marrow edema, subchondral sclerosis and cystic changes of the lunate (87 %) and or triquetrum (43 %) [12] (Fig. 2). Other theories have concentrated on the lunate shape such as the Antuna Zapico classification, determined by the angle made by the lateral scaphoid side with the proximal radial side of the lunate. In this analysis, it was found that a square or rectangular shape of the lunate bone may increase the risk for lunate necrosis compared to a trapezoidal shape, due to the Table 1 Lichtman staging as it pertains to appearance of the lunate bone on imaging Stage
Radiographs
MRI
I II IIIa IIIb IV
Normal Increased density w/o collapse Collapse w/o carpal instability Collapse with carpal instability Collapse with radiocarpal or midcarpal arthrosis
↓ T1 signal ↓ T1 signal ↓ T1 signal ↑ T2 signal ↓ T1 signal ↑ T2 signal ↓ T1 signal ↑ T2 signal
decreased number of articular surfaces with the radius and therefore different distribution of forces on the bone [1, 13]. However, most compelling are the theories that focus on the native carpal blood supply. Normally, the lunate receives dual blood supply on both the dorsal and volar sides, with a variable intraosseous branching pattern. It has been shown that up to 7 % of people may have only a single palmar vessel [14], and up to one-third of patients may have comparatively less intraosseous branching, both of which may predispose to ischemia, especially in the setting of acute or chronic repetitive microtrauma [14–16]. Other theories suggest there may be outflow obstruction in the draining venous plexus [14], supported by the finding of increased intraosseous pressure in necrotic lunate bones [17]. However, it is unknown whether the increased venous pressure is a cause or a result of the disease [15]. Theoretically any systemic disease that predisposes to thrombosis or vascular pathology such as sickle cell disease, septic emboli, systemic lupus erythematosus or exogenous steroids, would also increase the risk for Kienböck disease, although there does not seem to be a direct correlation [5, 14].
Staging The radiological-pathological staging system for Kienböck disease was developed by Stahl 1947 and modified by Lichtman in 1977 [18], based solely on radiographs. MRI is now routinely used and has added greatly to diagnostic sensitivity, especially in the early stages of the disease (Table 1). In stage I, the lunate appears normal on plain radiographs, but demonstrates mildly decreased signal on T1-weighted MR images. If there is corresponding increased lunate marrow signal on T2-weighted images, the changes likely represent marrow edema versus innate revascularization attempts [19]. If intravenous contrast is given, the lunate will often demonstrate intense enhancement at this stage [20]. Alternatively, corresponding decreased lunate signal on T2-weighted images represents completed necrosis [21], a finding which is more likely
Skeletal Radiol
Fig. 2 Ulnar abutment syndrome. PA radiograph (a) of the wrist demonstrating only mild radiocarpal joint space narrowing. Sequential coronal T2-weighted MR images (b–c) of the wrist, demonstrating cystic
changes of the lunate and bone marrow edema in the distal ulna. In this case the ulnar variance was neutral
to be seen in stage II. The radiologist should look for increased T2 signal in the distal ulna and/or triquetrum, which in the presence of ulnar positive variance, would suggest an alternative diagnosis of ulnar abutment syndrome (UAS). A bone scan is also abnormal in stage I, although this is not routinely incorporated in the work-up. In stage II, radiographs demonstrate increased lunate density and sclerosis, without change of lunate shape. When the disease progresses to stage III, there is collapse of the lunate bone with marked hyperintensity of the lunate marrow on T2-weighted imaging (Fig. 3). The degree of lunate collapse can be quantified using the Stahl index, defined as the height (superior-inferior dimension) of the lunate divided by its diameter (anterior-posterior dimension) as measured on a lateral radiograph, although in clinical practice most radiologists can simply visually assess this parameter. An alternative measurement for assessing carpal collapse is the carpal height ratio, defined as the carpal height divided by the length of the 3rd metacarpal as measured on the PA radiograph [10]. In stage IIIa disease, there is no evidence of carpal instability (radioscaphoid angle < 60°), while stage IIIb is defined as
having a radioscaphoid angle of > 60° [22]. The radioscaphoid angle is defined as the angle between a line drawn through the center of the radius and a line tangential to the volar aspect of the scaphoid on the lateral radiograph [23] (see Fig. 4a). While computed tomography (CT) is not routinely used in the workup of Kienböck disease, it can be used to better discriminate the degree of carpal collapse and associated arthritis, especially if the patient cannot undergo MRI [1]. It has been reported that up to 50 % of patients are upgraded to a higher stage of disease on CT compared initial radiographs [20]. Contrastenhanced MRI can also be helpful to discriminate between partial and complete lunate necrosis, as the dead portions of the bone will not enhance [19]. An additional stage IIIc was added by Lichtman in 2010 for cases with chronic coronal lunate fracture [19]. Stage IV disease includes the sequlae of chronic lunate collapse, with secondary radiocarpal and midcarpal arthrosis, manifested by joint space narrowing, subchondral sclerosis and osteophyte formation. An alternative three stage articular-based staging system based on arthroscopic findings was put forth by Bain in 2011 [24], but this is less widely used.
Fig. 3 PA radiograph (a) coronal T1-weighted (b) and coronal fluid sensitive (c) MR images of the wrist, demonstrating sclerotic and lytic changes of the lunate with lunate collapse and widening of the
scapholunate interval. There was no evidence of carpal instability. The lunate is hypointense on T1-weighted imaging with areas of T2 hyperintensity on the fluid sensitive sequence
Skeletal Radiol
Fig. 4 a Measurement of the radioscaphoid angleb Measurement of the scaphocapitate angle c Schematic of measurement of the carpal-ulnar distance, indicated by a solid yellow line
Mechanical unloading Treatment options for Kienböck disease depend on the stage of the disease, as well as the patient’s age, hand dominance and functional status. Stage I disease is treated conservatively, with noninvasive bracing and activity modification, aimed at reducing mechanical stress on the lunate bone [25, 26]. Surgical intervention is typically indicated for stage II–IIIb disease, primarily with the intent to reduce mechanical stress on the lunate, halt disease progression and ideally facilitate spontaneous lunate revascularization. Based on the older evidence linking negative ulnar variance with lunate necrosis, the
Fig. 5 a Pre-op PA and lateral radiographs of the right wrist, demonstrating lunate sclerosis and mild lunate collapse, without evidence of arthrosis. There is negative ulnar variance. b Post-op PA and lateral radiographs 2 months after radial shortening osteotomy, showing intact plate and
most common procedure for stage II–IIIa disease is still radial osteotomy, with or without radial shortening (Fig. 5) [10, 27]. Multiple studies have demonstrated success with radial osteotomy alone, again bringing into question the hypothesis that ulnar variance alone is responsible for lunate avascular necrosis [10, 11]. Some have theorized that the therapeutic effect of radial osteotomy results from the secondary increase in local blood flow related to healing of the osteotomy. Similarly, although less commonly performed, are various ulnar lengthening procedures. In either case, postoperative radiographs will show plate and screw fixation of the osteotomy site, with some degree of improvement of the negative ulnar variance if bone
screw fixation and healing callus at the osteotomy site. There is now neutral ulnar variance. In this case, there has been no increased lunate sclerosis
Skeletal Radiol
Fig. 8 Schematic demonstrating complete transfer of the pisiform into the defect in the lunate. The pisiform maintains its anatomic blood supply from a dorsal branch of the ulnar artery Fig. 6 Post-op PA (a) and lateral (b) radiographs after capitateshortening osteotomy alone. In this case there is no evidence of capitate migration or proximal carpal row translation. Images courtesy of Dr. Michael Richardson, Professor of Radiology at University of Washington
shortening or lengthening was performed. Bony union at the osteotomy site is usually seen within 3 months [10]. A similar procedure which is reported to have good clinical success in grades II–IIIa disease without altering the biomechanics of the wrist joint is metaphyseal core decompression. In this procedure, the surgeon creates a cortical window, then curettes and impacts the cancellous bone of the radial and sometimes also the ulnar metaphysis without actually removing any osseous tissue [9]. Although the biologic response to this procedure is not yet well understood, it again highlights that Kienböck disease cannot be purely mechanical in nature and that perhaps changes in regional vascular flow and pressure may be more important. A newer technique for unloading the lunate bone is capitateshortening osteotomy [28], in which a central bony wedge of the capitate is resected, followed by fusion of the two remaining
Fig. 7 Schematic demonstrating harvest of a DDR graft supplied by the 4th extensor compartment branch of the anterior interosseous artery (AIA), which is placed into the carved out defect in the lunate
capitate bone segments. The goal is to increase the scaphocapitate angle, defined as the angle between the longitudinal axes of the capitate and scaphoid on posterior-anterior radiograph (see Fig. 4b). Although there is no defined normal range for the scaphocapitate angle, an increased post-operative to pre-operative angle has been correlated with decreased lunate stress and subjective pain relief [28]. In contrast, if a decreased scaphocapitate angle is seen on post-operative
Fig. 9 a Pre-op PA and lateral radiographs in a patient with stage IIIa Kienböck disease, demonstrating lunate sclerosis and collapse. b Post-op PA and lateral radiographs 19 years after vascularized pisiform transfer demonstrating increased carpal height and no progression of disease. Images reproduced with permission from Elsevier Publishing from Daecke et al. [32]
Skeletal Radiol
imaging. It is also helpful to comment on the degree of bony fusion at the osteotomy site. Post-operative imaging in many patients shows improvement in cystic change, sclerosis and collapse of the lunate. However, while there is overall favorable clinical outcome reported for these joint-leveling and metaphyseal decompression procedures, many studies show some discrepancy in the radiological outcome, with a small portion of patients going on to show progressive lunate collapse even while their pain and functional status were improved [9]. As such, while the radiologist should comment on increased lunate sclerosis and collapse if it is seen post-operatively, it should be noted that this finding does not necessarily indicate procedure failure clinically. Fig. 10 Schematic illustrating bone and cartilage graft site on the medial femoral trochlea and orientation within the resection cavity along the proximal lunate
Revascularization
radiographs, it is indicative of further lunate collapse. As there is some risk of post-operative migration of the proximal carpal row, newer modified procedures may include capitate shortening in conjunction with capitohamate fusion and/or capitometacarpal fusion. In the latter case, the cartilage on the proximal 3rd metacarpal is stripped and a plate is placed across the capitometacarpal joint. In the case of capitohamate fusion, there has been some report of post-operative carpal collapse [28, 29], which is important to note on post-operative imaging if present (Fig. 6). Radiologist should look for proximal capitate migration, proximal carpal row translation and/or carpal collapse. It is also helpful to comment on the degree of bony fusion at the osteotomy site. After any mechanical unloading procedure, the lunate morphology should be preserved. The radiologist should look carefully for any evidence of proximal capitate migration, proximal carpal row translation or carpal collapse on post-operative
Historically, once lunate collapse occurred, the patient would be relegated to salvage procedures. However, understanding that the ultimate pathology results from vascular compromise to the lunate, there has been development of various novel treatment options intended to actively revascularize the ischemic bone. In general, these procedures involve resection of the dead portions of the lunate with subsequent insertion of a pedicled vascularized bone graft to reinstate the lunate blood supply. Cortical and cancellous bone graft from the radius (Fig. 7) [30] or scaphoid [31] can be used to partially replace the lunate, while retaining its anatomical blood supply. A similar procedure may be performed using the entire ipsilateral pedicled pisiform, which maintains its anatomical blood supply from a dorsal branch of the ulnar artery (Figs. 8 and 9) [32]. A graft from the 2nd metacarpal joint has so far only been performed in cadavers [33]. For all pedicled cases, the surgeon must be able to re-expand the intact lunate cartilage, meaning that it cannot be used in patients with stage IIIb or later disease.
Fig. 11 a Pre-op PA radiograph in a 30-year-old patient with stage IIIa Kienböck disease, demonstrating lunate sclerosis and fracture. b Post-op AP radiograph in the same patient 19.5 months after vascularized bone grafting from the MFT, demonstrating improved carpal height, lunate
substance and conversion to stage II disease. There is no carpal step-off in this patient. Images reproduced with permission of Thieme Publishing from Higgins and Burger [36]
Skeletal Radiol
Fig. 12 a Pre-op PA radiograph in a 23-year-old patient with stage IIIa Kienböck disease, with history of prior radial shortening osteotomy, who presented with recurrent pain and a new coronal fracture of the lunate. b Post-op AP radiograph in the same patient 13 months after vascularized
bone grafting from the MFT, demonstrating improved carpal height, lunate substance and conversion to stage II disease. Images reproduced with permission of Thieme Publishing from Higgins and Burger [36]
Lunate buttressing and replacement
plate, screw or temporary K-wires passed through scaphoid into the radial side of graft and through the triquetrum through ulnar side of graft. It should be noted that these grafts are dependent on the local blood supply, and may suffer nonunion. Because the cartilage of the medial femoral trochlea is usually much thicker than that of the native carpal cartilage, it is also normal to see an apparent step-off of the proximal carpal arc on post-operative radiographs [35] (Figs. 10, 11, and 12). The lunate may also be replaced with interposition of a silicone or pyrocarbon prosthesis [37, 38]. These procedures routinely require partial radial styloidectomy and partial resection of the scaphoid and head of the capitate, in addition to total lunectomy. The surgeon may also need to deepen the radial glenoid concavity or perform capsular plication. Postoperatively, it is important to check for lysis, densification or ossification of the interposition material, as well as to exclude subluxation, dislocation or impaction of the prosthesis. A unique procedure involving insertion of a palmaris longus tendon ball [39] has also been tried. In these cases, it is normal to see calcification or ossification in the defect filled by the tendon ball [40] (Fig. 13).
In the setting of a completely necrotic lunate bone, revascularization procedures are unlikely to be successful; however, there are other newer techniques that allow for replacement of the lunate. For example, in the pisiform transfer procedure described in the preceding, if there is not a sufficient amount of lunate cortex remaining after debridement, a modified procedure may be performed where the lunate is entirely replaced by the pedicled pisiform. However, in this case, some wrist function will likely be sacrificed. For cases where the lunate cartilage is inadequate, a free osteocartilaginous graft from a remote donor site such as the iliac bone [34] or medial femoral trochlea (MFT) [35, 36] can be used to restore a more natural lunate shape. In the latter case, the cartilage on the radial aspect of the lunate is replaced by cartilage from the knee, holding promise for patients with later-stage disease and radiocarpal arthrosis. Postoperatively, radiographs will demonstrate fixation of the graft with a mini-
Partial and complete fusion
Fig. 13 Schematic of a pyrocarbon prosthesis
For stage IIIb disease and beyond, the conventional treatment is proximal row carpectomy (PRC). PRC requires intact cartilage on the capitate head and in the lunate fossa, excluding it as an option for the most advanced cases [41]. Once there is collapse of the radiocarpal joint or midcarpal arthrosis, complete or partial wrist joint arthrodesis may be performed to relieve pain at the expense of some additional loss of wrist function. Modifications of the historical Graner fusion
Skeletal Radiol
Fig. 14 a Pre-operative PA radiograph in a patient with stage IIIb Kienböck disease, demonstrating lunate sclerosis and collapse. b–c Post-operative PA and lateral radiographs after scaphocapitate fusion,
demonstrating bony fusion between the capitate and scaphoid and preserved radioscaphoid alignment and carpal-ulnar distance
procedures [42] include scaphoid-trapezium-trapezoid fusion [43] and scaphoid-capitate fusion (Fig. 14) [44]. These are all designed to minimize mechanical stress and force transmission through the lunate bone. Scaphoid-lunate fusion is another option, although it carries a high risk of nonunion given the tenuous lunate blood supply at baseline. On follow-up imaging after wrist fusion, there should be correct radioscaphoid alignment (ideal radioscaphoid angle is between 30 and 57° [45]), without evidence of ulnar translation [46, 47]. Ulnar translation can be evaluated by measuring the distance from the center of the head of the capitate to a line drawn down the center of the ulna, also known as the carpalulnar distance (Fig. 4c). The ratio of the carpal-ulnar distance to
the length of the 3rd metacarpal is normally approximately 0.3, with a lower ratio indicating possible ulnar translation [48]. Comparison to pre-operative imaging should show no significant change in this measurement. It is also important to note union at the fusion site, and in the case of partial fusion, to evaluate for progression of radioscaphoid arthrosis, as these issues can relate to any persistent patient complaint of pain.
Palliative procedures Total wrist arthroplasty is now being explored as an alternative to total wrist arthrodesis, as it can preserve some wrist function. This option is typically reserved for the older, lower functional demand patient. As with implants throughout the rest of the body, the radiologist should familiarize themselves with the particular model and brand of implant, as well as evaluate for evidence of loosening, peri-implant fracture or Table 2 Checklist for evaluating Kienböck disease on initial diagnostic radiographs Pre-operative radiograph checklist ✓ Ulnar variance positive, negative or neutrala ✓ Lunate morphology (shape, height, Stahl index) ✓ Lunate sclerosis or cystic change ✓ Radioscaphoid angle (> or < 60°) ✓ Scaphocapitate angle ✓ Scapholunate interval (widened or normal) ✓ Radiocarpal or intercarpal arthrosis ✓ Evidence of carpal instability ✓ Lichtman stage ✓ Contralateral wrist radiograph (bilateral disease?)
Fig. 15 PA and lateral radiographs in a patient after total wrist arthroplasty. Images courtesy of Alice Ha, MD, Associate Professor at University of Washington
a Although the contribution of ulnar variance is controversial, it should still be mentioned as the surgeon may value this information
Skeletal Radiol Table 3 MRI
Checklist for evaluating Kienböck disease on initial diagnostic
Pre-operative MRI checklist
successfully evaluate pre-operative (Tables 2 and 3) and post-operative imaging (Table 4) in a way that adds the most possible value to clinical patient care.
✓ Lunate morphology (shape, height, Stahl index)
Compliance with ethical standards
✓ Lunate bone marrow signal ✓ Distal ulnar and triquietral bone marrow signal (distinguish from UAS) ✓ Scaphocapitate ✓ Scapholunate ligament (intact or torn) ✓ Enhancement of lunate (if IV contrast was administered) ✓ Evidence of carpal instability ✓ Lichtman stage
Funding This study did not require funding. Conflict of interest The authors declare that they have no conflict of interest. Research involving Human Participants and/or Animals This article does not contain any studies with human participants or animals performed by any of the authors. Informed consent This article does not contain patient data.
malalignment (Fig. 15). A wrist neurectomy may also be performed for symptom relief, either as a single procedure, or in conjunction with any of the aforementioned procedures [49]. The same post-operative imaging checklists apply, with the exception that after neurectomy alone, the patient is usually just followed clinically.
References 1.
2.
Conclusion 3.
Although the etiology of Kienböck disease is controversial, the pathogenesis is likely a combination of predisposing anatomy and exogenous factors such as trauma and/or medications, which ultimately compromise vascular supply to the lunate bone. Successful evaluation of post-operative imaging for Kienböck disease requires an understanding of the various surgical treatments, including the newer revascularization, buttressing and salvage techniques. While it is beyond the scope of this paper to compare effectiveness of the myriad of surgical treatments for Kienböck disease, the provided checklists and imaging examples can help the radiologist
4.
5. 6.
7.
8. Table 4 Checklist for evaluating radiographs after conservative or surgical treatment for Kienböck disease. Suggestions are provided for comments in comparison to pre-operative radiographs
9.
Post-operative radiograph checklist 10. ✓ Ulnar variance (negative or neutral)a ✓ Lunate morphology (shape, height, Stahl index) ✓ Lunate sclerosis or cystic change (increased, decreased or stable) ✓ Radioscaphoid angle (ideally 30–60°) ✓ Scaphocapitate agle (ideally increased compared to pre-op) ✓ Scapholunate interval (widened or normal)
11.
✓ Radiocarpal or intercarpal arthrosis (progressed or unchanged) ✓ Evidence of carpal row translation ✓ Fixation hardware (proper placement, any complication) ✓ Bony fusion at osteotomy/fusion site (usually seen by 3 months)
13.
a Although the contribution of ulnar variance is controversial, it should still be mentioned as the surgeon may value this information
12.
14. 15.
Arnaiz J, Piedra T, Cerezal L, Ward J, Thompson A, Vidal JA, et al. Imaging of Kienböck disease. AJR Am J Roentgenol. 2014;203(1): 131–9. van Leeuwen WF, Janssen SJ, Ter Meulen DP, Ring D. What is the radiographic prevalence of incidental Kienböck disease? Clin Orthop Relat Res. 2015. Yazaki N, Nakamura R, Nakao E, Iwata Y, Tatebe M, Hattori T. Bilateral Kienböck’s disease. J Hand Surg Br Eur Vol. 2005;30(2): 133–6. Seah KT, McEachan J, Davidson D. Bilateral Kienböck’s disease in association with type 1 diabetes. J Hand Surg Eur Vol. 2012;37(7): 697–8. Mok CC, Lau CS, Cheng PW, Ip WY. Bilateral Kienböck’s disease in SLE. Scand J Rheumatol. 1997;26(6):485–7. Wollstein A, Tantawi D, Wollstein R. Bilateral Kienbock’s disease concomitant with bilateral Legg-Calve-Perthes disease: a case report. Hand (New York, NY). 2013;8(1):120–2. Cerezal L, del Pinal F, Abascal F, Garcia-Valtuille R, Pereda T, Canga A. Imaging findings in ulnar-sided wrist impaction syndromes. Radiographics. 2002;22(1):105–21. Stahl S, Stahl AS, Meisner C, Hentschel PJ, Valina S, Luz O, et al. Critical analysis of causality between negative ulnar variance and Kienböck disease. Plast Reconstr Surg. 2013;132(4):899–909. Illarramendi AA, Schulz C, De Carli P. The surgical treatment of Kienböck’s disease by radius and ulna metaphyseal core decompression. J Hand Surg. 2001;26(2):252–60. Iwasaki N, Minami A, Oizumi N, Suenaga N, Kato H, Minami M. Radial osteotomy for late-stage Kienböck’s disease: wedge osteotomy versus radial shortening. J Bone Joint Surg Br Vol. 2002;84(5):673–7. Blanco RH, Blanco FR. Osteotomy of the radius without shortening for Kienböck disease: a 10-year follow-up. J Hand Surg. 2012;37(11):2221–5. Imaeda T, Nakamura R, Shionoya K, Makino N. Ulnar impaction syndrome: MR imaging findings. Radiology. 1996;201(2):495– 500. JM. AaZ. Malacia del semilunar. Valladolid, Spain: University of Valladolid; 1966. Beredjiklian PK. Kienböck’s disease. J Hand Surg. 2009;34(1): 167–75. Lutsky K, Beredjiklian PK. Kienböck disease. J Hand Surg. 2012;37(9):1942–52.
Skeletal Radiol 16.
Gelberman RH, Bauman TD, Menon J, Akeson WH. The vascularity of the lunate bone and Kienböck’s disease. J Hand Surg. 1980;5(3):272–8. 17. Schiltenwolf M, Martini AK, Mau HC, Eversheim S, Brocai DR, Jensen CH. Further investigations of the intraosseous pressure characteristics in necrotic lunates (Kienböck’s disease). J Hand Surg. 1996;21(5):754–8. 18. Lichtman DM, Mack GR, MacDonald RI, Gunther SF, Wilson JN. Kienböck’s disease: the role of silicone replacement arthroplasty. J Bone Joint Surg Am Vol. 1977;59(7):899–908. 19. Lichtman DM, Lesley NE, Simmons SP. The classification and treatment of Kienböck’s disease: the state of the art and a look at the future. J Hand Surg Eur Vol. 2010;35(7):549–54. 20. Schmitt R, Heinze A, Fellner F, Obletter N, Struhn R, Bautz W. Imaging and staging of avascular osteonecroses at the wrist and hand. Eur J Radiol. 1997;25(2):92–103. 21. Luo J, Diao E. Kienböck’s disease: an approach to treatment. Hand Clin. 2006;22(4):465–73. abstract vi. 22. Goldfarb CA, Hsu J, Gelberman RH, Boyer MI. The Lichtman classification for Kienböck’s disease: an assessment of reliability. J Hand Surg. 2003;28(1):74–80. 23. Larsen CF, Mathiesen FK, Lindequist S. Measurements of carpal bone angles on lateral wrist radiographs. J Hand Surg. 1991;16(5): 888–93. 24. Bain GI, Durrant A. An articular-based approach to Kienböck avascular necrosis of the lunate. Tech Hand Upper Extrem Surg. 2011;15(1):41–7. 25. Wollstein R, Wollstein A, Rodgers J, Ogden TJ. A hand therapy protocol for the treatment of lunate overload or early Kienböck’s disease. J Hand Ther. 2013;26(3):255–9. quiz 260. 26. Meena D, Saini N, Kundanani V, Chaudhary L, Meena D. Distraction histiogenesis for treatment of Kienböck’s disease: a 2to 8-year follow-up. Ind J Orthop. 2009;43(2):189–93. 27. Matsui Y, Funakoshi T, Motomiya M, Urita A, Minami M, Iwasaki N. Radial shortening osteotomy for Kienböck disease: minimum 10-year follow-up. J Hand Surg. 2014;39(4):679–85. 28. Fouly EH, Sadek AF, Amin MF. Distal capitate shortening with capitometacarpal fusion for management of the early stages of Kienböck disease with neutral ulnar variance: case series. J Orthop Surg Res. 2014;9(1):86. 29. Okamoto MAM, Shirai H, Ueda N, Kagawa Y. Capitate shortening for Kienböck’s disease. J Jpn Soc Surg Hand. 1999;15:674–6. 30. Kakar S, Shin AY. Vascularized bone grafting from the dorsal distal radius for Kienböck’s disease: technique, indications and review of the literature. Chir de la main. 2010;29 Suppl 1:S104–11. 31. Mir X, Barrera-Ochoa S, Lluch A, Llusa M, Haddad S, Vidal N, et al. New surgical approach to advanced Kienböck disease: lunate replacement with pedicled vascularized scaphoid graft and radioscaphoidal partial arthrodesis. Techn Hand Upper Extrem Surg. 2013;17(2):72–9. 32. Daecke W, Lorenz S, Wieloch P, Jung M, Martini AK. Vascularized os pisiform for reinforcement of the lunate in Kienböck’s disease: an average of 12 years of follow-up study. J Hand Surg. 2005;30(5): 915–22.
33.
Vilkki SK. A new concept and technique for custom-made microvascular lunate bone reconstruction. J Reconstr Microsurg. 2007;23(7):351–9. 34. Arora R, Lutz M, Deml C, Krappinger D, Zimmermann R, Gabl M. Long-term subjective and radiological outcome after reconstruction of Kienböck’s disease stage 3 treated by a free vascularized iliac bone graft. J Hand Surg. 2008;33(2):175–81. 35. Burger HK, Windhofer C, Gaggl AJ, Higgins JP. Vascularized medial femoral trochlea osteochondral flap reconstruction of advanced Kienböck disease. J Hand Surg. 2014;39(7):1313–22. 36. Higgins JP, Burger HK. Osteochondral flaps from the distal femur: expanding applications, harvest sites, and indications. J Reconstr Microsurg. 2014. 37. Bellemere P, Maes-Clavier C, Loubersac T, Gaisne E, Kerjean Y. Amandys® implant: novel pyrocarbon arthroplasty for the wrist. Chir de la main. 2012;31(4):176–87. 38. Bellemere P, Maes-Clavier C, Loubersac T, Gaisne E, Kerjean Y, Collon S. Pyrocarbon interposition wrist arthroplasty in the treatment of failed wrist procedures. J Wrist Surg. 2012;1(1):31–8. 39. Ueba Y, Nosaka K, Seto Y, Ikeda N, Nakamura T. An operative procedure for advanced Kienböck’s disease: excision of the lunate and subsequent replacement with a tendon-ball implant. J Orthop Sci. 1999;4(3):207–15. 40. Mariconda M, Soscia E, Sirignano C, Smeraglia F, Soldati A, Balato G. Long-term clinical results and MRI changes after tendon ball arthroplasty for advanced Kienböck’s disease. J Hand Surg Eur Vol. 2013;38(5):508–14. 41. Chim H, Moran SL. Long-term outcomes of proximal row carpectomy: a systematic review of the literature. J Wrist Surg. 2012;1(2):141–8. 42. Facca S, Gondrand I, Naito K, Lequint T, Nonnenmacher J, Liverneaux P. Graner’s procedure in Kienböck disease: a series of four cases with 25 years of follow-up. Chir de la main. 2013;32(5): 305–9. 43. Lee JS, Park MJ, Kang HJ. Scaphotrapeziotrapezoid arthrodesis and lunate excision for advanced Kienböck disease. J Hand Surg. 2012;37(11):2226–32. 44. Luegmair M, Saffar P. Scaphocapitate arthrodesis for treatment of late stage Kienböck disease. J Hand Surg Eur Vol. 2014;39(4):416– 22. 45. Minamikawa Y, Peimer CA, Yamaguchi T, Medige J, Sherwin FS. Ideal scaphoid angle for intercarpal arthrodesis. J Hand Surg. 1992;17(2):370–5. 46. Bain GI, McGuire DT. Decision making for partial carpal fusions. J Wrist Surg. 2012;1(2):103–14. 47. Budoff JE, Gable G. Ulnar translation of scaphocapitate arthrodeses in Kienböck’s disease: two case reports. J Hand Surg. 2005;30(1): 65–8. 48. Youm Y, McMurthy RY, Flatt AE, Gillespie TE. Kinematics of the wrist. I. An experimental study of radial-ulnar deviation and flexion-extension. J Bone Joint Surg Am Vol. 1978;60(4):423–31. 49. Buck-Gramcko D. Wrist denervation procedures in the treatment of Kienböck’s disease. Hand Clin. 1993;9(3):517–20.
