Fig. 1
The midcarpal articulation of two wrists : (a) Type I lunate, (b) Type II lunate (Reprinted with permission from Kapoutsis DV, Dardas A, Day CS. Carpometacarpal and scaphotrapeziotrapezoid arthritis: arthroscopy, arthroplasty, and arthrodesis. J Hand Surg Am. 2011;36:354–66)
It should be noted that most of the aforementioned studies have focused on adult patients or specimens. Important differences between the adult and pediatric population include the presence of the physis, increased remodeling potential in pediatric patients, and a potentially different vascular supply. Although these factors have not been comprehensively investigated, they present potential explanations as to why the natural history of Kienbock’s disease appears to be more favorable in pediatric patients.
Assessment of Kienbock’s Disease
Physical Examination
The characteristic presentation of Kienbock’s disease is vague, dorsal-sided wrist pain. The intensity of symptoms and associated findings such as weakness and decreased range of motion depend on disease severity. Lichtman et al. correlated patient presentation with disease stage and developed the “Lichtman classification” (Lichtman et al. 2010). The following section describes the latter radiographic classification and the associated physical findings.
Radiographic Assessment
Lichtman et al.’s radiographic staging system is the most common classification system used to describe the severity of disease. The system utilizes posteroanterior (PA) hand radiographs and is based on the lunate’s architecture, density, fracture, and collapse, as well as the anatomic relationship of the surrounding carpal bones (Lichtman et al. 1977; Fig. 2). It was a modification of Stahl’s original grading system (Stahl 1947). The classification system was developed in order to guide treatment and to enable a better comparison of clinical outcomes research (Lichtman et al. 2010). The system is also clinically useful, as it appears to reflect patient symptom severity. That is, patients with more severe symptoms tend to have a worse radiographic grade (Lichtman et al. 2010). The system has been found to have substantial interobserver reliability and excellent intraobserver reliability (Goeminne et al. 2010; Jafarnia et al. 2000), although some authors have found the inter- and intraobserver reliability to be fair to moderate at best (Jensen et al. 1996; Shin et al. 2011). Furthermore, some authors have suggested that the Lichtman classification is not applicable to children younger than 12 years of age due to the small size of the lunate and the favorable natural history of disease (Irisarri et al. 2010). The following is a description of the four disease stages:
Fig. 2
Lichtman’s modification of Stahl’s staging of Kienbock’s disease. (a) Stage I, lunate fracture without density or shape change. (b) Stage II, increased lunate density without marked change in lunate size or shape, and relationship of the bones is not changed. Marked fracture lines may be noted. (c) Stage IIIa, lunate collapse without carpal collapse. (d) Stage IIIb, static carpal collapse. (e) Stage IV, extensive osteoarthritic changes (Reprinted with permission from Elhassan BT, Shin AY. Vascularized bone grafting for treatment of Kienböck’s disease. J Hand Surg Am. 2009;34:146–54)
Stage I: This stage reflects the earliest phase of pathogenesis. Lunate architecture and density appear normal on plain radiographs, with the possible exception of a linear or compressive fracture. At the time this classification system was proposed in 1977, magnetic resonance imaging (MRI) had not become a standard diagnostic tool. However, MRI could distinguish stage I disease by allowing the visualization of a uniformly decreased signal on T1 or a uniformly increased signal on T2 imaging. Patients with stage I disease usually complain of mild pain that worsens with activities, particularly during extension and axial loading.
Stage II: In this stage, definite changes in lunate density are observable on plain radiographs, with evidence of sclerosis and lytic changes. The lunate maintains its structural integrity, however, and the relationship of the surrounding carpal bones, in particular the scaphoid and capitate, is maintained, without any evidence of wrist instability. Patients with stage II disease exhibit more persistent pain and swelling as a result of synovitis.
Stage III: The hallmark of this stage is lunate fracture and collapse. In addition, secondary proximal migration of the capitate might be observed. The original description by Lichtman et al. included stages IIIA (no disruption of carpal alignment) and IIIB (disruption of carpal alignment and instability evidenced by fixed flexion deformity of the scaphoid, apparent as the cortical ring sign on PA radiographs). Recently, IIIC has been proposed as an additional disease stage. This stage is characterized by a coronal lunate fracture and warrants a different treatment approach (Lichtman et al. 2010). Patients with stage III disease experience more intense and constant pain. The clinical picture resembles early degenerative arthritis.
Stage IV: This is the final stage of disease and is characterized by lunate collapse in addition to radiocarpal or midcarpal degenerative arthrosis. Patients typically present with symptoms of persistent wrist stiffness and pain more severe than stage III.
The carpal height ratio has sometimes been used to assess the radiographic extent of lunate collapse in Kienbock’s disease and offers another measurement option that allows objective comparison of preoperative and postoperative PA x-rays (Delaere et al. 1998). The carpal height ratio is calculated by dividing the height of the carpus by the height of the third metacarpal on the PA view (Fig. 3).
Fig. 3
The carpal height ratio is determined by dividing the carpal height (A) by the height of the third metacarpal (B) on the PA radiograph
Magnetic resonance imaging is often used when Kienbock’s disease is suspected, as it is capable of delineating early disease development that is not visualized on plain film. Increased signal changes on T2-weighted images in adults and children are encouraging signs as they reflect edema or revascularization and may be of prognostic value (Lichtman et al. 2010). MRI also useful to longitudinally track the patient’s response to treatment following direct and indirect revascularization procedures (Lichtman et al. 2010). It should be noted that in Kienbock’s disease, T1-weighted MRI signal intensity is decreased and diffusely involves the lunate in its entirety (Fig. 4), while in other pathological conditions such as a lunate bone bruise, ulnar impaction syndrome, enchondroma, or intraosseous ganglions, signal changes are more focal.
Fig. 4
A 17-year-old boy with Kienbock’s disease. (a) Preoperative x-ray shows sclerosis, fragmentation of the lunate, and decrease of the lunate height. (b) Preoperative MRI shows low signal intensity on T1-weighted images (Reprinted with permission from Ando Y, Yasuda M, Kazuki K, Hidaka N, Yoshinaka Y. Temporary scaphotrapezoidal joint fixation for adolescent Kienböck’s disease. J Hand Surg Am. 2009;34:14–9)
Owing to the diagnostic limitations of plain radiographic films, wrist arthroscopy has been used as an additional diagnostic and staging tool to help guide treatment (Bain and Begg 2006; Watanabe et al. 1995). The advantage of wrist arthroscopy is its ability to allow direct observation of the articular cartilage, intercarpal and extrinsic ligaments, synovium, and triangular fibrocartilage complex (TFCC) in order to hone in on the source of pain. An arthroscopic staging system was developed by Bain and Begg in order to help guide definitive treatment (Bain and Begg 2006). This classification system is based on the number of nonfunctional cartilaginous surfaces of the lunate, the lunate facet of the distal radius, and the proximal pole of the capitate.
Grade 0: Wrist arthroscopy identifies no abnormal cartilage surfaces, with a normal cartilaginous surface defined as normal glistening appearance or minor fibrillation, with normal, hard subchondral bone on palpation.
Grade 1: Involvement of one nonfunctional articular surface, which is typically the proximal articular pole of the lunate.
Grade 2A: Involvement of two nonfunctional articular surfaces, which are the proximal lunate and the lunate fossa of the distal radius.
Grade 2B: Involvement of two nonfunctional articular surfaces, which are the proximal articular surface as well as the distal articular surface of the lunate.
Grade 3: Involvement of three nonfunctional articular surfaces, which include the lunate facet of the distal radius, the proximal articular surface of the lunate, and the distal articular surface of the lunate.
Grade 4: Involvement of four nonfunctional articular surfaces, which is the same as grade 3, in addition to proximal capitate involvement.
Other Diagnostic Modalities
Less commonly used imaging modalities for suspected Kienbock’s disease include bone scan and computed tomography (CT) scans. Bone scan usually shows increased absorption of Tc-99 m, but its specificity is unclear for Kienbock’s disease. CT scan may reveal cystic changes and could confirm a more advanced stage of disease as compared to plain film and may be useful in particular when distinguishing stage II from stage IIIA, although these stages are usually treated the same way.
Outcome Tools Available for the Assessment of Kienbock’s Disease
Currently, there are no unique outcome tools used for the assessment of clinical results following the treatment of Kienbock’s disease. Several validated, generalized health questionnaires and focused hand and upper extremity patient-rated questionnaires have been used to convey the patient’s experience of disease and response to treatment. These include the Disabilities of the Arm, Shoulder and Hand (DASH) patient-rated functional questionnaire and the Patient-Rated Wrist Evaluation (PRWE) (Croog and Stern 2008; Watanabe et al. 2008). Unlike their utility in adult patients, however, the use of these specific questionnaires in pediatric patients may not be useful due to the inherent differences in the cognitive capacities of the two age groups. An alternative would be the use of the generic Pediatric Outcomes Questionnaire (also referred to as the Pediatric Outcomes Data Collection Instrument, PODCI). Parents are required to complete this questionnaire on behalf of their children between the ages of 2 and 10 years.
Conservative Treatment Options
Most, if not all, pediatric patients who have been diagnosed with Kienbock’s disease should undergo a trial of conservative treatment for at least 3–6 months. It is appropriate to consider conservative management in younger patients who present with advanced stage disease, including stage IIIB (lunate collapse with fixed flexion deformity of the scaphoid) given the enhanced remodeling potential of pediatric patients. In most cases, conservative treatment consists of wrist immobilization with a short arm cast for 6 weeks, followed by a removable wrist splint for an additional period of time depending on the patient’s symptoms. Activity modification is advisable in order to minimize strenuous activities that load the wrist. Parents are often asked to prevent their children from participating in sports activities during the treatment period. Nonsteroidal anti-inflammatory drugs (NSAIDs) may be used to relieve pain. A formal therapy program is generally not required (Table 1).
Table 1
Indications for nonoperative treatment of Kienbock’s disease
Indications | Contraindications |
|---|---|
Initial form of treatment for all stage I to stage IIIB disease, except advanced disease with evidence of severe deformity or arthritis | Failed conservative treatment or radiographic progression of disease |
Advanced disease: stage IV |
Treatment-Specific Outcomes
Given the rarity of this condition, most of the available literature on the conservative management of pediatric Kienbock’s disease is composed of case reports (Cvitanich and Solomons 2004; De Smet 2003; Docquier et al. 2009; Greene 1996; Herzberg et al. 2006). In general, outcomes of conservative treatment have been favorable. On the other hand, most reports that describe operative management usually indicate failed conservative treatment (Ferlic et al. 2003; Foster 1996). Very few case series have been dedicated exclusively to pediatric patients. Irisarri et al. (Irisarri et al. 2010) presented the outcomes of conservative treatment in 13 pediatric patients: 4 were 12 years of age or younger (disease entity labeled “infantile lunatomalacia”), and 9 were older than 12 years of age (“juvenile lunatomalacia”). Conservative treatment consisted of wrist immobilization for an unspecified duration. Based on patient age and response to conservative treatment, the authors distinguished between infantile lunatomalacia (IL) and juvenile lunatomalacia (JL) and suggested that this condition should be viewed as a distinct entity from Kienbock’s disease, which they suggested should be reserved to adults. All patients in the IL group responded positively to conservative management. One of the patients was a 9-year-old girl with 6 years of follow-up and although she demonstrated radiographic findings consistent with an abnormal appearing lunate, she remained asymptomatic. On the other hand, three out of nine patients in the JL group required operative intervention.
It should be noted that stage IIIB has been treated successfully with conservative measures alone. In a case report by Cvitanich and Solomons, an 8-year-old boy gymnast was successfully treated after an 11-month treatment period consisting of an initial 6-week period of a short arm cast, followed by the use of a removable thermoplastic splint (Cvitanich and Solomons 2004). MRI evidence of revascularization was noted as early as 2 months after treatment initiation. Following 16 months of treatment, the patient had regained full wrist motion and returned to gym participation.
Operative Treatment of Kienbock’s Disease
Failure of symptom resolution following conservative treatment should prompt the clinician to consider operative management. There is currently no consensus on the most effective surgical procedure for the treatment of Kienbock’s disease. Although numerous techniques have been described, most of the literature is based on Level IV studies, with very little comparative data to help establish superiority of one technique over another. In general, and with the exception of end-stage disease requiring salvage, most of the described techniques have afforded acceptable results regardless of disease stage and level of invasiveness. Most procedures that have been described could be grouped into treatment categories. These include (1) mechanical unloading and joint-leveling procedures aimed at decreasing load and strain across the lunate (e.g., radial shortening osteotomy, ulnar lengthening osteotomy, capitate shortening, radial wedge osteotomy), (2) revascularization procedures aimed at restoring the blood supply of the necrotic lunate, (3) miscellaneous procedures that seem to improve symptoms but do not conform to the former two categories (e.g., distal radius core decompression), and (4) salvage procedures aimed at eliminating pain once the disease has reached its end stages. Salvage procedures in young patients include principally proximal row carpectomy but also include wrist arthrodesis. In this section, a select, non-exhaustive number of surgical techniques are described based on the authors’ preferences and experiences.
Arthroscopic Debridement
Wrist arthroscopy could be used as a diagnostic adjunct to guide definitive treatment (Bain and Begg 2006) or as definitive treatment in the form of synovectomy and debridement for early as well as stages IIIA and IIIB disease (Menth-Chiari et al. 1999). The advantages of this technique include its minimally invasive nature and the ability to directly observe the lunate and surrounding carpus under magnification (Tables 2, 3, 4, and 5).
Table 2
Indications for arthroscopic treatment of Kienbock’s disease
Indications | Contraindications |
|---|---|
Stage I–IIIB disease that has failed conservative treatment | Stage IV disease |
Ulnar positive or ulnar negative wrists |
Table 3
Preoperative planning of arthroscopic debridement of for Kienbock’s disease
OR table: Standard wrist arthroscopy setup |
Position: Patient placed supine. Handheld in 4–5 kg of traction in a traction tower |
Equipment: Standard wrist arthroscopy setup |
Tourniquet: A nonsterile tourniquet is preferred |
Table 4
Surgical steps of arthroscopic debridement for Kienbock’s disease
Portal placement: Standard 3–4, 6R and midcarpal portals are used |
Using a probe, examine the status of the synovium and the articular surfaces of the lunate at the lunate fossa and the capitate. In particular, the quality of the subchondral bone deep to the articular surfaces of the lunate should be assessed for hardness and whether there are associated fractures of the lunate or floating chondral bodies. The scapholunate and lunotriquetral ligaments should also be assessed for laxity or rupture |
The lunate is debrided of any necrotic fragments until capillary bleeding is observed. Inflamed synovium should also be debrided |
Ensure that all loose bodies and debris have been removed from the radiocarpal and midcarpal joints prior to closure |
Table 5
Postoperative protocol for arthroscopic debridement of Kienbock’s disease
At the end of the procedure, a resting volar wrist splint is applied for comfort for 1 week |
Hand therapy to improve wrist and hand range of motion and strength is optional in pediatric patients |
If therapy is initiated, it should be started 1 week postoperatively. The therapy program should be tailored according to the patient’s symptoms and requirements |
Treatment-Specific Outcomes
Outcomes of arthroscopic debridement in pediatric patients are generally lacking. In adults, results are scarce but largely favorable. The advantage of this treatment modality is its minimally invasive nature and the flexibility imparted to the surgeon, with regard to the capacity of performing related future surgeries. Menth-Chiari et al. reviewed the outcomes of seven adult patients with stage III disease treated with arthroscopic debridement only and found synovitis in five out of seven patients and loose fragments in the radiocarpal or midcarpal joints in six out of seven patients. The lunate was variably softened, delaminated, or ulcerated. After a mean follow-up duration of 19 months, two patients had “excellent” results, four patients had “good” results, and one patient had a “fair” outcome according to the modified Mayo score.
Temporary Scaphotrapezial Fusion
Over the past several years, temporary scaphotrapezial (ST) pinning has gained a favorable reputation and has become a popular surgical option for the treatment of pediatric Kienbock’s disease (Fig. 5). In part, this has been due to the simple, minimally invasive nature of the procedure, which generally requires less operative time and less postoperative immobilization in comparison to other procedures used to treat pediatric Kienbock’s disease. Clinical outcomes, although sparse, have afforded consistently encouraging results and have also helped to popularize this surgical technique (Ando et al. 2009; Kazuki et al. 2006; Yasuda et al. 1998). In theory, temporary ST pinning unloads the lunate, by transferring strain to the adjacent carpal joints, thereby providing a favorable environment for the revascularization of the lunate. Indications typically include failed conservative treatment of stages I to IIIB disease, regardless of ulnar variance (Tables 6, 7, 8, and 9).
Fig. 5
Postoperative x-ray shows temporary ST joint fixation using three titanium wires (Reprinted with permission from Ando Y, Yasuda M, Kazuki K, Hidaka N, Yoshinaka Y. Temporary scaphotrapezoidal joint fixation for adolescent Kienböck’s disease. J Hand Surg Am. 2009;34:14–9)
Table 6
Indications for temporary scaphotrapezoidal pinning in Kienbock’s disease
Indications | Contraindications |
|---|---|
Ulnar negative or positive variance | Stage IV disease |
Failed treatment of stages I–IIIB disease |
Table 7
Preoperative planning for temporary scaphotrapezoidal pinning
OR table: Hand table |
Position/positioning aids: The patient is placed supine on a translucent operative table |
Fluoroscopy is required |
Tourniquet: A nonsterile well-padded pneumatic tourniquet is placed on the arm |
Table 8
Surgical steps of temporary scaphotrapezial joint pinning
Two or three K-wires are inserted in a retrograde fashion from the dorsal trapezoid to the scaphoid under fluoroscopic control in the maximally ulnar-deviated position of the wrist |
After satisfactory pinning, a splint is applied for an initial 2 weeks followed by a short arm cast or splint for an additional 2–4 months depending on the patient’s progress and symptoms |
The K-wires are typically removed 3–4 months after surgery, and this could be done in the office under local anesthesia. The timing of wire removal is decided by the findings of improving revascularization on x-ray or MRI |
Table 9
Postoperative protocol for temporary scaphotrapezial joint pinning
Following the initial 2-week postoperative immobilization period, patients are placed in a short arm cast or splint for an additional 2–4 months depending on medical progress. No formal therapy studies have been conducted to assess the efficacy of a regimented therapy protocol. It is reasonable to begin passive and active range of motion exercises as soon as a cast is removed. This should be followed by gradual strengthening exercises |
Treatment-Specific Outcomes
Yasuda et al. were among the first authors to describe their experience with temporary ST fusion in a pediatric patient. A 12-year-old girl with stage IIIB disease and 0.5 cm negative ulnar variance was treated for 4 months with ST fusion and a splint. Sixteen months following surgery, the patient had no pain, improvements in range of motion, and lunate revascularization as seen on MRI (Yasuda et al. 1998). Ando et al. reported the outcomes of temporary ST joint fixation in six patients, including four females and two males, with a mean age at surgery of 14 years (range, 9–17 years) (Ando et al. 2009). Four patients had an open physis, while two had a closed physis at the time of surgery, and all presented with either stage IIIA or IIIB disease. Postoperatively, patients were immobilized for variable periods of time ranging from 3 weeks to 4 months. Wires were removed between 3 and 6 months. The mean follow-up duration was 23 months (range, 7–48 months), and the results indicated significant improvements in flexion, extension, and grip strength in addition to reductions in pain severity. Furthermore, radiographic findings improved as evidenced by the absence of sclerosis and lunate fragmentation on x-rays and normalization of signal intensity on MRI. Complications occurred in two out of six patients and included pin migration, failed surgery requiring a second temporary ST fixation, and pin track infections.
Radial Shortening Osteotomy
Radial shortening osteotomy (RSO) is one of the most common procedures used to treat Kienbock’s disease in both the adult and adolescent/pediatric patient population. The aim of this joint-leveling procedure is to decrease the load on the injured lunate by diverting load to the surrounding carpus. Biomechanical studies have indicated that shortening the radius by 3 mm decreases lunate strain by an average of 70 % (Trumble et al. 1986). Horii et al. used a computer-simulated carpal model and found that joint-leveling procedures such as RSO and ulnar lengthening osteotomy resulted in a 45 % decrease of radio-lunate load with only moderate changes in force at the midcarpal and radio-scaphoid joints (Horii et al. 1990). Makabe et al. used osteoabsorptiometry to assess the effects of RSO on the subchondral bone density of the distal radius in seven patients with stage IIIA and IIIB disease (Makabe et al. 2011). Subchondral bone density was used as a surrogate marker of stress distribution across the distal radius articular surface. After a mean follow-up of 27 months, the authors found the mean high-density area ratio in each fossa to be significantly reduced from 0.253 to 0.096 in the scaphoid fossa and from 0.160 to 0.045 in the lunate fossa, indicating lunate unloading (Fig. 6). RSO is usually indicated in patients with stage I–IIIB disease and reserved for patients with positive ulnar variance, although it has also been performed in ulnar neutral wrists. In order to avoid ulnocarpal impaction, ulnar positive variance is a contraindication to RSO, as is advanced wrist arthritis seen in stage IV disease (Tables 10, 11, and 12).
