Kienbock's disease most often occur's between the ages of 20 and 40 years and is twice as common in men. The clinical features are dorsal wrist pain, usually associated with synovitis, limitation of movement and decrease in grip strength. In advanced disease the symptoms are those of established osteoarthritis of the wrist. Bilateral Kienbock's disease is uncommon (Morgan and McCue, 1983) and the condition is rarely associated with other disorders, although Rooker and Goodfellow (1977) described a number of cases in patients with cerebral palsy who had abnormally flexed wrists.
Several hypotheses have been suggested to explain the changes in the lunate. Factors that may be important are:
1.Interruption of the blood supply leading to secondary osteonecrosis and fracture;
2. Damage to the blood supply by a primary lunate fracture; or
3.Repetitive stressing of the lunate at the point where the relatively compliant triangular fibrocartilage is attached to the radius, leading to subcortical microfractures.
The blood supply of the lunate is probably the key factor in the pathogenesis of the disorder, and the extra-osseous and intra-osseous vascular anatomy has been well studied. The lunate may be supplied by a single vessel, either volar or dorsal, or there may be several vessels (Lee, 1963). Theoretically, the lunate that is supplied by a single vessel is more at risk of avascular necrosis due to interruption of its blood supply but that is not the whole story. Gelberman and colleagues (1980) studied the vascular anatomy in fresh bones and found that the extraosseous supply was usually extensive. The intra-osseous blood supply had three main patterns and there was a relatively avascular area in the subchondral bone adjacent to the radius. They concluded that the intra-osseous Supply might be damaged by repeated minor trauma.
Several investigators have confirmed Hulten's observation that the ulna was relatively shorter than the radius in 78% of patients with Kienbock's disease and only 23% of the normal population (Gelberman et al., 1975). However doubt has been cast on the significance of this observation (Kristensen et al., 1986a) and the importance of standardised radiological views when measuring u1nar variance must be emphasised (Palmer et al., 1982). Theoretically the "ulna-minus variant" could increase the shear stress on the lunate, although the striking step between the radius and ulna seen on radiographs is of course occupied by the radiolucent triangular fibrocartilage. Kienbock's disease does not inevitably occur when there is a gross discrepancy in length between radius and ulna (Nathan and Meadows, 1987) so other factors must be involved.
Kienbock's disease can be classed in four stages (Lichtman et al., 1982):
Stage 1. The radiographic appearance of the lunate is normal, although there may be a suggestion of a compression fracture.
Stage 11. The lunate exhibits increased density but its size and shape are unchanged.
Stage 111. The lunate has collapsed, allowing the capitate to migrate proximally. In stage IIIA the scaphoid maintains a normal position relative to the rest of the carpus but in stage IIIB it has moved into a position of fixed rotation, shown by the scaphoid "ring" sign.
Stage IV. Secondary degenerative changes are present in the carpus.
The patient who presents with wrist pain but no physical signs and a normal radiograph is a common problem. Does the patient have stage I Kienbock's disease? A radioisotope bone scan is sensitive in demonstrating disorders of the carpus but not specific in its features, so its diagnostic value is limited. In this issue of the Journal, Imaeda and colleagues report their experience with magnetic resonance imaging in Kienbock's disease. They establish criteria for diagnosis by M.R. and also find it useful in prognosis.
Many forms of operative treatment have been described for the various stages of Kienbock's disease but their effectiveness must be compared with the outcome of untreated disease. Kristensen et al. (1986b) reported a long-term review (mean follow-up 20 years) of 46 patients who received no treatment or only a short period of immobilisation. Around 80% of the wrists were pain-free and there was no correlation between pain and radiological appearance. Little has been written about treatment in stage 1, probably because the diagnosis is seldom made at that stage. Immobilisation is generally recommended, although progressive collapse can still occur when the wrist is immobilised since the compressive forces continue to act across it (Lichtman et al., 1977).
Treatment in stage 11, when there is avascularity but no collapse, is based on the current concepts of pathogenesis. The options are ulnar lengthening (Armistead et al., 1982) or radial shortening (Rock et al., 1991) both of which have been shown to be effective in altering stresses on the lunate (Trumble et al., 1986). Imaeda et al. report that M.R. signal intensity returns to normal after radial osteotomy. This is of great interest and lends support to the current concept of abnormal loading of the lunate being an important factor in causation. As an alternative to osteotomy, attempts have been made to restore the blood supply by arterial implantation (Hori et al., 1977) or vascularised bone graft at this stage but results have not been published.
Operations for stage II disease can also be used in stage IIIA. They cannot restore the height of the lunate and therefore are not recommended in stages IIIB and IV. Treatment in stage IIIB aims to maintain carpal height and unload the lunate. This is usually done by scapho-trapezio-trapezoidal fusion (Watson et al., 1985). Whether the lunate should also be excised and replaced by rolled tendon or silicone prosthesis is contentious. Concern about silicone synovitis has made many surgeons wary of using silicone replacements in young patients with high demands on the wrist (Alexander et al., 1990). Voche and colleagues in their paper in this issue of the journal report that silicone synovitis occurred in one of I I patients who underwent lunate replacement along with S.T.T. fusion. They provide a realistic assessment of the problems associated with partial carpal fusion.
In stage IV disease with severe symptoms a salvage procedure such as proximal row carpectomy or wrist fusion may have to be considered. Wrist denervation (Buck-Gramcko, 1977) has also been recommended. In my experience patients very seldom present with symptoms at this stage and the condition is usually picked up on films taken. for some other reason.
A striking feature of papers describing surgical treatment in Kienbock's disease is that most series include very few patients and comparison of treatments is uncommon. Now that the condition can be accurately staged and some treatments appear to have a rational basis,perhaps it is time to establish multicentre studies to enable sufficient patients to be recruited into comparative trials that will determine the effectiveness, or otherwise, of the various forms of treatment.