The use of arthroscopy in kienböck disease

Relevant anatomy and etiology

Dr. Robert Kienböck was a Viennese radiologist who described a series of patients presenting with radiographic signs of a collapsed lunate and attributed it to a traumatic disruption of the bone’s blood supply. Although more than 100 years have passed, the etiology of avascular necrosis (AVN) remains to be determined. After observing a negative ulnar variance in 74% of his patients with Kienböck disease, Hùlten postulated that this might increase the shear forces across the lunate and predispose patients to developing lunate AVN. D’Hoore et al. found no significant difference between the mean ulnar variance of 125 patients without Kienböck disease and 52 patients with the disease. Nakamura et al. measured the ulnar variance in 325 Japanese patients with normal wrists and 41 patients with Kienböck disease when the effects of sex and age were taken into account.

Gelberman et al. injected 35 cadaver wrists with latex and observed that there was both a palmar and dorsal blood supply in most specimens. Of those with a dual blood supply, 59% had a three-vessel anastomosis (Y pattern), 31% had a single palmar and dorsal vessel anastomosis (I pattern), and 10% had a four-vessel anastomosis (X pattern). There was a single vascular supply in approximately 7% of lunates. The authors concluded that Kienböck disease may be the result of compression fractures due to repeated trauma in poorly vascularized lunates. One criticism of this conclusion, however, is that none of the lunates showed signs of AVN. Others have proposed that lunates with a single vessel or with limited intraosseous branching may be at increased risk for AVN. Disruption of venous outflow has also been proposed as an etiology for Kienböck disease. Schiltenwolf et al. measured the lunate intraosseous pressure in 12 patients with normal lunates and 12 patients with necrotic lunates. The intraosseous pressure was recorded in neutral position and in 60 degrees of wrist extension under normotensive conditions and during venous stasis. In neutral position, no significant differences in pressure were found between normal lunates and capitates. A significant difference was found for venous stasis and extension. The intraosseous pressure rise in the normal lunate in extension was slightly higher than the pressure found during venous stasis. The intraosseous pressure differed significantly (by 56.9 mm Hg) between normal and necrotic lunates in extension. These data support the hypothesis that impairment of venous drainage plays a role in lunate necrosis. It is impossible, however, to know if these increased pressures were the cause or result of Kienböck disease.

Bain et al. have proposed that the AVN should be considered as consisting of three pathologic phases: vascular (early), osseous (intermediate), and chondral (late). The early vascular changes start with ischemia, followed by subsequent necrosis and revascularization. The osseous changes initially consist of subchondral sclerosis, which is followed by subchondral collapse, and possibly a coronal fracture and remodeling if the process is reversible. If the initial insult persists, the changes become irreversible. In the late chondral phase, the articular cartilage is often soft and can be indented, giving the impression that the articular surface has a false floor. Bain states that this softness is most likely due to loss of the subchondral bone. Subchondral bone collapse can then result due to extensive osseous necrosis, with intervening fibrous tissue and cysts that fill the osseous voids from reabsorption of the medullary canal.

Diagnosis

Kienböck disease often occurs in males, age 20 to 40 years old, who may present with the insidious onset of wrist pain and stiffness. There is usually no history of trauma. The pain is typically exacerbated by wrist extension and axial loading, and is relieved by immobilization. Patients may have symptoms of an associated carpal tunnel syndrome or flexor tenosynovitis secondary to protrusion of a volar lunate fragment. On examination, there is dorsal wrist swelling and synovitis, with tenderness over the lunate and decreased wrist motion and grip strength. Radiographs are diagnostic for Kienböck disease. Although early in the disease radiographs may be normal, diffuse sclerosis may be present. With progression, there may be linear compression fractures due to trauma or a weakened subchondral plate, cystic changes, lunate collapse, loss of carpal height, and perilunate arthritic changes.

Open treatment

The Lichtman classification is still in widespread use. In stage I, the plain radiographs are normal and there is no evidence of sclerosis or lunate collapse. With an MRI, the lunate shows a low-signal intensity on T1-weighted images. On T2-weighted images, the signal may be decreased or increased depending on whether bone edema is present or if revascularization is occurring ( Fig. 16.1 A–B). Immobilization is the initial treatment for patients with stage 1 disease and can be effective in relieving symptoms. This can be accomplished using a splint or cast, external fixator, or temporary intercarpal pinning for up to 3 months.

In Stage II, radiographs reveal increased lunate density without collapse, but there may be one or more fracture lines present. The T1-weighted signal is similar to those in stage I, but the T2-weighted images may now show larger areas of decreased signal. In stage IIIA there is maintenance of carpal alignment. Stages II and IIIA are generally the same in regard to treatment options. The goal of treatment at these stages is to restore vascularity to the lunate in the hope of preventing progression. This can be done by mechanically unloading the lunate either by a radial shortening osteotomy if there is an ulnar-minus variance, or a capitate osteotomy if there is an ulnar-positive variance. This may also be accomplished by bringing in a new blood supply directly through revascularization before articular collapse has occurred, which includes a vascularized pisiform transfer, vascularized pedicled bone grafts from the distal radius, vascularized metacarpal bone grafts, and free vascularized grafts. These procedures are most appropriate in the ulnar-neutral or ulnar-positive wrist, in which joint leveling may not be appropriate. They may be combined with additional procedures to unload the lunate, such as a capitate shortening ( Fig. 16.2 A–C), radius osteotomy ( Fig. 16.3 A–D), external fixation, or intercarpal fixation with pinning. Illaramendi et al. reported their experience with curettage of the radius and ulna ( Fig. 16.4 A–B) through small cortical windows for 22 patients with stage 1-IIIA Kienböck disease. At an average follow-up of 10 years (range, 6–16 yr) 17 patients did not show disease progression, 2 improved radiographically, and 3 showed disease progression.

In stage IIIB, there is lunate collapse along with a flexion deformity of the scaphoid due to carpal instability, which results in a dorsal intercalated segmental instability (DISI) pattern. This can be managed with some type of midcarpal fusion such as a scaphotrapeziotrapezoidal (STT) fusion or a scaphocapitate (SC) fusion, with or without excision of the lunate ( Fig. 16.5 A–D). Excision of the lunate and an interposition arthroplasty can be performed, and it may be combined with vascularized bone graft, external fixation, or a partial wrist arthrodesis. A proximal row carpectomy (PRC) is an option with a normal capitate. A wrist denervation can be performed alone or in combination with these procedures ( Fig. 16.6 A–B). Stage IV disease is defined by lunate collapse and secondary degenerative changes in the radiocarpal, STT, and/or midcarpal joints. The options include a total wrist fusion or total wrist arthroplasty. Illaramendi et al. also proposed a stage 0, which includes patients with intermittent wrist pain, normal radiographs, and a normal MRI. One way to identify these patients would be to perform a contrast-enhanced MRI immediately after repetitive axial loading of the wrist, In a stage 0 patient, there would be an increased signal in the lunate following this wrist stress test.