Résumé
We retrospectively compared results of 27 wrists with bicolumnar arthrodesis with mean follow-up of 67 months to 28 wrists with three-corner arthrodesis adding triquetral excision with mean follow-up of 74 months in 54 patients (55 wrists). Minimal follow-up was 2 years for all patients. Capitolunate nonunion occurred in three wrists with bicolumnar arthrodesis and six wrists with three-corner arthrodesis, and radiolunate arthritis developed in four wrists with three-corner arthrodesis. Among patients with bicolumnar arthrodesis, hamatolunate arthritis occurred in seven wrists, all with a Viegas type II lunate; and pisotriquetral arthritis occurred in three wrists. At mean 5 years after surgery, 45 wrists had not needed revision surgery, and both groups had similar revision rates. The wrists with three-corner arthrodesis and bicolumnar arthrodesis had similar functional outcomes, and range of wrist motion was not significantly different between the two groups. We concluded that bicolumnar arthrodesis results in greater longevity than three-corner arthrodesis for a type I lunate. We do not recommend bicolumnar arthrodesis for type II lunate. We also concluded that three-corner arthrodesis has a greater incidence of radiolunate arthritis and capitolunate nonunion.