Metatarsophalangeal Joint Arthrodesis after Failed Keller-Brandes Procedure

Background: Keller-Brandes resection arthroplasty for correction of symptomatic hallux valgus deformity can obtain early good results, but late complications, such as recurrence of the deformity and instability of the first ray, have been described. Arthrodesis of the first metatarsophalangeal, (MTP) joint can be done as a salvage procedure. The aim of this prospective study was to evaluate the clinical outcome of the arthrodesis and its effect on the biomechanics of the first ray. Methods: Between October, 1999, and December, 2002, arthrodesis of the MTP joint was done after a failed Keller-Brandes procedure in 28 feet of 26 consecutive patients. Twenty patients (22 feet) with a minimum of 24 months followup were available for clinical and radiographic assessment. Pedobarographic measurements were obtained at latest followup in 16 patients (17 feet). Results: Sixteen feet (72%) were pain-free and six feet (28%) had mild, occasional pain. The American Orthopaedic Foot and Ankle Society (AOFAS) forefoot score increased from a preoperative 44 (range 29 to 67) points to 85 (range 73 to 90) points at longest clinical followup (average 34 months, range 23 to 48, p < 0.001). The average hallux valgus angle was corrected from 24.0 (range 7 to 47) degrees preoperatively to 16.0 (range 0 to 40) degrees postoperatively (p < 0.001). Two feet had pseudoarthroses. Biomechanically, the MTP joint arthrodesis could not fully restore the function of the hallux but produced a significant improvement, allowing a more physiologic loading pattern under the hallux and the metatarsal heads. Conclusions: First MTP joint arthrodesis after a failed Keller-Brandes procedure is a technically safe and reliable technique. It resulted in a marked reduction of pain and gain of function that produced high patient satisfaction.