BACKGROUND: There have been few prospective studies that have documented the outcome of surgical treatment of hallux valgus deformities. The purpose of this investigation was to evaluate the effect of operative treatment of hallux valgus with use of a proximal crescentic osteotomy and distal soft-tissue repair on the first metatarsophalangeal joint. METHODS: All adult patients in whom moderate or severe subluxated hallux valgus deformities had been treated with surgical repair between September 1999 and May 2002 were initially enrolled in the study. Those who had a hallux valgus deformity treated with a proximal crescentic osteotomy and distal soft-tissue reconstruction (and optional Akin phalangeal osteotomy) were then invited to return for a follow-up evaluation at a minimum of two years after surgery. Outcomes were assessed by a comparison of preoperative and postoperative pain and American Orthopaedic Foot and Ankle Society scores; objective measurements included ankle range of motion, Harris mat imprints, mobility of the first ray (assessed with use of a validated calibrated device), and radiographic angular measurements. RESULTS: Of the 108 patients (127 feet), five patients (five feet) were unavailable for follow-up, leaving 103 patients (122 feet) with a diagnosis of moderate or severe primary hallux valgus who returned for the final evaluation. The mean duration of follow-up after the surgical repair was twenty-seven months. The mean pain score improved from 6.5 points preoperatively to 1.1 points following surgery. The mean American Orthopaedic Foot and Ankle Society score improved from 57 points preoperatively to 91 points postoperatively. One hundred and fourteen feet (93%) were rated as having good or excellent results following surgery. Twenty-three feet demonstrated increased mobility of the first ray prior to surgery, and only two feet did so following the bunion surgery. The mean hallux valgus angle diminished from 30 degrees preoperatively to 10 degrees postoperatively, and the mean first-second intermetatarsal angle decreased from 14.5 degrees preoperatively to 5.4 degrees postoperatively. Plantar gapping at the first metatarsocuneiform joint was observed in the preoperative weight-bearing lateral radiographs of twenty-eight (23%) of 122 feet, and it had resolved in one-third (nine) of them after hallux valgus correction. Complications included recurrence in six feet. First ray mobility was not associated with plantar gapping. There was a correlation between preoperative mobility of the first ray and the preoperative hallux valgus (r = 0.178) and the first-second intermetatarsal angles (r = 0.181). No correlation was detected between restricted ankle dorsiflexion and the magnitude of the preoperative hallux valgus deformity, the post-operative hallux valgus deformity, or the magnitude of hallux valgus correction. CONCLUSIONS: A proximal crescentic osteotomy of the first metatarsal combined with distal soft-tissue realignment should be considered in the surgical management of moderate and severe subluxated hallux valgus deformities. First ray mobility was routinely reduced to a normal level without the need for an arthrodesis of the metatarsocuneiform joint. Plantar gapping is not a reliable radiographic indication of hypermobility of the first ray in the sagittal plane.