We compared the chevron and the Wilson metatarsal osteotomy for hallux valgus in a prospective randomised trial on 87 feet in 51 patients, reviewed at averages of 5.5 and 38 months after operation. The patients in the chevron group returned to work earlier and mobilised faster, but, at the later review, those in the Wilson group had better functional results and were more satisfied with the appearance of the foot. Correction of the hallux valgus angle was better maintained in patients in the Wilson group and they had a better range of motion at the metatarsophalangeal joint; fewer complained of metatarsalgia.

Management of the failed total knee replacement is a formidable problem. In a 13-year period, 24 patients with above knee amputation following unsuccessful knee replacement were referred for prosthetic limb fitting. Adequate data was available for 23 patients. At review (or at the time of death), an average of 48.6 months after amputation, only seven of the 23 patients were regular daily walkers, although 10 patients had managed to walk for more than two years; 20 of the 23 used a wheelchair for some part of the day and 12 were confined to a wheelchair.

To detect sepsis in the painful joint, autologous granulocytes labelled with indium-111 were used to scan 60 patients with 74 prosthetic joints. All 18 patients with confirmed sepsis had positive scans, that is, migration of granulocytes into the region of the prosthesis. Of the 22 sterile arthroplasties 20 had negative scans. In 34 cases there was good correlation between the clinical impression and the result of the scan. We suggest that indium granulocyte scintigraphy is a reliable method of detecting an infected prosthesis.