We report the outcome of 58 knees with anteromedial osteoarthritis in which the Oxford unicompartmental arthroplasty was inserted. These were performed in a district general hospital by three surgeons.

All the knees had only anteromedial disease, an intact anterior cruciate ligament and correctable varus. The indication for replacement in all cases was pain. The mean follow up was 24.5 months (6–48). Outcome was assessed by patient satisfaction and the Oxford knee score. Complications, revisions, time to mobility and time to return to work were also noted.

The average age of the 26 women and 23 men at time of operation was 65 years. 31 of the patients were very happy with the outcome, 12 were happy, 5 were unhappy, and one was very unhappy. Mean pre-operative Oxford knee score was 43 (27–53) this improved post-operatively to 18 (12–45) a significant improvement (p< 0.005, paired t-test). Time taken to mobility was an average of 36 hours (24–72), 24 of the patients were in full or part time employment at the time of operation, all returned to their former posts at an average of 6 weeks (2–24).

Three patients have ongoing pain and are booked for revision to TKR. One patient had a dislocated femoral component and required this to be revised twice with a meniscus change at the same time; this patient is now happy. 2 further patients had revision of the meniscus to a larger size for meniscal dislocation. One patient had an infection treated with debridement and antibiotics; infection settled. Our results show that there is a learning curve; all of the insert revision occurred early in the series. Patient selection is important, those with disease in other compartments have continuing pain. Appropriate selection of patients and good surgical technique are the key to obtaining a good outcome.