Revision surgery for failed unicompartmental knee replacement: technical aspects and clinical outcome.

The aims of this study were to determine the complexity of surgery required to revise failed unicompartmental knee replacements and to evaluate the outcome following revision. Between 2000 and 2009, 494 cemented Oxford phase 3 medial unicompartmental knee replacements were implanted, with 24 (4.9%) requiring revision (mean age: 63.5 years; 58% male). Mean time to revision was 3.0 years. All cases were revised to a cemented total knee replacement, with primary components used in 67% and revision components in 33%. At a mean follow-up of 3.2 years the median Oxford knee score was 333% with one knee requiring re-revision (5-year survival 933%). Most failed unicompartmental knee replacements could be revised without the need for stemmed implants, augmentation, or bone allograft. When bone loss occured it was commonly on the tibial side. Good functional outcome for the revised unicompartmental knee replacement was achieved and was comparable to primary knee replacement.

Survival of autologous osteochondral grafts in the knee and factors influencing outcome.

The primary aim of this retrospective study was to determine the medium-term survival of autologous osteochondral grafts (mosaicplasty) in the knee. The secondary aims were to evaluate any reason for failure, and to ascertain if age, gender, BMI, previous or associated knee surgery, site or size of the osteochondral graft had any influence on outcome. Fifty-five patients undergoing an autologous osteochondral graft procedure between 1999 and 2008 were evaluated, with a mean follow-up of 5.9 years (range 0.5 to 10.5). Loosening, graft degeneration or subsequent surgery in the form of arthroplasty, revision autologous osteochondral graft or any other osteochondral procedure, and an Oxford knee score above 49.7 (see text) were considered as failure. Kaplan-Meier analysis demonstrated an 87.5% survival at 8 years (95% CI 72% to 97%). The mean Oxford score at follow-up was 163% (95% CI 10.6% to 22.1%) at follow-up. Two of 6 failures occurred in patients with varus malalignment. Linear regression analysis demonstrated an improved outcome in Oxford knee score in younger patients. Gender, BMI, previous or associated knee surgery, site and size of the graft had no influence on outcome. The authors consider autologous osteochondral grafting as a reliable method of treatment in the medium-term for young patients with small contained articular cartilage lesions up to 4 cm2. Careful attention should be paid to the mechanical axis (varus) prior to grafting. Long-term studies are necessary.

Assessment of acetabular retroversion following long term review of Salter's osteotomy.

Salter's innominate osteotomy may predispose to anterior over-coverage of the acetabulum. Over cover or retroversion has been demonstrated to be a cause of hip pain, impingement and subsequent osteoarthritis. We reviewed the long-term follow up of seventeen skeletally mature hips in sixteen patients who had previously undergone a Salter's osteotomy in childhood. The Salter pelvic osteotomy was performed at a mean average age of 5 years and follow up at a mean average age of 20 years. Patients were assessed by clinical examination for signs of impingement, Harris Hip Score and pelvic radiograph. Acetabular version was evaluated by the relationship between anterior and posterior walls of the acetabulum using templates applied to the pelvic radiograph as described by Hefti. The median acetabular cover averaged 17 degrees of anteversion with 2 patients (12%) demonstrating retroversion, neither of whom, had signs of impingement on examination. The mean average Harris Hip Score was 85 indicating a good outcome at long-term follow-up. We believe acetabular remodelling may occur with age after Salter's innominate osteotomy and have found good results in patients after skeletal maturation. Fears of long-term anterior over-coverage and retroversion with this operation may be unfounded.