Polymethyl methacrylate spacers are commonly used during staged revision knee arthroplasty for infection. In cases with extensive bone loss and ligament instability, such spacers may not preserve limb length, joint stability and motion.

We report a retrospective case series of 19 consecutive patients using a custom-made cobalt chrome hinged spacer with antibiotic-loaded cement. The “SMILES spacer” was used at first-stage revision knee arthroplasty for chronic infection associated with a significant bone loss due to failed revision total knee replacement in 11 patients (58%), tumour endoprosthesis in four patients (21%), primary knee replacement in two patients (11%) and infected metalwork following fracture or osteotomy in a further two patients (11%). Mean follow-up was 38 months (range 24–70). In 12 (63%) patients, infection was eradicated, three patients (16%) had persistent infection and four (21%) developed further infection after initially successful second-stage surgery. Above knee amputation for persistent infection was performed in two patients.

In this particularly difficult to treat population, the SMILES spacer two-stage technique has demonstrated encouraging results and presents an attractive alternative to arthrodesis or amputation.

The results for autologous chondrocyte implantation (ACI) in the treatment of osteochondral defects in the knee are encouraging. At present, two techniques have been described to retain the chondrocyte suspension within the defect. The first involves using a periosteal flap harvested from the distal femur and the second involves using a type I/III collagen membrane. To the authors' knowledge there are no comparative studies of these two techniques in the current literature.

A total of 68 patients with a mean age of 30.52 years (range 15 to 52 years) with symptomatic articular cartilage defects were randomised to have either ACI with a periosteal cover (33 patients) or ACI with a type I/III collagen cover (35 patients). The mean defect size was 4.54 cm² (range 1 to 12 cm²). All patients were followed up at 24 months.

A functional assessment using the Modified Cincinnati score showed that 74% of patients had a good or excellent result following the ACI with collagen cover compared with 67% after the ACI with periosteum cover at 2 years (p>0.05). Arthroscopy at 1 year also demonstrated similar results for both techniques. However, 36.4% of the periosteum covered grafts required shaving for hypertrophy compared with 1 patient for the collagen covered technique.

This prospective, randomised study has shown no statistical difference between the clinical outcome of ACI with a periosteal cover versus ACI with a collagen cover at 2 years. A significant number of patients who had the ACI with periosteum technique required shaving of a hypertrophied graft within the first year of surgery. We conclude that there is no advantage in using periosteum as a cover for retaining the chondrocytes within an osteochondral defect; as a result we advocate the use of an alternative cover such as a porcine-derived, type I/III collagen membrane.