To study the survival analysis of the Accord TKR and to analise the causes for its failure. 111 Accord knees were implanted in 106 patients between 1986 and 1996. All components were cemented. Eighty-seven patients were followed up and assessed according to the Knee Society Clinical Rating System.

Life table analysis of this implant using revision as the end point shows a survival of 25% at 11–12 years.

Of the 87 knees, only 31 were still in situ and of these 7 showed radiographic signs of severe loosening. However, the average knee score for these 31 was only 65/100 and average functional score was down to 42/100.

56 implants have been revised, 21 due to aseptic loosening, 11 due to gross valgus/varus instability, 9 due to deep infection, 8 due to loosening of the patella liner and 7 due to mobile bearing complications.

All 21 (24%) cases of aseptic loosening were found to have a loose femoral component; however, 56% of the total showed significant radiographic osteolysis around the stem of the tibial implants.

Retrieved implants in 5 patients showed significant delamination of the UHMWP at its margins and also revealed a track through the tibial baseplate into the medullary cavity of the tibia.

Conclusion: Early failure of this implant is due to increased debris formation from the UHMWP due to edge loading and early delamination probably as a consequence of the shape of its articulating surface. This situation was significantly aggravated by a design fault in the tibial baseplate, which encouraged unrestricted access of debris into the medullary cavity of the tibia and hence early and severe osteolysis.

Our aim was to review the efficacy of the wound vacuum-assisted closure (VAC) system in the treatment of deep infection after extensive instrumentation and fusion for spinal deformity in children and adolescents.

A total of 14 patients with early deep spinal infection were treated using this technique. Of these, 12 had neuromuscular or syndromic problems. Clinical and laboratory data were reviewed. The mean follow-up was 44 months (24 to 72). All wounds healed. Two patients required plastic surgery to speed up the process. In no patient was the hardware removed and there was no loss of correction or recurrent infection.

We believe that the wound VAC system is a useful tool in the armamentarium of the spinal surgeon dealing with patients susceptible to wound infections, especially those with neuromuscular diseases. It allows for the retention of the instrumentation and the maintenance of spinal correction. It is reliable and easy to use.

Introduction Kirschner wire (K-wire) pinning is a widely accepted technique for fixation of distal radius fractures. Potential exists for injury to the soft tissues. This cadaveric study evaluates the safety of percutaneous pinning of distal radius fractures using a technique of intrafocal placement of K-wires.

Methods Three K-wires (1.6 mm diameter) were inserted percutaneously into 18 cadaveric wrists 18 mm proximal to the radial styloid. A radial wire was placed between the first and second extensor compartments. A dorso-radial wire was placed proximal to Lister’s tubercle. A dorsoulnar wire was placed between the fourth and fifth compartments. The wrists were dissected along the wires to the bone.

Results The superficial radial nerve (SRN) was pierced twice (11%), the abductor pollicis longus six times (33%), the extensor pollicis brevis and the extensor carpiradialis brevis once (6%). Extensor pollicis longus (EPL) was entered five times (28%), the fourth compartment four times (22%) and the fifth compartment once (6%). Only four wrists (22%) escaped injury to any important structure.

Conclusions The incidence of SRN injury by percutaneous insertion of K-wires is similar to that reported for the mini-open approach of around 12%. Rupture of EPL has been reported at around two percent. Aetiology of rupture is unclear and a K-wire that is subsequently removed may not increase the risk of rupture.

Aim: To ascertain the accuracy of partial weight bearing.

Method: 6 healthy volunteers with a below knee plaster cast, 10 patients with uncemented hip replacements and 12 patients with lower limb fractures were trained to partial weight bear. They were asked to place the affected leg on a bathroom scale and to press on it till the prescribed limit. This process was repeated till the subject formed a mental image of the amount of load they must put through the limb. The ability to partial weight bear was tested in a gait lab by making them walk on a walkway incorporating a Bertec force platform. Exact magnitude of weight bearing was calculated from the vertical ground reaction forces produced.

Results: 4 out of 6 volunteers exerted mean weight of 20.3 kg above and the remaining 2 exerted 5.6 kg below that prescribed. Of the 22 patients, 19 exerted mean weight of 24.3 kg above and 3 patients exerted mean weight of 7.5 kg below that prescribed. As per Spearmanñs rank correlation test, the relationship between the prescribed weight bearing and the actual weight bearing was non-signiþcant (p=0.399) i.e., there is little relationship between the prescribed and actual weight bearing.

Conclusions: Neither patients nor healthy volunteers could partial weight bear to the extent required. They were either above or below the prescribed level of partial weight bearing. Current method of teaching partial weight bearing is inaccurate and has poor reproducibility. Such methods use static loading situations whereas walking is a dynamic activity. An inexpensive, easy to use, dynamic device is required to train patients to partial weight bear.