Introduction This study aims to improve knee arthroplasty prosthetic alignment by determining if an algorithm based on establishing the most prominent points on the medial and lateral malleolion 3D CT scans can be used to establish the true center of the ankle joint.

Methods Axial, coronal and sagittal multi-planar reconstructions were generated on 20 ankles. Two observers independently identified the most prominent medial and lateral malleolar points, in the coronal plane, and the highest talar dome point, in the sagittal plane. Ratios were calculated comparing total intermalleolar distance to distance to medial and lateral malleolus, and the ratio of medial to lateral distance. The distance from the true center of the joint, in the sagittal plane, to the computer calculated center was determined. Statistical analysis using ANOVA, paired t-tests and regression analysis was performed. There were 17 normal ankles, two arthritic ankles, and one previously fractured ankle.

Results In the coronal plane there was a strong correlation between the measurements of each observer. The mean intermalleolar distance was 70.2 mm (95% CI 68.3–72.0). The strongest correlation was seen in the ratio of lateral distance to total distance (r=0.728) which was 0.57 in normal ankles (95% CI 0.55–0.58). The ratio for arthritic ankles was 0.48 (95% CI 0.46–0.50) and for the fractured ankle 0.57 (95% CI 0.15–0.99). These were significantly different at the five percent level (p< 0.02). The normal ankle ratio was substantiated by regression analysis. There was a poor correlation between the individual measurements in the sagittal plane (r=0.218). The mean distances from the calculated line to the true center were not statistically different with the true center always lying posterior to the calculated line (4.2 mm (95% CI 2.5–5.9) and 2.8 mm (95% CI 1.7–3.8) posterior. For the combined data this means that the mean distance that the true center of the ankle joint’s from a line joining the medial and lateral malleoli is 3.2 mm (95% CI 2.3–4.0 mm). The data was reproducible with a small standard deviation in each plane. Assuming a 300 mm tibial length, angular error in tibial alignment generated by a computer navigation system is less than one degree in both planes.

Conclusions The algorithm presented can give accurate measurements of normal ankle joints in knee navigation surgery.

Introduction Unicompartmental knee replacement (UKR) is an increasingly utilised alternative to tibial osteotomy and total knee arthroplasty in patients with single compartment degenerative disease. We report on four fractures of the medial tibial plateau following UKR.

Methods We retrospectively reviewed four cases with periprosthetic tibial plateau fractures following unicompartmental knee replacement. Each arthroplasty, performed between 1999 and 2002, was done in a community teaching hospital by a single orthopaedic surgeon and a senior level assistant. All patients had medial compartment osteoarthritis confirmed both radiographically and arthroscopically prior to arthroplasty surgery. The arthroplasties were performed by four different surgeons and three different arthroplasty systems were used. All cases were reviewed using the documented chart histories and x-ray evaluation. Each surgeon was contacted individually for the relevant case history and x-rays. The study population was composed of four females, and no males with a mean age of 63.5 years (range 58 to 68). Two patients (50%) had simultaneous bilateral UKRs performed. The remaining two patients had unilateral procedures, involving one right and one left knee. Two patients were clinically obese, and one patient had had a previous ipsilateral high tibial osteotomy.

Results The total number of fractures was four, involving three left knees and one right knee. Of the bilateral arthroplasties each patient sustained a unilateral fracture of the left knee. The patient with the previous tibial osteotomy sustained an ipsilateral fracture. Two fractures involved traumatic falls, the remaining fractures had no history of trauma. The mean post-operative period to fracture was 95.75 days with a range of 5 to 195 days. Two patients had revision surgery to total knee arthroplasty. One patient underwent internal fixation of the fracture with retention of the original prosthetic components and exchange of the polyethylene bearing. The remaining patient underwent revision of the tibial component with concurrent internal fixation and was subsequently revised to total knee arthroplasty as the result of failure. Subsequent to the described surgery all fractures have healed with no further surgical intervention.

Conclusions This series, whilst small, demonstrates that tibial periprosthetic fracture following UKR is a previously unreported but important cause of failure. Revision surgery to total knee replacement appears to be a reasonable salvage option.