Abstract
INTRODUCTION
The major loss of articular cartilage in medial osteoarthritis occurs in a central band on the distal femur, and in the center of the tibial plateau (Figure). This is consistent with varus deformity due to cartilage loss and meniscal degeneration, together with the sliding regions in walking. Treatment at an early stage such as KL grade 2 or 3, has the advantages of little bone deformity and cruciate preservation, and could be accomplished by resurfacing only the arthritic areas with Early Intervention (EI) components. Such components would need to be geometrically compatible with the surrounding bearing surfaces, to preserve continuity and stability. However because of the relatively small surface area covered, compared with total knees and even unicompartmentals, it is hypothesized that EI components will be an accurate fit on a population of knees with only a small number of sizes, and that accuracy can be maintained without requiring right-left components. We examined this hypothesis using unique design and methodology.
METHODS
Average femur and tibia models, including cartilage, were generated from MRI scans of 20 normal males. The images were imported into Geomagic software. Surface point clouds based on least squares algorithms produced the average models. Averages were also produced from different numbers to determine method validity. Average arthritic models were also generated from 12 KL 1–2 cases, and 13 KL 2–3 cases. The 3 averages were compared by deviation mapping. Using the average from the 20 knees, femoral and tibial implant surfaces were designed using contour matching to fit the arthritic regions, maintaining right-left symmetry. A 5 size system was designed corresponding to large male, average male, small male/large female, average female, small female. For the 20 knees, the components were fitted based on the best possible matching of the contours to the surrounding bearing surfaces. For the femoral component the target was 1 mm projection at the center, matching at the ends. The accuracy of reproducing the cartilage surfaces was then determined by mapping the deviations between the implant surfaces and the cartilage surfaces.
RESULTS & DISCUSSION
The average femur and tibia from the 20 knees (Figure) was almost identical no matter what groupings were used to produce the average. Likewise the 2 arthritic and the normal averages were almost identical. The accuracy of fit (Figure) averaged for the 20 normal knees was well below 1mm either above or below the original cartilage surfaces (see table below). This study indicates that such Early Intervention components are a viable method for resurfacing cases with early arthritis, and are likely to show almost normal mechanics due to preserving the original normal geometry.
Deviations between tibial cartilage and implant (mm)
Above implant mean 0.5 SD 0.2. Max deviation 1.5 SD 0.7
Below implant mean 0.7 SD 0.2. Max deviation 2.0 SD 0.1
Deviations between femoral cartilage and implant (mm)
Above implant mean 0.3 SD 0.1. Max deviation 0.8 SD 0.1
Below implant mean 0.3 SD 0.1. Max deviation 0.8 SD 0.3
