Patellar Morphology and Osteoarthritis: A Cadaveric Analysis.

Abstract

Previous studies have attempted to characterize patellar morphology and its relation to patellofemoral osteoarthritis but have utilized relatively small samples. We designed a large-scale cadaveric analysis to elucidate any potential relationship. Patellar morphology in 1,032 patellae was characterized using patellar facet concavity patterns initially developed by Wiberg and using the ratio of the lateral and medial facet lengths. Patellofemoral arthritis grading for specimens was previously completed based primarily on osteophytic burden. The average age at time of death was 56 ± 10 years. Wiberg's grading returned kappa values of 0.43 for intrarelator reliability and 0.27 for interrelator reliability. Using multiple regression analysis, Wiberg's type II was found to be correlated with patellofemoral arthritis (p = 0.003). Lateral to medial facet ratio measurements returned intraclass coefficients of 0.91 for intrarelator reliability and 0.85 for interrelator reliability. There was no association between lateral to medial facet ratio and osteoarthritis. Regression analysis between Wiberg's typing and lateral-to-medial ratio (L:M) resulted in a coefficient of determination of 0.14. To our knowledge, this is the largest study evaluating the relationship between patellar morphology and arthritis. Although there was a relationship between Wiberg's type II and patellofemoral arthritis, the Wiberg systems was not easily reproducible in our study and returned a correlation with arthritis different then anatomically expected. A more reproducible system of lateral to medial facet ratio was investigated but did not find an association with arthritic changes. Given these findings, we did not find a patella morphology grading system which is both reliable and valid in regard to association with patellofemoral arthritis.