What should be considered in using standard knee radiographs to estimate mechanical alignment of the knee?

Abstract

Objective: Anatomical tibiofemoral angle (anatomical TFA) of the knee measured on standard knee radiographs is still widely used as proxy for mechanical tibiofemoral angle (mechanical TFA), because of the practical and economic limitations in using full-limb radiographs. However, reported differences between anatomical and mechanical TFAs show wide variations. The first aim of this study was to determine whether gender, the presence of advanced osteoarthritis (OA), and history of total knee arthroplasty (TKA) influence the differences between anatomical and mechanical TFAs. The second aim was to identify anatomical features that cause divergences between anatomical and mechanical TFAs, and the final aim was to determine whether anatomical TFA measured using reference points more distant from the knee provides more accurate estimates of mechanical TFA. Design: In 102 knees with advanced OA before and after TKAs and 99 control knees with no/minimal OA, we assessed the differences between two anatomical TFAs, namely, anatomical TFA1 and anatomical TFA2, which were based on conventional or more distant proximal and distal reference points on standard knee radiographs, respectively, and the mechanical TFA measured on full-limb radiographs. These differences were investigated for women vs men, no/minimal OA vs advanced OA, and for knees before vs after TKA. Regression analyses were performed to determine associations between femoral and tibial anatomical characteristics and the differences between mechanical and anatomical TFAs. Results: The OA group showed significantly greater differences between mechanical and anatomical TFAs than the control group for both genders. In OA and TKA group, women were more likely to have greater mean differences between mechanical and anatomical TFAs than men. However, TKA did not significantly affect these differences. Femoral and tibial bowing angles, particularly of the femur, were found to be the major contributors to divergences between mechanical and two anatomical TFAs. Furthermore, anatomical TFA2 was found to provide more accurate estimates of mechanical TFA. Conclusions: We found that the differences between mechanical and anatomical TFAs depend on gender and the presence of advanced OA, but not on a history of TKA. These findings indicate that prediction of mechanical TFA based on anatomical TFA is dependent on study population characteristics. This study also shows that the presence of lateral bowing of the femur is a major cause of mechanical TFA to anatomical TFA variations associated with gender and advanced OA. To reduce the adverse effects of anatomical variations on estimations of mechanical TFA based on an anatomical TFA method, more distant proximal and distal reference points are recommended to determine anatomical TFA value on standard knee radiographs. © 2010 Osteoarthritis Research Society International.