Can Demographic Variables Accurately Predict Component Sizing in Primary Total Knee Arthroplasty?

Abstract

As health care reform drives providers to reduce costs and improve efficiencies without compromising patient care, preoperative planning has become imperative. The purpose of this study is to determine whether height, weight, and gender can accurately predict total knee arthroplasty (TKA) sizing. A consecutive series of 3491 primary TKAs performed by 2 surgeons was reviewed. Height, weight, gender, implant, preoperative templating sizes, and final implant sizes were collected. Implant-specific dimensions were collected from vendors. Using height, weight, and gender, a multivariate linear regression was performed with and without the inclusion of preoperative templating. Accuracy of the model was reported for commonly used implants. There was a significant linear correlation between height, weight, and gender for femoral (R² = 0.504; P < .001) and tibial sizes (R² = 0.610; P < .001). Adding preoperative templating to the regression analysis increased the overall model fit for both the femoral (R² = 0.756; P < .001) and tibial sizes (R² = 0.780; P < .001). Femoral and tibial sizes were accurately predicted within 1 size of the final implant 71%-92% and 81%-97% using demographics alone or 85%-99% and 90%-99% using both templating and demographics, respectively. This novel TKA templating model allows final implants to be predicted to within 1 size. The model allows for simplified preoperative planning and potential implementation into a cost-savings program that limits inventory and trays required for each case.