Anatomical alignment, but not goniometry, predicts femorotibial cartilage loss as well as mechanical alignment: data from the Osteoarthritis Initiative

Abstract

To determine how frontal plane lower limb alignment obtained using a new femorotibial angle (FTA) measurement and non-radiographic handheld goniometry, predict femorotibial cartilage thickness loss in varus and valgus knees, compared with the gold standard mechanical axis (hip-knee-ankle [HKA]). 934 Osteoarthritis (OA) Initiative knees with radiographic OA had the above alignment measures and 3T knee MRIs acquired. The new FTA measure was compared to the gold standard, with and without adjusting FTA for the sex-specific varus shift. Changes in medial (MFTC) and lateral femorotibial (LFTC) cartilage thickness were quantified over 1-year and 2-years. Adjusted odds ratios (adjORs) were used to compare how the different alignment measures predict medial and lateral cartilage loss in varus and valgus knees. Pearson correlation coefficients between 2-year MFTC/LFTC cartilage loss and alignment measures were small to moderate, and were similar for FTA (r=0.28/-0.30) and for HKA (r=0.28/-0.29). Using the adjusted FTA measure, varus and valgus predicted MFTC progression (adjOR=3.73) and LFTC progression (adjOR=2.55) as well as HKA (adjOR=3.16 and 2.31) over 1-year, and this relationship was also observed over 2-years. Goniometry was a weak predictor for MFTC and LFTC progression (adjOR1-year=1.65 and 1.71; adjOR2-year=0.68 and 1.24). After adjustment, the new FTA measure obtained from short (fixed-flexion) knee films was as good as the gold standard in predicting medial and lateral cartilage loss over 1- or 2-years, without need for obtaining long-limb radiographs for determining the mechanical axis. Goniometry and non-adjusted FTA measures, in contrast, were poor predictors of cartilage loss.