87 WALKING LOADING AT THE KNEE PREDICTS MRI-DERIVED CARTILAGE THICKNESS CHANGES IN MEDIAL COMPARTMENT KNEE OSTEOARTHRITIS

Abstract

Purpose: Knee osteoarthritis (OA) occurs in a substantial portion of the population over the age of 50. However, there is limited information on the underlying causes of OA and the reasons why the rate of disease progression varies between patients. In medial compartment knee OA, the peak knee adduction moment has been reported to predict disease progression as evaluated by radiographic JSW. However, mechanistic interpretation of radiographic measures of cartilage thinning is limited, and it is difficult to distinguish between femoral and tibial cartilage loss as well as assess if changes occur in specific walking load-bearing regions of the knee joint. Magnetic resonance (MR) imaging allows the investigation of specific regional changes in cartilage. We tested the hypothesis that baseline knee loading during gait predicts local MRIderived cartilage loss in the medial compartment in patients with knee OA at a five year follow-up. Further, we hypothesized that increases in KAM over 5 years lead to focal increased cartilage loss. Methods: Fourteen medial compartment OA knees (6 male, 8 female; age: 64.2+/-8.2 yrs) with an average baseline Kellgren-Lawrence grade of 2 were tested twice using MRI and gait analysis, with an average time between testing of 55 months following written consent in accordance with the Institutional Review Board. MR images were acquired with a 1.5T scanner (General Electric) using a 3D spoiled gradient-echo sequence in the sagittal plane. Images were manually segmented and 3D cartilage thickness maps were created. For local thickness analysis, the weightbearing medial femoral cartilage was divided into three sub-regions: external, central, and internal. Similarly, the medial tibia cartilage was divided in five sub-regions: central, anterior, external, posterior, and internal. Subjects performed 3 walking trials at a self-selected normal speed in their personal shoes. Kinematic and kinetic data were captured using a 10-camera optoelectronic motion capture system (Qualisys), and the first peak knee adduction moment was calculated using inverse dynamics. Linear regression was used to test for a correlation between changes in cartilage thickness and joint loading. The level of significance was set at <0.05, with trends defined as <0.1. (Table presented) Results: At the five-year follow-up, baseline peak KAM significantly predicted cartilage thinning in the medial femoral external region (P = 0.01), with trends towards correlations for the medial central (P = 0.06) and total medial (P = 0.07) weight-bearing regions (Table 1). Only one significant association between the baseline peak KAM and thinning was seen on the medial tibia posterior region (P = 0.04), with a trend for the central medial tibia (P = 0.08). When the change in adduction moment (DELTAKAM) was considered there was a broader range of regions on the medial femur where cartilage thickness changes over 5 years were related to changes in KAM. An increase in DELTAKAM was significantly associated with an increase in thinning for the external, central, and total weight-bearing medial femur (P = 0.02, 0.002, and 0.02, respectively). No significant associations were seen on the medial tibia. Conclusions: These results suggest that over 5 years baseline KAM can predict OA progression, and suggest that the effect of the KAM on cartilage thinning may be region dependent, with more associations found on the medial weight-bearing femur. Further, an increased loss of cartilage thickness is associated with an increase in dynamic joint loading over 5 years, which may further accelerate the rate of OA progression, and suggest the KAM should be reduced to slow disease progression.