WEIGHT BEARING 3-D JOINT SPACE WIDTH DISTRIBUTION AT THE KNEE VARIES ACCORDING TO LOCATION AND EXTENT OF MENISCAL EXTRUSION: A MOST INVESTIGATION

Abstract

Quantitative 3-D measurement of joint space width (JSW) from weight bearing computed tomography (WBCT) allows analysis of knee tibiofemoral compartments in a loaded stance without subregional constraints on analysis. The relationship between 3-D JSW distribution and meniscal extrusion is yet to be established. To demonstrate topographic variations in 3-D JSW distribution at the knee derived from WBCT imaging data according to meniscal extrusion location and extent. WBCT knee imaging was acquired at the 144-month visit of the Multicenter Osteoarthritis Study (MOST). A subset of individuals had medial and lateral meniscal extrusion grading according to the MRI Osteoarthritis Knee Score to identify medial extrusion of the medial meniscus (MMmed), anterior extrusion of the medial meniscus (MMant), anterior extrusion of the lateral meniscus (LMant), and lateral extrusion of the lateral meniscus (LMlat). The extent of extrusion was graded as 0: <2 mm; 1: 2 to 2.9 mm, 2: 3 to 4.9 mm; and 3: >5 mm. After semi-automatic segmentation of 663 baseline knees, joint space mapping was performed to create 3-D JSW maps for each knee registered to a template surface. Statistical parametric mapping (SPM) was performed using a general linear model to test the dependence of 3-D JSW distribution on each of meniscal extrusion type in turn. SPM results were plotted on the canonical joint surface, with unmasked regions demonstrating the topographic difference in JSW for each increment in extrusion score. 568 knees from 330 participants had meniscal extrusion grading available. Participants (178 female) had mean ± SD age 62.0 ± 9.3 years, body mass 82.2 ± 17.4 kg, height 170 ± 9.5 cm, and BMI 28.3 ± 4.8 kg/m2. Analysed knees were bilateral for 238 participants and unilateral for 92 participants. For MMmed (cohort score range 0-3), the central-to-posterior medial compartment had significantly lower JSW by up to 0.4 mm for each increment in meniscal extrusion grade. Both MMmed and MMant (score range 0-3) were associated with a small region of similarly significantly lower JSW along the medial margin of the lateral compartment, but also with significantly greater JSW at the anterior medial compartment of up to 0.6 mm per grade. The posterior medial joint space was significantly narrower (up to 3 mm) comparing knees with LMant scores of 0 vs 1. This same magnitude of narrowed JSW was detected along the periphery of the lateral joint space comparing grades in the lateral tibiofemoral compartment, but this was not statistically significant. LMlat (score range 0-2) was associated with a trend for greater reduction in JSW in the central lateral compartment of up to 0.5 mm per grade, but this was not statistically significant. Direction of meniscal extrusion is represented by the black arrowhead in the figure. Both medial and anterior extrusion of the medial meniscus are associated with significant reduction in JSW at the medial margin of the lateral joint space and widening of the anterior margin, likely from a posteromedial shift of the femoral condyles with respect to the tibia. Medial extrusion of the medial meniscus is also associated with narrowed JSW in the central to posterior medial joint space, likely as a direct effect of the loss of meniscal hoop tension. Anterior (but not lateral) extrusion of the lateral meniscus is associated with significantly lower JSW in the posterior medial joint space. These findings confirm that the extent and location of meniscal extrusion are key factors in weight bearing 3-D JSW distribution.