A NOVEL APPROACH USING UPRIGHT OPEN MRI TO MEASURE MENISCAL EXTRUSION IN WEIGHT BEARING - FEASIBILTY AND RELIABILITY...

Meniscal extrusion, defined as abnormal projection of the meniscus beyond the edge of the tibial plateau, has been strongly linked to the progression and severity of knee OA. This link is likely due to the inability of the extruded meniscus to properly distribute load, resulting in greater stress on the cartilage. Meniscal extrusion is often the result of a meniscal tear, and can be corrected as part of surgical repair, however the effectiveness of current surgical approaches is not known. While extrusion can be detected radiologically with MRI before and after surgery, current assessment with clinical MRI scanners requires the knee to be in an unloaded and extended position. Since menisci distribute loads across a range of knee motion, measuring extrusion in weightbearing and flexed positions would give better insight into the mechanical changes caused by injury and, crucially, whether current repair methods better restore knee mechanics to normal. Upright open MRI allows for knee imaging in weightbearing and flexed positions, but the tradeoff is worsened image resolution due to lower field strengths. Therefore, it is important to establish the reliability of meniscal extrusion measurements in weightbearing and flexed postures using upright open MRI.

1) Develop an upright open MRI protocol for studying meniscal extrusion in weightbearing and flexed positions, and 2) Assess the reliability of measuring meniscal extrusion.

We recruited four healthy participants (3 female/1 male, ages 24-28) and scanned their right and left knees individually in four different postures with a 0.5T upright open MRI scanner (Figure 1). Each participant had two identical scanning sessions for reliability analysis. Dixon sequences were used to obtain coronal images of the knee (TR = 430.7 s, TE = 7 s, FOV = 17 × 17 × 6 cm, slice thickness = 3.5 mm, slice gap = 0.5 mm, voxel dimensions = 0.56 × 0.41 × 3.5 mm, flip angle = 40°, bandwidth = 158.8 Hz/pixel, total scan time = 3 min 10 s). The postures were: a) supine with knees extended; b) unloaded seated at 90° of knee flexion; c) weightbearing standing with knees extended; and d) weightbearing assisted squat with knees flexed to 90°. An MRI-compatible sloped seat supported participants in assisted squatting while still allowing for joint loading and muscle activation. Medial and lateral meniscal extrusions were calculated on a midcoronal slice as the distance between the edge of the tibial plateau and the external edge of the meniscus. Test-retest reliability was reported using intra-class correlation coefficients (ICC), standard errors of measurement (SEM), and smallest detectable changes with 95% confidence (SDC₉₅).

Mean medial extrusion was 2.5 mm for supine, 1.9 mm for sitting, 2.9 mm for standing, and 2.1 mm for squatting. Mean lateral extrusion was 0.7 mm for supine, 1.2 mm for sitting, 0.4 mm for standing, and 0.7 mm for squatting. ICCs ranged from 0.75-0.96, SEM ranged from 0.3-0.4 mm, and SDC₉₅ ranged from 0.7-1.0 mm (Table 1).

We successfully developed a protocol for measuring in vivo meniscal extrusion in weightbearing and flexed knee positions. All participants could tolerate the four postures for the time required to acquire the images, and the images contained minimal motion artefact. Pathological extrusion is often defined as extrusion >3 mm beyond the edge of the tibial plateau in clinical, non-weightbearing images. Both medial and lateral menisci with posterior root tears average approximately 1 mm greater extrusion than healthy menisci. Therefore, our upright open MRI protocol is sufficiently reliable to detect the clinically significant differences in meniscal extrusion in weightbearing and flexed knee positions, making it feasible to study effects of meniscal injury and repair techniques.