MENISCAL EXTRUSION IN ASYMPTOMATIC KNEES: THE 3MM CUTOFF MUST BE REVISITED

Abstract

Meniscal extrusion is frequently encountered in OA, and an arbitrary cutoff value of 3 mm is commonly accepted for a significant extrusion of the body of the medial meniscus (MM). Different methods of measurement have also been described. We aim to provide the normal range of MM extrusion in different zones in asymptomatic volunteers without radiographic and/or MRI OA and to compare different measurement methods on coronal and sagittal 2D images to those performed on 3D images in radial planes. This is a post-hoc analysis of knee MRIs of 93 participants (39.6±16.0 years; mean ± SD) with asymptomatic, healthy knees (KOOS score 97.42±3.22 and WOMAC score 97.41±7.54) from the Lausanne Knee Study. Inclusion criteria: age between 18-35 or 45-70 years, no knee symptoms in the past 12 months, no history of severe lower limb injury, no impairment that might affect gait, BMI≤30. Imaging protocol included weight-bearing Schuss and lateral radiographs of both knees and 3T MRI of a randomly selected knee. MRI protocol included 2D fat-suppressed FSE IW images (sagittal, coronal, and transverse planes), isotropic 3D DESS, and high-resolution 3D T1. One radiologist measured the maximal extrusion of the MM in the radial plane using 3D FSE T1 images in 4 anatomical subregions: zones II-a (anterior), II-b (anteromedial), III (medial), and IV (posterior). Measurements were also performed on coronal and sagittal 2D images using two different methods: on the image where the maximal extrusion was seen (methodMax) and on an image selected based on anatomical landmarks (methodLandmark). Another musculoskeletal radiologist graded knee MRIs for all structural lesions included in the MOAKS criteria. In addition, meniscal extrusion was measured in the sagittal and radial planes to assess interobserver agreement. Measurements on 2D images were compared to those on the radial plane, considered as the reference, using the Bland and Altman method. Sixty-one (65.5%) participants (41.4±16.6 years) who did not have either radiographic OA (KL<2) or MRI-OA (Hunter 2011 criteria) were used to provide the normal range of meniscal extrusion. Among the 61 participants without radiographic or MRI OA, the mean extrusion of MM was 2.39mm±1.41 (zone II-a), 3mm±0.85 (zone II-b), 2.4mm±0.81 (zone III), and 0.3mm±0.9 (zone IV). The 95th percentile of the reference interval for meniscal extrusion in the radial plane was 4.7mm for zone II-a, 4.4mm (zone II-b), 3.7mm (zone III), and 2mm (zone IV). By applying a cutoff value of >3mm for significant meniscal extrusion on measurements in the radial plane, 28/61 (46%) would have been categorized as having extrusion in zone II-b and 12/61 (20%) in zone III. In the coronal plane, 13/61 (21%) or 18/61 (30%) would have been categorized as having meniscal extrusion depending on whether measurements were performed using methodLandmark, or methodMax respectively. In both sagittal and coronal planes, methodMax overestimated meniscal extrusion (95%CI for systematic bias=0.5, 1 in the sagittal plan, 0.1, 0.3 in the coronal plane), while there was no systematic bias for methodLandmark. The width of limits of agreement was similar in both planes and for both methods. Interobserver agreement was moderate for measurements in the radial (ICC=0.67) and sagittal (0.68) plane. This study provided the normal range of meniscal extrusion in asymptomatic knees without radiographic/MRI OA. Applying a 3mm cutoff value for the diagnosis of significant meniscal extrusion would carry a risk to categorize normal menisci as being extruded. Methods of measurement based on anatomical landmarks correlated better with measurements in the radial plane. This work was supported by the Swiss National Science Foundation, Switzerland (SNSF Grant #CRSII5_177155). None to declare. CORRESPONDENCE ADDRESS: [email protected]