Despite contemporary surgical treatments of posterior meniscus root tears, there is a low rate of healing and an incidence of residual meniscus extrusion approaching 30%. Here, we characterized and compared the behavior of the medial and lateral meniscus in response to axial compression load and real-time dynamic motion using a cadaveric model. There were no differences in the amount of meniscus extrusion between the medial and lateral meniscus with a competent posterior root (0.338mm vs. 0.235mm; p-value = 0.181). However, posterior root detachment resulted in a consistently increased amount of meniscus extrusion for the medial meniscus compared to the lateral meniscus (2.233mm vs. 0.4705mm; p-value < 0.0001). Moreover, the medial and lateral menisci responded differently to the effects of knee flexion angle. Detachment of the posterior root of the medial meniscus resulted in an increase in extrusion at all angles of knee flexion, but was most pronounced (4.00mm ± 1.26mm) at 30-degrees of knee flexion. In contrast, the lateral meniscus only extruded (1.65mm ± 0.97mm) in full extension. Furthermore, only the medial meniscus extruded during dynamic flexion after posterior root detachment. These findings suggest that while meniscus extrusion occurs in both the medial and lateral meniscus, the functional consequences of extrusion are more significant for the medial meniscus than that of the lateral meniscus. Limiting the amount of meniscus extrusion after posterior root tear in response to static compressive forces and dynamic motion is necessary to recapitulate native meniscus function.

Equipment and set-up

Custom fabricated grips were attached to an MTS 858 Bionics servo-hydraulic test system. The lower grip was fabricated using an axle system to release the knee rotational (medial/lateral) degree of freedom. The axle system was then mounted on two crossed roller slides (Deltron NBT-6160A) in an X-Y arrangement to allow degree of freedom release of the medial/lateral and anterior/posterior planes. Data acquisition included the specimen load, actuator displacement, and meniscus displacement, using a Lord Microstrain model M-DVRT-6 transducer (M-DVRT-6, 6.0mm stroke micro-miniature DVRT linear displacement Transducer, and DEMOD-DC, Miniature DC input, DC output signal conditioner, LORD Corporation, 459 Hurricane Lane, Suite 102, Williston, Vermont 05495). Data were acquired using a Data Translation 16-bit digital to analog converter. The force generated during each testing protocol was recorded in real time. The upper grip was fabricated to allow positioning of the knee in the range of zero to one hundred degrees of flexion for mechanical testing during dynamic range of motion.

Testing protocols

Protocols simulated clinical knee activities during weight-bearing loading and passive non-weight-bearing dynamic range of motion. The amount of load delivered across the knee simulated that of a 75-kg person. Specimens were tested in the following sequence: posterior root intact and posterior root detached. Axial compression load: Testing was performed in load control in the axial direction. A ramp to a constant load to simulate a load from 25-N preload to full weight-bearing (1800-N) at 0.1-second ramp to set load, with manual return to preload performed in full knee extension (0-degrees of knee flexion). A sinusoidal cyclic load to simulate a walking gait at 20-cycle compressive ½ cycle sine wave, 25-N to set load with 1-sec cycle (0.3-second down, 0.3-second up, 0.4-second pause to stimulate swing phase of gait). This test was performed across 0-, 30-, 45-, 60- and 90-degrees of knee flexion. Dynamic non-weight-bearing range of motion: Each knee was taken through a range of motion from 0- to 100-degrees of flexion. Meniscus extrusion was linked to range of motion data detected with a calibrated custom electronic goniometer in real-time.

Statistical analysis

Means and standard deviations or standard error of the means were calculated. Between group comparisons were performed using a Student’s t-test for two groups and analysis of variance (ANOVA) for more than two groups. Pairwise comparisons and post-hoc adjustments were performed if the ANOVA model was statistically significant. Linear regression was used to model the relationship between knee flexion and meniscus extrusion. A p-value < 0.05 was considered statistically significant. Calculations were performed using GraphPad Prism 9.0.1 for Mac (San Diego, CA, USA).