Dynamic Restraints of the Medial Side of the Knee: The Semimembranosus Corner Revisited

Little is known about the dynamic restraints of the semimembranosus muscle (SM). The goal of the present study was to elucidate the role of (1) passive and (2) active restraints to medial-side instability and to analyze (3) the corresponding tightening of the posteromedial structures by loading the SM. It was hypothesized that points 1 to 3 will significantly restrain medial knee instability. This will aid in understanding the synergistic effect of the semimembranosus corner. Controlled laboratory study. Nine knees were tested in a 6 degrees of freedom robotic setup and an optical tracking system. External rotation (ER; 4 N·m), internal rotation (4 N·m), anteromedial rotation (4-N·m ER and 89-N anterior tibial translation), and valgus rotation (8 N·m) were applied at 0°, 30°, 60°, and 90°, with and without an SM load of 75 N. Sequential cutting of the medial collateral ligament and posterior oblique ligament was then performed. At the intact state of the knee and after each cut, the aforementioned simulated laxity tests were performed. The medial collateral ligament was found to be the main passive stabilizer to ER and anteromedial rotation, resulting in 9.3° ± 6.8° (P < .05), 8.1° ± 3.6° (P < .05), and 7.6° ± 4.2° (P < .05) at 30°, 60°, and 90°, respectively. Conversely, after the posterior oblique ligament was cut, internal rotation instability increased significantly at early flexion angles (9.3° ± 3.2° at 0° and 5.2° ± 1.1 at 30°). Loading the SM had an overall effect on restraining ER (P < .001) and anteromedial rotation (P < .001). This increased with flexion angle and sectioning of the medial structures and resulted in a pooled 2.8° ± 1.7° (not significant), 5.4° ± 2° (P < .01), 7.5° ± 2.8° (P < .001), and 8.3° ± 4.4° (P < .001) at 0°, 30°, 60°, and 90° when compared with the unloaded state. The SM was found to be a main active restraint to ER and anteromedial rotation, especially at higher flexion angles and in absence of the main passive medial restraints. The calculated tensioning effect was small in all flexion angles for all simulated laxity tests. A complete semimembranosus avulsion may indicate severe medial knee injury, and refixation should be considered in multiligament injury