Abstract
Background:
Previous research demonstrated that the attachment of the anterolateral ligament (ALL) to the lateral meniscus is stiffer and stronger in its tibial attachment than its femoral attachment. How this relates to anterolateral knee stability and lateral meniscal function is unknown.
Hypothesis/Purpose:
The hypothesis was that the ALL acts as a peripheral anchor to the lateral meniscus, aiding in anterolateral rotatory stability, and that the inframeniscal fibers of the ALL will provide greater anterolateral rotatory stability because of their greater tensile properties. The purpose was therefore to compare the difference in kinematics of the anterior cruciate ligament (ACL)–deficient knee between the infra- and suprameniscal ALL-sectioned states.
Study Design:
Controlled laboratory study.
Methods:
Eight paired fresh-frozen cadaveric knees were tested in a 5–degree of freedom loading jig under the following loading conditions: 5-N·m internal rotation at 15° incremental angles of flexion and combined 5-N·m internal rotation moment, 10-N·m valgus moment, and 88-N anterior translation force representing a pivot shift test at 0°, 15°, and 30° of flexion. The knees were tested under intact, ACL-deficient, and ACL-/ALL-deficient conditions, with the pairs of knees being randomized to either supra- or inframeniscal ALL sectioning. Resultant joint kinematics and tibiofemoral translations were measured and compared with a 2-way mixed repeated measures analysis of variance.
Results:
Internal rotation increased by 3° after sectioning of the ACL at 0° of knee flexion (P = .035). At 45° of knee flexion, internal rotation increased significantly by 2° between the ACL-deficient and the ACL-/ALL-deficient conditions (P = .049). Secondary kinematics of valgus and anterior translation were observed in response to the 5-N·m load after ACL and ALL sectioning. Analysis of the pivot shift showed increases in tibiofemoral translation after sectioning of the ACL, with further translations after sectioning of the ALL. No differences were observed between supra- and inframeniscal ALL sectioning under any of the loading conditions.
Conclusion:
An injury to the ALL, coexisting with ACL deficiency, results in only minor increases in knee joint patholaxity. No differences in pivot-shift kinematics or tibiofemoral rotations were observed between the supra- and inframeniscal sectioning of the ALL in the ACL-deficient knee
Clinical Relevance:
Tears of the midbody and/or posterior root attachment of the lateral meniscus are often observed at the time of ACL reconstruction. Increased anterolateral rotatory laxity has been observed in both lateral meniscus– and ALL-deficient states in combination with an ACL injury. While no significant functional relationship was found between the ALL and lateral meniscus, ALL sectioning did result in increased knee joint patholaxity, as demonstrated by composite tibiofemoral rotations.
The anterolateral complex (ALC) of the knee has recently received considerable attention regarding its role in providing anterolateral rotatory stability, particularly in the setting of an anterior cruciate ligament (ACL) injury.6,13,18 Of the structures forming the ALC, the anterolateral ligament (ALL) has generated the most controversy surrounding its role in knee joint stability. The importance of this structure and whether it should be reconstructed in conjunction with the ACL remains controversial.7,14,16,23
A number of dissection studies described the ALL as a distinct structure that spans the joint from a point proximal and posterior to the lateral epicondyle to a point midway between the fibular head and Gerdy tubercle on the lateral tibial plateau1,2,5,12,17; others described it as a thickening of the anterolateral joint capsule. 10 Regardless of the exact anatomic description, several studies showed that the ALL may have a role in providing secondary support to the iliotibial band (ITB) in controlling anterolateral rotation.18,22,23 However, some of these studies demonstrated only modest increases in internal rotation after sectioning of the ALL,22,23 as compared with much larger increases after sectioning of the ITB. 18
Helito et al 11 and Corbo et al 3 further evaluated the structure and function of the ALL, and they identified 2 distinct fiber bundles: suprameniscal fibers, which span the femoral insertion to the lateral meniscus, and inframeniscal fibers, which extend from the lateral meniscus onto the tibial insertion. Corbo et al also observed that, when isolated, the inframeniscal fibers were stronger and stiffer than the suprameniscal fibers, despite relatively similar histological properties. Furthermore, a recent biomechanical study highlighted the intricate relationship between the ALL and the lateral meniscus posterior root. In this sectioning study, these tissues demonstrated a synergistic effect in controlling rotatory laxity, which was dependent on the knee flexion angle. The meniscal root provided greater stability closer to extension (0°-30°), and the ALL provided greater stability at flexion angles >30°. 19
These recent studies highlight a complex interaction between the ALL and lateral meniscus in controlling anterolateral rotation. However, to date, there is a lack of objective biomechanical data on the stabilizing effect of the infra- and suprameniscal fibers of the ALL.
Therefore, the purpose of the study was to compare the difference in kinematics of the ACL-deficient knee between the infra- and suprameniscal ALL-sectioned (ALL–) states. Our hypothesis was that the ALL acts as a peripheral anchor to the lateral meniscus, aiding in anterolateral rotatory stability, and that the inframeniscal fibers of the ALL will provide greater anterolateral rotatory stability, owing to their greater tensile properties.
Methods
Specimen Preparation
Eight paired fresh-frozen cadaveric knee specimens (−20°, midfemur to midtibia; Science Care Inc) were procured for use in this study (n = 16, mean ± SD age = 60 ± 3.4 years, height = 1.76 ± 0.17 m, weight = 84.37 ± 15.66 kg). The study protocol was reviewed and approved in adherence with the tissue use and ethical guidelines of de-identified cadaveric tissue (approval MW 030217). The specimens, visually free of any bone or soft tissues disorders, were thawed at room temperature for approximately 18 hours, and were then prepared by stripping soft tissues from the proximal femur and the distal tibia to allow for potting; all soft tissues surrounding the knee joint were left intact. The tibia was potted into a 3-cm section of ABS tubing (9-cm diameter) secured with dental cement (Denstone Dental Cement; Hereaus Holdings GmbH), and the fibula was sectioned such that it was not potted and provided no load-bearing capacity. Once set, the specimen was inverted and maintained in extension while the proximal femur was potted into a section of ABS tubing (6-cm diameter). A 10-N compressive load was applied through the tibia during this process to ensure that the loads would be transmitted through the knee’s center of rotation when the specimen was placed into the robot and subsequently loaded.
Once the specimens were potted, a custom-designed 5–degree of freedom joint motion simulator was used to apply the physiologically relevant loading protocols (Blokker AM. “Development and Assessment of a Micro-CT Based System for Quantifying Loaded Knee Joint Kinematics and Tissue Mechanics.” Electronic Thesis and Dissertation Repository 5420; 2018). Briefly, the simulator consists of a flexion jig that accepts the potted femur and allows the knee flexion angle to be passively adjusted by rotating a crossbar superiorly while translating it inferiorly along guide rails (Figure 1). The entire jig can also translate horizontally to accommodate the knee’s flexion arc. The tibia is secured to the active component of the simulator where internal/external and varus/valgus moments can be applied, in addition to compression/tension and anterior/posterior forces. This is a closed-loop load control system where a series of motors (SM23165D SmartMotor; Moog Animatics) actuates the separate motions until the desired load targets are met, as measured from a 6–degree of freedom load cell (MC3A-6-250; Advanced Mechanical Technology Inc). These forces can be applied independently or as a combination of loads, and the simulator was shown to be accurate to within 1% of the target force. Two optical tracking marker clusters (Optotrak Certus; Northern Digital Inc) were inserted into the tibia and femoral diaphysis via rigid orthopaedic bone pins to track the motion of the tibia relative to the femur within an accuracy of 0.1 mm. A series of anatomic landmarks were subsequently digitized to allow for the creation of anatomic coordinate systems and the calculation of joint kinematics per the joint coordinate system method of Grood and Suntay. 9
(A) Active and (B) passive motion components of the custom-designed knee joint simulator that is capable of applying active internal and external rotation, varus and valgus rotation, anterior and posterior translation, and compression loads while maintaining passive flexion and extension angles. (C) Experimental setup of a cadaveric specimen at 0° of flexion within the simulator; also shown are the locations of the tibia and femur optical tracking marker sets (Optotrak Certus).
Experimental Conditions
All specimens were tested in the intact state (which included creation and repair of a 2-cm lateral parapatellar arthrotomy) and after transection of the ACL (ACL–). After this, matched-pair randomization was performed such that half of the specimens (n = 8) had the suprameniscal fibers of the ALL transected and the other half (n = 8), the inframeniscal fibers. To transect the ALL, a second 2-cm lateral arthrotomy was made through the skin and the ITB to provide access to the anterolateral capsule. Because of the difficulty in identifying the ALL without fully dissecting away the ITB, 4 the inframeniscal fibers were transected starting just anterior to the lateral collateral ligament to a position adjacent to the posterior margin of the Gerdy tubercle; by definition, this would result in cutting the inframeniscal portion of the ALL.1,17 The suprameniscal fibers were divided in a similar manner, with an incision starting anterior to the lateral collateral ligament and continuing to the Gerdy tubercle on the superior surface of the meniscus. The arthrotomy through the ITB and the skin were sutured closed before testing. Testing was performed for each state at knee flexion angles from 0° to 90° in 15° increments.
Loading Protocol
For each specimen, an initial passive trial was performed at each knee flexion angle, where the loads in all axes were maintained at zero. This trial was used as the baseline position for all kinematic measurements. For all combinations of knee angle and condition, a 5-N·m internal rotation moment was applied with a 10-N compressive load. The loads in the remaining axes were maintained at zero, and the joint simulator was free to move to accommodate these zero-load conditions.
A pivot shift was also simulated at 0°, 15°, and 30° where a combined loading protocol was utilized, which included a simultaneous application of a 5-N·m internal rotation moment, a 10-N·m valgus moment, and an 88-N anterior translation force. 22 Each load pattern was applied 3 times for all knee angle–condition combinations, and the mean of the 3 cycles was used for subsequent analysis.
Data Analysis and Statistics
The degree of internal rotation at the 5-N·m internal rotation moment was extracted from the moment-angle curve as the primary measure of stability. In addition, the magnitude of varus/valgus and anterior translation that occurred in response to the 5-N·m internal rotation moment was calculated and analyzed. The magnitude of internal rotation, valgus rotation, and anterior translation that resulted from the simulated pivot shift was also determined. Furthermore, given the combined loading associated with the pivot-shift test, a metric was developed that would capture the coupling associated with the simultaneous response of internal rotation, valgus rotation, and anterior translation (ie, anterolateral tibiofemoral translations). The surface area of a quadrilateral formed by the digitized points of the medial and lateral femoral epicondyles and the widest medial and lateral points of the tibial plateau was calculated. This is similar to a measurement reported by Noyes et al, 21 who calculated the internal rotation and anterior translation of the tibia by comparing the vectors created by the most medial and lateral points of the tibial plateau before and after a simulated pivot-shift test. In the current investigation, this method was adapted to account for the 3-dimensional nature of the motions that occur in response to the pivot shift. The surface area at the endpoints of the simulated pivot shift was extracted and expressed as the change from the passive zero-load condition at each flexion angle. Therefore, a greater change in the surface area would indicate an increase in the total motion of the knee, suggesting greater knee laxity. These data were further reduced and analyzed as the difference from the intact condition and are presented as a percentage of the intact trials (Figure 2).
The region used to calculate the surface area changes (red) between the (A, C, E) unloaded and (B, D, F) pivot-shift conditions after the sectioning of the intrameniscal anterolateral ligament. (C, D) The quadrilateral in the frontal plane; (E, F) an axial view. (C-F) The femur has been removed for clarity.
A 2-way mixed repeated measures analysis of variance (3 sectioning conditions × 2 ALL sections) was used to determine the statistical effect of these independent variables on the kinematic and surface area variables. The sectioning condition was the within-specimen variable, and the sectioning of the infra- or suprameniscal ALL fibers was the between-specimen variable. These analyses of variance were performed separately at each knee angle with SPSS statistical software (v 23; IBM). Post hoc analysis was performed with a Bonferroni correction, and alpha was set at 0.05 for all statistical tests.
Results
Internal Rotation
At 0° of flexion, the mean internal rotation increased significantly when the ACL was sectioned (13.7° ± 4.5°) as compared with the intact condition (10.8° ± 5.3°; P = .035) (Figure 3A). At 45° of knee flexion, internal rotation increased significantly from 19.8° ± 9.8° to 22.0° ± 9.8° between the ACL– and the ACL–/ALL– conditions, respectively (P = .049) (Figure 3D). When the knee was flexed to 60°, the mean internal rotation during the intact condition increased significantly from 16.0° ± 8.4° to 18.0° ± 8.5° during the ACL–/ALL– condition (P = .041) (Figure 3E). There was also a significant increase (P = .019) between the ACL– (14.8° ± 9.1°) and the ACL–/ALL– (17.98° ± 8.52°) conditions. Similarly, at 75° the mean internal rotation was significantly greater when the ALL was sectioned (ACL–/ALL–; 26.7° ± 8.2°) versus the intact (22.3° ± 10.2°; P = .030) and ACL– (24.1° ± 8.8°; P = .018) conditions (Figure 3F). This indicates that sectioning the ACL alone at these 2 flexion angles had a limited effect on internal rotation stability. There were no significant main effects of tissue sectioning at 15° (P = .09), 30° (P = .44), or 90° (P = .08) of knee flexion (Figure 3, B, C, G). There were also no significant differences in internal rotation between the specimens that had the inframeniscal fibers sectioned and those that had the suprameniscal fibers sectioned, and there were no significant interaction effects at any knee angle.
Comparison of the mean ± SD internal rotation across the different sectioning conditions and between the infra- and suprameniscal fiber sectioning. *A significant main effect of tissue sectioning at P < .05. Each graph represents the results from each knee angle tested. ACL–, anterior cruciate ligament transection; ALL–, anterolateral ligament transection.
Varus/Valgus in Response to Internal Rotation Moment
At 0° and 15° of knee flexion, there was increased valgus rotation in response to the 5-N·m torque between the intact and ACL– conditions (P = .013 and P = .011, respectively) and between the intact and ACL–/ALL– conditions (P = .01 and P = .005, respectively) (Table 1). Additionally, there were significant tissue sectioning × ALL fiber interactions at 45° and 90°. At 45°, a significant increase in valgus rotation (P = .001) occurred from the intact to ACL–/ALL– conditions among the specimens that had the suprameniscal fibers sectioned. At 90°, the ACL–/ALL– condition also demonstrated significantly greater valgus rotation as compared with the intact condition (P = .022) and the ACL– condition (P = .001) when the suprameniscal fibers were sectioned.
Valgus Rotation and Anterior Translation Across Conditions and Between the Infra- and Suprameniscal Fiber Conditions in Response to the 5-N·m Internal Rotation Load a
Values are presented as mean ± SD. ACL–, anterior cruciate ligament transection; ALL–, anterolateral ligament transection.
P < .05 vs the intact arthrotomy condition.
Represents the presence of a main effect that could not be identified by post hoc testing.
P < .05 vs the ACL– condition.
Anterior Translation in Response to Internal Rotation Moment
When the knee was flexed to 15°, the ACL– and ACL–/ALL– conditions resulted in statistically significant increases in anterior translation—1.6 mm (P = .004) and 1.3 mm (P = .042), respectively—as compared with the intact condition (Table 1). Finally, at 75°, anterior translation increased significantly from 1.9 ± 2.3 mm for the intact condition to 3.3 ± 2.8 mm for the ACL–/ALL– condition (P = .005). There were no other significant main effects at any other knee flexion angle, and no interaction effects were revealed.
Pivot-Shift Kinematics
At 0° of knee flexion, the anterior translation increased significantly between the intact and ACL– conditions (P = .013) and between the intact and ACL–/ALL– conditions (P = .031) (Figure 4B). Internal rotation was not significantly affected by tissue sectioning (P = .481).
Comparison of the mean ± SD (A) internal rotation, (B) anterior translation, and (C) valgus rotation that occurred in response to the simulated pivot shift at 0° of knee flexion. *A significant main effect of tissue sectioning at P < .05. ACL–, anterior cruciate ligament transection; ALL–, anterolateral ligament transection.
At 15°, the anterior translation increased by 4.07 mm between the intact and ACL– conditions (P = .001) and 4.55 mm between the intact and ACL–/ALL– conditions (P = .002) (Figure 5B). Valgus rotation also increased significantly at 15° of flexion from 4.4° ± 2.2° during the intact condition to 7.1° ± 3.6° after sectioning of the ACL (ACL–; P = .04) (Figure 5C).
Comparison of the mean ± SD (A) internal rotation, (B) anterior translation, and (C) valgus rotation that occurred in response to the simulated pivot shift at 15° of knee flexion. *A significant main effect of tissue sectioning at P < .05. ACL–, anterior cruciate ligament transection; ALL–, anterolateral ligament transection.
Finally, at 30° there were 4.5- and 5.0-mm increases in anterior translation (P = .001; effect size = 0.376; power = 0.971) after sectioning of the ACL (ACL–; P = .006) and ALL (ACL–/ALL–; P = .003), respectively, versus the intact condition (Figure 6B). Across all knee flexion angles for the simulated pivot shift, there was no significant main effect regarding which fibers of the ALL were sectioned, and there were no significant interactions.
Comparison of the mean ± SD (A) internal rotation, (B) anterior translation, and (C) valgus rotation that occurred in response to the simulated pivot shift at 30° of knee flexion. *A significant main effect of tissue sectioning at P < .05. ACL–, anterior cruciate ligament transection; ALL–, anterolateral ligament transection.
Pivot-Shift Surface Area
At 0°, 15°, and 30° of knee flexion, there were significant differences between the intact and ACL– conditions (0°, P = .008; 15°, P = .001; 30°, P = .013) and the intact and ACL–/ALL– conditions (0°, P = .011; 15°, P = .002; 30°, P = .034) (Figure 7); the change in surface area was greater when the ACL and ALL (ACL–/ALL–) were sectioned. There was no main effect related to which fibers of the ALL were sectioned and no significant interaction effects.
Comparison of the mean ± SD change in the surface area in response to the simulated pivot shift at (A) 0°, (B) 15°, and (C) 30° of knee flexion. *A significant main effect of tissue sectioning at P < .05. Data are presented as a percentage of the intact condition. ACL–, anterior cruciate ligament transection; ALL–, anterolateral ligament transection.
Discussion
To our knowledge, this is the first study to investigate the kinematic effect of a combined injury to the supra- and inframeniscal fibers of the ALL with a concomitant ACL injury. The most important finding of this study was that no significant differences were identified between the suprameniscal ALL injury and the inframeniscal ALL injury in an ACL-deficient knee.
The purpose of the study was therefore to compare the difference in kinematics of the ACL-deficient knee between the infra- and suprameniscal ALL-sectioned states. The assertion is that, with the lateral meniscus, the inframeniscal fibers of the ALL would provide greater rotational control to the knee. 3 Lateral meniscus radial tears are often observed with ACL injury, located at the ALL insertion on the midbody of the lateral meniscus. 20 This suggests that the lateral meniscus is tethered at this position and may succumb to shear stresses during a significant anterolateral subluxation episode. The kinematic results presented in this study do not support our hypothesis, with similar changes in rotational stability regardless of the location of ALL injury. Therefore, the clinical significance of the different locations of ALL injury is most likely negligible.
The data presented in this study also evaluated the clinical significance of an ALL injury combined with an ACL injury. To better investigate the role of the ALL, a model was developed to measure tibiofemoral rotations as a result of soft tissue injury. As reported in other studies, tibiofemoral kinematics consist of coupled anterior translation, valgus, and internal rotations in response to pivot-shift testing. 15 The described technique of using a 3-dimensional surface area is a novel method for describing this joint motion coupling and may be a promising measure to better describe the complex rotational and translation changes that occur between the femur and tibia. The present study showed that changes in the surface area—calculated from the trapezoidal region between medial and lateral femoral condyles and the medial and lateral tibial plateau—clearly describe the tibiofemoral translations. However, further work is required to determine the relationship of this novel measure with clinically relevant assessments. This is similar to the results presented by Noyes et al 21 ; however, it also considers the contribution of the motions that occur in the other planes.
Noyes et al 21 also questioned the significance of the ALL in controlling anterolateral rotation, concluding that it acted as secondary stabilizer to the ACL, only after the ACL was cut. Thein et al 24 made similar conclusions, finding the ALL to engage only after tibiofemoral subluxations that were outside the pathophysiologic range of the ACL. Furthermore, Kittl et al 18 clearly showed that the contribution to internal rotation control of the ALL was small (<15%) in comparison with the superficial and deep components of the ITB. Specifically, they found that the capsulo-osseous layer had a more significant contribution to internal rotation control through the knee flexion arc. Geeslin et al 6 most recently compared knee kinematics after sectioning the ALL or the distal Kaplan fibers of the ITB. Similar to Kittl et al, they found that the ITB and its Kaplan fiber attachment on the distal femur had a greater role in controlling internal rotation and anterior translation. These results are relatively consistent with other studies that indicated an approximately 2° increase in internal rotation when the ALL is sectioned in an ACL-deficient knee; the clinical significance and application of these small changes remain unclear.6,15,17,19,22,23
The current research agrees with the fact that while the ALL does have a role in controlling rotation after ACL injury, the magnitude of its clinical effect is questionable and, as reported previously, likely acts as a component of the ALC. A recent consensus group meeting concluded that the ALC functions synergistically with a number of structures (eg, the ITB and ALL) that contribute to providing control of tibiofemoral rotation, namely the pivot shift. 8 It is still undetermined whether an ALL reconstruction or lateral tenodesis should be used in primary ACL reconstruction and, if so, for what specific indications.
On the basis of the results of this work, the concept of an additional point of fixation of the lateral meniscus at the ALL insertion, similar to that of posterior root repair, cannot be supported. However, a number of studies clearly showed an attachment point of the ALL to the lateral meniscus1,3,11,17; therefore, the question of its functional role remains unanswered. As such, further research is warranted to determine the interaction between the ALL and the lateral meniscus. While it may not play a significant role in stabilizing the lateral compartment in the ACL-deficient knee, it may add to the stability of the lateral meniscus itself and therefore add to its chondroprotective role in meniscal repair or meniscus transplantation techniques.
The strengths of this study include the use of paired cadaveric specimens with no evidence of joint injury or degeneration, as is sometimes encountered in cadaveric work. The loading conditions were applied with a validated 5–degree of freedom robot that provides much less variability than previously used testing rigs.19,23 Last, the addition of combined loads to test the pivot shift, with the subsequent development of a surface area metric to quantitate tibiofemoral rotation, is novel and may prove to be of significant use in future studies. Limitations of the study include the relatively small sample size. However, the numbers of specimens used is consistent with the literature and is representative of the limitations placed by time and research funding for this type of work. The subsequent post hoc sample size calculation, which indicated that a much larger sample would be required to achieve statistical significance, is also indicative of a comparison that is unlikely to be clinically significant.
Conclusion
An injury to the ALL, coexisting with ACL deficiency, results in only minor increases in knee joint patholaxity. No differences in pivot-shift kinematics or tibiofemoral rotations were observed between the supra- and inframeniscal sectioning of the ALL in the ACL-deficient knee. While there is a clearly described attachment of the ALL to the lateral meniscus, its functional significance is yet to be determined and warrants further research.
Footnotes
One or more of the authors has declared the following potential conflict of interest or source of funding: Funding was received from the Western University Bone and Joint Institute (Catalyst Grant) and Smith & Nephew. T.A.B. received salary support from Smith & Nephew. A.G. receives consulting fees from Smith & Nephew, Ossur Inc, Joint Restoration Foundation, and Collagen Solutions. AOSSM checks author disclosures against the Open Payments Database (OPD). AOSSM has not conducted an independent investigation on the OPD and disclaims any liability or responsibility relating thereto.