Abstract
Objectives:
The anterolateral ligament (ALL) is an important structure for knee stability and has roles in maintaining internal knee rotation and anterior-posterior stability in anterior cruciate ligament (ACL)- insufficient knees. There is a growing body of evidence demonstrating superior clinical outcomes including reduced graft failure rate and residual pivot shift with combined ACL and ALL reconstruction compared to isolated ACL reconstruction. One common complication of this procedure is convergence of the two tunnels used for ACL and ALL graft passage. Several studies have aimed to determine the optimal angles to drill these tunnels that minimizes collision risk, but most of the literature is limited by few angle combinations or manipulation of the ALL tunnel exclusively. One study examined how changing the orientation of the femoral ACL tunnel affected collision risk, but they limited their investigation to the ACL tunnel angle in relation to a line perpendicular to the femoral axis. We aim to investigate the relation between the ACL tunnel angle in coronal and axial planes and risk of collision with the ALL tunnel as measured by the minimum distance between tunnels (MDBT). We hypothesize that there is a correlation between the minimum distance between ALL and ACL tunnels and the orientation of the ACL tunnel in both coronal and axial planes.
Methods:
This study was determined to be non-human subjects research by the Institutional Review Board. A 3D-CT reconstruction of a knee determined to be anatomically normal by a radiologist was de- identified and exported into the 3D modeling software Autodesk Fusion 360. The model was prepared for analysis based on previous 3D-CT reconstruction studies of the knee. The central longitudinal axis of the femur was set as the long axis, and the femur model was rotated so that the trans-epicondylar line was oriented with the x-axis. The model was then rotated so that the long axis was parallel to the z-axis in the sagittal plane. Based on the results of multiple prior studies evaluating the optimal ALL tunnel orientation, a “safe” ALL tunnel angle of 40° anterior in the axial plane and 10° proximal in the coronal plane was used for all ALL-ACL combinations. The ALL attachment site was designated as 8mm proximal and 4.3mm posterior to the lateral epicondyle. The ALL tunnel was modeled as a cylinder 30mm long and 6mm in diameter with the cylinder center placed at the attachment site. The ACL tunnel was modeled as a cylinder 8mm in diameter with a 4mm diameter cylinder at one end to represent an EndoButton tunnel. The cylinders were then rotated in 15° increments up to 60° in both coronal and axial planes until 25 unique ACL tunnel orientations were created. (Figure 2) For each orientation, the center of the large cylinder face was placed at the anatomic ACL footprint, and the cylinder was extended until the Endobutton tunnel penetrated the lateral cortex. The MDBT was then measured, and a 2mm clearance between tunnels was designated as satisfactory to avoid tunnel convergence. A collision was recorded if the cylinders overlapped in the 3D model. Linear regression was used to determine the relation between the dependent variable of MDBT and the independent variables of axial and coronal angles. Significance was set to p < 0.05.
Results:
A collision occurred in 36% of orientations, with unsatisfactory clearance between tunnels in 20% of orientations and satisfactory clearance in the remaining 44%. The axial angle producing the largest average MDBT was 0°, with an average MDBT of 9.47mm and zero collisions across all coronal angles. The coronal angle producing the largest average MDBT was also 0°, with an average MDBT of 3.93mm across all axial angles. An axial angle of 60° produced the lowest average MDBT across all coronal angles at 1.26mm and three collisions. A coronal angle of 15° produced the lowest average MDBT across all axial angles at 1.26mm and three collisions. The linear regression analysis had an R2 of 0.424. There was a negative association between ACL tunnel axial angle and MDBT indicating that as the ACL tunnel axial angle increases, MDBT decreases (β = –0.654 ,p < 0.001). There was no evidence of a relation between ACL tunnel coronal angle and MDBT (β =0.209, p=0.19).
Conclusions:
Despite using an ALL tunnel determined in previous studies to be optimal for avoiding collisions, tunnel collision or dangerously close tunnel proximity occurred in over half of the ACL orientations tested. These findings help to reinforce that the risk of tunnel collision in combined ACL- ALL reconstruction is high if not properly accounted for. They also raise awareness that even when using an ALL tunnel deemed optimal to avoid collisions, there remains a large proportion of ACL tunnel orientations that create a high risk for tunnel convergence. The linear regression model demonstrates that with coronal orientation held constant, a more anteriorly angled tunnel in the axial plane decreases the MDBT and increases risk for tunnel collision. Further work should involve performing similar methodology on a larger scale with more patients, in order to determine an optimal ACL tunnel angle that minimizes collision risk across a population. These findings also suggest a role for 3D-CT modeling in preoperative planning at the individual level, given the high risk of tunnel collision and anatomic differences between patients.
