Background: Anterior cruciate ligament (ACL) reconstruction tunnel placement is often evaluated by radiographs. This study examines the interobserver reliability of various radiographic measurements of ACL tunnels.
Hypothesis: When ideal radiographic views are obtained, the interobserver reliability of the measurements among experienced surgeons would be good to excellent.
Study design: Descriptive laboratory study.
Methods: Tunnels for single-bundle ACL reconstruction were drilled and filled with metal interference screws or a tibial reamer on 73 cadaveric knees. Ideal fluoroscopic radiographs were obtained. Three independent reviewers performed 18 measurements including a modification of the grid method. For the grid method analysis, reviewers fit a 16 × 12 grid to the lateral knee radiograph, and the center of the femoral tunnel was marked. Interobserver reliability of the measurements was performed using intraclass correlation coefficients (ICCs). A precision grouping analysis was performed for the grid measurements to calculate the mean radius and standard deviation grouping distances.
Results: The ICCs were excellent (>.75) for the tibial tunnel angles and tunnel measurements, the clock face measurement, and the Aglietti et al and Jonsson et al measurements. The ICCs were good (.4-.75) for an estimation of graft impingement, Harner et al measurements, and notch height. The mean radius for grid measurements was 0.6 ± 0.4 units (range, 0-2.36 units), with each unit being 1 box in the 16 × 12 grid. When a circle was constructed with a 1.3-unit radius, 95% of the 3 surgeons' measurements would be included in the area of that circle.
Conclusion: Reliability of ACL tunnel measurements was good to excellent under ideal circumstances for the majority of measurements. The modified grid method demonstrated very acceptable reliability.
Clinical relevance: Measurements with good to excellent reliability can be used to evaluate ACL tunnel placement when ideal radiographic views are obtained.