Abstract
Introduction
There is a challenge to detect partial tear of the ACL, the number of bundle injured and the proportion of fibers torn. The MRI was shown efficient to individualize the two anteromedial (AM) and posterolateral (PL) bundles of the ACL. The purpose of this study was to assess the ability of the MRI to detect partial tears of the ACL on axial views to display the AM and PL bundles.
Materials and methods
This retrospective study included 48 patients (19 partial tears of the ACL, 16 complete rupture of the ACL and 13 normal knee) who underwent both arthroscopy and MRI examinations of the knee. The conventional MRI protocol included one sagittal T1- weighted sequence and 3 proton-density fat sat. The images from MRI were analysis by a radiologist specialized in musculoskeletal imaging who was blinding to the arthroscopic findings. The criteria for the analysis of MRI were divided into primary (those involving the ACL himself) and secondary signs (associated abnormalities). The primary signs included the horizontalisation of the ACL (ACL axis), the global ACL signal intensity and the signal intensity of each AM and PL bundle. The secondary signs included bone bruise, osteochondral impaction, popliteus muscle injury, medial collateral ligament injury and joint effusion. The ACL was classified as normal, partially or totally torn. The rupture of the AM and PL bundle was specified.
Results
In our study, MRI was found to have a 75% sensitivity and a 73% specificity for the diagnosis of partial tears of the ACL. The sensitivity to detect AM bundle lesion was 88% but the specificity was 50%. The lack of horizontalisation of the ACL was a very good sign, for partial tears of the ACL, with a sensitivity of 84% and a specificity of 81% (p<0.05). Regarding the secondary signs, there was no significant difference to distinguish partial and complete tear. However, we found that there was a greater probability to find a partial tear of the ACL with the lack of joint effusion, bone bruise and medial collateral ligament injury. Quantifying the proportion of injured fibers was unsuccessful and was found as a failure of the MRI.
Conclusions
MRI exhibited lower sensitivity and specificity for partial tear than for complete rupture of the ACL. However 3 important arguments can guide us: the lack of horizontalisation of the ACL, a continuous ACL signal, the display of one of two bundles on the axial view.
