Patterns of Radiological Misinterpretation of Complete Anterior Cruciate Ligament Tears on Routine MRI Reports: An Arthroscopic Validation Study

Background: Magnetic resonance imaging (MRI) is the primary noninvasive technique in the diagnosis of anterior cruciate ligament (ACL) injuries, and its performance is well described under ideal circumstances; however, the diagnostic accuracy in routine clinical practice is inadequately described. Objective: To determine the diagnostic accuracy of routine radiological MRI reports in the diagnosis of complete ACL tears when arthroscopy is used as the reference standard. Methods: This was a retrospective study, which involved 100 consecutive patients (28.8±7.8 years, and all males) who underwent knee arthroscopy with an intraoperative finding showing complete ACL tears. Preoperative MRI reports obtained from on-duty radiologists were reviewed and classified as complete, partial tear, or normal ACL. We calculated sensitivity for diagnosing complete tears in the ACL, and an exploratory analysis was done to examine the relationship between the MRI-arthroscopy interval and the accuracy of the diagnosis. Results: Routine MRI interpretation correctly detected complete ACL rupture in 61 cases with a sensitivity of 61.0% (95% CI: 51.2–70.0). It is significant to note that 32% of complete tears were underestimated as partial injuries, and 7% were misinterpreted as normal. The MRI-arthroscopy interval significantly influenced diagnostic outcomes (p=0.038); however, subgroup analyses were underpowered and remain hypothesis-generating. Conclusions: In routine clinical use, MRI interpretation achieved a suboptimal sensitivity (61%) in identifying arthroscopically proved complete ACL tears, and a significant proportion of cases were underestimated. To improve diagnostic accuracy, it is necessary to combine clinical examinations with imaging and employ radiologist-focused quality improvement programs.