Timing of magnetic resonance imaging affects the accuracy and interobserver agreement of anterolateral ligament tears detection in anterior cruciate ligament deficient knees.

PURPOSE We aimed to identify the anterolateral ligament (ALL) tears in anterior cruciate ligament (ACL)-deficient knees using standard 1.5-Tesla magnetic resonance imaging (MRI). METHODS We included all patients who underwent primary ACL reconstruction at our center between 2012 and 2015. Exclusion criteria included patients with multiple ligament injuries, lateral collateral ligament, posterolateral corner, and infections, and patients who underwent MRI more than 2 months after their injury. All patients (n = 148) had ACL tears that were subsequently arthroscopically reconstructed. The magnetic resonance (MR) images of the injured knees performed within 2 months of injury were reviewed by a musculoskeletal radiologist and an orthopedic surgeon. The patients were divided into two groups. The first group of patients had MRI performed within 1 month of injury. The second group of patients had MRI performed 1-2 months after the index injury. Both assessors were blinded and the MR mages were read separately to assess the presence of ALL, presence of a tear and the location of the tear. Based on their readings, interobserver agreement (kappa statistic (K)), sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and accuracy were compared. RESULTS The ALL was identified in 100% of the patients. However, there was a discrepancy of up to 15% in the identification of tear of the ALL. In the first group in which MRI scans were performed within 1 month of injury, the ALL tear was identified by the radiologist in 92% of patients and by the surgeon in 90% of patients (Κ = 0.86). In the second group in which MRI scans were performed within 1-2 months of the injury, the ALL tear was identified by the radiologist in 78% of patients and by the surgeon in 93% of patients (K = 0.62). CONCLUSION The ALL can be accurately identified on MRI, but the presence and location of ALL tear and its location cannot be reliably identified on MRI. The accuracy in identification and characterization of a tear was affected by the interval between the time of injury and the time when the MRI was performed. LEVEL OF EVIDENCE Diagnostic, level IIIb, retrospective.