Bone marrow oedema (BME) is a common finding following severe acute knee trauma, like those responsible for anterior cruciate ligament (ACL) tears. Magnetic resonance imaging (MRI) is the method of choice to study bone bruise and different evaluation scores have been developed to detect and quantify BME extent. Nevertheless, the most reliable method to detect and quantify bone bruise is still unknown. The purpose of this study was to identify the most reliable method to evaluate bone bruise in knees with ACL injury.
One-hundred knee MRIs of patients with an ACL lesion were retrieved from the institutional Picture Archiving and Communication Systems (PACS) database between 2010 and 2018 and selected for the purpose of this study. Bone bruise evaluation in each knee MRI was assessed by applying the most clinically used semi-quantitative scores, WORMS and MOAKS, and by area measurement as quantitative evaluation by different evaluators: two musculoskeletal radiologists, one senior and one young orthopedics. The obtained values were analyzed to obtain the inter-rater and intra-rater reliability assessment.
WORMS and MOAKS showed a moderate agreement among the all operators with a Fleiss-K value of 0.459 and 0.45 respectively, while area measurement resulted in a poor agreement resulting in ICC ≤ 0.130. Higher agreement rates have been obtained for intra-test assessment performed by the same medical researcher. Specifically, WORMS and MOAKS resulted in almost perfect agreement while the area measurement showed a good agreement.
WORMS and MOAKS represent suitable tools to assess BME in patients affected by ACL injury, presenting good results in term of reliability. On the other hand, area measurement showed worse results in term of reliability compared to the semi-quantitative parameters. Further high-quality studies are needed to confirm these results and the best approach for the assessment of bone bruise in knees affected by ACL lesion.