The Determination of Interobserver and Intraobserver Reliability of a Magnetic Resonance Imaging Based Classification System for Ulnar Collateral Ligament Injury
Ramkumar PN, Frangiamore SJ, Navarro SM, Lynch TS, Kaar SG, Akhavan S, Moutzouros V, Westermann RW, Farrow LD, and Schickendantz MS. The Determination of Interobserver and Intraobserver Reliability of a Magnetic Resonance Imaging Based Classification System for Ulnar Collateral Ligament Injury. J Shoulder Elbow Surg 2019; 28(8):e287-e288.
J Shoulder Elbow Surg
Background: Despite improvements in the biomechanics and surgical options for UCL tears, there remains a need for a reliable classification of UCL tears that has the potential to guide clinical decision-making. Purpose: The purpose of this cross-sectional study was to assess the intraobserver and interobserver reliability of the newly proposed MRI-based classification to UCL tears. Secondary objectives included assessing the impact of additional views, discrimination between distal and non-distal tears, and correlation of imaging reads with intraoperative findings of the UCL. Methods: Nine fellowship-trained specialists from seven institutions independently completed four series surveys consisting of 60 total elbow MRIs with UCL tears using a newly proposed six-stage classification system. The first and third surveys contained a total of 60 coronal MRI images, while the second and fourth contained the same MRI images with both coronal and axial views presented in a random order to assess intraobserver variability using the weighted kappa value and impact of additional imaging views. Weighted kappa values were also calculated for each of the four surveys to acquire interobserver reliability. Reliability analysis was repeated using a two-group classification analysis for distal and non-distal disease. Observer readings were compared to intraoperative UCL findings. Results: For the newly proposed six-stage MRI-based classification, intraobserver and interobserver reliability demonstrated near perfect and substantial agreement, respectively. These values only increased when sub-stratified into the two-group distal and non-distal disease classification (p<0.05). The additional axial view did not statistically improve the agreement between and among readers. Observer readings were accurate for tear grade (partial and complete), proximal location, and distal location, but not midsubstance tears, when compared to intraoperative findings from 30 elbows. Conclusion: Our newly proposed six-stage MRI-based classification utilizing grade and location of the injury was found to have substantial to near perfect agreement between and within fellowship-trained observers. The results of this study provide a foundation for future validation studies, in which the classification system may be associated with clinical decision-making and patient outcomes.