Reliability of ulnar collateral ligament findings in magnetic resonance imaging.

Abstract

Ulnar collateral ligament (UCL) tears are common in baseball players. Although magnetic resonance imaging (MRI) is the most used imaging modality for UCL injuries, the inter-rater reliability of MRI analyzing specific characteristics of UCL pathology has not been defined. The purpose of this study was to define the inter-rater reliability of MRI for evaluating UCL characteristics. Three surgeons analyzed 45 noncontrast MRIs: 15 without UCL tears, 15 with intraoperatively confirmed partial-thickness tears, and 15 with intraoperatively confirmed full-thickness tears. Findings evaluated included ligament characteristics (periligamentous or osseous edema, ligament hypertrophy, calcifications, and partial- or full-thickness tearing) and location of the pathology (proximal, midsubstance, or distal). Fleiss' κ coefficients were used to assess interobserver reliability. There were high rates of agreement for midsubstance (κ: 0.50) and distal (κ: 0.69) ligament edema and proximal osseous edema (κ: 0.48). There were low rates of agreement for proximal ligament edema (κ: 0.08), ligament thickening (κ: -0.02 to 0.20), and proximal to midsubstance calcifications (κ: -0.04 to 0.10). Midsubstance (κ: 0.55) and distal full-thickness tears (κ: 0.63) demonstrated higher agreement than proximal (κ: 0.29). Proximal partial-thickness tears (κ: 0.45) had higher agreement than midsubstance (κ: -0.02) or distal (κ: -0.02). In our study, there was high inter-rater reliability regarding proximal osseous edema, midsubstance, and distal ligament edema and full-thickness tears. There was no agreement for midsubstance and distal partial-thickness UCL tears, but fair agreement for proximal tears. UCL tissue characteristics including ligament thickening and calcifications demonstrated low agreement. Challenges remain in effectively evaluating UCL tissue characteristics on noncontrast MRIs.