Reliability of MRI in Acute Full-thickness Proximal Hamstring Tendon Avulsion in Clinical Practice

In clinically suspected acute full-thickness proximal hamstring tendon avulsions, MRI is the gold standard for evaluating the extent of the injury. MRI variables such as full-thickness free tendon discontinuity, extent of tendon retraction (>20 mm), and continuity of the sacrotuberous ligament with the conjoint tendon (STL-CT) are used in treatment decision-making. The objective was to assess the intra- and inter-rater reliability of these relevant MRI variables after acute full-thickness proximal hamstring tendon avulsion. Three musculoskeletal radiologists assessed MRIs of 40 patients with an acute full-thickness proximal hamstring tendon avulsion. MRI variables included assessment of free tendon discontinuity and continuity of the STL-CT and extent of tendon retraction. Absolute and relative intra- and inter-rater reliability were calculated. Intra- and inter-rater reliability for the assessment of tendon discontinuity was substantial (Kappa [ĸ]=0.78;0.77). For the retraction measurement of the conjoint and semimembranosus tendons, intra-rater reliability was moderate and poor (Intraclass correlation coefficient (ICC)=0.74;0.45), inter-rater reliability was moderate (ICC=0.73;0.57). Intra- and inter-rater reliability of the STL-CT continuity assessment was substantial and fair (ĸ=0.74;0.31). In conclusion, MRI assessment for full-thickness free tendon discontinuity is reliable. However, assessment of extent of tendon retraction and STL-CT continuity is not reliable enough to guide the treatment decision-making process.