Can the dimensions of artificial tendon lesions be predicted ultrasonographically? A cadaveric study

Objective. To prove the reliable assessment of artificial partial tendon le sions in cadavers. Methods. The tendon of the tibialis anterior muscle was cut twice (transversely, not standardized depths) along its dorsal surface. The dimensions were measured by using a 10-MHz linear probe and a gliding caliper. The lesions were classified as type A tears (< 60%, n = 15) and ty pe B tears (> 60%; n = 28). Results. The depths of type A tears had high co rrelation (r = 0.827), and those of type B tears had low correlation (r = 0 .415). This low correlation was due to subtotal lesions (50%-70%; r = 0.699 ). High-resolution ultrasonography allowed the correct detection of 14 type A tears and 21 type B tears (kappa = 0.63; accuracy, 81%; type B tears. se nsitivity 75%; specificity, 93%). Conclusions. High-resolution ultrasonogra phy is able to assess the dimensions of artificial partial tendon lesions i n cadavers and would be an accurate tool for determining the lesion percent ages in patients with partial tendon tears.