Comparative Sensitivity and Specificity of Static and Dynamic High-Resolution Ultrasound in Diagnosis of Small Gaps in Repaired Flexor Tendons: A Cadaveric Study

Abstract

To compare the sensitivity and specificity of high-resolution static and dynamic ultrasound (US) for diagnosing intact repairs and small, clinically relevant gaps (≥4 mm) in repaired flexor digitorum profundus tendons within zone 2 and, secondarily, to evaluate the effect of suture artifact from 3 commonly used suture types. Eighty-eight fresh-frozen cadaveric digits (thumbs excluded) were randomized to either an intact repair (0-mm gap) or repairs using a locked 4-strand suture repair with either 4-0 Prolene, Ethibond, or FiberWire and gaps of 2, 4, or 6 mm and no suture in which 2-, 4-, or 6-mm gaps were created without a suture crossing the repair site. Gap widths were estimated by a blinded musculoskeletal ultrasonographer in static and dynamic modes. Both static and dynamic modalities tended to overestimate actual gap sizes. For the suture gaps, both modalities had poor sensitivity (29% static; 42% dynamic) for accurately diagnosing a clinically intact repair (<4 mm), but better specificity (83% static; 75% dynamic) for diagnosing a clinically failed repair (≥4-mm gap). Although suture presence decreased the sensitivity of gap width measurement for both modalities, no differences were seen between suture types. Static and dynamic US have poor sensitivity for diagnosing clinically intact repairs (gaps < 4 mm) because they typically overestimate gap size. The ability to diagnose failed repairs (gap ≥ 4 mm), based on greater specificity, is much better, but still suboptimal. Based on a receiver operating characteristic analysis cutoff of 5 mm, if a gap of 5 mm or larger is identified with US when evaluating a zone 2 flexor digitorum profundus tendon repair, a failed repair is likely in about 80% of cases. A gap measurement of less than 5 mm may miss a high percentage of repairs that are clinically failed.