Ultrasound localization of lacerated flexor tendon ends in the hand: a cadaveric study

Abstract

Flexor tendon repair currently requires extensive exposure to locate and repair tendons. Ultrasound (US) has been used to identify lacerated tendon ends with little information on accuracy. This study was designed to measure the accuracy of US to localize tendon ends in zone II flexor tendon lacerations in a cadaveric model. US was used to locate tendon ends in zone II lacerations of fingers of six cadaveric hands (96 tendon ends) by a musculoskeletal radiologist. The distance of each tendon end relative to the laceration was recorded. Specimens were dissected and tendon position was compared to US position. The radiologist correctly identified full-thickness lacerations of both superficial and deep tendons 99.0% (n = 05/96 tendons) of the time. The average difference between mean US predicted retraction and anatomic confirmed retraction for all digits all tendons was 3.5mm of underestimation. US correctly identified the position of all tendon stumps to within 10mm 92.7% (n = 89/96 tendons) of the time and 69.8% (n = 67/96 tendons) of the time to within 5mm. Error tended to underestimate (61.5%; 59/96 tendons) rather than overestimate retraction (29.2%; 28/96 tendons). This fresh cadaveric study has demonstrated that with an experienced radiologist, there was 99.0% accuracy identifying a completed tendon tear and locating the tendon ends with US to within 1cm was 92.7% accurate.