Rubin D A, Kneeland J B, Kitay G S, Naranja R J
Department of Radiology, Hospital of the University of Pennsylvania, Philadelphia, 19104, USA.
AJR Am J Roentgenol. 1996 Mar;166(3):615-20. doi: 10.2214/ajr.166.3.8623637.
Treatment of flexor tendon lacerations of the finger partly depends on the degree of injury, which is difficult to determine clinically. We used a cadaver model to investigate the potential of MR imaging in evaluating these injuries.
A scalpel was drawn transversely across the volar surface of four cadaver hands, producing various flexor tendon injuries. MR imaging of each hand was performed using axial two-dimensional spin-echo and three-dimensional gradient-recalled-echo sequences. The three-dimensional data sets were interactively reformatted along the long axis of each tendon. The hands were then dissected; injury to each digit was categorized, measured, and compared with the prospective MR interpretations.
Twelve high-grade flexor tendon tears (10 complete tears, with 1- to 14-mm separation of the torn ends, and two partial tears involving 50% or more of the total tendon cross-sectional area) and two partial tears of less than 50% of tendon area were produced; four tendons were not injured. Using MR imaging, we diagnosed 11 of the 12 high-grade lesions (those involving at least 50% of the total tendon cross-sectional area); the MR images did not show one complete tear whose separation measured 2 mm long at dissection. All intact tendons were correctly identified. We underestimated the extent of five lesions but overestimated none. Using the reformatted images, we reduced the number of errors that we would have made interpreting the transverse images alone.
In this cadaver model, using MR imaging we accurately distinguished different degrees of flexor tendon tears. The potential of this technique for noninvasively diagnosing flexor tendon injury in patients awaits clinical studies.
手指屈肌腱撕裂伤的治疗部分取决于损伤程度，而损伤程度在临床上难以确定。我们使用尸体模型研究磁共振成像（MR成像）在评估这些损伤方面的潜力。
用手术刀横向划过四只尸体手的掌面，造成不同程度的屈肌腱损伤。对每只手进行MR成像，使用轴向二维自旋回波序列和三维梯度回波序列。将三维数据集沿每条肌腱的长轴进行交互式重新格式化。然后解剖这些手；对每个手指的损伤进行分类、测量，并与预期的MR解释结果进行比较。
造成了12处严重屈肌腱撕裂（10处完全撕裂，撕裂端分离1至14毫米，2处部分撕裂累及肌腱总横截面积的50%或更多）以及2处累及肌腱面积小于50%的部分撕裂；4条肌腱未受伤。通过MR成像，我们诊断出了12处严重损伤中的11处（那些累及肌腱总横截面积至少50%的损伤）；在解剖时发现，MR图像未显示一处撕裂端分离2毫米长的完全撕裂。所有完整的肌腱均被正确识别。我们低估了5处损伤的程度，但没有高估任何一处。使用重新格式化后的图像，我们减少了仅解读横向图像时可能出现的错误数量。
在这个尸体模型中，使用MR成像我们准确地区分了不同程度的屈肌腱撕裂。这项技术在无创诊断患者屈肌腱损伤方面的潜力有待临床研究。
