Dirschl Douglas R, Ferry Scott T
Department of Orthopaedics, University of North Carolina School of Medicine, Chapel Hill, NC 27599-7055, USA.
J Trauma. 2006 Dec;61(6):1463-6. doi: 10.1097/01.ta.0000202484.23607.ce.
Many orthopedic classification systems, including those for tibial plafond fractures, are either unvalidated or have demonstrated problems with interobserver reliability. Classification of tibial plafond fractures according to a rank-order method has shown excellent interobserver reliability with several observers. The purpose of this study is to determine the reliability of a rank order classification of plafond fractures with a large number of observers.
A radiographic review study was completed by 69 orthopedists of varying training levels. Observers ranked 10 fractures of the tibial plafond based on anteroposterior and lateral ankle radiographs. Fractures were ranked in increasing severity from 1 to 10. No instructions were given regarding determination of severity. Agreement between rankings was analyzed by the intraclass correlation coefficient (ICC).
Rankings were performed by viewing prints at the annual Orthopaedic Trauma Association meeting and through the Orthopaedic Trauma Association website using digital images. The overall ICC was 0.62. There was no difference in the ICC between traumatologists and general orthopedists (p > 0.5). Eleven observers commented that the radiographs did not represent the full spectrum of injury severity.
The interobserver reliability of the rank-order classification in this study was fair to good, which is better than previously reported for plafond fracture classification systems. It remains to identify and validate a series of tibial plafond fractures that represent a full spectrum of injury and can be ranked with excellent interobserver reliability. A series of cases such as this may then serve a measurement standard for severity of bony injury against which individual cases may be reliably compared.
许多骨科分类系统，包括胫腓骨远端骨折的分类系统，要么未经验证，要么已证明存在观察者间可靠性问题。根据排序方法对胫腓骨远端骨折进行分类，已显示出在多个观察者之间具有出色的观察者间可靠性。本研究的目的是确定大量观察者对胫腓骨远端骨折排序分类的可靠性。
69名不同培训水平的骨科医生完成了一项影像学回顾研究。观察者根据踝关节正侧位X线片对10例胫腓骨远端骨折进行排序。骨折按严重程度从1到10递增排序。未给出关于确定严重程度的指导。通过组内相关系数（ICC）分析排序之间的一致性。
排序是在年度骨科创伤协会会议上查看打印件，并通过骨科创伤协会网站使用数字图像进行的。总体ICC为0.62。创伤科医生和普通骨科医生之间的ICC没有差异（p>0.5）。11名观察者评论说，X线片没有代表损伤严重程度的全谱。
本研究中排序分类的观察者间可靠性为中等至良好，优于先前报道的胫腓骨远端骨折分类系统。仍有待确定和验证一系列代表损伤全谱且观察者间可靠性极佳的胫腓骨远端骨折。这样的一系列病例然后可以作为骨损伤严重程度的测量标准，据此可以可靠地比较个体病例。
