Niglis L, Collin P, Dosch J-C, Meyer N, Kempf J-F
Service de chirurgie orthopédique du membre supérieur, CCOM, 10, avenue Achille-Baumann, 67400 Illkirch, France.
CHP Saint-Grégoire, 6, avenue de la Boutière, 35760 Saint-Grégoire, France.
Orthop Traumatol Surg Res. 2017 Oct;103(6):835-839. doi: 10.1016/j.otsr.2017.06.006. Epub 2017 Jun 24.
The long-term outcomes of rotator cuff repair are unclear. Recurrent tears are common, although their reported frequency varies depending on the type and interpretation challenges of the imaging method used. The primary objective of this study was to assess the intra- and inter-observer reproducibility of the MRI assessment of rotator cuff repair using the Sugaya classification 10years after surgery. The secondary objective was to determine whether poor reproducibility, if found, could be improved by using a simplified yet clinically relevant classification.
Our hypothesis was that reproducibility was limited but could be improved by simplifying the classification.
In a retrospective study, we assessed intra- and inter-observer agreement in interpreting 49 magnetic resonance imaging (MRI) scans performed 10years after rotator cuff repair. These 49 scans were taken at random among 609 cases that underwent re-evaluation, with imaging, for the 2015 SoFCOT symposium on 10-year and 20-year clinical and anatomical outcomes of rotator cuff repair for full-thickness tears. Each of three observers read each of the 49 scans on two separate occasions. At each reading, they assessed the supra-spinatus tendon according to the Sugaya classification in five types.
Intra-observer agreement for the Sugaya type was substantial (κ=0.64) but inter-observer agreement was only fair (κ=0.39). Agreement improved when the five Sugaya types were collapsed into two categories (1-2-3 and 4-5) (intra-observer κ=0.74 and inter-observer κ=0.68).
Using the Sugaya classification to assess post-operative rotator cuff healing was associated with substantial intra-observer and fair inter-observer agreement. A simpler classification into two categories improved agreement while remaining clinically relevant.
II, prospective randomised low-power study.
肩袖修复的长期效果尚不清楚。复发性撕裂很常见，尽管其报告的发生率因所使用的成像方法的类型和解读挑战而异。本研究的主要目的是评估术后10年使用 Sugaya 分类法对肩袖修复进行MRI评估时观察者内和观察者间的可重复性。次要目的是确定如果发现可重复性差，是否可以通过使用简化但临床相关的分类法来改善。
我们的假设是可重复性有限，但可以通过简化分类来改善。
在一项回顾性研究中，我们评估了观察者内和观察者间对肩袖修复术后10年进行的49次磁共振成像（MRI）扫描解读的一致性。这49次扫描是从609例接受重新评估和成像的病例中随机选取的，这些病例用于2015年肩袖全层撕裂修复10年和20年临床及解剖学结果的SoFCOT研讨会。三位观察者中的每一位都在两个不同的场合阅读了这49次扫描中的每一次。每次阅读时，他们根据Sugaya分类法将冈上肌腱分为五种类型进行评估。
Sugaya类型的观察者内一致性较高（κ=0.64），但观察者间一致性仅为中等（κ=0.39）。当将五种Sugaya类型合并为两类（1-2-3类和4-5类）时，一致性得到改善（观察者内κ=0.74，观察者间κ=0.68）。
使用Sugaya分类法评估术后肩袖愈合情况时，观察者内一致性较高，观察者间一致性中等。简化为两类的分类法在保持临床相关性的同时提高了一致性。
II级，前瞻性随机低功率研究。
