Sitzman Thomas J, Allori Alexander C, Matic Damir B, Beals Stephen P, Fisher David M, Samson Thomas D, Marcus Jeffrey R, Tse Raymond W
Cleft Palate Craniofac J. 2018 Jul;55(6):871-875. doi: 10.1597/16-186. Epub 2018 Feb 22.
Oronasal fistula is an important complication of cleft palate repair that is frequently used to evaluate surgical quality, yet reliability of fistula classification has never been examined. The objective of this study was to determine the reliability of oronasal fistula classification both within individual surgeons and between multiple surgeons.
Using intraoral photographs of children with repaired cleft palate, surgeons rated the location of palatal fistulae using the Pittsburgh Fistula Classification System. Intrarater and interrater reliability scores were calculated for each region of the palate.
Eight cleft surgeons rated photographs obtained from 29 children.
Within individual surgeons reliability for each region of the Pittsburgh classification ranged from moderate to almost perfect (κ = .60-.96). By contrast, reliability between surgeons was lower, ranging from fair to substantial (κ = .23-.70). Between-surgeon reliability was lowest for the junction of the soft and hard palates (κ = .23). Within-surgeon and between-surgeon reliability were almost perfect for the more general classification of fistula in the secondary palate (κ = .95 and κ = .83, respectively).
This is the first reliability study of fistula classification. We show that the Pittsburgh Fistula Classification System is reliable when used by an individual surgeon, but less reliable when used among multiple surgeons. Comparisons of fistula occurrence among surgeons may be subject to less bias if they use the more general classification of "presence or absence of fistula of the secondary palate" rather than the Pittsburgh Fistula Classification System.
口鼻瘘是腭裂修复术后的一种重要并发症，常用于评估手术质量，但瘘管分类的可靠性从未得到检验。本研究的目的是确定个体外科医生之间以及多个外科医生之间口鼻瘘分类的可靠性。
利用腭裂修复患儿的口腔内照片，外科医生使用匹兹堡瘘管分类系统对腭瘘的位置进行评分。计算腭部每个区域的评分者内和评分者间可靠性得分。
八位腭裂外科医生对从29名儿童获得的照片进行评分。
在个体外科医生中，匹兹堡分类各区域的可靠性从中度到几乎完美不等（κ = 0.60 - 0.96）。相比之下，外科医生之间的可靠性较低，从一般到实质性不等（κ = 0.23 - 0.70）。软硬腭交界处的外科医生间可靠性最低（κ = 0.23）。对于继发腭瘘的更一般分类，外科医生内和外科医生间的可靠性几乎完美（分别为κ = 0.95和κ = 0.83）。
这是首次关于瘘管分类的可靠性研究。我们表明，匹兹堡瘘管分类系统由个体外科医生使用时是可靠的，但在多个外科医生之间使用时可靠性较低。如果外科医生使用“继发腭有无瘘管”这种更一般的分类，而非匹兹堡瘘管分类系统，那么外科医生之间瘘管发生率的比较可能偏差较小。
