背景:
順利氣管內管置入是全身麻醉的首先要件；一旦置入困難甚至無法插管，可能會造成缺氧傷害，甚至死亡。美國麻醉醫師學會針對封閉性索賠分析發現17%嚴重呼吸事故是困難插管造成，而其中的85%會導致腦傷或死亡，可見困難插管發生的嚴重性。
方法:
本研究採前瞻性觀察研究法，從西元2008年8月至2012年12月，收集接受一般非緊急和非心臟手術且需全身麻醉與氣管內管置入處置的病患。針對每位病患收集其術前病患特性資料、詳細的氣道身體檢查與氣道結果。本次研究的非預期困難插管定義主要是用插管次數超過三次以上。主要針對東方人種，對可能造成困難插管的眾多因子進行評估，希望找出最適切的指標危險因子，我們探討東方人種預測非預期困難插管的各因子的最佳切點與預測模型是否有與西方研究不同。本研究主要分析方法為多變量邏輯斯回歸分析和多變量ROC曲線分析。
結果:
本次研究共收集1300位病患，，非預期困難插管發生機率於研究對象中只佔1.46%(19位)。與文獻建議比較，本研究單因子最佳切點，舌骨頦骨距離(最佳切點≦4.2cm)明顯小於西方文獻所指，身高(最佳切點≦158cm) 明顯小於西方體型。依研究對象的各種可能預測因子量測值所建立的最佳預測模型為(張口寬度/甲狀頦骨距/身高/ Mallampati張口吐舌)的四因子組合模型。其組合的變項為身高(最佳切點≦158cm)、張口寬度(最佳切點≦4.5cm)、甲狀頦骨距離(最佳切點≦6cm)，和Mallampati張口吐舌(最佳切點≧grade 2)；有高敏感度(0.857)與不錯的特異度(0.745)，漢好的平衡準確度為0.801，AUC為0.77。。之使用二疊交叉驗證方法，仍有相同的結果(平均AUC=0.727)，是具有區別力的最佳模型。
此外針對肥胖與否去做預測UDI單因子最佳切點分析，發現肥胖組值得關注的是甲狀頦骨與舌狀頦骨距離比率(TMD/HMD)，有高敏感度，可作為不錯的預測UDI風險的單因子。非肥胖者張口寬度(AUC=0.769)，具高敏感度，也是不錯的單因子預測指標。
結論:
因此本研究發現西方與東方因體型的不同，預測非預期困難插管的危險因子可能不同，尤其是各因子的最佳切點。此外，肥胖與非肥胖者可能也需要分開界定不同預測UDI的因子。
Background:
Smooth tracheal intubation is the first important document for general anesthesia. Once unanticipated difficult tracheal intubation (UDI) or failing tracheal intubation had happened, patients who were received general anesthesia would suffer possible hypoxia injuries, or even deaths. In a closed-claim analysis of American Society of Anesthesiologists, difficult tracheal intubation accounted for approximately 17% of adverse respiratory events and the outcome of 85% of these cases was either brain damage or death.
Methods:
We performed a prospective observational study to enroll adult patients presenting the hospital for general anesthesia with tracheal intubation for any type of nonemergency and noncadiac surgical procedures from August 2008 to December 2012. For each patient, preoperative patient characteristics, detailed airway physical exam, and airway outcome data were collected. Difficult tracheal intubation was defined as 3 or more attempts at placing the endotracheal tube. We investigated the possible different best cutpoint of predictors and predicting models of UDI between Easterners and Westerners. Multivariate logistic analysis and multivariate ROC curve analysis were performed.
Results:
A total of 1300 adult patients were enrolled in the study. The incidence of UDI was 1.46% (19 cases). The best cutpoint of two predictors (Hyomental distance ((HMD)≦4.2cm) and height (≦158cm)) suggested in our study was smaller than that reported for Westerners. Multivariate analysis revealed that the best predicting model of UDI in our study population consisted of four predictors (interincisor gap(IG)≦4.5cm, thyromental distance (TMD)≦6cm, height≦158cm, modified Mallapati score≧grade 2) had high sensitivity (85.7%) , specitity (74.5%) and good balance accuracy (80.1%). The area under the receiver operating characteristic curve (AUC) was 0.77. Two-fold cross validation also showed the similar result (mean AUC=0.73). Moreover, the ratio of TMD to HMD (TMD/HMD) showed high sensitivity and may be a not bad single predictor in obese patients. We also found IG (AUC=0.769) had high sensitivity and also a good single predictor in non-obese patients.
Conclusions:
Predictors of UDI need to have different concerns in Westerns and Easterners, especially in terms of the best cutoff values. Moreover, it seems that obese and non-obese patients have difficult pridictors of UDI.
