Eberhart Leopold H J, Arndt Christian, Cierpka Thomas, Schwanekamp Judith, Wulf Hinnerk, Putzke Caroline
Department of Anesthesiology and Critical Care Medicine, Philipps-University Marburg, Baldingerstrasse 1, 35043 Marburg, Germany.
Anesth Analg. 2005 Jul;101(1):284-9, table of contents. doi: 10.1213/01.ANE.0000154535.33429.36.
Recently, a new bedside screening test to predict the occurrence of a difficult laryngoscopy has been developed as a substitute for the Mallampati classification. The Upper-Lip-Bite test (ULBT) evaluated the patient's ability to reach or completely cover the upper lip with the lower incisors. It is often accepted that new predictive tools should undergo an external evaluation before the tool is used in clinical practice. Thus, we evaluated this test with respect to applicability, interobserver reliability, and discriminating power and compared it with the Mallampati-score (using Samsoon and Young's modification). The ULBT could not be applied in 12% of all patients (Mallampati score, <1%). However, the interobserver reliability was better for the ULBT (kappa = 0.79 versus kappa = 0.59). The discriminating power to predict a patient with difficult laryngoscopy was evaluated in 1425 consecutive patients. Both tests were assessed simultaneously in these patients by two specially trained independent observers. After the induction of anesthesia, the laryngoscopic view was assessed by the attending anesthesiologist using the classification of Cormack and Lehane. A grade I or II was called easy laryngoscopy and grade III and IV difficult laryngoscopy. The discriminating power for both tests was low (0.60 for the ULBT [95% confidence interval, 0.57-0.63] and 0.66 [0.63-0.69]) for the Mallampati score), indicating that both tests are poor predictors as single screening tests.
最近，一种用于预测困难喉镜检查发生情况的新型床旁筛查试验已被开发出来，以替代Mallampati分级。上唇咬试验（ULBT）评估患者用下切牙触及或完全覆盖上唇的能力。人们普遍认为，新的预测工具在临床实践中使用之前应进行外部评估。因此，我们评估了该试验的适用性、观察者间可靠性和鉴别力，并将其与Mallampati评分（采用Samsoon和Young的改良版）进行比较。在所有患者中，12%无法进行ULBT（Mallampati评分<1%）。然而，ULBT的观察者间可靠性更好（kappa = 0.79，而Mallampati评分为kappa = 0.59）。在1425例连续患者中评估了预测困难喉镜检查患者的鉴别力。两名经过专门培训的独立观察者在这些患者中同时评估这两种试验。麻醉诱导后，主治麻醉医生使用Cormack和Lehane分类法评估喉镜视野。I级或II级称为喉镜检查容易，III级和IV级为困难喉镜检查。两种试验的鉴别力都较低（ULBT为0.60 [95%置信区间，0.57 - 0.63]，Mallampati评分为0.66 [0.63 - 0.69]），这表明作为单一筛查试验，这两种试验都是较差的预测指标。
