The Comparison of Glidescope® and the Direct Laryngoscopy in the Insertion of the Double Lumen Endotracheal Tube

The study protocol was approved by the Institutional Research Board of the Memorial Hospitals of Kaohsiung Medical University and all patients included in the study gave written informed consent. Sixty patients with American Society of Anesthesiologists physical status I-III, ≧ 18 years of age, and needed double-lumen endotracheal tube (DLT) for thoracic surgery were enrolled in this study. Exclusion criteria were increased risk for regurgitation and pulmonary aspiration, history of gastroesophageal reflux, and pregnancy. Furthermore, when a tracheostomy or prolonged ventilation on ICU was planned, patients were also excluded. All tracheal intubations in the study were performed by two experienced (> 10 year) Board Certified anesthesiologists who each had experience of at least 300 tracheal intubations with each device before starting the study.

After the patient had entered the operating room, standard monitoring including ECG, invasive blood pressure measurement on the radial artery and pulse oximetry were established. Patients were randomly assigned to direct MacIntosh laryngoscope group (Group M, n = 30) or GVL group (Group G, n = 30) by opening a sealed envelope after they had entered the operating theatre. Anesthesia was induced with propofol 2-3 mg.kg-1, fentanyl 2μg.kg-1, and rocuronium 0.6mg.Kg-1. After achieving adequate muscle relaxation and anesthetic depth, DLT intubation was performed with direct laryngoscope or GVL. The bronchial and the tracheal cuff of the DLT were lubricated with SURGILUBE® sterile, surgical lubricant (E. FOUGERA \& CO. Melville, New York, A division of Nycomed US Inc.) thinly before intubation.

Measurements The demographic characteristics of both groups included age (years), gender (female/male), weight (kg), height (cm), body mass index (BMI), and ASA classification were recorded. The airway characteristics included the modified Mallampati classification, mouth opening (active and passive), and the thyromental distance were also recorded. Patients were measured of thyromental distance and incisor gap on maximal opening of the mouth and grading of The modified Mallampati classification in the operative room by another anesthesiologist not involved in the study. Passive mouth opening was measured after induction of anesthesia. After pre-oxygenation with oxygen 100%, patients received thiamylal 5 mg.kg-1 and fentanyl 2μg.kg-1 for anesthesia induction and facilitated DLT intubation with rocuronium 0.8 mg.kg-1 and propofol 1.0 mg.kg-1.

The total time to calculate DLT insertion (T3) was separated into T1 and T2, respectively, defined as started after jaw opening when the GVL or the direct laryngoscope passed the patient's lips and stopped when three complete waves of end-tidal CO2 showed on the monitor of capnography and measured by an independent observer. T1 time interval was calculated from the jaw opening, placement of GVL or standard Macintosh laryngoscope, to the view of glottis opening. T2 time interval was calculated from intubator holding DLT to three complete waves of end-tidal CO2 showed on the monitor of capnography. The procedures to position of DLT were declared in brief as following: the distal 10-12cm of the DLT was curved to approximately 90°; a conventional laryngoscope in group M and a Glidescope in group G to assist position DLT into trachea [4]; the tip of the DLT bronchial lumen was passed through the vocal cords and advanced into the glottis; to remove stylet of DLT; 180° counterclockwise rotation to facilitate the insertion of the bronchial cuff and tracheal lumen past the vocal cords; an additional 90° clockwise rotation to fit the tube into left main bronchus [5]. Grading of laryngo-pharyngeal tissues viewed by intubator according to the Cormack and Lehane classification (grades I-IV) that defined as grade I- complete glottis visible, grade II- anterior glottis not seen, grade III- epiglottis seen, but not glottis, and grade IV- epiglottis not seen [6]. The intubator determined the grades of laryngeal viewing either by direct laryngoscope or by Glidescope® with video monitor. The backwards upwards rightwards pressure (BURP) maneuver to optimize the glottis view was also recorded and the pressure grading was defined as none, light, and heavy.

After the blade of laryngoscope or GlideScope® was removed, we examined blood coating on its surface. The oral cavity, pharynx and larynx for signs of lacerations or bleeding provoked by the intubation was also examined by another otolaryngologist 5 minutes later, who was not aware of the chosen technique to complete intubation.

Hemodynamic parameters were recorded in the operating theatre after starting standard monitoring. Measurement of hemodynamic parameters included of time interval of baseline, intubation and post-intubation 1, 3 and 5 min, respectively. Anesthesia was maintained with sevoflurane. At the end of surgery, patients were trained to breathe spontaneously. The DLT was removed when spontaneous ventilation reached a tidal volume of 8 ml/kg and the patient could respond to simple verbal commands. The patients were interviewed by an independent investigator to determine whether they were suffering from a sore throat or dysphagia the next day of post-operation. We quantified the degree of soreness of the throat and dysphagia by using a visual analogue scale ranging from 0 to 10, with VAS 0 being rated as 'none', 1-3 as 'mild', 4-6 as moderate' and 7-10 as 'severe'.

Assuming a likely difference in mean times to intubate of 10 s with a SD of 10 s, an a priori power analysis revealed a group size of n = 27 was needed to detect a difference with a power of 0.8 at α-level of 0.05. Statistical analysis was performed using the Mann-Whitney U-test for non-parametric and univariate ANOVA for parametric data. Duncan's post-hoc test was performed where appropriate. Hemodynamic data were analyzed with ANOVA for repeated measures. Data are expressed as mean (SD) where applicable or actual numbers. SPSS 16.0.1 software (Apache Software Foundation, Forest Hill, MD, USA) was used for all statistical analyses.