This was an open-label, single-centre, randomized controlled non-inferiority trial with two parallel groups. This study was approved by the Ajou Institutional Review Board (AJIRB-DEV-OBS-17-247; date of registration: 30/11/2017) and was registered at ClinicalTrials.gov (NCT03392766; date of registration: 08/01/2018). All the study procedures were performed in accordance with the relevant guidelines and regulations. Written informed consent for the study was obtained from each patient participating in the study. Patients aged 19–75 years with an American Society of Anaesthesiologists physical status of 1 and 2 who were scheduled to undergo elective thoracic surgery requiring one-lung ventilation (OLV) were enrolled in this study. Patients were excluded if they met one of the following criteria: (1) had an anatomical anomaly or intraluminal mass in the upper airway tract; (2) had a history of gastroesophageal reflux or recent upper airway tract infection; (3) had a risk of aspiration; or (4) had a body mass index greater than 35 kg/m2. Equal numbers of eligible patients were randomly allocated to the SLT group or DLT group by an investigator not involved in this study, using random numbers generated in Microsoft Excel 2010 (Microsoft Corp., USA). Patients were randomized to their study group in the operating room after airway assessment.
On arrival in the operating room, all patients were monitored using an electrocardiogram and pulse oximeter, and non-invasive blood pressure and bispectral index (BIS) were measured. After preoxygenation with 100% oxygen for 3 min, fentanyl 1.0–1.5 μg/kg and thiopental sodium 4.0–5.0 mg/kg were administered. On loss of consciousness, muscles were relaxed using rocuronium 0.6 mg/kg and anaesthesia was maintained with sevoflurane in oxygen. The thyromental distance mouth opening was measured 2 min after muscle relaxation, and the laryngeal view was evaluated with direct laryngoscopy according to the modified Cormack–Lehane classification. Mouth opening and modified Cormack–Lehane grade were rechecked after a semi-rigid foam neck collar (Philadelphia cervical collar) was placed on all patients. All tracheal intubations were performed with a flexible fiberoptic bronchoscope (PortaView® LF-GP; Olympus Optical Company, Tokyo, Japan) with an outer diameter of 4.1 mm by the same investigator (DH Kim, an anaesthesiologist with over 20 years of experience with fiberoptic bronchoscope-guided intubation). In both groups, the polyvinyl chloride SLT or silicone DLT was preloaded before intubation. One investigator introduced the fiberoptic bronchoscope into the trachea via the patient’s mouth at the head of the bed, while the other performed jaw-thrust manoeuvre to clear the airway and provide sufficient space for fiberoptic bronchoscope passage. After confirming that the tip of the fiberoptic bronchoscope was on the carina, the preloaded tube was railroaded along the fiberoptic bronchoscope. If the railroading met resistance, the tracheal tube was pulled back 2–3 cm, rotated 90° counter-clockwise, and then reinserted. If the resistance was not resolved after this, the degree of counter-clockwise rotation was increased to more than 120° or clockwise rotation was attempted until successful intubation.
In the SLT group, a standard bevelled polyvinyl chloride Portex® endotracheal tube (Smith Medical, Hythe, UK, Fig. 1) was used. The size of the endotracheal tube was chosen according to the sex of the patient: 7.0 mm internal diameter and 9.6 mm outer diameter SLT for females and 8.0 mm internal diameter and 10.9 mm outer diameter SLT for males in the SLT group. The preloaded polyvinyl chloride SLT was introduced along the fiberoptic bronchoscope (railroading over the fiberoptic bronchoscope) with the bevel of the tube oriented towards the patient’s left side. After intubation, the bronchial blocker (Coopdech Endobronchial Blocker Tube, Daiken medical Co., LTD, Japan) was inserted into the proper mainstream bronchus under fiberoptic bronchoscope control with a smaller fiberoptic bronchoscope (PortaView® LF-GP; Olympus Optical Company, Tokyo, Japan; outer diameter 3.1 mm).
In the DLT group, a HumanBroncho® silicone left-sided DLT (Insung Medical, Seoul, Korea, Fig. 1) was used according to the sex of the patient: size 35 Fr DLT (internal diameter short/long, 4.5 mm/7.0 mm; outer diameter short/long, 10.0 mm/13.3 mm) for female patients and 37 Fr DLT (internal diameter short/long, 4.9 mm/7.5 mm; outer diameter short/long, 10.5 mm/14.3 mm) for male patients. As described above, the fiberoptic bronchoscope was introduced into the trachea and then the preloaded silicone DLT was railroaded over the fiberoptic bronchoscope with the concave curvature facing left. The DLT was inserted into the left main bronchus until resistance was felt; then, the correct position of the DLT was confirmed by fiberoptic bronchoscope reinsertion through tracheal lumen after withdrawing it from the bronchial lumen.
All procedures were recorded by another investigator using a video camera for measurement of the following time points: (1) insertion time of fiberoptic bronchoscope, defined as the time from which the fiberoptic bronchoscope started to pass the tooth and reached above the carina; (2) railroading time, the time from positioning the fiberoptic bronchoscope above the carina until the tube was placed over the carina, with the tube not needing to be in the final position; (3) time to tracheal intubation, insertion time of fiberoptic bronchoscope plus the railroading time; and (4) total time for correct tube and bronchial blocker positioning, the time from which the fiberoptic bronchoscope started to pass the tooth till confirming the proper position of the DLT or bronchial blocker for adequate OLV. The difficulty of railroading was evaluated by the investigator (DH Kim) as follows: (grade I) the railroading was easily done following the natural curvature of the SLT or DLT; (grade II) failure of grade I railroading, necessitating tube readvancement after 90° counter-clockwise rotation; (grade III) failure of grade II railroading, requiring another manipulation like 120° counter-clockwise rotation, 90° clockwise rotation, rerotation, or external manipulation; (grade IV) failure of grade III railroading, needing direct laryngoscopy or removal of anterior cervical neck collar. If the time to tracheal intubation was over 120 s, it was recorded as a failed case and intubation was performed with another method after removing neck collar.
In the post-anaesthesia care unit (PACU) after surgery, the incidence of sore throat, difficulty in swallowing, and hoarseness were evaluated by an investigator who was not involved in the study.
The primary outcome of this study was the time to tracheal intubation between the two groups, designed as a non-inferiority test. The sample size required for non-inferiority was calculated based on an assumed standard deviation of 18 s with a non-inferiority margin of 10 s9. A sample size of 40 patients per group was obtained with an alpha error of 0.05 and 80% power. The secondary outcomes include the time to insert the fiberoptic bronchoscope, the railroading time, the total time for correct tube and bronchial blocker positioning, the difficulty of railroading over a fiberscope bronchoscope, and the incidence of sore throat, hoarseness, and swallowing difficulties.
Statistical analysis was performed using SPSS version 21 (SPSS Inc., Chicago, IL, USA). Continuous data between groups were analysed using Student’s t-test or Mann–Whitney U-test as appropriate. Normality of the data distribution was tested using the Kolmogorov–Smirnov test, and results are presented as means with standard deviations and medians with interquartile ranges (IQR) as appropriate. Categorial data between groups were analysed using Chi-square or Fisher’s exact tests as appropriate. Results are presented as numbers and percentages. Continuous and categorial data before and after applying neck collar were analysed using paired t-test or Wilcoxon signed-rank test. A p value of < 0.05 was considered statistically significant. To demonstrate non-inferiority of the primary outcome, the primary outcome was assessed by a two-sided 95% confidence interval using Hodges–Lehmann method. If the upper bound of the 95% confidence interval in the time to endotracheal tube intubation between groups was below the non-inferiority margin of 10 s, non-inferiority would be declared.