The Regional Ethics Committee Capital Region (Address: Kongens Vænge 2, 3400 Hillerød, Denmark) approved the trial the 24th of February 2010 (Protocol: H-2-2009-148). The trial was registered at ClinicalTrials.gov (Identifier: NCT01090726) prior to inclusion of the first patient. Written informed consent was obtained from the parents of all participating patients.
Patients were eligible if they were between 1 month and 2 years of age, were cleft lip/palate surgical patients with an ASA Physical Scale class 1–2, and had an indication of intubation for general anesthesia. We excluded patients if a history of difficult intubation was present, if the child had breathing difficulties during sleep or a dysmorphic feature with apparent micro- or retrognathia.
Patients were randomized 1:1 according to a computer generated list (GraphPad Software®, Inc., La Jolla, California, USA). A total of 10 envelopes were prepared for the trial, by staff with no other involvement in the trial. The patients were randomized to be intubated by either the SVL or the AOL.
The patients were monitored with a three lead ECG, non-invasive blood pressure measurement, pulse oximetry, capnography and temperature. Induction of anesthesia was done with sevoflurane and an intravenous access was established. Intravenous fentanyl (2.5 microgram/kg) and cisatracurium (0.1 mg/kg) was given. Mask ventilation and sevoflurane inhalation were continued for 4 minutes after which the time measurements were done and the endotracheal intubation was performed. Intubation was preceded by a crossover design with three laryngoscopies performed with three different laryngoscopes. Between each laryngoscopy the patient was ventilated by face mask. First, laryngoscopy was performed with a Macintosh blade and Cormack-Lehane (CL)-score  and timing to CL-score were registered. Then laryngoscopy was performed with either AOL or SVL depending on randomization and the CL-score with this device was registered. Time to CL-score and time to “ready to intubate” position of the endotracheal tube just in front of the glottis were also registered. Finally the patient was intubated using the other device (AOL or SVL). The CL-score with this device was registered along with time to CL-score, time to position the endotracheal tube in “ready to intubate” in front of glottis and time to complete endotracheal intubation, defined as passage of the vocal cords with the endotracheal tube. The evaluation of the airway was performed using a Macintosh blade, which is our standard method in children weighing more than 3 kg. A laryngoscope blade size 0 was used in children younger than six months otherwise a size 1 blade was used. When using the AOL the size was chosen from the recommendation on the package. AOL “infant” (size 0) for endotracheal tubes 3.0 and 3.5, and AOL “pediatric” (size 1) for size 4.0. With the SVL a “Miller-like” blade size 0 was used. The blade resembles an ordinary Miller blade, but is lower and half a cm longer. In all cases with all three devices the tip of the laryngoscope blade was placed in the vallecula and external pressure was applied to the neck if necessary. Preformed oral Microcuff ® endotracheal tubes were used for intubation in all cases. When using the SVL a malleable stylet with the tip bent as a hockey-stick was placed in the tube. Only indirect laryngoscopy, using the video monitor, was attempted with the SVL. When using the AOL only the distal half of the preformed tube was kept in the tube conduit during advancement of the tube as described by Xue et al . The experiment was finished upon completed intubation, defined by verification of ETT placement. Only two intubation attempts were accepted. After this the experimental setup would be considered finished. In case of continued failure to intubate the responsible anesthesiologist would determine how to proceed. All investigations were performed by one experienced pediatric anesthesiologist, who had performed 20 or more endotracheal intubations with each of the new devices (AOL, SVL) in small children prior to inclusion of patients in the trial. Both the AOL and the SVL were connected via their monitors to a laptop computer. The video-feed was integrated in the software program Final Cut Pro 7 (Apple Inc., California, U.S.A.) with live audio from the room. An equipment rack was built for the purpose with monitors, laptop and cables securely clamped. This ensured that the setup would be the same for all patients. A switch made it possible to easily control the source of video-input to the computer, without affecting the ease of changing devices, the time recording, or the safety of the patient. The recording was started when the first laryngoscopy with the Macintosh laryngoscope was initiated. Time recording was done with both video and audio recording and by having the anesthesiologist mark the endpoints with a clear voice. The integrated recording was analyzed after the procedure. Endpoints were measured and evaluated independently two times, and then they were compared to identify potential discrepancies. If a discrepancy was found, the shortest time or best rating was chosen.
The primary endpoint was the success rate, defined as intubation in first attempt. Secondary endpoints were CL, time from start of laryngoscopy to CL, time to ETT positioning in front of the glottis, and time to endotracheal intubation. Time measurement was started from the point where the device was lifted off the table. Time to CL-score was recorded, and it was defined as the time point when the best possible view was achieved including the use of external manipulation. If a secondary intubation attempt was needed, all time recordings were based on the secondary attempt alone.
No sample size calculation has been performed. An objective of this pilot-study was to determine the baseline data, on which a sample size calculation could be made for similar but larger studies.
The Wilcoxon test was used for statistical analysis of paired data. A P-value of less than 0.05 was considered statistical significant. All statistical calculations were done using SAS statistical software, Version 9.1 (SAS Institute, Cary, NC, USA).