This study was approved by the Institutional Review Board (approval obtained on 25th October 2012) at the Quanzhou Women’s and Children’s Hospital, Fujian Province, China (clinical trials registration number: NCT01858467). At the time of the study, the majority of Cesarean sections at Quanzhou Women’s and Children’s Hospital were performed under general anesthesia, with routine airway management using the SLMA in both elective and urgent cases19. The hospital Cesarean section rate is 35% and the SLMA technique for general anesthesia is used in about 2000 deliveries annually. In this hospital, ETT airway management is employed as an alternative airway management for Cesarean section.
We recruited parturients with singleton pregnancies aged 18 to 50 years old with healthy or well-controlled medical conditions (American Society of Anesthesiology; ASA 2 ) undergoing elective Caesarean section under general anesthesia at Quanzhou Women’s and Children’s Hospital, China between May 2013 and July 2014. Written informed consent was obtained from every participants. Parturients with body mass index (BMI) ≥ 35 kg/m2, potentially difficult airway (modified Mallampati grade 4, upper respiratory tract or neck pathology) or gastro-esophageal reflux disease (self-reported) were excluded from the study. All parturients were fasted for at least 6 hours. Once enrolled in the study, the parturients were randomized into 2 groups using sealed opaque envelopes prepared by a statistician not involved in the study recruitment. The 2 groups were (1) SLMA group; and (2) ETT group. The parturients were blinded to their assigned group, but the investigators managing the airway were not blinded.
Premedication with intravenous ranitidine was administered to all parturients. Electrocardiography, pulse oximetry, capnography and non-invasive blood pressure measurements were applied. After preoxygenation for three minutes, RSI was carried out with cricoid pressure applied. Anesthesia was induced with intravenous propofol (2-3mg/kg) and 100mg succinylcholine. After the airway device was inserted, rocuronium (0.5mg/kg) was used to maintain muscle relaxation and fentanyl (100mcg) was administered for intraoperative analgesia after the fetus was delivered. All cases regardless of assigned study groups were induced in similar fashion.
The size of SLMA used was based on manufacturer’s recommendations, however, at the discretion of the attending anesthesiologist, a more appropriate size could be selected based on parturient’s weight, BMI and mouth opening. ETTs of intraluminal diameters between 6.5 to 7.0 millimeters were used, chosen at the investigators’ discretion. Three investigators (Yao, Li, and Yuan), each with more than 5 years of experience in the use of both SLMA and ETT as well as difficult airway management in general anesthesia, managed the airway for all parturients enrolled in this study.
In the SLMA group, the SLMA was inserted using the single-handed rotational technique recommended by the manufacturer until resistance was met with the cricoid pressure maintained. The cuff was then inflated with air to a pressure of 60 cmH2O as measured by an intracuff pressure monitor, and the volume of air needed to achieve this pressure was recorded. Next, the ability to ventilate was confirmed as evidenced by the presence of auscultation of bilateral air entry and carbon dioxide trace on capnography. The cricoid pressure was then released. After successful placement, a pre-mounted #14 orogastric tube was advanced through the gastric drainage aperture and placement was confirmed by: (1) aspiration of gastric contents; or (2) injection of air into orogastric tube via the large lumen whilst auscultating the epigastrium for a “swoosh”. The number of orogastric tube insertion attempts and failure to pass the orogastric tube were recorded. Suctioning of the orogastric tube was performed at the beginning of the surgery and at the end before emergence. Finally, the seal pressure was determined by recording the peak airway pressure after closing the adjustable pressure-limiting valve and insufflating the closed breathing system with 3L/min of fresh gas flow.
The investigators were allowed to use additional maneuvers (chin lift, jaw thrust, head extension) or reposition the SLMA if necessary to achieve airway patency. If successful placement could not be achieved (1) after 2 attempts, (2) within 60 secs, or (3) before desaturation occurred (oxygen saturation < 92%), the airway would then be secured using direct laryngoscopy and endotracheal intubation. The surgical procedure was allowed to proceed if the following criteria were met: (1) a square-wave capnograph tracing was present; (2) the pilot cuff was inflated to 60cmH2O and checked with a manometer; (3) the bite block of the SLMA was sitting between the incisors; (4) the gastric tube was inserted into the drain tube, and the position confirmed using insufflation of 5mL of air and auscultation over the epigastric region, followed by active/passive suction and then passive drainage of the gastric tube; and (5) the seal pressure was checked and ≥ 20cmH2O.
In the ETT group, oral endotracheal intubation via direct laryngoscopy was performed using a Macintosh blade (size 3-4 based on anesthesiologist’s preference). The cuff was inflated with air to a pressure of about 25 cmH2O as measured by an intracuff pressure monitor, and the volume of air needed to achieve this pressure was recorded. Cricoid pressure was then released after ability to ventilate was confirmed, as evidenced by the presence of auscultation of bilateral air entry and carbon dioxide trace on capnography. A #14 orogastric tube was inserted, and the position was confirmed as above.
For all parturients, successful placement of the airway device was confirmed by auscultation and the presence of end-tidal carbon dioxide on the capnogram. The number of attempts required to achieve successful placement, with an attempt defined as insertion and complete withdrawal of the device from the airway was recorded. The time to effective airway placement, defined as the interval from when the device was picked up until appearance of the first end-tidal carbon dioxide waveform, was also measured and recorded.
Anesthesia was maintained with 1.5 to 2.0% sevoflurane and 50% nitrous oxide in oxygen. All parturients were placed in the left lateral tilt position using a wedge. During maintenance of anesthesia, complications including loss of airway, desaturation, inadequate ventilation and bleeding into the airway device were recorded. The tidal volume was set from 6 to 10mL/kg, and the respiratory rate ranged from 10 to 16 breaths/min to maintain an end-tidal carbon dioxide concentration of 30 to 40 mmHg. If there were signs of aspiration (perioperative hypoxemia, wheezing, crepitations upon lung auscultation, postoperative dyspnea), the parturient would be investigated with bronchoscopy or chest X-rays.
The obstetricians were instructed to reduce fundal pressure during delivery regardless of device used. Upon completion of surgery, muscle paralysis was reversed and the orogastric tube was suctioned and removed. After return to spontaneous respiration and consciousness (defined as when parturient was able to follow instructions to open eyes and mouth) was achieved, the airway devices were removed and inspected for blood. The incidence of sore throat and hoarseness were assessed by an independent assessor before discharge from the post anesthesia care unit.
Our primary outcome was the first insertion attempt success rate of the airway devices. Secondary anesthetic outcomes included:
- time to effective ventilation; seal pressure;
- ventilation parameters (tidal volume, respiratory rate, peak airway pressure) to maintain effective oxygenation and ventilation, defined as the ability to maintain SpO2 ≥ 92% and an end-tidal carbon dioxide concentration of < 50mmHg, using inspired oxygen concentration ≤ 0.5 with respiratory rate 10 to 16 breaths/min and tidal volume 6 to 10 mL/kg; and
- hemodynamic parameters (heart rate and blood pressure) for 6 minutes after induction.
The amount and pH of gastric aspirate; incidence of regurgitation (clear or bile stained fluid seen during procedure or removal of airway device); incidence of aspiration (bile stained fluid seen in the lung during bronchoscopy or postoperative radiological evidence) and pH of the laryngeal surface of the SLMA and ETT cuff were also recorded. Apart from neonatal weight, the obstetric outcomes recorded were neonatal Apgar score at 1 and 5 min and umbilical cord venous pH. Maternal satisfaction with the anesthetic experience (question asked “how would you rate your overall satisfaction with your anesthetic care from 0 to 100%”), the presence of sore throat and voice hoarseness were assessed one day after surgery by an independent assessor.
Statistical Analysis
The primary outcome measure of first attempt insertion success rate in SLMA insertion and secondary outcomes, namely incidence of regurgitation, aspiration, blood stain on SLMA and hoarseness were treated as categorical data, with categories “yes” or “no”. All demographic, anesthetic and clinical categorical variables were summarized as frequency with corresponding proportion; and continuous variables were expressed as mean (standard deviation (SD)) or median [interquartile range (IQR)], whichever applicable with respect to two study groups SLMA and ETT. Difference in outcomes between ETT and SLMA were represented as risk ratio and mean difference with 95% confidence interval for categorical and continuous data respectively. Difference in categorical and continuous variables between SLMA and ETT were tested using Chi-Square test and Student’s t-test or Mann-Whitney U test, whichever applicable, respectively. Significance level was set at p-value < 0.05 and all tests were two sided. Data analysis was generated using SAS 9.3 software (SAS Institute Inc., Cary, NC, USA).
We performed a sample size calculation based on our previous experience estimate of 98% in first attempt insertion success rate of SLMA insertion19. An equivalence boundary of 0.034 i.e. (96.3%, 99.7%) as the maximum difference between maximum likelihood estimates of the true proportions in the first attempt insertion success rate in SLMA and ETT use was considered as clinical equivalence. A sample size of 916 parturients (458 per group) would achieve 80% power with the above-mentioned equivalence limits, based on a one sided test with a significance level of 2.5% and retrospective estimate in the first attempt success rate of 98%. We aimed to recruit a total of 920 parturients in this study.