Laryngospasm is a serious complication of general anaesthesia and often occurs during recovery from anaesthesia with the use of supraglottic devices. We hypothesised that the timing of sugammadex administration may affect the postoperative incidence of laryngospasm.1 To test this, we conducted a prospective randomised controlled study to evaluate the timing of sugammadex administration after the end of surgery by administering sugammadex before administration of sevoflurane attenuation was stopped or after recovery of consciousness after administration of sevoflurane had been discontinued. The primary hypothesis was tested by assessing the incidence of laryngospasm between early and late sugammadex administration.
The research ethics committee of Hokusetsu General Hospital approved this study (approved on 29 January 2015; Head: Masaya Kino). We also registered this study in the UMIN Clinical Trials Registry (UMIN000016419). From February to July 2015, 49 patients aged 20 to 75 years who were scheduled to undergo general anaesthesia were assessed for eligibility to participate. Seven patients were excluded and two refused to participate. After obtaining written informed consent, 40 patients were assigned randomly using an envelope method to either the early phase sugammadex administration group (early sugammadex group; 20 patients) or the late phase sugammadex administration group (late sugammadex group; 20 patients). Exclusion criteria were any contraindications for the use of supraglottic devices or a recent (within 7 days) history of an upper respiratory tract infection).2
Muscle relaxant levels were measured using a TOF-Watch monitor (Nihon Kohden, Tokyo, Japan). Without any premedication, anaesthesia was induced with propofol 1 to 2 mg kg-1 and remifentanil 0.3 to 0.5 μg kg-1 min-1. Rocuronium 0.6 mg kg-1 was administered as a muscle relaxant. An i-gel supraglottic airway of either size 3 or 4 was selected according to the manufacturer's guidelines.
After successful insertion of the i-gel, mechanical ventilation was immediately performed, and anaesthesia was maintained by inhalation of sevoflurane 33–40% in oxygen, and the remifenanil infusion. The lungs were ventilated with a tidal volume of 8 ml kg-1 at 8–12 breaths per min to keep the end-tidal CO2 at 4.0 to 4.7 kPa (30 to 35 cmH2O). Rocuronium was continuously administered at an injection rate of 4 μg kg-1 min-1 until the end of surgery. The dose of rocuronium was determined in a preliminary study which found the train-of-four ratio to be more than 1/4 at this dose.
At the end of surgery, the total flow of oxygen and air was set to 6 l min-1. Sugammadex 2 mg kg-1 was administered before sevoflurane discontinuation in the early sugammadex group. Specifically, sugammadex was administered immediately after surgery and administration of sevoflurane was stopped 2 min later. For the late sugammadex group, administration of sevoflurane was discontinued attenuated immediately after surgery, and sugammadex 2 mg kg-1 was administered after recovery of consciousness. In the late sugammadex group, patients with a train-of-four (TOF) count of 0 or 1 were excluded as recovery of patients with this level of anaesthesia was considered unethical. The preliminary study confirmed that no patient with a TOF count more than 1 had any memories of discomfort after complete recovery of consciousness.
Laryngospasm was considered to have occurred when peak inspiratory pressure (PIP) suddenly increased by over 30 cmH2O without i-gel displacement. We also distinguished laryngospasm from coughing of the patient. We did not touch the i-gel so as to not affect the relative position of the i-gel and pharynx. In cases of laryngospasm, anaesthesiologists considered the bispectral index (BIS) and PIP values, and awakened the patient by calling to the patient or providing stimulation. We also asked patients if they experienced immobilisation with consciousness.
Statistical analysis was performed using JMP 11 (SAS Institute Inc., Cary, North Carolina, USA). The primary measurement was incidence of laryngospasm, which was tested using the χ2 test. Data are presented as mean ± SD with 95% confidence intervals. P less than 0.05 was considered statistically significant. As for the sample size calculation, the incidence of laryngospasm was approximately 50% in the early sugammadex group and 10% in the late sugammadex group. To detect this difference with 80% power at a 5% significance level, 18 patients would be necessary in each group. Therefore, we recruited 20 patients for each group to adjust for missing data.
None of the participants dropped out or were lost to follow-up. Patients’ characteristics were similar between the groups. All patients in the late sugammadex group had TOF counts more than 1 at the end of surgery. The incidence of laryngospasm was significantly higher in the early sugammadex group (9/20 cases) than in the late sugammadex group (0/20 cases) (P < 0.001). Average BIS and PIP values were 74.0 ± 5.7 and 34.1 ± 3.5 cmH2O, respectively, in the nine patients with laryngospasm in the early sugammadex group. All patients with laryngospasm were treated by calling and/or mechanical stimulation and recovered uneventfully. None of the patients in either group complained of immobilisation or discomfort with consciousness.
In the present study, the incidence of laryngospasm was significantly higher in the early sugammadex group than in the late sugammadex group. Our results suggest that sugammadex administration after recovery of consciousness may be safer in that it prevents laryngospasm.
There are several limitations to this study. First, the continuous infusion of rocuronium we employed may not be a common practice when using supraglottic devices. Second, although none of our patients complained of discomfort after recovery of consciousness in either the preliminary or main trials, some discomfort may have remained in the implicit memory.
In conclusion, the incidence of laryngospasm was significantly higher in the early sugammadex group compared with the late sugammadex group. BIS and TOF should be carefully monitored during recovery from anaesthesia when using supraglottic devices to avoid laryngospasm.
Acknowledgements relating to this article
Assistance with the study: none.
Financial support and sponsorship: funding was provided by our institution and department.
Conflicts of interest: none.
1. Olsson GL, Hallen B. Laryngospasm during anaesthesia. A computer-aided incidence study in 136,929 patients. Acta Anaesthesiol Scand
2. Komasawa N, Nishihara I, Tatsumi S, Minami T. Prewarming i-gel® to 42°C facilitates successful insertion and ventilation efficacy with muscle relaxation: a randomized study. J Clin Anesth