In vitro experiment
In order to achieve the even distribution of lidocaine around the cuff, we modified a dual-cuff endotracheal tube and conducted a dyeing experiment in vitro. We evenly punctured small holes through the outer cuff as dye liquid delivery channel. As the tracheal diameter ranges between 8.9–17 mm in women[12], a tube with an internal diameter of 14 mm was used to simulate the trachea and graph paper was placed on the internal surface of the tube to record the outcome of the experiment. The dual-cuff ETT was placed into the 14 mm tube lined with graph paper and the internal cuff pressure was maintained at 25 cmH2O. To optimize the number of holes and the volume of dye liquid, we assessed the difference between 8 and 16 holes by using 1–4 ml dye liquid used to simulate lidocaine; this experiment was performed in triplicate. The dye liquid was fully transferred from the outer cuff by injecting 20 ml air into the outer cuff. This optimum combination was subsequently used in vivo experiment.
In vivo experiment
1) Study Design
This one-centre, double-blind, randomized study was conducted at the Department of Anesthesiology, Guangdong Provincial Hospital of Traditional Chinese Medicine in Guangzhou, China, from 09 / 10 / 2020 to 30 / 04 / 2021.
2) Patients
All 98 ASA I–II females, aged 18–65 years, with a height of 140–180 cm and weight of 40–85 kg, planned for elective thyroidectomy due to unilateral thyroid neoplasm (T1 /T2 N0 M0 in Tumor-Node -Metastasis System/American Joint Committee on Cancer). The exclusion criteria included patients using the NIM standard reinforced EMG endotracheal tube, signs of a difficult airway, perioperative aspiration, psychosis, acute respiratory tract infection lasting less than 2 weeks, smoking lasting less than 48 h, allergic reaction to lidocaine, beta-adrenergic receptor line, and serious cardiovascular, pulmonary, hepatic, or renal disease.
3) Randomization
Patients were randomly allocated in a 1:1 ratio to either 4 ml 2%lidocaine in the outer cuff (Group L) or predicted effect-site concentration (Ce) of TCI remifentanil 2.0 ng/ ml (Group R), according to a computer-generated centralized random table with no block size and stratification factors. A unique randomization number was took out from the sequentially numbered containers on the morning of surgery. The first anesthetist generated the random allocation sequence, enrolled participants, and assigned participants to interventions.
4) Blind evaluation
Three anesthetists and 3 anesthetist nurses participated in this study to conduct a blinded evaluation. The first anesthetist was the only one with knowledge of the group. The other anesthetists, anesthetist nurses, and patients were blinded to the groups. The first anesthetist completed the following tasks at the beginning of skin suture: shielding the TCI pump from other anesthetists, control of the TCI pump, and handling 4 ml 2% lidocaine or 4 ml saline solution to the second anesthetist who was blinded to the actual treatments. The second anesthetist completed the intraoperative and postoperative care, except for the control of the TCI pump. The third anesthetist and 3 anesthetist nurses recorded variables in the post-anesthesia care unit (PACU).
5) Intraoperative and Postoperative Care
Basic monitoring was performed at 3 min intervals. All patients received induction medication including propofol 1.5 mg/ kg, cisatracurium 0.2 mg/ kg, predicted effect-site concentration of TCI remifentanil (Ce 3.5 ng/ ml) and sufentanil 0.4 ug/ kg to facilitate tracheal intubation. Remifentanil was administered as a TCI pump (targeted effect-site TCI, Minto model and CONCERT-III, Guangxi VERYARK Technology Co., Ltd, China). All patients received a 7.0 mm modified dual-cuff ETT, and the internal cuff pressure was maintained at 20–30 cmH2O. Maintenance with sevoflurane (1.2–2.5%) and TCI remifentanil (Ce 2–5 ng/ ml) kept the mean arterial pressure (MAP) and heart rate (HR) within 20% of baseline (values at one day before surgery). Nasopharyngeal temperature was maintained at 36–37 ℃. The tidal volume and ventilatory frequency were adjusted using end-tidal carbon dioxide (ETCO2) at 34–45 mmHg.
At the beginning of the skin suture, sevoflurane was maintained at 1.4–1.5 an end-tidal concentration and remifentanil was maintained at 2 ng/ ml. At the same time, ketorolac 0.5 mg/ kg was administered to relieve pain and tropisetron 4.48 mg was administered to prevent nausea and vomiting. Neostigmine (50 ug/kg) and atropine (15 µg/kg) were administered to reverse the neuromuscular block to confirm a train-of-four response greater than 90%. Patients with ETT were randomized to receive either 4 ml 2% lidocaine in the outer cuff (Group L) or 4 ml saline solution (placebo) in the outer cuff (Group R), followed by 20 ml air into the outer cuff at the beginning of skin suture (each time the internal cuff pressure was less than 30 cmH2O). Remifentanil in group R was maintained at 2 ng/ ml and turned off at tracheal extubation. Remifentanil in group L and sevoflurane in the two groups were turned off at the end of the skin suture (the end of surgery).
Patients with ETT were transferred to the post-anesthesia care unit (PACU) after confirmation of stable hemodynamics. During the transfer of patients (the phase from operating table to PACU), remifentanil 2.0 ng/ ml by TCI was maintained for Group R, whereas remifentanil was stopped in Group L. It took 4 min during the phase from the end of surgery to the beginning of transfer and 2 min during patient transfer. The shielded TCI pump from all the patients was taken to the PACU. Recovery profiles were recorded in the PACU by video after obtaining the patients’ consent. Manual ventilation was provided until the patient breathed spontaneously. Mild hypercapnia (end-tidal carbon dioxide, ETCO2 45–55 mmHg) was permitted to promote spontaneous respiration. Continuous verbal stimuli were used to prompt the patients to open their eyes without any other stimuli. After opening their eyes, the patients were asked to breathe deeply and nod their head. Extubation was completed when adequate breath on command (ETCO2 < 50 mmHg and ventilatory frequency > 12 beats/min), nodding and handshake on command, and spontaneous deglutition were achieved.
6) Primary Outcomes
The primary outcome was the incidence of cough during the recovery period. A cough was defined as any evidence of irritation from a tube in the trachea[13]. Cough was recorded while patients transfer (the phase from the operating table to the PACU), 1 min before extubation and extubation.
7) Secondary Outcomes
Five time points were defined as follows: T0, one day before surgery (baseline); T1, end of surgery; T2, 1 min before extubation; T3, 5 min after extubation; and T4, 20 min after extubation. The secondary outcomes were the following variables: MAP and HR at the above five time points, elevation of blood pressure and heart rate at T2, T3 and T4 (MAP and HR increase by 30% from their respective values at baseline), eye opening time (time period from the end of surgery to first eye opening on command), consciousness recovery time (time period from the end of surgery to nodding on command), spontaneous respiration recovery time (time period from the end of surgery to ETCO2 ༜ 55 mmHg and ventilatory frequency ༞ 10 beats/ min), extubation time (time period from the end of surgery to extubation), duration of PACU stay (the time period from the end of surgery to leaving from PACU), the Richmond Agitation-Sedation Scale (RASS) [14] at T2, and the Critical-Care Pain Observation Tool (CPOT) [15] at T2.
In addition, the secondary outcomes were a composite of the following variables: hypoxemia at T3 and T4 (less than 95% of Spo2), sedation grading system (SGS) [16] at T3, postoperative pain at T4 (more than 5 points on the visual analog scale), residual sedation at T4 (less than Grade 2 on the SGS), nausea and vomiting at T4 (need for drug), and pharyngalgia at T4 (swallowing with more than 3 points on the visual analog scale).
8) Sample size calculation
The incidence of cough was 76% during recovery from general anesthesia[1]. Based on the assumption that TCI remifentanil Ce 2.0 ng/ ml could suppress cough by 90%[17] and 4 ml of 2% lidocaine used with the dual-cuff endotracheal tube could suppress cough by 50%[16, 18], 49 patients in each group would be required for 80% power at a two-sided α of 0.05, with 20% lost to follow-up.