This observational study included consecutive MG patients who underwent thoracoscopic thymectomy with Univent tube intubation under general anesthesia without NMBAs and combined with airway topical anesthesia between January 2016 and December 2019. The investigators focus on intubation conditions, surgical conditions, intraoperative respiratory, and airway complications

This prospective study was conducted on MG patients who underwent thoracoscopic thymectomy under general anesthesia without NMBAs using Univent tube at Vietnam Military Hospital 103 from January 2016 to December 2019. MG was definitively diagnosed based on clinical features, electromyography results, and/or circulating antibodies against acetylcholine receptors and classified according to the Osserman Classification. Preoperative evaluations included CT scans, pulmonary function tests, and routine exams. The operative risk and airway assessments were evaluated by expert anesthesiologists and excluded patients with preexisting hoarseness or sore throat. This study was approved by Hanoi Medical University's ethics committee (No.168-QĐ/ĐHYHN) and Military Hospital 103. All patients were anesthetized and intubated by one anesthesiologist assigned to the thoracic surgery unit.

Anesthesia procedure

In the operating theatre, the standard monitoring included electrocardiogram (ECG), invasive arterial blood pressure (IABP), peripheral oxygen saturation (SpO2), end-tidal capnography (EtCO2), ventilation peak and plateau pressures, continuous spirometry, and entropy (response entropy-RE and state entropy-SE; Datex Omeda S/5 Advance, GE, USA). Neuromuscular monitoring was recorded in the left forearm using TOF (train-of-four)-WatchsSX (Organons, Oss, the Netherlands).

Patients were pre-oxygenated with 100% oxygen for 5 minutes before induction with sufentanil at 0.5 µg.kg-1 and propofol at 2-3 mg.kg-1. When patients lost their eyelash reflex, lidocaine 10% was initially sprayed in the oral cavity, oropharynx, and upper part of the larynx. Subsequently, lidocaine 2% was administered using The MADgic Laryngo-Tracheal Mucosal Atomization Device (Teleflex, Wayne, Pennsylvania, USA) to cover the oral cavity, oropharynx, larynx, vocal cords, and as deep as possible into the trachea.

Once the SE value reached 50 or lower, a Univent tube (Fuji Systems, Tokyo, Japan) was intubated using conventional direct laryngoscopy. The tube size was standardized based on the patient's sex, with an 8.0-mm inner diameter (ID) used for men and a 7.0-mm ID used for women. Depending on the side of the chest for approaching the anterior mediastinum, the endobronchial blocker was carefully guided into either the right (for a right-sided approach) or left mainstem bronchus (for a left-sided approach) using flexible fiberoptic bronchoscopy. The tracheal and blocker cuffs were gently inflated, ensuring that the intracuff pressure stayed below 30 cmH2O, as measured by a noninvasive manometer. Following the secure positioning of patients in a 45° lateral decubitus orientation for surgery, with their heads firmly fixed, the correct placement of the blocker was reassessed through bronchoscopy.

Anesthesia maintenance was guided by Entropy, utilizing either propofol TCI or sevoflurane to keep the SE index within the 40-60 range throughout the surgical procedure. Additionally, a continuous infusion of sufentanil at 0.2 µg.kg-1.hr-1 was administered, and the patient's temperature was carefully regulated with a heating blanket.

For two-lung ventilation, the ventilator was initially set in A/C mode (FiO2 = 60%, respiratory rate of 14 breaths.min-1, tidal volume of 8-10 ml.kg-1, inspiration-expiration ratio of 1:2). When establishing one-lung ventilation, investigators adjusted the tidal volume (to 5-7 ml.kg-1) and respiratory rate to maintain EtCO2 between 35-40 mmHg and Ppeak < 35 cmH2O; oxygen was given at 100% and gradually reduced by 10% while aiming for SpO2 > 95%. Trocars were placed and the tumor was dissected and removed. Following the insertion of the drainage tube, the collapsed lung was inflated and re-ventilated at the end of the surgery. Patients meeting extubation criteria were extubated in the operating room; otherwise, they were transferred to the postoperative care unit.

Data collection and criteria

Characteristics of the study population included demographic data (age, gender, BMI), disease characteristics (classification, disease duration, symptoms, diagnostic tests), laboratory data, preoperative pulmonary function status, surgical characteristics (tumor features, surgical side), and duration of surgery and anesthesia.

Intubating conditions and intubating variables

The intubating score was assessed following the consensus conference guidelines on Good Clinical Research Practice in Pharmacodynamic Studies of Neuromuscular Blocking Agents. This evaluation included laryngoscopy difficulty (categorized as easy, fair, or difficult), vocal cord position (described as abducted, intermediate/moving, or closed), and the patient's reaction to tracheal tube insertion and cuff inflation (evaluated based on diaphragmatic movement/coughing, categorized as none, slight, or vigorous/sustained). These criteria were used to assign an overall intubation condition score of excellent, good, or poor.

Additionally, the following intubation-related variables were recorded: Mallampati score, mask ventilation categorized as "easy" (successful mask ventilation with or without an oral airway), "difficult" (inadequate or unstable ventilation, requiring two practitioners), or "impossible" (unable to ventilate without a muscle relaxant), the number of intubation attempts, the time required for intubation and bronchial blocker placement (measured in seconds, from the start of laryngoscopy to successful bronchial blocker positioning), and the TOF (train-of-four) value at the time of intubation (%).

Device conditions and surgical variables

During the surgical procedure, the malposition of the bronchial blocker were noted. After opening the pleura and directly examining the lungs, the quality of lung collapse were recorded as follows: 1 - spontaneous; 2 - assisted with suction; or 3 - manual. The surgeon was asked to provide an overall assessment of lung isolation and rated the surgical conditions as follows: 1 - excellent (complete collapse with perfect surgical exposure); 2 - fair (total collapse, but the lung still has residual air); 3 - poor (no collapse or partial collapse with interference in surgical exposure).

Respiratory evaluations

The lung mechanical ventilation parameters, oxygenation, and capnography parameters were documented. Arterial blood gas analysis was also performed during two-lung ventilation and after 30 minutes of one-lung ventilation. Hypoxemia during surgery was defined as a drop in SpO2 below 90% at any point during the intraoperative period.

Postoperative airway injuries, sore throat, and hoarseness After surgery, all patients underwent bronchoscopy examination while still under anesthesia to assess their bronchial lesions. Following successful extubation, a flexible laryngoscopy was performed to evaluate potential vocal cord injuries. Results from both examinations were recorded and analyzed by an experienced ear, nose, and throat specialist. Airway injuries were classified as redness, edema, or hematoma.

Postoperative sore throat and hoarseness were assessed immediately after anesthesia emergence and on days 1, 2, and 3 post-surgeries. Sore throat was defined as continuous throat pain and postoperative hoarseness was categorized as a change in acoustic quality from the patient's previous voice (yes or no). Daily follow-up examinations were conducted until complete resolution.

Inclusion Criteria:

• Myasthenia gravis patients
• Undergoing thoracoscopic thymectomy
• Undergoing general anesthesia

Exclusion Criteria:

• Mallampati IV
• Surgical complications

MG patients who underwent thoracoscopic thymectomy under general anesthesia without NMBAs using Univent tube. MG was definitively diagnosed based on clinical features, electromyography results, and/or circulating antibodies against acetylcholine receptors and classified according to the Osserman Classification