Application of right bronchial blocker under articial pneumothorax in thoracoscopic enucleation of oesophageal leiomyoma

Background Oesophageal leiomyomas are one of the most common benign oesophageal tumours. This study summarized and analysed the clinical experience of thoracoscopic enucleation of oesophageal leiomyoma. Methods A total of 36 patients who underwent thoracoscopic enucleation of oesophageal leiomyoma at Peking Union Medical College Hospital between 2014 and 2020 were retrospectively analysed. Fifteen patients received single-lumen endotracheal intubation combined with a right bronchial blocker (SLET-B group), and twenty-one patients received double-lumen endotracheal intubation (DLET group). Clinical data, surgical variables, and postoperative complications were analysed and compared.


Introduction
Oesophageal leiomyomas are the most common benign oesophageal tumours, accounting for approximately 70% − 80% of oesophageal submucosal tumours (1). In recent years, video-assisted thoracoscopic enucleation of oesophageal leiomyoma has been shown to have better postoperative recovery than the thoracotomy approach (2). Compared with double-lumen endotracheal intubation anaesthesia, single-lumen endotracheal intubation combined with a bronchial blocker under arti cial pneumothorax is a new type of anaesthetic technology. Arti cial pneumothorax achieves better lung collapse and surgical eld exposure and decreases blood loss, which are more convenient for tumour enucleation with less morbidity (3). This study retrospectively analysed a total of 36 patients diagnosed with oesophageal leiomyoma undergoing thoracoscopic surgery at Peking Union Medical College Hospital from 2014 to 2020. We present surgical outcomes of video-assisted thoracoscopic enucleation of oesophageal leiomyoma and evaluate the differences between traditional double-lumen endotracheal intubation and single-lumen endotracheal intubation combined with a bronchial blocker under arti cial pneumothorax.

Anaesthetic and surgical procedures
All patients underwent a right thoracoscopic side approach, and a levin tube was placed before the operation. General anaesthesia was adopted in all patients. In the SLET-B group, a 7-Fr or 7.5-Fr singlelumen endotracheal tube was inserted, and a bronchial blocker was inserted into the right main bronchus under the guidance of a bre bronchoscope. The patient was arranged in a left semi-prone position inclined 45 degrees. Arti cial pneumothorax was created by CO 2 insu ation with a pressure of 8 mmHg.
In the DLET group, a left double-lumen endobronchial intubation was inserted. The parameters of the anaesthetic machine and anaesthetic drugs were the same as those in the SLET-B group.
In the operation, after the tumour was detected, the muscular layer of the oesophagus was opened by a coagulating hook. The tumour was separated from the muscular layer, carefully protecting the mucosa from injuries. After tumour enucleation, the muscular layer and the parietal pleura were closed using continuous 3 − 0 V-loc stitches. The resected tumour was placed in a bag and removed from the chest. Air in ation by a levin tube was used to con rm whether there was oesophageal mucosal injury. If mucosal injury was found, the oesophageal mucosa was sutured by interrupted absorbable 3 − 0 stitches. A 28-French thoracic drain was left in the right chest. If no oesophageal mucosa injury was found during the operation, the levin tube was removed by the rst day after the operation, and the oral diet was resumed the second day. If mucosal injury was found and repaired during the operation, patients received nothing by mouth with total enteral nutrition for 7 days, oesophagography was performed to con rm mucosal integrity, and the levin tube was removed after con rmation. The thoracic drainage tube was removed after drainage was less than 100 ml and there was no air leak after oral diet feeding.

Variables collected
We collected data on the baseline characteristics and surgical and postoperative characteristics in this study.

Statistical analysis
Analysis was performed using Statistical Product and Service Solutions 22.0 statistical software. The continuous variables are expressed as the means (x ± s) or medians (P25, P75). We use Student's t-test or nonparametric test to compare the means between groups according to the normal distribution test (K-S test). The chi-squared test was used to compare the frequencies of the categorical variables data. P value < 0.05 were considered signi cant.

Results
Patient characteristics are shown in Table 1. There were 23 men and 13 women, with an average age of 43.67 ± 10.55 years. The majority of the tumours arose in the middle (19/36 [52.8%]). Clinical symptoms were found in 16 patients, and 20 (55.6%) patients were asymptomatic, with their tumours discovered incidentally by routine physical examination. The most common symptom that patients reported was dysphagia (10/36 [27.8%]). Most patients had chronic symptoms persisting for months to years before surgical management. The average tumour size in all patients was 4.31 ± 1.96 cm. Among symptomatic patients, the average tumour size was 5.08 ± 2.02 cm, compared with an average tumour size of 3.71 ± 1.72 cm in asymptomatic patients (P = 0.035).  Table 2. The SLET-B group had a signi cantly shorter operation time, shorter duration of chest tube and total hospital stay than the DLET group. No mucosal injury, conversion to thoracotomy, or other operative complications occurred in the SLET-B group. Two cases of mucosal injury occurred in the DLET group and were repaired during the operation. The patients were discharged from the hospital after conservative treatment. The period of follow-up was 6 months to 6 years. In the follow-up, no recurrence, dysphagia, or regurgitation was found in any of the patients.

Discussion
Oesophageal leiomyomas are the most common benign tumours of the oesophagus. It has been reported that the disease mainly occurs in the middle part of the oesophagus, followed by the lower and upper parts of the oesophagus (4). The tumour grows slowly, and small oesophageal leiomyomas are usually asymptomatic. With the growth of the tumour, patients may present acid regurgitation, chest pain, dysphagia and dyspepsia. With the development of gastroscopy and ultrasonic gastroscopy technology, an increasing number of asymptomatic small oesophageal leiomyomas have been detected and treated (5). However, in symptomatic patients, oesophageal leiomyomas are usually larger, which increases the di culty in treatment.
At present, surgical treatment is the preferred choice for oesophageal leiomyoma, and enucleation is widely accepted as an adequate treatment (6). However, surgical indications for oesophageal leiomyoma are still controversial. Oesophageal leiomyoma may grow gradually, which may lead to surrounding tissue compression and potential malignant transformation. It is also di cult to differentiate leiomyoma from stromal tumour and leiomyosarcoma before surgery. Therefore, it is suggested that once diagnosed, surgical treatment should be performed regardless of symptoms (7). Codipilly et al summarized the clinical data of 105 patients with submucosal tumours and found that small leiomyomas grew very slowly, with an average growth rate of 0.5 mm every 6 years (8). At the same time, the rate of dysphagia in patients with larger leiomyoma surgery was higher than that in the conservative observation group, indicating that the risk of digestive tract obstruction caused by larger leiomyoma is lower than that caused by scar stenosis after surgery. However, there are some shortcomings in this study, such as a high rate of loss to follow-up and retrospective design. In our study, the average size of the tumour was 4.31 ± 1.96 cm, and the proportion of symptomatic patients was 44.4% (16/36). Therefore, we indicate surgical enucleation for patients with symptomatic or larger oesophageal leiomyomas (larger than 2 cm), while patients with smaller and asymptomatic oesophageal leiomyomas could be observed and then treated when they were obviously enlarged or had the possibility of malignancy.
Surgery consists of enucleation of smaller tumours. Oesophageal resection is advocated for large lesions, circular tumours, or if the tumour is densely adhesive to the mucosa. Compared with traditional thoracotomy, video-assisted thoracoscopic surgery has the advantages of minimal scarring, less pain, better postoperative respiratory function with fewer perioperative respiratory complications, and an enhanced fast recovery. It is generally considered that a right thoracic approach is used for middle and upper oesophageal tumours, and a left thoracic approach is used for lower oesophageal tumours (9). In our study, all patients completed tumour resection through the right thoracic approach, indicating that the right thoracic approach can complete all operations. Traditional double-lumen endotracheal intubation can achieve single-lung collapse to provide enough surgical vision, but due to the complex intubation process, the incidence of intraoperative tube displacement is high, and there are many postoperative respiratory complications. Thoracoscopic surgery in the prone position has been adopted in the majority of minimally invasive surgeries for oesophagus because it is easy to mobilize the oesophagus (10). Single-lumen endotracheal intubation combined with a bronchial blocker under arti cial pneumothorax has the following advantages: 1. Single-lumen endotracheal intubation is relatively simple, and a bronchial blocker can provide an effective seal of the bronchus with minimal trauma to achieve singlelung ventilation. 2. Single-lumen endotracheal intubation causes less damage to the airway mucosa and respiratory tract, which can reduce postoperative pharyngeal discomfort or pain. 3. Arti cial pneumothorax (8 mmHg) can obtain good lung collapse, cause capillaries to collapse and reduce bleeding, maintain a better surgical eld and reduce side injury. In our study, complete tumour enucleation was achieved in patients in the SLET-B group, with a larger average tumour size and no mucosal damage or other complications and a shorter postoperative hospital stay than patients in the DLET group. Thoracoscopic operation in the prone position with arti cial pneumothorax with CO 2 insu ation may add merits to the conventional decubitus position.
Oesophageal mucosal injury is the most common complication, especially when the tumour is large and close to the oesophageal mucosa. According to the author's experience, the main operation approach should be designed based on the tumour location, such as the 4th or 5th intercostal space for high and middle thoracic oesophageal leiomyomas and the 6th or 7th intercostal space for lower thoracic oesophageal leiomyomas. Blunt dissection should be used to dissect the tumour, and energy instruments should be avoided near the oesophageal mucosa side to prevent intraoperative oesophageal mucosal injury. After the enucleation of the tumour, the mucosal wound should be carefully examined. Air in ation of the oesophagus through endoscopy or a Levin tube should be used to check the dissected area to con rm the integrity of the mucosa. Absorbable sutures should be used to repair the injured oesophageal mucosa. For large mucosal injuries, the mediastinal pleura should be used to cover the injured part after repair, and the levin tube should be reserved. After 7 days of fasting, upper oesophagography should be performed to determine whether the injury had healed.
According to previous research, carbon dioxide arti cial pneumothorax under low pressure (< 8 mmHg) has no signi cant effect on respiration and circulation (11). In our study, we did not observe any respiratory or haemodynamic disorders perioperatively. Another disadvantage is the inability of continuous sucking of the operative eld, which may cause di culty in the haemostatic process and increase the operation time. In our study, however, the operation time was signi cantly shorter in the SLET-B group. Therefore, we believe that this shortcoming can be overcome through more experience with this surgical technique.
Nevertheless, several limitations in our study are noted. First, as our data were retrospectively collected, patient selection bias may exist even though baseline characteristics were comparable. The SLET-B group has only been applied since 2017, thus, more experience with surgical technique may contribute to a better outcome in this group. In addition, the sample size was small; thus, larger-scale studies, especially multicentre collaboration, are needed.

Conclusions
In conclusion, thoracoscopic enucleation of oesophageal leiomyomas is a feasible and safe procedure. The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki (as revised in 2013). The study was approved by the independent medical ethical committee of the Peking Union medical College and all the patients signed extensive informed consents.

Consent for publication
Not applicable.

Availability of data and materials
The datasets supporting the conclusions of this article are included within the article and its additional les.

Competing interests
Not applicable.

Funding
Not applicable.
Authors' contributions LZ, LL performed the conception and design. LZ, LL analyzed and interpreted the patient data. LZ, CG, JH, ZJH, LL drafted the manuscript and all authors read and approved the nal manuscript.