Complications After Lymph Node Dissection Along Bilateral Recurrent Laryngeal Nerves Through Right Thoracic Approach in the Patients With Thoracic Esophageal Cancer

Background: Extensive lymph nodes dissection can improve the accuracy of tumor staging and prognosis of the patients with thoracic esophageal cancer, palsy of recurrent laryngeal nerve (RLN) caused by the lymph node (LN) dissection along RLN chain also increase postoperative complications and may affect the prognosis. This study aimed to evaluate the associated postoperative complications after LN dissection along RLNs in the patients with thoracic esophageal squamous cell cancer (ESCC). Methods: 339 eligible patients with thoracic ESCC who underwent radical McKeown or Ivor-Lewis esophagectomy by open or VATS procedures through right thoracic approach with LN dissection along bilateral RLNs were included in this study. Univariate and multivariate logistic regression analysis were conducted to assess the correlation of RLN paralysis (RLNP) with other post-operative complications. Results: 39 of the 339 patients were diagnosed with RLNP (11.5%) postoperatively. The incidence of RLNP in three-eld (3FL) LN dissection was signicantly higher than that in the two-eld (2FL) LN dissection ( 24.0% vs 8.0%, P(cid:0)0.001). Compared with the patients without RLNP, the patients with it had a signicantly higher incidence of postoperative anastomotic leakage (P=0.029), pulmonary complications (P=0.001) and much longer hospital stay (P=0.001). Two patients died of respiratory failure within 30 days caused by RLNP and were treated by reintubation. Conclusion: RLNP after LN dissection along bilateral RLN in thoracic ESCC was associated with much higher morbidity such as pulmonary complications, anastomotic leakage, and much longer hospital stay. New technologies are required to reduce RLNP incidence and its associated complications.


Introduction
For locally advanced esophageal carcinoma, esophagectomy with radical lymphadenectomy is still the mainstay of the treatment [1]. Complete 3FL-LN dissection, especially the LN dissection along bilateral RLNs, can not only increase the accuracy of pathological staging, but also improve the prognosis for the patients with thoracic ESCC [2]. However, it can also increase the risk of RLNP and associated with high incidence of complications, such as aspiration, respiratory complication, anastomotic leakage, et al.
Permanent RLNP also compromises patients' life quality. Therefore, how to balance the pro t and disadvantage from LN dissection along RLNs, which has been debated and is still unclear until present. During extensive LN dissection along RLNs, thermal injury, stretching, compression, or vascular damage of the RLN may cause RLNP [3]. The incidence of RLNP after esophagectomy was reported ranging from 10.0%-58.9% [4][5][6][7][8][9]. Which clinicopathological factors affect the incidence of RLNP during esophagectomy and whether RLNP compromise the patients' survival and life quality has not been clari ed. Therefore, this retrospective study aimed to investigate the risk factors of RLNP and its associated post-operative complications as well as prognostic role in the patients with ESSC.

Inclusion criteria
From March 2015 to December 2018, totally, 1686 patients with thoracic ESCC underwent esophagectomy in our hospital. The patients were selected into this study according to the following criteria: (1) R0 resection;(2) pathologically con rmed squamous cell carcinoma (3) with LNs dissection along bilateral recurrent laryngeal nerves ;(4) negative pathological resection margin ;(5) through right chest;(6) without preoperative suspected distant metastasis. Finally, 339 patients in total were selected into this study.
High resolution and enhanced chest/abdominal CT scans, bone scanning, brain MRI/CT, neck ultrasonography; FOE and EUS were performed preoperatively in all patients in order to exclude distant metastasis and make a precise clinical TNM staging. All specimens were examined pathologically at the Department of Pathology in our institution. The seventh edition UICC esophageal cancer staging criteria was used for pathological TNM classi cation.

Principle Surgical Procedure
All 339 patients underwent McKeown or Ivor-Lewis esophagectomy via either open procedures through conventional three incisions (thoracotomy + midline laparotomy + left neck incision) or minimally invasive procedures (thoracoscopy/laparoscopy/left neck incision). Lymph nodes in the chest including periesophageal lymph nodes, bilateral recurrent laryngeal nerve lymph nodes, subcarinal lymph nodes, and left tracheobronchial lymph nodes were all dissected. The abdominal lymph nodes consisting of paracardial, lesser curvature, and left gastric artery, splenic artery, common hepatic artery were all dissected. Next, the stomach conduit was made and pulled to the neck or the apex of the right thorax, a handsewn or stapled gastro-esophageal anastomosis was carried out.
Preoperative assessment and Diagnosis of RLNP Chest computed tomography, bone scanning, brain MRI, and abdominal ultrasonography were performed in all patients preoperatively to exclude distant metastasis. All specimens were pathologically diagnosed at the Department of Pathology in our institution. The seventh edition UICC esophageal cancer staging criteria was used for TNM classi cation, and pathology results determined the T and N stage. RLNP was diagnosed based on clinical symptoms such as hoarseness/aspiration/di cult in coughing.

Statistical analysis
All statistical analysis was performed using IBM SPSS Statistics ver. 22.0(SPSS Inc, Chicago, IL, USA). Chi-square test or Fisher's exact test was used to compare categorical data. Student's t test was used for continuous data. OS was created using the Kaplan-Meier method and compared between two groups with log-rank test. Univariate Cox regression analysis was used to estimate the hazard ratios of OS. Twosided P values less than 0.05 were considered statistically signi cant.

Clinical and Pathologic characteristics of 339 patients
The clinical and pathologic characteristics of the study population are summarized in Table 1. Ages ranged from 40-80 years (average 60.6 years). There were 282 men and 57 women. 74 tumors were in the upper third segment of the thoracic esophagus, 188 in the middle third, 77 in the lower third. All 339 patients underwent subtotal esophagectomy through a right thoracotomy. 75 underwent 3-FL lymph node dissection, and the other 264 underwent 2-FL lymph node dissection. According to the seventh edition UICC esophageal cancer staging criteria, there were 101 cases in T1, 61 cases in T2, 160 cases in T3, 17 cases in T4;43 cases of well differentiated, 181 cases of moderately differentiated, 115 cases of poorly differentiated.

Correlation of RLNP and clinicopathologic factors
In the univariate analysis, the incidence of RLNP was signi cantly correlated with the extent of lymph node dissection (3FL 24.0% versus 2FL 8.0%, P < 0.001). Based on the number of metastatic LNs along RLNs, the 339 patients were divided into three groups. the incidence of RLNP increased as the number of metastatic LNs along RLN increased, while no signi cant difference among each group was observed (

Discussion
Esophagectomy with radical lymphadenectomy still plays the most important role in the multimodality treatment for locally advanced esophageal carcinoma. In recent years, complete 2FL-or 3FL-LN dissection along the esophagus, especially along the bilateral RLNs has been emphasized in the surgical treatment for ESCC. It can not only increase the accuracy of pathological staging, but also improve the prognosis for the patients with thoracic ESCC [2]. However, extensive LN dissection also increase the risk of RLNP and the incidence of associated complications, such as aspiration, respiratory complication, anastomotic leakage, et al. Permanent RLNP also compromises patients' life quality. As reported in the literatures, the incidence of RLNP after esophagectomy ranged from 10.0-58.9% [4][5][6][7][8][9]. These disparities may be attributed to variation in the extent of lymph node dissection, usage of surgical technique, the Tstage of the primary tumor, and number of metastatic lymph nodes, and surgeons' experience [10][11]. Patients with RLNP may present with symptoms from hoarseness, dyspnea during speech, aspiration, di culty with coughing and expectation of sputum, and may even suffocate in case of bilateral damage. Consequently, dysfunction of RLNs may result in other post-operative complications such as pneumonia, anastomotic leakage, which prolong hospital-stay and severely compromised the post-operative life quality of the patients with esophageal cancer [12][13]. Our study demonstrated that extensive lymph nodes dissection along recurrent laryngeal nerves increased the risk of RLN injury, and subsequently resulted in high incidence of pneumonia and anastomotic leakage, this is consistent with what reported in the literature [12][13]. The patients with temporary or permanent RLNP after esophagectomy usually have di culties with swallowing and coughing, suffer from the associated pulmonary complications, they can't recover smoothly and need much longer treatment and hospital stay. Their life quality was deteriorated and medical cost signi cantly increased. These results stress the importance of preserving the RLN for both short-and long-term outcomes.
The incidence of RLNP after esophagectomy with radical 2FL or 3FL LN dissection was 11.5% in this series, which is similar to the results of previous studies [4][5][6][7][8][9], Our diagnosis of RLNP mainly based on the symptoms such as hoarseness, aspiration, di culty with coughing except a few by bronchoscopy or esophagoscopy. And most RLNPs were temporary, and most of them recovered within 6 months, indicating that injury of the RLNs might be caused by compression or stretching or compromised blood supply of the nerves other than structure damage. In this study, univariate analysis showed that patients with extensive 3FL lymph node dissection had a higher RLNP rate than that those with 2FL (24.0% vs 8.0%, P 0.001) and the incidence of RLNP increased as the number of metastatic LNs along RLN increased. This suggested that more extensive LN dissection could lead to higher RLNP rate and neoadjuvant therapy was needed for patients with suspicious RLN LNs metastasis.
Pulmonary complication was reported to be the most common postoperative complication after esophagectomy [14][15]. Our study showed that RLNP was an independent risk factor for postoperative pulmonary complication. In our study, 8 patients were transferred to ICU due to respiratory failure, 7 of the 8 patients were reintubated, 1 patient had tracheotomy eventually. And 2 of them died within 30 days postoperatively, both of them experienced RLNP. Patients with RLNP could not close the glottis completely, which resulted in pulmonary infection leading to weak coughing, ineffective expectoration aspiration and retention of secretion, which nally result in pulmonary infection even respiratory failure [16]. These results stress the importance of preserving the RLN to decrease the incidence of postoperative pulmonary complications.
Anastomotic leakage is also a common and serious complication after esophagectomy. which does not prolong hospital stay and treatment, but also increase medical expense signi cantly. [17][18][19]. Our results showed that the incidence of anastomotic leakage was signi cantly correlated with RLNP (17.9% vs 7.0%, P=0.029). This may be due to the inability of coughing in patients with RLNP, which lead to the rising of pressure in the thoracic cavity and was transmitted to the anastomosis site when patient tried to cough and expect the sputum hardly [8]. Another possible reason may be poor nutritional status caused by di culty in swallowing, which affects tissue healing [12].
Considering harmful consequences of RLNP, it is impressive to prevent RLNP during esophagectomy. It was reported in literatures that noninvasive intraoperative neurological monitoring (IONM) is safe and feasible, which can help avoid intraoperative RLN injury and reduce the incidence of RLNP [20]. IONM has been a routine application during thyroid surgery to avoid RLN injury, and its positive effects have been proven [21]. Therefore, IONM may enable surgeons to identify and preserve the RLN during lymph nodes dissection near bilateral RLN. The usage of energy instruments such as electric knife, electric hook, and ultrasonic knife brings great convenience to the operation, which can signi cantly reduce surgical hemostasis and operation time [22]. However, thermal injury during the use of energy instruments is a substantial reason of RLNP. Therefore, direct contact with the RLN should be avoided and a safe distance of more than 3mm should be maintained when using energy instruments [23]. And it has been proven that the use of non-energy devices such as scissors and hemostatic clips can reduce the incidence of RLNP [24][25][26].
One limitation of this study is that this is a retrospective study and all patients included were from a single center and operated by a group of surgeons with difference in surgical skills and experience. Another limitation is that diagnosis of RLNP were mainly based on clinical symptoms instead of laryngoscopy.
RLNP after esophagectomy is signi cantly associated with an increased incidence of pulmonary complication, anastomotic leakage and much longer hospital stay. It is impressive that renovation and usage of new technologies are needed to reduce RLNP.