Abstract
Background/Aim: Recurrent laryngeal nerve paralysis (RLNP) induces aspiration pneumonia and reduces the patient’s quality of life. To decrease the incidence of RLNP, we performed intraoperative neural monitoring (IONM) during thoracoscopic surgery for esophageal cancer and evaluated its usefulness. Patients and Methods: A total of 737 consecutive patients who underwent thoracoscopic surgery for esophageal cancer were enrolled in this study. Between May 1995 and March 2016, thoracoscopic esophagectomies were performed using video-assisted thoracoscopic surgery (VATS) with a small incision, whereas from April to June 2023, we used positive pressure pneumothorax with port placement only [minimum invasive esophagectomy (MIE)]. A total of 110 consecutive patients who underwent thoracoscopic surgery with IONM (IONM group) were retrospectively compared with those who underwent VATS or MIE without IONM (No-IONM group). Results: The incidence of RLNP [Clavien–Dindo (CD) classification of ≥1] on postoperative day (POD) 5 was 13.9% in the IONM group, which was significantly lower than that of the no-IONM group (31.2%, p<0.001). Even when comparing only patients who underwent MIE, the incidence of RLNP on POD5 was 13.9% in the IONM group, which was significantly lower than that in the no-IONM group (26.2%, p=0.035). The incidence of postoperative pneumonia (CD ≥2) was 10.9% in the IONM group, which was significantly lower than that in the no-IONM group (26.1%, p=0.005). Bilateral RLNP did not occur in any of the IONM groups. Conclusion: IONM is a useful tool for reducing RLNP incidence and postoperative pneumonia after thoracoscopic surgery for esophageal cancer.
- Intraoperative neural monitoring
- recurrent laryngeal nerve paralysis
- thoracoscopic esophagectomy
- esophageal cancer
- postoperative pneumonia
Esophageal cancer is the seventh leading cause of cancer, and the sixth cause of cancer death worldwide (1). Surgery is the mainstay of treatment for resectable, locally advanced esophageal cancer. In Japan, squamous cell carcinoma accounts for approximately 90% of cases (2), and neoadjuvant chemotherapy and surgery are standard treatments for resectable locally advanced esophageal cancer (3, 4). In thoracic esophageal cancer, metastases to the lymph nodes around the recurrent laryngeal nerve (RLN) often occur, and upper mediastinal lymph node dissection is commonly performed (5, 6). However, recurrent laryngeal nerve paralysis (RLNP) is associated with lymphadenectomy around the RLN.
In 1992, thoracoscopic surgery for esophageal cancer was reported (7) and introduced worldwide. In Japan, approximately 70% of patients with thoracic esophageal cancer undergo thoracoscopic surgery (8). However, the Japanese National Clinical Database has reported that thoracoscopic surgery exhibits a significantly higher RLNP incidence than conventional open surgery (9, 10). Additionally, RLNP induces aspiration pneumonia and reduces the patient’s quality of life (11), with such postoperative complications resulting in poor long-term prognosis (12, 13). As intraoperative neural monitoring (IONM) has been introduced in recent years in thyroid surgery to avoid RLN injury (14, 15), we introduced IONM during thoracoscopic surgery for esophageal cancer and evaluated its usefulness in reducing the incidence of RLNP.
Patients and Methods
Patients. A total of 737 consecutive patients who underwent thoracoscopic surgery for esophageal cancer from May 1995 to June 2023 at Osaka Metropolitan University Hospital were enrolled in this study.
Surgical procedure. At our hospital, thoracoscopic surgery is performed in the left lateral decubitus position, which facilitates quick transition to thoracotomy, in order to ensure safety. From May 1995 to March 2016, video-assisted thoracoscopic surgery (VATS) with a small incision of less than 5 cm between the fifth intercostal space was performed (VATS group, n=539). From April 2016, thoracoscopic surgery under positive pressure pneumothorax with only port placement without small incisions (minimum invasive esophagectomy; MIE) (MIE group, n=198) was performed. To ensure safety, we performed thoracoscopic surgery in line with how it is implemented at our hospital, that is, in the left lateral decubitus position, which enables a quick transition to thoracotomy. From May 1995 to March 2016, VATS was performed using a small incision of <5 cm in the fifth intercostal space (VATS group, n=539). From April 2016 to June 2023, thoracoscopic surgery was performed under positive pressure pneumothorax with port placement only and without small incisions [minimum invasive esophagectomy (MIE)] (MIE group, n=198).
Data collection. We collected clinical data from all patients and evaluated the progression of esophageal cancer according to the Union for International Cancer Control TNM staging system (8th edition). The severity of postoperative complications was evaluated using Clavien–Dindo (CD) classification.
IONM. IONM was performed using the NIM response® system 3.0 (Medtronic Inc. Jacksonville, FL, USA). The patients who underwent IONM were intubated under general anesthesia using the NIM TriVantageTM® Electromyography (EMG) tube (Medtronic Inc.). A balloon blocker was placed in the right main bronchus and thoracoscopic surgery was performed with left unilateral lung ventilation. IONM was not performed unless left unilateral ventilation using a blocker was implemented. No muscle relaxants were used during thoracoscopic surgery. Nerve stimulation was performed by intermittently using a hand-held monopolar stimulator (Medtronic Inc.). The RLN position was investigated at 2.0-3.0 mA and confirmed upon stimulation at 0.5 mA (Figure 1). Loss of signal (LOS) with electrical stimulation over the entire length or on the opposite RLN side may indicate intubation tube dislocation and disappears when the anesthesiologist adjusts the position of the intubation tube. When the EMG activity with electrical stimulation remained on the peripheral, but not on the central, RLN side, LOS was confirmed, and neural injury was suspected. IONM was performed before, during, and after the dissection of the lymph nodes around the right and left RLN, as well as at the end of thoracoscopic surgery. The IONM and no-IONM groups consisted of 110 and 627 patients, respectively. Patients in the MIE group who did or did not undergo IONM were further divided into the IONM-MIE (n=110) and no-IONM-MIE (n=88) groups, respectively.
Operative field of intraoperative neural monitoring with a hand-held monopolar stimulator using the intermittent method. A) Electrical stimulation with a hand-held monopolar stimulator allows for the determination of the right recurrent laryngeal nerve (RLN) (white arrow) and its esophageal branch (black arrow). B) A case in which electrical stimulation revealed bifurcation of the right RLN (white arrows). C) The location of the left RLN can be estimated by electrical stimulation in the surgical field before the left RLN is exposed. D) Electrical stimulation facilitates the differentiation of the left RLN (white arrow) and left cardiac branch of the vagus nerve (black arrow).
RLNP determination. Postoperative RLNP was determined using laryngoscopy at the time of extubation on the day after surgery. If RLNP determination was challenging at that time, it was repeated up to five days after surgery. If judging RLNP continued to be challenging, we asked otolaryngologists to make a judgment.
Statistical analysis. All statistical analyses were performed using SPSS® version 27 (IBM, Armonk, NY, USA). The frequencies of the categorical variables were compared using Pearson’s chi-squared test, whereas continuous variables were compared using the Mann–Whitney U-test. Cumulative overall survival rates were identified using the Kaplan–Meier method, and log-rank statistics were used for comparisons between the groups. A p-value of <0.05 denoted statistical significance.
Results
IONM was attempted in 115 patients; however, it could not be performed in five patients (4.3%), in whom left unilateral lung ventilation with a blocker was infeasible. RLN could be identified bilaterally in all cases using a hand-held monopolar stimulator. Twenty-one (19.1%) of the 110 patients had LOS during the thoracic procedure, of whom only 12 (10.9%) developed postoperative RLNP, 89 (80.9%) had no LOS, and two (2.2%) developed postoperative RLNP (sensitivity, 85.7%; specificity, 90.6%; positive predictive value, 57.1%; negative predictive value, 97.8%). In patients with RLNP, postoperative pneumonia (CD ≥2 or CD ≥3) was found in 38.7% and 24.0%, respectively, which was significantly higher than that in patients without RLNP (12.3% and 4.7%, respectively; p<0.001). The postoperative hospital stay of patients with RLNP (days, median/range) was 35/4-407, which was significantly longer than that of patients without RLNP (29/11-389 days; p=0.001).
Clinical characteristics of the IONM group and no-IONM groups are shown in Table I. Although there was no significant difference in sex and pathological type, there were significant differences in age, tumor location, degree of progression, prior treatments, clinical stage (cStage) and presence of preoperative RLNP. All patients in the IONM group underwent MIE, and most patients in the no-IONM group underwent VATS.
Patient characteristics of intraoperative neural monitoring (IONM) group and No-IONM group.
The clinical characteristics of the IONM and no-IONM groups are shown in Table I. Although there were no significant differences in sex and pathological type, there were significant differences in age, tumor location, degree of progression, prior treatments, clinical stage (cStage), and presence of preoperative RLNP. All patients in the IONM group underwent MIE, and most patients in the no-IONM group underwent VATS.
As clinical characteristics differed greatly between the IONM and no-IONM groups, the MIE- and VATS-group clinical characteristics are presented in Table II. Similarly, no significant difference was observed in sex or pathological type, whereas significant differences were found in age, tumor location, degree of cStage, prior treatments, and presence of preoperative RLNP.
Patient characteristics of MIE group and VATS group.
Table III displays the postoperative results in the IONM and no-IONM groups. Significant differences were uncovered in the degree of pathological stage (pStage), incidence of all postoperative RLNP, late-onset RLNP, RLNP up to POD5 cases in which recovery from RLNP could be evaluated, and postoperative pneumonia (CD ≥2). Significant differences were also observed in thoracic operation times, amount of bleeding during the thoracic procedures, and number of mediastinal- and superior-mediastinal-dissection lymph nodes.
Surgical outcomes of IONM group and No-IONM group.
Table IV shows the postoperative results of the MIE and VATS groups. Similarly, significant differences were detected in the degree of pStage, incidence of all postoperative RLNP, late-onset RLNP after POD6, RLNP up to POD5, number of cases in which recovery from RLNP could be evaluated, thoracic operation time, amount of bleeding during thoracic procedures, as well as number of mediastinal- and superior-mediastinal-dissection lymph nodes. There was no significant difference in the incidence of postoperative pneumonia (CD ≥2) between the MIE and VATS groups. A significant difference in the days of recovery from RLNP without resection of RLN was observed between the MIE and VATS groups, albeit not between the IONM and no-IONM groups.
Surgical outcomes of MIE group and VATS group.
Table V shows the clinical characteristics and postoperative results of the IONM-MIE and no-IONM-MIE groups. No significant differences were observed in sex, age, tumor location, prior treatments, degree of cStage, degree of pStage, resection of RLN, late-onset RLNP after POD6, cases in which recovery from RLNP could be evaluated, recovery from RLNP without resection of RLN, days of recovery from RLNP without resection of RLN, thoracic operation time, amount of bleeding during thoracic procedure, or number of mediastinal- and superior-mediastinal-dissection lymph nodes. However, there was significantly more adenocarcinoma in the no-IONM-MIE group (p=0.019), as well as significant differences in RLNP incidence (CD ≥1) up to POD5 without RLN resection or postoperative pneumonia (CD ≥2). No significant difference in recovery or days of recovery from RLNP without RLN resection, or postoperative hospital stay was detected. Bilateral RLNPs were fewer in the IONM-MIE (no patient) than those in the no-IONM-MIE group, albeit not significantly (p=0.082).
Patient characteristics and surgical outcomes of IONM-MIE group and No-IONM-MIE group.
In the IONM-MIE group, RLNP (CD ≥1) up to POD5 without RLN resection occurred in 8/34 patients (23.5%) after the introduction of IONM and decreased significantly in only six cases (9.6%) after the 35th patient (Table VI). In the no-IONM-MIE group during the same period, RLNP (CD ≥1) on POD5 without resection of RLN occurred in nine (31.0%) and decreased in 11 (20.0%) patients, albeit not significantly (p=0.076).
Incidence of postoperative RLNP within POD5 without RLN resection in MIE group.
Figure 2 shows the Kaplan–Meier overall survival curves for each RLNP site. Regarding bilateral RLNP, the overall survival curves for patients with right-only (p=0.048) and left-only (p=0.006) RLNP were significantly lower than those of patients without RLNP (p=0.002).
Overall survival curve of patients with recurrent laryngeal nerve paralysis (RLNP). The Kaplan–Meier overall survival curves for each RLNP site. The overall survival curve of patients with bilateral RLNP differed between those without RLNP (p=0.002) and those with right-only RLNP (p=0.048) or left-only RLNP (p=0.006) for whom it was significantly lower.
Discussion
Esophageal cancer surgery is a particularly invasive gastrointestinal surgical procedure. To reduce its invasiveness, we have introduced since 1995 thoracoscopic surgery for esophageal cancer and reported respiratory function improvement without observing any long-term changes in the results (16-18). However, the Japanese National Clinical Database reported that the incidence of RLNP remains significantly higher than that of open thoracotomy (9, 10). Due to differences in evaluation methods among institutions, the incidence of postoperative RLNP varies widely (10-70%) (9, 10, 19-27). Additionally, RLNP has been reported to induce postoperative pneumonia (11, 27). In this study, postoperative pneumonia (CD ≥2 and CD ≥3) was significantly more common in patients with RLNP, who also had a significantly longer postoperative hospital stay than those without RLNP. Based on these results, from 2016 to 2023, we performed thoracoscopic surgery in the left lateral decubitus position with port placement only under artificial positive pressure pneumothorax without incision to further reduce surgical invasiveness and IONM.
Presently, the IONM effectiveness in cancer esophagectomy remains controversial. IONM for esophageal cancer surgery has been reported (20-27) and there have been only six reports with 30 or more cases (22-27). Among these, Hikage et al. (23) with 51 cases, and Yuda et al. (25) with 41 cases showed no significant decrease in RLNP incidence, whereas Takeda et al. (26) with 83 cases showed a significant reduction in RLNP incidence (only CD ≥2). Similarly, Zhong et al. (22) with 54 cases, Kobayashi et al. (24) with 31 cases, and Zhao et al. (27) with 70 cases showed significant decreases in RNLP incidence. In this study of 110 cases, that is the largest number ever reported, the incidence of RLNP was significantly reduced in the IONM group than in the no-IONM group (Table III).
There were large differences between the MIE and the VATS groups, as shown in Table II and Table IV. Notably, the VATS group consisted of cases from May 1995 to March 2016, and the MIE group of cases from April 2016 to June 2023. In our opinion, a comparison between the No-IONM group and no-IONM group consisting mostly of the VATS group would not lead to a rigorous evaluation of IONM. The NIM response system, which is necessary for performing IONM, is often used for thyroid surgery or neurosurgery in our hospital, thus, IONM could not be performed in all patients who underwent MIE. As propensity score matching failed to align the surgical method with the historical background, we divided the MIE group into two groups (the IONM-MIE and no-IONM-MIE group) and reexamined the patients. Upon reexamination of the MIE group, as shown in Table V, the incidence of postoperative RLNP, excluding cases of combined RLN resection and late-onset RLNP, as well as postoperative pneumonia (CD ≥2), were significantly reduced in the IONM-MIE group. Therefore, IONM is considered a particularly useful option during thoracoscopic surgery for esophageal cancer.
In this study, the postoperative hospital stay was significantly shorter in the IONM group compared to the no-IONM group, which included a larger proportion of the VATS group; however, no difference was found in the postoperative hospital stay between the IONM-MIE and no-IONM-MIE groups. We reasoned that the longer postoperative hospital stay in the VATS group compared to that in the MIE group may reflect historical factors. IONM significantly reduced the incidence of RLNP but did not shorten the postoperative hospital stay. According to Wang et al.’s meta-analysis, IONM did not lead to a significant reduction in postoperative hospital stay either (28). Our results showed no significant difference in the frequency of severe pneumonia (CD ≥3). We argue that this could explain why the postoperative hospital stay was not shortened.
Of the 110 patients who underwent IONM, 21 (19.1%) had LOS during thoracoscopic surgery, but nine patients (8.2%) had no RLNP on the next day. The LOS positive predictive value in this study was only 57%, whereas for Kobayashi et al. (24) it was 100%, for Takeda et al. (26) 60%, for Hikage et al. (23) 78% (albeit limited to the left side), and for Yuda et al. (25) 75%.
Seddon et al. classified peripheral nerve paralysis into neurotmesis, axonotmesis, and neurapraxia and proposed neurapraxia to be a short-lived paralysis that could not be explained by true regeneration (29). Carballo et al. stated that neurapraxia was the mildest type of peripheral nerve injury commonly induced by focal demyelination or ischemia, and neurapraxic injuries generally had a good prognosis (30). Therefore, when performing IONM, neurapraxia should be considered and careful surgical manipulation should be continued, bearing in mind that even if LOS is observed during surgery, it may be temporary.
In this study, late-onset RLNP was observed in nine patients: seven patients (3.5%) in the MIE group and two patients (0.4%) in the VATS group. These 7/9 patients had anastomotic leakage and RLNP on the left side near the anastomosis site. We considered that late-onset RLNP was due to the spread of infection from this anastomotic leakage. In the VATS group, late-onset RLNP and recovery from RLNP without RLN resection may have occurred at a low rate due to a lack of clear medical record documentation of older patients. Additionally, the number of dissected lymph nodes of the mediastinum and upper mediastinum was lower in the IONM group than that in the no-IONM group, which mainly consisted of the VATS group. A possible reason for this could be that our esophageal surgery team members changed when MIE was introduced. Thus, the number of dissected abdominal lymph nodes in the MIE group was lower than that in the VATS group (Table IV) and there was no significant difference in the number of dissected lymph nodes of the mediastinum between the IONM-MIE and no-IONM-MIE groups (Table V).
A learning curve also exists for IONM. The incidence of RLNP significantly decreased in up to 34 cases (14 months) after IONM was introduced, as well as in subsequent cases (Table VI). We reasoned that a certain level of experience and skill was required to successfully use IONM. In addition, when evaluating the no-IONM-MIE cases 14 months after the introduction of IONM, we identified no significant difference between the 34 cases up to the introduction of IONM and the cases after that, and the RNLP incidence did not decrease significantly (p=0.076). Nevertheless, we were interested in investigating which procedures are dangerous without IONM usage by acquiring experience with IONM.
Regarding bilateral RLNP, patient prognosis was poor for those without RLNP and for those with unilateral RLNP (Figure 2). In addition, apart from the decreased RLNP incidence after thoracoscopic surgery for esophageal cancer, none of patients in the IONM group had bilateral RLNP. This suggests that IONM may be useful in suppressing the occurrence of bilateral RLNP.
This study has the limitation of being a single-center, retrospective observational study. Future multicenter prospective studies are required to clarify the usefulness of IONM in thoracoscopic surgery for esophageal cancer.
Conclusion
In conclusion, the incidences of postoperative RLNP and pneumonia were significantly lower in the IONM group. IONM is considered useful during thoracoscopic surgery for esophageal cancer. It is important to be aware of the presence of neurapraxia and to continue to perform careful surgical manipulation if LOS occurs during IONM.
Acknowledgements
The Authors would like to thank Editage (www.editage.com) for English language editing.
Footnotes
Authors’ Contributions
S.L. conceived and designed the study, acquired, analyzed, and interpreted the data, confirmed the authenticity of the data, and drafted the manuscript. Y.F., K.G., T.Ta. assisted with the surgery and analyzed the data. T.To., Y.M., M.Y., H.K., T.F. and M.S. acquired and analyzed the data. H.O. made substantial contributions to the conception and design of the study. K.M. contributed to the conception and design of the study and revised the manuscript critically. All Authors read and approved the final manuscript.
Conflicts of Interest
There are no financial or other interests with regard to the submitted manuscript that might be regarded as conflicts of interest.
Funding
No external funding was received for this study.
- Received September 24, 2023.
- Revision received November 29, 2023.
- Accepted November 30, 2023.
- Copyright © 2024 International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.
This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY-NC-ND) 4.0 international license (https://creativecommons.org/licenses/by-nc-nd/4.0).
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.