Abstract
Objective
Thymomas sometimes are located in the cervical region, and strategies are needed to safely remove these tumors. The purpose of this study was to evaluate the feasibility and safety of subxiphoid robotic thymectomy (SRT) for anterior mediastinal tumors located in cervical region.
Methods
This was a retrospective database review of patients who underwent SRT from January 2011 to April 2024. Of the 81 patients who underwent SRT for anterior mediastinal tumors, 79 patients were included, excluding 2 patients who underwent reconstruction using artificial blood vessels.
Results
In total, 41 patients in whom part of the tumor was located above or in contact with the innominate vein were classified as group A, and 38 patients in which the tumor was located caudal to the innominate vein were classified as group B. The patients in group A were significantly younger (median, 54 years vs 63 years; P = .035). There were no differences in perioperative data between the 2 groups: operative time (median, 192 vs 188 minutes; P = .961), intraoperative blood loss (median: 5 vs 5 g; P = .235), combined resection rate (17.1% vs 21.1%; P = .776), duration of thoracic drain insertion (median: 1 vs 1 day; P = .221), postoperative hospital stay (median: 4 vs 4 day; P = .694), and postoperative complications (9.8% vs 5.3%; P = .677). Conversion to other approaches and perioperative mortality were not observed in either group.
Conclusions
SRT, which enables full visualization of the innominate vein from the midline view, may offer increased safety for complicated thymectomies, including those requiring combined resection of the innominate vein, and may demonstrate safety comparable with that of procedures performed for cases caudal to the innominate vein.
Key Words: subxiphoid, robotic, thymectomy, mediastinal, minimally invasive surgery
Subxiphoid robotic thymectomy for tumors located in cervical region.
Central Message.
Subxiphoid robotic thymectomy can be safely performed for anterior mediastinal tumors located in cervical region, including combined resection of the innominate vein, pericardium, and lung.
Perspective.
Subxiphoid robotic thymectomy, in which the entire innominate vein can be seen from the midline view, allows for the safe treatment of anterior mediastinal tumors located in cervical region. We believe that subxiphoid robotic thymectomy will play a major role in minimally invasive surgery for anterior mediastinal tumors because of its ability to perform combined resection of invaded organs safely.
Surgery for anterior mediastinal tumors has shifted from median sternotomy to minimally invasive procedures such as video-assisted thoracoscopic surgery (VATS) and robot-assisted thoracic surgery (RATS).1, 2, 3 Surgery for anterior mediastinal tumors sometimes requires complicated procedures, including combined resection of adjacent organs and reconstruction of the pericardium and blood vessels.
Currently, VATS and RATS for anterior mediastinal tumors are generally performed via a lateral thoracic intercostal approach.4,5 In 2015, we developed subxiphoid robotic thymectomy (SRT), in which a camera is inserted through a subxiphoid incision and the arms are inserted through the intercostal spaces on both sides.6 SRT facilitates bilateral phrenic nerve and neck visualization using a camera inserted through the midline of the body below the xiphoid process (Figure 1). The superior maneuverability of the robotic system enables more complicated and minimally invasive procedures, such as combined resection of the pericardium, lung, innominate vein, and superior vena cava.7 With the lateral thoracic approach, if the tumor is in contact with or located above the innominate vein, the tumor and thymus can obstruct the surgeon's view, making it impossible to visualize the innominate vein distal to the tumor. Moreover, clamping the distal innominate vein becomes impossible if bleeding occurs. Therefore, the lateral thoracic approach is not recommended for tumors in contact with or located cranial to the innominate vein. In contrast, the field of view of the SRT camera, which is inserted from the midline of the body, enables observation of the entire innominate vein, making it possible to tape or block both the central and peripheral sides of the tumor, resulting in safer surgery. In this study, we hypothesized that SRT, which is easy to operate and allows for observation of the entire innominate vein, would be as safe as in cases of tumors on the caudal side, which are not in contact with the innominate vein, a relatively easy procedure to perform. The aims of this study were to compare the perioperative outcomes of cases in which SRT was performed for anterior mediastinal tumors that were in contact with the innominate vein or located cranial to the innominate vein with the outcomes of cases in which the tumor was not in contact with the innominate vein, and to verify the safety of SRT for anterior mediastinal tumors that progress to the cervical region.
Figure 1.
Midline view. A, ∗Right brachiocephalic artery, ∗∗trachea, and ∗∗∗innominate vein from the midline view. B, Right phrenic nerve from the midline view. C, Left phrenic nerve from the midline view.
Methods
This study was designed in accordance with the Declaration of Helsinki (revised 2013), and it was approved by the Ethics Committee of Fujita Health University on August 10, 2021 (approval number HM20-566). Written informed consent for publication of their study data was obtained from all patients.
A total of 338 thymectomies were performed at Fujita Health University Hospital (January 2011 to May 2020) and Fujita Health University Okazaki Medical Center (June 2020 to April 2024). Of these, 23 patients underwent median sternotomy, 13 underwent VATS via the lateral intercostal approach, 221 underwent subxiphoid uniportal thymectomy (SUT),8 and 81 underwent SRT. At our hospital, the indications for thymectomy approaches are as follows: for anterior mediastinal tumors that have not invaded other organs or for thymectomy for myasthenia gravis, we apply the most minimally invasive SUT, whereas for cases in which the tumor is in contact with the innominate vein or was located cranial to the innominate vein or in which it is difficult to confirm the location of the phrenic nerve because of the tumor, we apply SRT, which is easier to operate. Tumor size is not an indication for SUT or SRT. Unlike the intercostal incision, which is difficult to enlarge, the subxiphoid incision is easily enlarged, enabling the removal of even large tumors.
The patient's body mass index was not considered in the selection of surgical approach. However, open thoracotomy was selected in cases of suspected arterial invasion, such as the subclavian artery or aorta, and in cases of suspected invasion into the subclavian vein. Of the 81 cases in which SRT was performed, 79 were included in this study, excluding 2 cases in which artificial vessel replacement was performed. We did not consider myasthenia gravis without thymoma to be suitable for SRT because we performed SUT. Patients whose tumors were in contact with the innominate vein or were located cranial to the innominate vein were categorized into group A, and those whose tumors were not in contact with the innominate vein but were located caudal to the innominate vein were categorized into group B, to compare the perioperative results (Figure 2). Tumor localization was determined using computed tomography. An example from group A is shown in Figure 3 and Video 1. If the vein is located cranial to the superior margin of the innominate vein, the difficulty of the procedure increases as a result of the need for caution to avoid damaging the innominate vein. In group A, there were 18 cases (43.9%) in which the innominate vein was located cranially to the upper border of the innominate vein. The evaluation items included operative time, blood loss, combined resection, drainage period, postoperative hospital stay, postoperative complications of Clavien-Dindo classification,9 and perioperative death.
Figure 2.
Flowchart demonstrating patient inclusion and exclusion. VATS, Video-assisted thoracoscopic surgery; SUT, subxiphoid uniportal thymectomy; SRT, subxiphoid robotic thymectomy.
Figure 3.
Subxiphoid robotic thymectomy for thymic cyst that has extended to the neck. A, The tumor displaced the trachea to the right side. B, The innominate vein was displaced caudally by the tumor.
Surgical Procedures
The SRT was performed as described in our previous articles.6,10,11 The technique is described in brief in the section that follows. All surgeries were performed with the patient receiving general anesthesia. Robotic surgery was performed using either the da Vinci Si or da Vinci Xi surgical system (Intuitive Surgical), with 3 types of surgeries: 3-port, 4-port, and uniportal (Figure 4). A 3-cm incision was made below the xiphoid process, and a Lapsingle (Sejong Medical Co) was placed. When SRT was initiated, it was performed with 3 ports by inserting an 8-mm port on each side of the sixth intercostal anterior axillary line in addition to the subxiphoid wound; however, from April 2018, the surgery was performed with 4 ports by adding an 8-mm da Vinci port in the right sixth intercostal space because the surgical field deployment with a retraction arm allowed the surgical operation to be safer and more complicated. In September 2022, we started preforming subxiphoid uniportal robotic thymectomy using the da Vinci Xi surgical system, in which a 4-cm longitudinal incision was made under the xiphoid process, and 3 arms were used to operate on 1 wound. In surgery for tumors that were in contact with or located cranial to the innominate vein, the central and peripheral sides of the innominate vein were taped with a thread before dissection around the innominate vein.
Figure 4.
A, Four-port subxiphoid robotic thymectomy. B, Subxiphoid uniportal robotic thymectomy.
Statistical Analysis
Categorical variables were reported as counts and percentages of patients. Continuous data are presented as medians and interquartile ranges (IQRs). Categorical variables were analyzed using the Fisher exact test, and continuous variables were analyzed using the Wilcoxon rank-sum test. Heterogeneity in baseline characteristics between the 2 groups was balanced by performing 1:1 propensity score matching (PSM) with a caliper width of 0.20. The caliper value of 0.20 was adopted to adequately reduce bias while maintaining a sufficient number of matched cases.12 Propensity scores were estimated using a logistic regression model that included age, sex, body mass index, Brinkman index, performance status, myasthenia gravis, tumor size, surgical procedure, pathologic diagnosis, and surgical method. Group comparisons were performed using the Fisher exact test or the Wilcoxon rank-sum test, and logistic regression was applied as a sensitivity analysis. Multivariable regression was performed as part of the sensitivity analysis in the full cohort, and a separate analysis was conducted in the propensity score−matched population. Because this study was an exploratory design, adjustments for multiple comparisons and imputation of missing data were not conducted. All tests were 2-tailed. Statistical analyses were conducted using the SAS software (version 9.4, SAS Institute Inc).
Results
The patient backgrounds of the 2 groups are presented in Table 1. There were 41 and 38 patients in groups A and B, respectively. The median age was 54 years (IQR, 43-65 years) in group A and 63 years (IQR, 51-70 years) in group B. The patients in group A were significantly younger (P = .035). There were no differences between the 2 groups in terms of sex, body mass index, Brinkman index, performance status, myasthenia gravis, tumor size, surgical procedure, surgical method, pathologic diagnosis of the anterior mediastinal tumor, or the Masaoka classification of thymoma. The operative and postoperative outcomes are shown in Table 2. There were no differences in perioperative data between the 2 groups: operative time (median [IQR], 192 [138-235] minutes vs 188 [136-234] minutes; P = .961), intraoperative blood loss (median: 5 [5-10] g vs 5 [5-30] g; P = .235), combined resection rate (17.1% vs 21.0%; P = .776), duration of thoracic drain insertion (median [IQR]: 1 [0-1] days vs 1 [0-1] days; P = .221), postoperative hospital stay (median [IQR]: 4 [3-6] days vs 4 [3-6] days; P = .694), postoperative complications (9.8% vs 5.3%; P = .677), and number of analgesic tablets (median [IQR]: 54 [36-98] tablets vs 42 [27-81] tablets; P = .272).
Table 1.
Patient background before and after 1:1 propensity score matching
| Characteristics | Before propensity score matching |
After propensity score matching |
||||||
|---|---|---|---|---|---|---|---|---|
| Group A (n = 41) | Group B (n = 38) | P value | Absolute standardized difference | Group A (n = 27) | Group B (n = 27) | P value | Absolute standardized difference | |
| Age, y, median (IQR) | 54 (43-65) | 63 (51-70) | .035 | 0.563 | 56 (50-76) | 58 (49-68) | .917 | 0.107 |
| Sex | .501 | 0.169 | 1.000 | 0.075 | ||||
| Male | 16 (39.0) | 18 (47.4) | 11 (40.7) | 12 (44.4) | ||||
| Female | 25 (61.0) | 20 (52.6) | 16 (59.3) | 15 (55.6) | ||||
| BMI, kg/m2, median (IQR) | 21.9 (19.4-23.6) | 22.9 (20.2-25.9) | .109 | 0.365 | 22.3 (19.5-24.3) | 22.5 (19.7-26.1) | .556 | 0.177 |
| Brinkman index, median (IQR) | 0 (0-200) | 0 (0-480) | .807 | 0.183 | 0 (0-200) | 0 (0-260) | .839 | 0.026 |
| Comorbidity, n (%) | .165 | 0.326 | 1.000 | 0.000 | ||||
| + | 11 (26.8) | 16 (42.1) | 10 (37.0) | 9 (33.3) | ||||
| − | 30 (73.2) | 22 (57.9) | 17 (63.0) | 18 (66.7) | ||||
| PS | .737 | 0.233 | 1.000 | 0.000 | ||||
| 0 | 40 (97.6) | 36 (94.7) | 26 (96.3) | 26 (96.3) | ||||
| 1 | 0 (0.0) | 1 (2.6) | 0 (0.0) | 0 (0.0) | ||||
| 2 | 0 (0.0) | 0 (0.0) | 0 (0.0) | 0 (0.0) | ||||
| 3 | 1 (2.4) | 1 (2.6) | 1 (3.7) | 1 (3.7) | ||||
| Myasthenia gravis, n (%) | 4 (9.8) | 4 (10.5) | 1.000 | 0.026 | 3 (11.1) | 2 (7.4) | 1.000 | 0.128 |
| Tumor size, cm, median (IQR) | 5.0 (3.5-6.3) | 5.0 (3.0-7.4) | .750 | 0.112 | 5.0 (3.4-7.0) | 4.5 (2.6-8.0) | .782 | 0.000 |
| Surgical procedure, n (%) | 1.000 | 0.026 | 1.000 | 0.128 | ||||
| Thymectomy | 37 (90.2) | 34 (89.5) | 24 (88.9) | 25 (92.6) | ||||
| Extended thymectomy | 4 (9.8) | 4 (10.5) | 3 (11.1) | 2 (7.4) | ||||
| Pathologic diagnosis, n (%) | .231 | 0.481 | .816 | 0.329 | ||||
| Thymoma | 20 (48.8) | 25 (65.8) | 17 (63.0) | 19 (70.4) | ||||
| Thymic carcinoma | 3 (7.3) | 4 (10.5) | 3 (11.1) | 1 (3.7) | ||||
| Thymic cyst | 9 (22.0) | 6 (15.8) | 4 (14.8) | 5 (18.5) | ||||
| Others | 9 (22.0) | 3 (7.9) | 3 (11.1) | 2 (7.4) | ||||
| Masaoka stage, n (%) | 1.000 | 0.061 | .598 | 0.370 | ||||
| I | 9 (39.1) | 11 (36.7) | 8 (42.1) | 5 (25.0) | ||||
| II | 12 (52.2) | 16 (53.3) | 9 (47.4) | 12 (60.0) | ||||
| III | 2 (8.7) | 3 (10.0) | 2 (10.5) | 3 (15.0) | ||||
| WHO classification, n (%) | .985 | 0.178 | .965 | 0.048 | ||||
| A | 2 (8.7) | 3 (10.0) | 1 (5.3) | 2 (10.0) | ||||
| AB | 4 (17.4) | 5 (16.7) | 3 (15.8) | 3 (15.0) | ||||
| B1 | 6 (26.1) | 5 (16.7) | 5 (26.3) | 4 (20.0) | ||||
| B2 | 6 (26.1) | 9 (30.0) | 5 (26.3) | 5 (25.0) | ||||
| B3 | 2 (8.7) | 4 (13.3) | 2 (10.5) | 4 (20.0) | ||||
| Thymic cancer | 3 (13.0) | 4 (13.3) | 3 (15.8) | 2 (10.0) | ||||
| Surgical method, n (%) | .383 | 0.338 | 1.000 | 0.000 | ||||
| Uniportal | 4 (9.8) | 7 (18.4) | 4 (14.8) | 4 (14.8) | ||||
| 3 ports | 4 (9.8) | 6 (15.8) | 4 (14.8) | 4 (14.8) | ||||
| 4 ports | 33 (80.5) | 25 (65.8) | 19 (70.4) | 19 (70.4) | ||||
IQR, Interquartile range; BMI, body mass index; PS, performance status; WHO, World Health Organization.
Table 2.
Operative and postoperative outcomes before and after 1:1 propensity score matching
| Variables | Before propensity score matching |
After propensity score matching |
||||
|---|---|---|---|---|---|---|
| Group A (n = 41) | Group B (n = 38) | P value | Group A (n = 27) | Group B (n = 27) | P value | |
| Operative time, min, median (IQR) | 192 (138-235) | 188 (136-234) | .961 | 183 (133-241) | 184 (136-290) | .736 |
| Intraoperative blood loss, g, median (IQR) | 5 (5-10) | 5 (5-30) | .235 | 5 (4-10) | 5 (5-40) | .072 |
| Conversion | 0 (0) | 0 (0) | ||||
| Combined resection, n (%) | 7 (17.1) | 8 (21.1) | .776 | 4 (14.8) | 6 (22.2) | .728 |
| Pericardium | 5 (12.2) | 6 (15.8) | 4 (14.8) | 5 (18.5) | ||
| Lung | 3 (7.3) | 6 (15.8) | 1 (3.7) | 5 (18.5) | ||
| Innominate vein | 6 (14.6) | 3 (7.9) | 3 (111.1) | 3 (11.1) | ||
| Phrenic nerve | 1 (2.4) | 1 (2.6) | 1 (3.7) | 1 (3.7) | ||
| Duration of chest tube, d, median (IQR) | 1 (0-1) | 1 (0-1) | .221 | 1 (0-1) | 1 (0-1) | .459 |
| Postoperative hospital stay, d, median (IQR) | 4 (3-6) | 4 (3-6) | .694 | 4 (3-5) | 4 (3-7) | .993 |
| Postoperative complications, n (%) | 4 (9.8) | 2 (5.3) | .677 | 1 (3.7) | 0 (0) | 1.000 |
| Postoperative bleeding | 1 (2.4) | 0 (0) | 0 (0) | 0 (0) | ||
| Chylothorax | 1 (2.4) | 0 (0) | 0 (0) | 0 (0) | ||
| Atrial fibrillation | 1 (2.4) | 0 (0) | 1 (3.7) | 0 (0) | ||
| Recurrent nerve paralysis | 1 (2.4) | 0 (0) | 0 (0) | 0 (0) | ||
| Respiratory failure | 0 (0) | 1 (2.6) | 0 (0) | 0 (0) | ||
| Phrenic nerve paralysis | 0 (0) | 1 (2.6) | 0 (0) | 0 (0) | ||
| Number of analgesic tablets, tablets, median (IQR) | 54 (36-98) | 42 (27-81) | .272 | 45 (33-75) | 45 (21-81) | .721 |
| Perioperative death | 0 (0) | 0 (0) | ||||
IQR, Interquartile range.
The number of combined resections was 7 in group A and 8 in group B. Five patients in group A and 4 in group B required combined resection of multiple organs. In group A, 1 case required combined resection of 4 organs (innominate vein, phrenic nerve, pericardium, and lung); 1 case required combined resection of 3 organs (innominate vein, pericardium, and lung); 2 cases required combined resection of the innominate vein and pericardium; and 1 case required combined resection of the innominate vein and lung. In group B, there were 3 cases of combined resection of the innominate vein, pericardium, and lung and 1 case of combined resection of the pericardium, lung, and phrenic nerve. In group B, the innominate vein combined resection involved a tumor that did not contact the innominate vein but was located just caudal to it; the innominate vein was removed as a precaution to ensure safe dissection of the thymic vein. Postoperative complications occurred in four patients in group A and 2 patients in group B. The complications were as follows: recurrent laryngeal nerve paralysis was grade 1; respiratory failure, postoperative bleeding, and atrial fibrillation were grade 2; and chylothorax and phrenic nerve paralysis were grade 3b. Phrenic nerve paralysis occurred in group B, but this was not due to technical reasons but rather to combined resection as the result of tumor invasion. Reoperation was required for ligation of the thoracic duct for chylothorax in group A and diaphragmatic plication for phrenic nerve palsy in group B. Among the cases in which combined resection was performed, the only complication was chylothorax in a patient in group A who underwent combined resection of the innominate vein. In group B, respiratory failure was observed in patients who underwent combined resection of the innominate vein, pericardium, and lung, and phrenic nerve paralysis was observed in patients who underwent combined resection of the phrenic nerve. The other cases of complications were those in which no combined resection was performed. An increase in complications is feared when performing complicated surgeries, such as combined resection. However, the complication rates in cases in which combined resection was performed were 14.3% and 12.5% in groups A and B, respectively, in this study. It cannot be said that combined resection necessarily affected the complication rate. We hypothesize that minimally invasive surgery using a robot has made it possible to perform combined resection more safely. All patients with thymic tumors had a resection status of R0. Resection margins were negative in all cases, and no patients received postoperative adjuvant therapy. Conversion to other approaches and perioperative mortality were not observed in either group.
Four surgeons performed the operations described in this study. Surgeon A performed 55 total cases (69.6%), 30 group A cases (73.2%), and 25 group B cases (65.8%). Surgeon B performed 8 total cases (10.1%), 3 group A cases (7.3%), and 5 group B cases (13.2%). Surgeon C performed 15 total cases (19.0%), 8 group A cases (19.5%), and 7 group B cases (18.4%). Surgeon D performed 1 group B case (1.3% of total; 2.6% of group B). Complication rates by surgeon are 9.1% (5 of 55 cases) for surgeon A, 12.5% (1 of 8 cases) for surgeon B, and 0% for surgeons C and D. No difference between surgeons was observed in incidence of complications (P = .432).
After PSM, each group was narrowed to 27 patients. None of the patient background factors, including age (P = .917), differed between the 2 groups. Perioperative data between the 2 groups after PSM showed no significant difference in operative time (median [IQR]: 183 [133-241] minutes vs 184 [136-290] minutes; P = .736), intraoperative blood loss (median [IQR]: 5 [4-10] g vs 5 [5-40] g; P = .072), combined resection rate (14.8% vs 22.2%; P = .728), duration of chest drain insertion (median [IQR]: 1 [0-1] days vs 1 [0-1] days; P = .459), postoperative hospital stay (median [IQR]: 4 [3-5] days vs 4 [3-7] days; P = .993), postoperative complications (3.7% vs 0%; P = 1.000), or number of analgesic tablets (median [IQR]: 45 [33-75] tablets vs 45 [21-81] tablets; P = .721). No conversion to other approaches or perioperative deaths were observed. No significant differences in postoperative outcomes were observed between the 2 groups in the sensitivity analyses (Table E1). A tendency toward a greater odds ratio of postoperative complications was observed in group A (odds ratio, 3.11; 95% confidence interval, 0.11-84.8; P = .501).
Discussion
In this study, we examined the safety of SRT for the treatment of anterior mediastinal tumors located in the cervical region. SRT demonstrated no serious complications or conversion to other approaches, regardless of tumor localization. Thus, SRT can be safely performed for anterior mediastinal tumors located in the cervical region.
Resection of anterior mediastinal tumors located in the cervical region is generally performed using a median sternotomy. This approach is typically selected because it is challenging to safely perform thoracoscopic surgery without damaging the innominate vein, and such surgeries are highly complicated. When an anterior mediastinal tumor is in contact with the innominate vein, great care must be taken to avoid damaging the vein during dissection. For safe surgery, the peripheral and central part of the innominate vein should be taped with a thread or tape to allow for clamping in case of injury to the innominate vein. However, using the lateral thoracic approach, the contralateral innominate vein of the tumor cannot be identified because of the presence of the tumor and thymus, and taping the contralateral innominate vein is therefore infeasible. By contrast, with the subxiphoid approach, the view from the midline of the body allows for observation of the entire innominate vein, and central and peripheral taping of the vein becomes feasible, even when the tumor is in contact with it. It has also been reported that the superior pole of the thymus cannot be sufficiently thoroughly removed using a lateral thoracic approach13,14 because the lateral thoracic approach does not provide sufficient view of the superior pole of the thymus. It is thus thought that tumors located cranial to the innominate vein cannot be safely removed without sufficient visibility. It has been reported that SRT can be safely performed in lieu of the lateral thoracic approach for thymomas located cranial to the innominate vein to achieve a clear visual field,15 but no reports, to our knowledge, describe resection using the lateral thoracic approach alone.
If the tumor is located cranial to the innominate vein, the transcervical approach may be considered. However, it is difficult to safely resect a tumor that is adjacent to or invades the innominate vein when viewed from the transcervical incision. It has also been reported that total thymectomy is difficult using only a transcervical view and requires a combination with other approaches, such as a subxiphoid approach.16 In addition, an incision wound on the neck remains visible and may be undesirable from a cosmetic standpoint. With SRT, a view comparable with that of a median sternotomy can be achieved from the midline of the body, allowing the visualization of innominate veins on both the central and peripheral sides of the tumor. This capability enables surgery to be performed as safely as in cases where the tumor is not in contact with the innominate vein. We did not consider unilateral VATS for tumors abutting or located cranial to the innominate vein to be indicated because we believe that VATS via the lateral thoracic approach does not allow for safe surgery. This is because both the proximal and distal sides of the tumor cannot be encircled or clamped. Therefore, we did not compare unilateral VATS with SRT, which is more commonly used in group A.
Postoperative complications rates of SRT in this study were 9.8% and 5.3% in groups A and B, respectively. These results are comparable with those reported for robot-assisted thymectomy using the lateral thoracic approach, which did not include cases with cervical extension, with complication rates ranging from 5.6% to 19.0%.17,18 In this study, there was no significant difference in operative time, blood loss, or postoperative complications in patients with cervical extension compared with that of those without cervical extension, and SRT can safely be used to treat cervical extension, which is more difficult to resect.
Recently, the subxiphoid approach has attracted attention as a surgical approach for thymectomies. The subxiphoid approach includes multiport VATS, SUT, multiport SRT, and uniportal SRT.6,8,19,20 SUT, without intercostal passage, does not cause chronic pain because of intercostal neuropathy and offers significant benefits to the patient. However, it is more difficult to perform because of limited maneuverability. In the future, if subxiphoid uniportal RATS, which does not pass through the intercostal space, is introduced using a single-port robotic system already on the market, minimally invasive and highly difficult surgery will become possible. SRT is expected to expand the indications of minimally invasive surgery in the future.
The present study has several limitations. These include its retrospective and single-center design, the small number of cases, the absence of a comparison with conventional techniques, and the lack of randomization. In addition, there is a learning curve, variability between surgeons, and potential bias introduced by evolving techniques and new technologies. Although PSM was used to reduce measured confounding, the influence of unmeasured confounders remains a potential limitation of this analysis. In the future, the safety of surgery should be investigated in a multicenter prospective study of SRT. Future studies are needed to compare SRT with established conventional surgical approaches for anterior mediastinal tumors that have extended to the cervical region, to conduct randomized multicenter trials, and to investigate long-term outcomes. Short-term perioperative results were good; however, long-term results are unknown. Therefore, the long-term prognosis of SRT for anterior mediastinal tumors should be studied further.
In conclusion, SRT, which enables full visualization of the innominate vein from the midline view, may offer increased safety for complicated thymectomies, including those requiring combined resection of the innominate vein, and may demonstrate safety comparable to that of procedures performed for cases caudal to the innominate vein. Further studies are required to investigate the long-term prognosis of SRT for anterior mediastinal tumors located in the cervical region.
Conflict of Interest Statement
The authors reported no conflicts of interest.
The Journal policy requires editors and reviewers to disclose conflicts of interest and to decline handling or reviewing manuscripts for which they may have a conflict of interest. The editors and reviewers of this article have no conflicts of interest.
Appendix E1
Table E1.
Logistic regression results
| Variables | Multivariate analysis |
Univariate analysis after propensity score matching |
||||||
|---|---|---|---|---|---|---|---|---|
| Group A (n = 41) | Group B (n = 38) | Adjusted odds ratio∗ (95% CI) | P value | Group A (n = 27) | Group B (n = 27) | Odds ratios (95% CI) | P value | |
| Operative time, min, n (%) | 1.12 (0.14-9.00) | .912 | 0.72 (0.15-3.57) | .686 | ||||
| ≥120 | 36 (87.8) | 33 (86.8) | 23 (85.2) | 24 (88.9) | ||||
| <120 (ref) | 5 (12.2) | 5 (13.2) | 4 (14.8) | 3 (11.1) | ||||
| Intraoperative blood loss, mL, n (%) | 1.19 (0.32-4.38) | .794 | 0.61 (0.20-1.89) | .394 | ||||
| ≥10 | 13 (31.7) | 13 (31.2) | 8 (29.6) | 11 (40.7) | ||||
| <10 (ref) | 28 (68.3) | 25 (65.8) | 19 (70.4) | 16 (59.3) | ||||
| Conversion | 0 (0) | 0 (0) | NA | 0 (0) | 0 (0) | NA | ||
| Combined resection, n (%) | 7† (17.1) | 8† (21.0) | 0.33 (0.06-1.76) | .195 | 4† (14.8) | 6† (22.2) | 0.61 (0.15-2.46) | .486 |
| Pericardium | 5 (12.2) | 6 (15.8) | 4 (14.8) | 5 (18.5) | ||||
| Lung | 3 (7.3) | 6 (15.8) | 1 (3.7) | 5 (18.5) | ||||
| Innominate vein | 6 (14.6) | 3 (7.9) | 4 (14.8) | 3 (11.1) | ||||
| Phrenic nerve | 1 (2.4) | 1 (2.6) | 1 (3.7) | 1 (3.7) | ||||
| Duration of chest tube, d, n (%) | 1.45 (0.09-24.61) | .797 | 1.00 (0.13-7.67) | 1.000 | ||||
| ≥2 | 3 (7.3) | 3 (7.9) | 2 (7.4) | 2 (7.4) | ||||
| <2 (ref) | 38 (92.7) | 35 (92.1) | 25 (92.6) | 25 (92.6) | ||||
| Postoperative hospital stay, d, n (%) | 0.91 (0.25-3.32) | .884 | 1.00 (0.34-2.96) | 1.000 | ||||
| ≥4 | 26 (63.4) | 21 (55.3) | 16 (59.3) | 16 (59.3) | ||||
| <4 (ref) | 15 (36.6) | 17 (44.7) | 11 (40.7) | 11 (40.7) | ||||
| Postoperative complications, n (%) | 4 (9.8) | 2 (5.3) | 2.04 (0.33-12.48)‡ | .439 | 1 (3.7) | 0 (0) | 3.11 (0.11-84.8)‡ | .501 |
| Postoperative bleeding | 0 (0) | 1 (2.6) | 0 (0) | 0 (0) | ||||
| Chylothorax | 1 (2.4) | 0 (0) | 0 (0) | 0 (0) | ||||
| Atrial fibrillation | 1 (2.4) | 0 (0) | 1 (3.7) | 0 (0) | ||||
| Recurrent nerve paralysis | 1 (2.4) | 0 (0) | 0 (0) | 0 (0) | ||||
| Respiratory failure | 1 (2.4) | 0 (0) | 0 (0) | 0 (0) | ||||
| Phrenic nerve paralysis | 0 (0) | 1 (2.6) | 0 (0) | 0 (0) | ||||
| Number of analgesic tablets, n (%) | 0.53 (0.16-1.82) | .316 | 1.38 (0.45-4.17) | .574 | ||||
| ≥60 | 20 (48.8) | 13 (34.2) | 11 (40.7) | 9 (33.3) | ||||
| <60 (ref) | 21 (51.2) | 25 (65.8) | 16 (59.3) | 18 (66.7) | ||||
| Perioperative death, n (%) | 0 (0) | 0 (0) | NA | 0 (0) | 0 (0) | NA | ||
CI, Confidence interval; NA, not available.
Odds ratios were adjusted by age, sex, body mass index, Brinkman index, comorbidity, performance status, tumor size, surgical procedure, pathologic diagnosis, and surgical method.
Contains duplicates.
Analyzed by Firth's logistic regression model.
Supplementary Data
Four-port SRT performed for a thymic cyst located in the cervical region. SRT showed good visibility of the neck. SRT, Subxiphoid robotic thymectomy. Video available at: https://www.jtcvs.org/article/S2666-2507(25)00454-7/fulltext.
References
- 1.Pennathur A., Qureshi I., Schuchert M.J., et al. Comparison of surgical techniques for early-stage thymoma: feasibility of minimally invasive thymectomy and comparison with open resection. J Thorac Cardiovasc Surg. 2011;141(3):694–701. doi: 10.1016/j.jtcvs.2010.09.003. [DOI] [PubMed] [Google Scholar]
- 2.Cerfolio R.J., Bryant A.S., Minnich D.J. Operative techniques in robotic thoracic surgery for inferior or posterior mediastinal pathology. J Thorac Cardiovasc Surg. 2012;143(5):1138–1143. doi: 10.1016/j.jtcvs.2011.12.021. [DOI] [PubMed] [Google Scholar]
- 3.Detterbeck F.C., Kim A.W., Zielinski M. Looking in from above and up from below: new vistas in thoracic surgery. Innovations (Phila) 2012;7(3):161–164. doi: 10.1097/IMI.0b013e31826145b1. [DOI] [PubMed] [Google Scholar]
- 4.Kneuertz P.J., Kamel M.K., Stiles B.M., et al. Robotic thymectomy is feasible for large thymomas: a propensity-matched comparison. Ann Thorac Surg. 2017;104(5):1673–1678. doi: 10.1016/j.athoracsur.2017.05.074. [DOI] [PubMed] [Google Scholar]
- 5.Casiraghi M., Galetta D., Borri A., et al. Robotic-assisted thymectomy for early-stage thymoma: a propensity-score matched analysis. J Robot Surg. 2018;12(4):719–724. doi: 10.1007/s11701-018-0816-3. [DOI] [PubMed] [Google Scholar]
- 6.Suda T., Tochii D., Tochii S., Takagi Y. Trans-subxiphoid robotic thymectomy. Interact Cardiovasc Thorac Surg. 2015;20(5):669–671. doi: 10.1093/icvts/ivv001. [DOI] [PubMed] [Google Scholar]
- 7.Negi T., Morota M., Tochii D., Tochii S., Suda T. Initial results of uniportal and robot-assisted subxiphoid thymectomy. J Thorac Dis. 2024;16(10):6778–6788. doi: 10.21037/jtd-24-914. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Suda T., Sugimura H., Tochii D., Kihara M., Hattori Y. Single-port thymectomy through an infrasternal approach. Ann Thorac Surg. 2012;93(1):334–336. doi: 10.1016/j.athoracsur.2011.08.047. [DOI] [PubMed] [Google Scholar]
- 9.Dindo D., Demartines N., Clavien P.A. Classification of surgical complications: a new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg. 2004;240(2):205–213. doi: 10.1097/01.sla.0000133083.54934.ae. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Suda T. Subxiphoid thymectomy: single-port, dual-port, and robot-assisted. J Vis Surg. 2017;3:75. doi: 10.21037/jovs.2017.05.07. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Suda T., Morota M., Negi T., Tochii D., Tochii S. Subxiphoid uniportal robotic thymectomy using da Vinci Xi system†. Eur J Cardiothorac Surg. 2025;67(4) doi: 10.1093/ejcts/ezaf127. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Austin P.C. Optimal caliper widths for propensity-score matching when estimating differences in means and differences in proportions in observational studies. Pharm Stat. 2011;10(2):150–161. doi: 10.1002/pst.433. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Shigemura N., Shiono H., Inoue M., et al. Inclusion of the transcervical approach in video-assisted thoracoscopic extended thymectomy (VATET) for myasthenia gravis: a prospective trial. Surg Endosc. 2006;20(10):1614–1618. doi: 10.1007/s00464-005-0614-7. [DOI] [PubMed] [Google Scholar]
- 14.Zielinski M., Hauer L., Hauer J., Pankowski J., Nabialek T., Szlubowski A. Comparison of complete remission rates after 5 year follow-up of three different techniques of thymectomy for myasthenia gravis. Eur J Cardiothorac Surg. 2010;37(5):1137–1143. doi: 10.1016/j.ejcts.2009.11.029. [DOI] [PubMed] [Google Scholar]
- 15.Park J.H., Na K.J., Kang C.H., Park S., Park I.K., Kim Y.T. Robotic subxiphoid thymectomy versus lateral thymectomy: a propensity score-matched comparison. Eur J Cardiothorac Surg. 2022;62(1) doi: 10.1093/ejcts/ezac288. [DOI] [PubMed] [Google Scholar]
- 16.Yokoyama S., Hayashida R., Yoshiyama K., et al. Ectopic cervical thymoma excised through a transcervical approach combined with video-assisted thoracoscopic surgery: a case report. Ann Thorac Cardiovasc Surg. 2015;21(3):293–297. doi: 10.5761/atcs.cr.14-00277. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Raja S.M., Guptill J.T., McConnell A., Al-Khalidi H.R., Hartwig M.G., Klapper J.A. Perioperative outcomes of thymectomy in myasthenia gravis: a thoracic surgery database analysis. Ann Thorac Surg. 2022;113(3):904–910. doi: 10.1016/j.athoracsur.2021.06.071. [DOI] [PubMed] [Google Scholar]
- 18.Seo Y.J., Christian-Miller N., Aguayo E., Sanaiha Y., Benharash P., Yanagawa J. National use and short-term outcomes of video and robot-assisted thoracoscopic thymectomies. Ann Thorac Surg. 2022;113(1):230–236. doi: 10.1016/j.athoracsur.2021.02.003. [DOI] [PubMed] [Google Scholar]
- 19.Lee J.H., Park T.H., Kim H.K. Robotic thoracic surgery using the single-port robotic system: initial experience with more than 100 cases. J Thorac Cardiovasc Surg. 2024;168(6):1513–1522.e2. doi: 10.1016/j.jtcvs.2024.03.005. [DOI] [PubMed] [Google Scholar]
- 20.Suda T., Ashikari S., Tochii D., Tochii S., Takagi Y. Dual-port thymectomy using subxiphoid approach. Gen Thorac Cardiovasc Surg. 2014;62(9):570–572. doi: 10.1007/s11748-013-0337-y. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Four-port SRT performed for a thymic cyst located in the cervical region. SRT showed good visibility of the neck. SRT, Subxiphoid robotic thymectomy. Video available at: https://www.jtcvs.org/article/S2666-2507(25)00454-7/fulltext.