Robotic resection of mediastinal tumors: surgical approach and procedure

Abstract

Background

Mediastinal tumors (MTs) develop in various sites within the thoracic cavity, and the robotic surgical approach for MTs varies depending on the tumor location. This study aimed to assess the optimal approach for robotic surgery for MTs.

Methods

From April 2012 to May 2023, 141 cases of MTs removed by robotic surgery were noted. Of these, 130 cases, excluding those with combined lung resection, combined chest wall resection, and biopsy, were investigated for the MT location, surgical approach, operative time, and console time.

Results

Of the participants, 61 were male and 69 were female, with a median age of 60 (16–85) years and a median tumor diameter of 23 (2.5–150) mm. Additionally, 5 upper MTs, 99 anterior MTs, 22 middle MTs, and 4 posterior MTs were observed. The median operative and console time was 146 (38–371) and 76 (14–239) min, respectively. All cases, except for one case of the upper and middle MTs, were operated in the lateral position via a lateral approach. Of the 99 anterior MTs, 87, 9, and 3 were operated via lateral approach, subxiphoid approach, and single incision, respectively. Of the 87 patients who underwent the lateral approach, 78 and 12 underwent surgery in the 30° lateral decubitus position and lateral positions, respectively.

Conclusions

Standardizing the robotic surgery approach for MTs on the basis of tumor location enhances procedural safety and feasibility.

Background

Since the National Health Insurance in Japan had covered robotic-assisted thoracoscopic surgery (RATS) for malignant lung tumors and malignant/benign mediastinal tumors (MTs) in 2018, a steady increase in the number of robotic surgeries has been observed.

The anatomical complexity of the mediastinum renders robotic surgical systems particularly suitable for treating mediastinal lesions. Robotic systems are highly beneficial for MTs because the robotic instrument can be deployed at any angle by the robotic arm even in confined spaces [1, 2]. Several studies on patient position and port placement have been conducted [3,4,5,6]. However, the optimal patient position and port placement during RATS resection for various MTs remained unconfirmed.

This study aimed to report patient positioning and port placement during RATS resection for various MTs in our department and examined the optimal patient positioning and port placement.

Methods

A total of 267 patients underwent surgery for MTs from April 2012 to May 2023. Of these cases, there were 38 cases of thoracotomy, 88 cases of video-assisted thoracic surgery (VATS), and 141 cases of RATS. Of the 141 RATS cases, 130 were investigated, and 11 patients who underwent lung resection combined chest wall resection and biopsy were excluded from the study.

All procedures were performed using the daVinci® Surgical System. The daVinci® S was employed from April 2012 to February 2018, daVinci® Si on March 2018, and daVinci® Xi since April 2018. All procedures were performed with carbon dioxide (CO2) insufflation.

The Human Ethics Committee of Tokyo Women’s Medical University approved the study protocol (approval number: 5164). All participants provided informed consent.

Results

Overall, between April 2012 and May 2023, 130 consecutive patients (61 males and 69 females) underwent RATS resection for various MTs. The patients’ characteristics are presented in Table 1. The median age of all patients was 60 years (range, 16–85), and all patients had a performance status of ≤ 1. Preoperative comorbidities included a history of malignancy in 37 patients, hypertension in 35 patients, diabetes mellitus in 13 patients, collagen disease in 10 patients, arrhythmia in 6 patients, chronic kidney disease in 5 patients, a history of kidney transplantation in 5 patients, coronary artery disease in 3 patients, and myasthenia gravis in 1 patient. The median tumor diameter was 23 (range, 2.5–150) mm. The median operative and console time was 146 (range, 38–381) and 76 (range, 14–239) min, respectively. For upper, anterior, middle, and posterior MTs, the median operative time was 136.5, 146.0, 162.5, and 108.0 min, respectively, and the median console time was 91.5, 75.0, 91.0, and 49.5 min, respectively. Conversion to open thoracotomy or VATS was not noted. The median duration of chest tube placement was 2 days (range, 1–16 days), and the median length of hospital stay was 7 days (range, 4–112 days). Postoperative complications included two cases of chylothorax, and one case each of middle and anterior MTs. All MTs were completely resected. Pathological findings are shown in Table 2.

Table 1 Patient characteristics

Table 2 Histopathological tumor entities (n = 99)

Upper MTs

Under general anesthesia, the patients were intubated using a double-lumen endotracheal tube and positioned in lateral decubitus with the affected side up. The patient cart was docked on the patient’s left side. Portal RATS was performed with CO2 insufflation at a pressure setting of 5–8 mmHg. First, an approximately 12-mm port was inserted anterior to the fourth intercostal space (ICS) as an assist port for CO2 infusion. To position the other incisions from within the thoracic cavity, an 8-mm rigid 30° oblique viewing robotic endoscope was employed. Three 8-mm robotic trocars were inserted into the seventh ICS along the midaxillary line as port 2 for the robotic camera, the seventh ICS posteriorly along the posterior axillary line as port 1, and the fifth ICS along the anterior axillary line as port 3 (Fig. 1a). The robotic camera, fenestrated bipolar forceps, and permanent cautery spatula/Maryland bipolar forceps were placed in ports 2, 1, and 3, respectively. The surgeon selectively employed a vessel sealer extend for intraoperative robotic instrument exchange. An additional movie file further illustrates these findings (Additional file 1). The intraoperative positions used were lateral in five cases and 30° lateral decubitus in one case (Table 3). The pathological findings included neurinoma in four of the five cases and lipoma in one case.

Fig. 1

Schematic illustrations of port placement in robotic surgery for mediastinal tumors

(a) Upper mediastinal tumor (MT). Patients are positioned lateral decubitus with the affected side up. The patient cart is docked on the patient’s left side. An approximately 12-mm port is inserted anterior to the fourth intercostal space (ICS) as the assist port (orange circle) for carbon dioxide (CO2) infusion. Three 8-mm robotic trocars are inserted into the seventh ICS along the midaxillary line as port 2 (black circle) for the robotic camera, the seventh ICS posteriorly along the posterior axillary line as port 1 (yellow-green circle), and the fifth ICS along the anterior axillary line as port 3 (yellow-green circle)

(b–d) Anterior MT. (b) The lateral approach. Patients are positioned 30° lateral decubitus position with the affected side up at a 30° angle, and the ipsilateral arm is placed low. The camera trocar (black circle) is initially inserted at the fifth ICS along the anterior axillary line. The robotic arms (yellow-green circle) are placed in the second (8 mm) and fifth (12 mm) ICSs along the anterior axillary line, and the fifth (8 mm) ICS along the midclavicular line. An assist port (orange circle) is inserted into the fourth ICS (12 mm) along the midaxillary line. The patient cart is docked from the left side of the patient. (c) The subxiphoid approach. Patients are positioned supine, a 4-cm skin incision is created above the subxiphoid, and a wound protector (Gel POINT Mini Advanced Access Platform; Applied Medical, Rancho Santa Margarita, CA, USA) is affixed to an 8-mm rigid 30° oblique viewing robotic endoscope (black circle). The robotic endoscope is inserted with CO2 gas injection at a pressure of 5–8 mmHg using an AirSeal® Intelligent Flow System (ConMed, Largo, FL, USA). Two 8-mm ports (yellow-green circle) for the da Vinci robotic arm are bilaterally inserted into the sixth ICS along the midclavicular line. The da Vinci camera scope is mounted through the subxiphoid port. (d) Single incision. Patients are positioned lateral decubitus. A 5-cm skin incision is created at the fifth ICS along the anterior axillary line, and a wound protector (Gel POINT Mini Advanced Access Platform; Applied Medical, Rancho Santa Margarita, CA, USA) is placed at the subcutaneous muscular layer. Through the wound protector, an 8-mm camera port (black circle) is placed at the fifth ICS along the anterior axillary line as the second arm. Two 8-mm assisted ports (yellow-green circle) are inserted at the fourth and sixth ICSs on the anterior side as the first and third arms, respectively. The distance between each port is approximately 3 cm

(e) Middle MT. Patients are positioned lateral decubitus. Three trocars are inserted posterior (yellow-green circle) to the eighth ICS, at the posterior axillary line (black circle), and at the midaxillary line (yellow-green circle), and the remaining trocar (yellow-green circle) is inserted anterior to the seventh ICS. An 8-mm 30° robotic camera is inserted through a trocar (black circle) on the posterior axillary line at the eighth ICS

(f) Posterior MT. Patients are positioned lateral decubitus. Port placements are variable, with the robotic camera placed in the sixth or seventh ICS along the midaxillary line. The robotic instruments are placed on either side of the robotic camera (black circle), with the left hand placed dorsal to the posterior axillary line at the seventh ICS (yellow-green circle), and the right hand placed anterior to the anterior axillary line at the sixth or seventh ICS (yellow-green circle)

Table 3 Intraoperative position and approach

Anterior MTs

For anterior MT resection, the following three approaches were employed: lateral approach, subxiphoid approach, and single incision.

(1) Lateral approach

Regarding upper MTs, surgery was performed under general anesthesia, employing single-lung ventilation via double-lumen endotracheal intubation. Patients were positioned 30° lateral decubitus position and the ipsilateral arm was placed low. A total of four incisions were created for the endoscope, instruments, and assist. The camera trocar was initially inserted into the fifth ICS along the anterior axillary line. The robotic arms were placed in the second (8 mm) and fifth (12 mm) ICSs along the anterior axillary line, and the fifth (8 mm) ICS along the midclavicular line. An assist port was inserted into the fourth ICS (12 mm) along the midaxillary line. The patient cart was docked from the left side of the patient. The robotic camera was inserted into the fifth ICS (Fig. 1b). In principle, in both 8-mm ports, a spatula was used as the right arm, and bipolar fenestrated grasping forceps was used as the left arm. Intraoperatively, the devices of the right and left arms were replaced. An additional movie file further illustrates these findings (Additional file 2).

(2) Subxiphoid approach

In the supine position, patients were placed under general anesthesia and one-lung ventilation, a 4-cm skin incision was created above the subxiphoid, and a wound protector (Gel POINT Mini Advanced Access Platform; Applied Medical, Rancho Santa Margarita, CA, USA) was affixed for an 8-mm rigid 30° oblique viewing robotic endoscope. The robotic endoscope was inserted with CO2 injection at a pressure of 5–8 mmHg using an AirSeal® Intelligent Flow System (ConMed, Largo, FL, USA). Two 8-mm ports for the da Vinci robotic arm were bilaterally inserted into the sixth ICS along the midclavicular line. Through the subxiphoid port, the da Vinci camera scope was mounted (Fig. 1c). A monopolar spatula and a pair of bipolar fenestrated grasping forceps were mounted on the right and left arms, respectively. The spatula was replaced with a bipolar vessel sealing system. An additional movie file further illustrates these findings (Additional file 3).

(3) Single incision

A 5-cm skin incision was created at the fifth ICS along the anterior axillary line, and a wound protector (Gel POINT Mini Advanced Access Platform; Applied Medical, Rancho Santa Margarita, CA, USA) was placed at the subcutaneous muscular layer. Through the wound protector, an 8-mm camera port was inserted into the fifth ICS along the anterior axillary line as the second arm. Two 8-mm assisted ports were inserted into the fourth and sixth ICSs along the anterior side as the first and third arms, respectively (Fig. 1d). The distance between each port was approximately 3 cm. After the da Vinci® Xi (Intuitive Surgical Inc., Sunnyvale, CA, USA) was positioned, a robotic arm was mounted to each port using bipolar fenestrated grasping forceps on the first arm and a monopolar spatula on the third arm. A rigid 30° oblique viewing endoscope was employed. CO2 was insufflated at a set pressure of 5 mmHg. The robotic instruments in the left and right arms were interchanged depending on the intraoperative situation, and robotic energy devices, including the da Vinci® Vessel Sealer Extend instrument, were also employed. An additional movie file further illustrates these findings (Additional file 4).

Of the 99 cases of anterior MT, 87 (87.9%), 9 (9.1%), and 3 (3.0%) were operated via the lateral, subxiphoid, and single-incision approaches, respectively. Of the 87 cases who underwent the lateral approach, 78 (89.7%) and 9 (10.3%) underwent surgery in the 30° lateral decubitus position and lateral position, respectively. Additionally, the subxiphoid approach and single-incision approaches were performed in the supine and lateral positions, respectively. Two (2.3%) of the eighty-seven cases who underwent the lateral approach underwent surgery via bilateral approaches (Table 3).

The pathological findings included 34, 32, 8, and 25 cases of cystic disease, thymoma, thymic cancer, and other diseases such as metastatic mediastinal lymph nodes and thymic hyperplasia, respectively. Of the 32 cases of thymoma, one was complicated with myasthenia gravis and was operated via the subxiphoid approach.

Middle MTs

As with surgery for upper MTs, surgery for middle MTs was performed in the lateral decubitus position with the affected side using one-lung ventilation. CO2 was also insufflated through the trocar. The console surgeon held the robotic camera in the middle, fenestrated bipolar forceps in the surgeon’s left hand, and a permanent cautery spatula in the right hand. The robotic instruments were intraoperatively replaced with either Maryland bipolar forceps, monopolar curved scissors, vessel sealer extend, or SynchroSeal at the surgeon’s discretion. Only one case underwent single-incision RATS.

All cases, except for one, utilized the lateral decubitus position; the exception employed 30° lateral decubitus position (Table 3). In two cases, the pelvic-style docking we previously reported, the “pelvic” setting on the visual pad of the patient cart was employed.

The pathological findings encompassed eight cases of lymph node metastasis from other organ cancer, eight cases of cystic disease, two cases of schwannoma, and one case each of solitary fibrous tumor, liposarcoma, and epithelioid granuloblastoma.

Posterior MTs

As with surgery for upper MTs, surgery for posterior MTs was performed in the lateral decubitus position with the affected side up using one-lung ventilation. Port placements were variable, with the robotic camera placed in the sixth or seventh ICS along the midaxillary line. The robotic instruments were placed on either side of the robotic camera, with the left hand placed dorsal to the posterior axillary line at the seventh ICS and the right hand placed anterior to the anterior axillary line at the sixth or seventh ICS. Subsequently, the fenestrated bipolar forceps and the Maryland bipolar forceps were placed in the surgeon’s left and right hands, respectively. As with other MT resections, surgery was performed under CO2 insufflation, and robotic instruments were exchanged as needed intraoperatively. In only one case, the pelvic style was used.

Discussion

The da Vinci® surgical system offers other advantages, including allowing surgeons to use intuitive maneuvers, providing wide visibility by high-quality three-dimensional images, and enabling complicated operations to be performed using delicate instruments. The da Vinci® surgical system provides surgeons with greater instrument maneuverability and a wider range of motion than VATS, which requires long straight instruments [1, 2]. The mediastinum, particularly the upper mediastinum, is a narrow space, which is tightly packed with intertwined organs, such as nerves and blood vessels [7], and the posterior mediastinum is also a narrow space surrounded by the heart [8], descending aorta, and vertebral body; therefore, RATS is particularly useful for the mediastinal region. RATS mediastinal surgery demonstrated comparable safety to VATS, as evidenced by its shorter operative time, shorter postoperative hospital stay, and significantly lower intraoperative blood loss [9, 10]. RATS became covered by insurance in 2018, and at our hospital, RATS is now the first choice for MTs that are deemed resectable by VATS. If it is determined that a malignant tumor has invaded surrounding organs such as the brachiocephalic vein, median sternotomy is also an option.

MTs encompass several tumors, including thymomas, germ cell tumors, lymphomas, and neurogenic tumors, and occur in various locations, such as the upper, anterior, middle, and posterior mediastinum. The approach, position, and port placement should be selected on the basis of the tumor size and location. For intraoperative patients’ position for anterior MTs, the subxiphoid approach is performed in the supine position, and a single incision is performed with the patient in a lateral decubitus position, with the thymic lobe containing the tumor superior. Thymectomy and extended thymectomy for thymoma are performed using a lateral approach with the patient in a 30° right lateral decubitus position, irrespective of tumor location. In cases where partial thymectomy is appropriate, such as cystic disease, the patient is placed in the lateral position with the thymic lobe containing the tumor superior.

The subxiphoid approach provides a better operative view in the upper mediastinal region than the lateral approach and is therefore recommended for patients with tumors located above the innominate vein. When tumor invasion of major blood vessels, particularly the left brachiocephalic vein, is suspected, we prefer the subxiphoid approach [11]. It should be noted that the subxiphoid approach causes significantly longer operative minus console times than the lateral approach [12]. Therefore, for robotic surgeries, the authors prefer the lateral approach for anterior MTs, as the robotic instruments provide precise manipulation. It has been reported that a single incision for RATS may reduce postoperative pain, accelerate recovery [13]. We believe a single incision offer cosmetic advantages.

For RATS thymectomy, selecting between right- and left-sided approaches has been controversial [14]. The right-sided approach provides greater operative exposure, thereby minimizing the risk of iatrogenic cardiac injury. Although the left-sided approach requires caution when inserting the port to prevent cardiac damage, the left-sided approach enables a direct view of the left-sided thymic horns, and bilateral thymic horn resection is frequently easier from the left side. The left-sided approach offers another benefit of enhanced visualization of the right phrenic nerve owing to its linear course. In the case of thymectomy in patients with myasthenia gravis, more frequently achieved good outcomes as evaluated by the Myasthenia Gravis Foundation of America than the right-sided approach [15]. The reason for this finding is that in the autopsy study, incomplete resection was slightly more frequent with the right-sided approach [16]. Moreover, 10% of ectopic thymic tissues was observed in the AP window and was frequently located along the phrenic nerves, particularly on the left side [17]. The left-sided approach enables the resection of a larger amount of thymus.

In our cases, RATS for upper, middle, and posterior MTs was performed using the lateral approach. Of these, two cases of middle MTs and one case of posterior MTs underwent tumor resection using a pelvic-style approach. Although several methods for docking the robotic patient cart have been described, the approach and docking must be considered for tumors occurring in the lower-middle mediastinum. We have previously reported tumor resection using the pelvic-style docking in robotic surgery for lower-middle MTs, particularly those occurring below the inferior pulmonary veins [6]. As the patient cart of the da Vinci Xi robotic system is equipped with a visual pad that surgeons can use to set the type of procedure they are performing, we select the “pelvis” setting for patients with lower-middle MTs, as the robotic procedures in the pelvis, such as prostate or colon procedures, would have a similar robotic instrument insertion orientation to procedures in the lower-middle mediastinum.

Conclusions

RATS is advantageous for MTs. Tumor location dictates the optimal approach and port placement for RATS.