Robotic Excision of Posterior Mediastinal Neurogenic Tumours: Technique and Surgical Outcomes

Figure 1: 

MRI images of a left side apical posterior mediastinal neurogenic tumour. MRI: Magnetic resonance imaging

Surgical technique

All patients were operated under general anaesthesia with lung isolation achieved via placement of a left-sided double-lumen endotracheal tube under bronchoscopic guidance.

Position and port placement

All the patients were operated in a lateral decubitus position with 20° anterior tilt [Figure 2a]. We used a 4-port technique with three ports (one 12 mm and two 8 mm ports) for robotic instruments and an additional 12 mm port for assistance. The camera port was marked first. A small incision was made, and a 5-mm port placed directly opposite the target lesion. CO2 insufflation was started at a pressure of 8 cm of H₂0. Two additional ports were then placed under thoracoscopic vision, at a distance of 8–9 cm from the camera port allowing a good triangulation to target lesion. A 12 mm assistant port was placed between the camera and the cranial port [Figure 2b]. The initial 5-mm port was finally converted to a 12 mm port for the robotic camera. The port placement varies slightly according to the location of the tumour. We decide the port placement as per the tumour’s location. For tumours in the extreme apex of the thoracic cavity, the patients are positioned in classical lateral decubitus and ports are placed like that described for lobectomy by Cerfolio et al. except that the fourth arm is not used.^[12] The patient cart is docked from the head end at an angle of 15° anteriorly [Figures 2c and 3a]. For tumours in the middle and lower third of the thoracic cavity, the arc is rotated counter-clockwise to achieve a better triangulation towards the tumour [Figure 3b]. For tumours in the upper 1/3^rd of the thoracic cavity but not in the extreme apex, the patients are placed in lateral decubitus [Figure 2a] with anterior tilt and the ports are placed to achieve an optimum triangulation to the tumour. The robotic cart is docked from over the patient’s shoulder at an angle, as shown in Figure 3c. This modification helps in optimal movements of the robotic arms with minimal or no external clashing. All patients were operated using the Da Vinci Si Surgical System.

Figure 2: 

Figure 3: 

Port placement and patient cart docking for PMTs. Black object depicts the location of the tumour. The ports are positioned, facing and triangulating towards the tumour. (a) Apical PMTs with ports placed in lateral thoracotomy position. (b) Middle-lower third tumours and (c) Upper 1/3^rd tumours with ports placed in lateral decubitus position with 20° anterior tilt. PMT: Posterior mediastinal tumour

We used a 30° scope in the down position for all cases. A Cadiere forceps was used in the left hand and a cautery hook/Maryland Bipolar forceps in the right hand. The pleura was incised a centimetre away from the tumour [Figure 2d]. With a combination of blunt and sharp dissection, the tumour was gradually mobilised and lifted off its bed. Care was taken not to breach the tumour capsule at any stage of the dissection. Branches from the intercostal and other feeding vessels to the tumour, encountered during dissection were either cauterised or clipped. It is vital to maintain perfect haemostasis and proceed slowly, as any accumulation of blood around the tumour, decreases vision drastically, making dissection difficult. After complete mobilisation, the tumour was placed in a bag and retrieved through the camera port enlarging it as needed. The largest tumour in our series was a 9 cm mid-thoracic tumour. Specimen retrieval for such large tumours is challenging. We usually retrieve large tumours by placing in a specimen bag and retrieving through a separate incision in the 10^th intercostal space. The incision is made in the anterior most part of the space under thoracoscopic guidance and extended posteriorly as needed. Due to 11^th rib being a floating rib the space can be widened significantly more enabling the removal of large tumours easily.^[13] Complete haemostasis at the tumour bed was ensured. A single 24 F chest tube was placed followed by an intercostal block using 0.125% Bupivacaine with epinephrine from 2^nd to 10^th intercostal space. All patients were extubated on the table.

RESULTS

A total of 20 patients underwent RP-PMT-3 and were included in the study. There were 14 males and 6 females. The median age was 41.2 years (range 12–62 years). The most frequent presentation was chest pain (n = 08, 40%) followed by referral due to incidental finding (n = 12, 60%). Out of 20 cases, 11 had right-sided lesions, while 9 were on the left side. There were 10 lesions in the upper chest, 6 in mid-chest, 2 in the lower chest below inferior pulmonary vein and 2 were at the apex. The histopathological diagnosis is listed in Table 1. Schwannoma arising from the nerve sheath was the most common histopathological diagnosis, present in 50% of the cases. One patient was reported to have a malignant nerve sheath tumour.

We had to convert to thoracotomy in 2 cases. The reason was dense pleural symphysis in one patient (he had tubercular empyema in childhood) and haemorrhage obscuring the vision in the other. In this patient, the large size of the tumour prevented adequate visualisation to achieve haemostasis; hence, thoracotomy was performed for safe tumour removal. Average total operative time was 110 min. The average blood loss was 30 mL. The average tumour size was 4.8 cm (2–9 cm). The post-operative hospital stay was 2.4 days, with 18 of the 20 patients discharged on the second post-operative day. One patient developed Horner’s syndrome post-operatively while 1 had lung atelectasis on the operated side, which was managed conservatively. Adjuvant radiotherapy was given to the patient whose lesion was reported as malignant nerve sheath tumour. These results are summarised in Table 1. With a median follow-up of 36 months (6–48 months), all patients were recurrence-free, except the patient with malignant nerve sheath tumour who developed local recurrence 13 months after the surgery with no other lesions elsewhere. An open surgical re-excision along with wedge resection of the adherent left upper lobe was done, and adjuvant chemotherapy was also given to her this time. At last follow-up, 6 months after the second surgery, she was disease-free.

DISCUSSION

Posterior Mediastinal Neurogenic Tumours account for 10%–34% of all the mediastinal tumours.^[14] Spinal nerve roots, sympathetic chain and branches of the intercostal nerves are the typical sites of origin. They are classified as Schwannoma or Neurofibroma when they arise from the nerve sheath. On the other hand, ganglioneuroma and paraganglionic tumours arise from the nerve cells. Most of these tumours are benign, produce no symptoms and are an incidental finding on imaging done for other reasons.^[1] When present, symptoms result either from a neurologic compromise of the spinal canal or from compression of adjacent structures.^[15-17] The common symptoms are back and chest pain, cough, dyspnoea and neurologic symptoms. The presence of neurologic symptoms should raise suspicion of intraspinal extension. The intraspinal extension should be actively sought for in all patients, even if asymptomatic.

All posterior mediastinal tumours, even when asymptomatic need surgical resection, to prevent problems secondary to the continued growth of the tumour and rule out the possibility of malignancy.^[18] Transthoracic open approach has been the standard of care for excision of these tumours till recently. Nowadays, most of the posterior mediastinal neurogenic tumours without intraspinal extension are being removed by thoracoscopy. This approach has gained acceptance as an alternative to open techniques because of its safety and feasibility together with lower morbidity, shorter hospital stay, lesser blood loss and similar survival.^[18,19] However, conventional thoracoscopy has limitations. These include a 2D vision and difficulty in using long rigid non-articulating instruments in the narrow space. These limitations make thoracoscopic removal difficult in some patients because these tumours are known to be densely adherent and going around the tumour with non-articulating instruments may be challenging.^[20] These limitations become more pronounced in case of large tumours and those located at either extremes of the thoracic cavity. The robotic surgical system provides the surgeon with capabilities that overcome many of the limitations of conventional thoracoscopy. It offers a superior 12-fold magnified stereoscopic 3D vision and enhanced dexterity due to articulating EndoWrist robotic instruments with 7° of freedom, motion scaling and tremor filtration making dissection around the tumour easier.

The first report of the use of the Da Vinci Surgical System for posterior mediastinal mass was by Yoshino et al. in 2002.^[21] Since then, several reports have been published.^[20,22-24] Nearly all of them report feasibility and safety. Li et al.^[23] did a comparative analysis between thoracoscopic and robotic approaches for posterior mediastinal tumours.^[22] The surgical time, blood loss and hospital stay were significantly lesser in the robotic group. On the other hand, no significant difference was noted between the robotic and VATS group in terms of duration of the chest tube, the volume of drainage or post-operative adverse effects or complications.

Selection of cases for robotic excision of posterior mediastinal tumours requires careful consideration of factors such as the size of the tumour, its location and presence or absence of features suggesting malignancy. The thoracoscopic or robotic approach should never be undertaken in patients with significant pre-operative concern for malignancy. No established guidelines are there suggesting a size cut-off limit for minimally invasive excision of these tumours. In a series of 58 patients treated by thoracoscopy, Li and Wang established a cut off of 6 cm while others have reported excision of even 7–8 cm size tumours via thoracoscopic surgery.^[25,26] We did not consider the large size of the tumour to be a contraindication as long as the pre-operative imaging assessment showed maintained planes. For tumours >8 cm, we do a diagnostic thoracoscopy and look for two things in order to decide whether to proceed with the robotic approach or convert. Firstly, we evaluate whether an adequate view of the entirety of the tumour is possible and secondly whether there will be enough space for movement of the robotic instruments under vision, without hitting any of the vital structures nearby. If these two criteria are fulfilled, we proceed with robotic excision. The robotic system provides better dissection abilities for larger tumours owing to its articulating instruments and helps surgeons take up larger tumours than they would generally accept for a thoracoscopic procedure. However, very large tumours at the apex or costophrenic junction may not be amenable to minimally invasive approaches as the view of the adjacent structures gets impaired.

Tumours located at either extreme of thoracic cavity are challenging to resect using conventional thoracoscopy because limited space severely restricts the vision. With critical structures like subclavian vessels, inferior border of the brachial plexus, azygos vein, oesophagus, descending aorta and diaphragm in the vicinity, the dissection by thoracoscopy is challenging.^[24] The Robotic system is probably better for these tumours at either extreme, as it provides a superior 3D vision along with the ability to perform a more controlled and precise dissection due to articulating instruments.

Proper planning of the ports and docking is essential for comfortable and safe surgery. We used a semi lateral decubitus position for all patients except for the tumours located in the apex. It allows for an optimal operative view and simpler docking of the patient cart from over the patient’s back. The anterior tilt allows the lung to fall away from the operative field naturally and keeps the operative field clear due to gravity-assisted drainage of any bleeding. Also, since we placed the ports anteriorly where the intercostal spaces are wider as compared to posterior intercostal spaces, less pressure was exerted on the intercostal neurovascular bundle. Moreover, if a conversion is required, it is relatively straight forward. For the tumours in the apex, a full lateral position and port placement similar to CPRL-4 technique, afford a relatively better overall view of the apex.^[12]

There were no significant complications except for temporary Horner’s syndrome after removal of an apical PMT which recovered on conservative treatment in 4 weeks. Horner’s syndrome is a known complication after excision of apical neurogenic PMTs and occurs due to injury/oedema of the stellate ganglion. Some authors have recommended an intracapsular enucleation of the apical neurogenic tumour followed by careful excision of the capsule as a safer method to reduce the chance of injury to surrounding structures.^[27] However, we do not recommend it as it is against oncological principles. Whether we use open, thoracoscopic or robotic approach, the oncological principles must be strictly adhered to. Any compromise of oncological principles during thoracoscopic or robotic procedures is an indication for immediate conversion to open surgery. For the apical tumours, Al-Mufarrej et al. have recommended mobilising the apical area of the tumour last so that traction can be applied on the tumour to make dissection away from stellate ganglion and subclavian vessels more comfortable and safer.^[20]

Two cases required conversion in this series. One was an elective conversion due to dense pleural symphysis in a patient with a history of childhood tubercular empyema. Dense adhesions precluded the creation of space for port placement and hence the procedure was converted. The other conversion was in a large mid-thoracic PMT (7 cm) immediately adjacent to descending thoracic aorta where an injury to posterior intercostal artery led to bleeding and obscuration of the field.

The most frequently quoted disadvantages of the robotic system include lack of haptic feedback, high cost and requirement of a specialised surgical non portable equipment. A single company with patent and low usage has been the main reasons for the high cost. Other manufacturers are developing newer robotic systems that are now in the market, and their entry is likely to reduce the cost and enable widespread usage.^[28] Lack of haptic feedback remains a problem till date. Technological development may change this in future. However, the superior 3D stereoscopic view compensates for the lack of haptic feedback significantly.

Retrospective nature, small numbers and lack of a controlled arm are the main limitations of this study. A prospective study with larger numbers comparing the thoracoscopic and robotic approaches will be ideal for establishing the exact benefit of the robotic approach in the management of PMT.

CONCLUSION

Our study demonstrates the feasibility and safety of robotic surgery for posterior mediastinal neurogenic tumours. It may offer advantages over conventional thoracoscopy in terms of improved visualisation and dexterity, which may allow a broader range of these tumours to be resected by a minimally invasive approach. However, more comparative studies especially comparison with VATS (uniportal/conventional) is needed to establish short as well as long term benefits.

For the reference original article published on: Journal of Minimal Access Surgery