Robotic Nerve Sheath Tumor Resection With Intraoperative Neuromonitoring: Case Series and Systematic Review

BACKGROUND:

Retroperitoneal nerve sheath tumors present a surgical challenge. Despite potential advantages, robotic surgery for these tumors has been limited. Identifying and sparing functional nerve fascicles during resection can be difficult, increasing the risk of neurological morbidity.

OBJECTIVE:

To review the literature regarding robotic resection of retroperitoneal nerve sheath tumors and retrospectively analyze our experience with robotic resection of these tumors using a manual electromyographic probe to identify and preserve functional nerve fascicles.

METHODS:

We retrospectively analyzed the clinical courses of 3 patients with retroperitoneal tumors treated at the National Institutes of Health by a multidisciplinary team using the da Vinci Xi system. Parent motor nerve fascicles were identified intraoperatively with a bipolar neurostimulation probe inserted through a manual port, permitting tumor resection with motor fascicle preservation.

RESULTS:

Two patients with neurofibromatosis type 1 underwent surgery for retroperitoneal neurofibromas located within the iliopsoas muscle, and 1 patient underwent surgery for a pelvic sporadic schwannoma. All tumors were successfully resected, with no complications or postoperative neurological deficits. Preoperative symptoms were improved or resolved in all patients.

CONCLUSION:

Resection of retroperitoneal nerve sheath tumors confers an excellent prognosis, although their deep location and proximity to vital structures present unique challenges. Robotic surgery with intraoperative neurostimulation mapping is safe and effective for marginal resection of histologically benign or atypical retroperitoneal nerve sheath tumors, providing excellent visibility, increased dexterity and precision, and reduced risk of neurological morbidity.

ABBREVIATIONS:

EBL

estimated blood loss

EMG

electromyography

NCI

National Cancer Institute

NIH

National Institutes of Health

NF1

neurofibromatosis type 1

SNB

surgical neurology branch

SUV

standardized uptake value

Despite being early adopters of robotic technology in the operating room, neurosurgeons have until recently found relatively limited applications for robotics.¹ Their use in cranial procedures has been primarily restricted to single-channel utilities, such as navigation for biopsies, laser lesion ablation, or electroencephalography lead and neurostimulator electrode implantation. In the spine, it has largely assisted in pedicle screw placement, although multiport surgical robots have shown promise in the transoral resection of odontoid pathology or transperitoneal approach to sacral tumors.^2,3 Although the theoretical advantages of multiarmed, minimally invasive surgical robotics in neurosurgery are obvious vis-à-vis improved visibility, tremor reduction, and motion scaling, these advantages are tempered by the need for multiple ports, lack of haptic feedback, and the need for relatively large proximal range of motion for instrument arms.⁴

The da Vinci Surgical System (Intuitive Surgical) is a multiarmed telesurgical robot primarily used in retroperitoneal and pelvic urological and gynecological procedures. Although abdominopelvic retroperitoneal nerve sheath tumors are rare in the general population, they occur with higher frequency in patients with neurofibromatosis type 1 (NF1). Their location makes resection through traditional open techniques difficult and potentially associated with significant morbidity, including injury to retroperitoneal vascular, urological, and neural tissues. Existing case reports of retroperitoneal nerve sheath tumors resected laparoscopically using the da Vinci Surgical System^3,5-7 suggest that a collaborative approach between neurological and urological surgeons using such techniques is a viable option.

Here, we present our experience with 3 cases in which the da Vinci robot was successfully used by a multidisciplinary team to resect retroperitoneal nerve sheath tumors with minimal blood loss or neurological morbidity. This adds to a growing body of literature demonstrating the utility of robot-assisted surgical technology in addressing neurosurgical pathologies, particularly tumors arising in the retroperitoneal and pelvic spaces, where robot-assisted techniques have already demonstrated utility in addressing urologic pathology.^6,7 In addition, these cases build on earlier reports of electromyographic (EMG) stimulation mapping during robot-assisted resection of retroperitoneal nerve sheath tumors to facilitate marginal extracapsular tumor resection sparing parent motor nerve fascicles, reducing the risk of postoperative motor deficits.^8,9

METHODS

Patient Cohort

We retrospectively analyzed the clinical outcomes of 3 patients treated with a robotic approach at the National Institutes of Health (NIH) Clinical Center between 2017 and 2019. The patients and/or their guardians consented to be evaluated and treated under a neurosurgical protocol (clinicaltrials.gov identifier NCT00060541) reviewed and approved by the NIH Combined Neuroscience Institutional Review Board (IRB). Two of the patients with NF1 and/or their guardians also consented to and were enrolled in a National Cancer Institute (NCI) NF1 natural history study (clinicaltrials.gov identifier NCT00924196) and phase II trials of a MEK inhibitor in NF1 (clinicaltrials.gov identifiers NCT02407405 and NCT01362803), approved by the NCI IRB.

Training

The urologic surgeons were credentialed to independently use the da Vinci robotic platform (Intuitive Surgical) at the NIH Clinical Center by the Clinical Center Credentialing Committee. The neurosurgical team underwent guided training under the supervision of credentialed urologic surgeons before each procedure.

Intraoperative Equipment

The da Vinci Xi system was used with standard urologic setup and instruments, and a disposable bipolar stimulator (Neurosign V4) was used for stimulation mapping of the surgical field.

Literature Search

A comprehensive literature search was conducted using the PubMed-NCBI and EMBASE electronic bibliographic databases. The following key words were used: ((retroperitoneal) or (pelvic)) and (“schwannoma” or “neurofibroma” or “nerve sheath tumor”)) and ((robot) or “da vinci”). No date restrictions were imposed, and only English language articles were reviewed. The results included published reports from 2009 to May 2021; conference abstracts and unpublished articles were excluded.

This search identified 20 articles, which were assessed for relevance by screening titles and abstracts, resulting in a final selection of 17 articles reporting data on 24 patients (Figure 1). Included articles were case reports or case series using robot-assisted laparoscopic surgery for the treatment of retroperitoneal or pelvic nerve sheath tumors. An article documenting an additional case by the present authors was included, for a total of 18 articles reporting data on 25 patients.¹⁰

FIGURE 1.

Preferred Reporting Items for Systematic Reviews and Meta-Analyses flow diagram of search.

RESULTS

Patient Demographics

Three patients underwent a combined neurosurgical–urological surgical team approach to the resection of retroperitoneal peripheral nerve sheath tumors (Table) using the da Vinci robot (Intuitive Surgical). Indications included function-limiting pain, tumor growth, fluorodeoxyglucose (FDG) uptake on positron emission tomography (PET), or risk of malignant transformation in patients with retroperitoneal nerve sheath tumors necessitating a transperitoneal approach.

TABLE.

Summary of Patient Characteristics

Patient	Sex	Age	BMI	Diagnosis	Location	Size (cm)	Tumor burden
1	Female	24	23.32	NF1	Pelvis; between right external iliac vein and obturator nerve	4.1 × 3.0	Multiple pelvic neurofibromas, with multiple plexiform lesions in bilateral axilla, retroperitoneum, thorax, and upper extremities
2	Female	48	22.67	Sporadic Schwannoma	Right psoas	3.9 × 3.6	No previous history
3	Male	15	25.42	NF1	Right psoas	7.1 × 5.5	Paravertebral lumbosacral and right inguinal neurofibroma, abdominopelvic plexiform lesion, and multiple subcutaneous nodules in upper extremities

BMI, body mass index; NF1, neurofibromatosis type 1.

Surgical Technique

Before surgery, imaging was reviewed by a multidisciplinary team, including neurosurgery, urological surgery, vascular surgery, and transplant surgery. A consensus opinion regarding the utility of a robotic approach was reached during the multidisciplinary review. Perioperative antibiotic prophylaxis was administered according to the American College of Surgeons and American Urologic Association guidelines for minimally invasive surgery not involving bowel entry, and all patients received prophylactic heparin. Vascular surgery and transplant surgery teams were available as necessary for emergent conversion to open laparotomy to address vascular or renal injury.

After induction of anesthesia with short-acting paralytics, subdermal needle EMG leads were placed in the iliopsoas, vastus lateralis, adductor longus, tibialis anterior and medial gastrocnemius muscles to monitor femoral, obturator, fibular, and tibial nerves of the affected lower extremity. EMG was monitored for spontaneous activity and intraoperative nerve stimulation. A temporary stent was placed under cystoscopic guidance in the right ureter for identification and protection of the ureter, followed by placement of a Foley catheter.

Patients were positioned depending on their tumor location relative to the 3 retroperitoneal zones described in the general surgery and trauma surgery literature (Figure 2).¹¹ For lesions in zones 1 and 2, as in patients 2 and 3 (Figure 3), standard flank positioning (lateral decubitus) is used. For lesions in zone 3 (pelvis), as in patient 1 (Figure 3), patients are positioned supine with Trendelenburg. After prepping and draping, peritoneal access was obtained with a Veress needle. After insufflation to 12 mm Hg, trocars were placed in a configuration appropriate for the tumor location, including trocars for 3 robotic instruments, the camera, and 2 assistant ports (Figure 2).

FIGURE 2.

Surgical anatomy of retroperitoneal nerve sheath tumors. A, Retroperitoneal zones. Zone 1 is the midline retroperitoneum and contains vascular structures, such as the abdominal aorta, inferior vena cava, proximal superior mesenteric artery, and proximal renal arteries, as well as viscera, such as the pancreas and duodenum. Zone 2 is located lateral to zone 1 on both sides and contains renal arteries and veins, kidneys and adrenals glands, and proximal ureters. Zone 3 is located in the pelvis and contains the iliac arteries and veins. Standard flank positioning and trocar configuration (inset, top) are used for tumors located in retroperitoneal zones 1 and 2, whereas supine positioning and trocar configuration (inset, bottom) are used for lesions located in retroperitoneal zone 3. The 10 to 12-mm trocar is indicated in red, 8 mm in gray, and 5 mm in yellow. B, Intraoperative scheme. The bipolar stimulating probe through a manual port for mapping tumor surface and the psoas muscle.

FIGURE 3.

Preoperative magnetic resonance images of included patients. Case 3: Coronal A and axial B T2 series demonstrating a plexiform neurofibroma (arrowheads) located in zone 2. Case 2: Contrast-enhanced coronal C and axial D T1 series showing a sporadic schwannoma (arrowheads) located in zone 2. Case 1: Contrast-enhanced coronal E and axial F T2 series showing a plexiform neurofibroma located in zone 3.

Anesthesia was maintained with intravenous agents while avoiding paralytic agents. The robot was then docked, and inspection and initial dissection was performed by the urological surgery team. Pelvic or retroperitoneal viscera and vasculature were identified and mobilized, and the ureter was identified and protected.

Once the tumor was located, a bipolar stimulating probe was inserted through a manual port to map tumor and psoas muscle surfaces, confirm continuity of the parent nerve, and identify motor nerve fascicles during fascicular dissection (Figure 2). These fascicles were then dissected from the tumor capsule, and marginal extracapsular resection with occasional bipolar stimulation was performed until the tumor could be completely separated from the parent nerve. The tumor was removed, and the port sites were closed by the urological surgery team after hemostasis and irrigation with an estimated blood loss (EBL) less than 50 mL for all cases. Specific and salient features of each case are presented below.

Case 1

A 25-yr-old woman with NF1 and a large total body tumor burden, currently enrolled in a phase II clinical trial of a MEK inhibitor presented with new diffuse right lower extremity pain and weakness in the context of an enlarging inguinal mass. Her pain was progressive, severe, and sufficiently functionally limiting that she required a walker to ambulate. She denied paresthesias, hypesthesia, or urinary retention or incontinence.

Magnetic resonance imaging (MRI) demonstrated multiple pelvic lesions, including a 4.1 × 3.0-cm right pelvic mass involving the lumbosacral plexus and a 3.7 × 2.3-cm right inguinal mass. FDG PET demonstrated increased avidity of both lesions when compared with imaging acquired 19 months before, with a standardized uptake value (SUV) = 6.69 in the inguinal mass (previously 4.55) and SUV = 4.4 in the pelvic mass (previously 2.63). She was offered robotic laparoscopic resection of the pelvic tumor to prevent further functional decline and to prevent malignant transformation, with concurrent resection of the FDG-avid inguinal tumor.

After placement of camera and robotic trocar ports and docking of the da Vinci Surgical System, retroperitoneal dissection proceeded with exposure of the right pelvic tumor beneath the right external iliac vein (Video 1). The tumor was located lateral to the obturator nerve, displacing it medially. Extracapsular dissection was performed circumferentially, and a distinct intracapsular margin between tumor and parent nerve was identified, permitting en bloc resection. Intraoperative nerve stimulation confirmed the presence of the adjacent obturator nerve and assured the anatomic continuity and functional integrity of the lumbar plexus nerves. The inguinal mass was resected without complication, with an operative time of 265 min. Pathology confirmed both masses were benign neurofibromas.

VIDEO 1.

Download video file ^{(277.2MB, mp4)}

Annotated intraoperative video of patient 1.

She tolerated surgery well and recovered uneventfully with discharge at postoperative day 3. At the last follow-up 13 mo postoperatively, she was ambulating without an assistive device and her pain had resolved completely.

Case 2

A 48-yr-old woman with an unremarkable medical history was referred to the surgical neurology branch (SNB) clinic for evaluation of a biopsy-confirmed right psoas intramuscular schwannoma discovered during workup for several years of progressive, intermittent, lancinating right lower extremity pain. She also described intermittent severe pain with hip flexion but denied weakness, paresthesias, or urinary retention or incontinence. Prior evaluation at an outside institution concluded that the lesion was unresectable, and she was referred to the NIH Clinical Center.

A contrasted MRI revealed a 3.9 × 3.6-cm T2-hyperintense mass within the right psoas muscle, with heterogeneous contrast enhancement. There was evidence of enlargement when compared with previous images provided by the patient. Given her progressive symptoms and lesion enlargement, she was offered robot-assisted resection.

After placement of camera and robotic trocar ports and docking of the da Vinci Surgical System, retroperitoneal dissection proceeded to the right psoas muscle, which was grossly dilated. Stimulation at the surface of the psoas muscle yielded no activation within the lumbar plexus, and the muscle was split longitudinally without evidence of nerve irritability. Extracapsular dissection proceeded circumferentially around the schwannoma, and the tumor was removed, with an operative time of 152 min. Pathology confirmed schwannoma.

She tolerated surgery well with an uneventful recovery and was discharged on postoperative day 2. At the last follow-up 2 weeks postoperatively, she was ambulating without difficulty and reported improved pain.

Case 3

A 16-yr-old boy with NF1 and lumbosacral and pelvic plexiform neurofibromas who was enrolled in a phase II clinical trial of a MEK inhibitor was referred to the SNB clinic for progressive growth of a right retroperitoneal biopsy-confirmed atypical neurofibroma arising from a pelvic plexiform neurofibroma. He denied pain, paresthesia, weakness, or other symptoms referable to the lesion.

Imaging revealed an extensive plexiform neurofibroma involving the right lumbosacral plexus with an associated retroperitoneal 7.1 × 5.5-cm distinct nodular-appearing lesion. Given the progressive enlargement of the lesion and the patient's young age, preemptive robot-assisted resection was offered to prevent malignant transformation of the atypical neurofibroma.

After placement of camera and robotic trocar ports and docking of the da Vinci Surgical System, the right kidney was mobilized to uncover the psoas muscle, which was split longitudinally to reveal a tumor attached to a nerve-like structure (Video 2). Stimulation of this structure yielded no activation within the lumbar plexus, and extracapsular dissection proceeded circumferentially. The tumor was delivered from the muscle and removed. Total operative time was 208 min. Pathology confirmed intraneural neurofibroma.

VIDEO 2.

Download video file ^{(288.4MB, mp4)}

Annotated intraoperative video of patient 3.

He tolerated surgery well with an uneventful recovery and was discharged on postoperative day 3.

DISCUSSION

Most schwannomas and neurofibromas are benign, well-circumscribed tumors arising from the Schwann cells surrounding peripheral nerves. Abdominopelvic retroperitoneal nerve sheath tumors are rare in the general population but occur with higher frequency in patients with NF1, typically as histologically benign plexiform neurofibromas. Plexiform neurofibromas are at risk for transformation to highly aggressive malignant peripheral nerve sheath tumors (MPNSTs), which require complete surgical resection with wide margins. A subset of patients with NF1 develop atypical neurofibromas, which are characterized by a distinct nodular appearance on MRI, FDG avidity on FDG PET, and heterozygous loss of CDKN2A/B. We and others have demonstrated that atypical neurofibromas are precursor lesions to MPNST, marginal resection is safe, and such a surgical strategy offers durable prevention of recurrence.^12-14 Both patients with NF1 were confirmed to have benign neurofibromas with no evidence of atypia, suggesting that early resection of symptomatic or growing tumors can preempt atypical or malignant transformation to MPNST.¹⁴

Although schwannomas tend to arise eccentrically to their parent nerve, thereby displacing normal nerve fascicles, neurofibromas may arise intraneurally, causing a fusiform dilation of the parent nerve and incorporating normal fascicles.¹⁵ Retroperitoneal and pelvic locations are exceptionally rare for these tumors. Schwannomas represent only 1% to 3% of all retroperitoneal tumors, and fewer than 4% of all schwannomas are retroperitoneal,¹⁶ whereas only 8 cases of retroperitoneal or pelvic neurofibromas have been reported in the previous literature.¹⁷

In this study, a total of 28 patients (25 patients in the previous literature and 3 in this study) with retroperitoneal nerve sheath tumors treated with robotic surgical resection were reviewed. There was no gender predilection. Of these patients, pathology confirmed schwannoma in 23 (82.1%), whereas 5 (17.9%) were diagnosed with neurofibromas. The average age of patients with schwannomas was 45.7 years (standard deviation [SD] = 14.18 years) while the average age of patients with neurofibromas was 29.2 years (SD = 11.38 years). Retroperitoneal neurofibromas were almost uniformly associated with NF1. Tumors in zone 2 were more common, composing 16 (57.1%) of the tumors in the present review, whereas 11 (39.3%) were in zone 3. Only 1 patient (3.6%) with a neurofibroma had a tumor in zone 1. Of the patients with pelvic tumors, 8 were diagnosed with schwannomas and 3 with neurofibromas. Tumors were more likely to arise from the right (67.9%) than from the left (32.1%). Based on available data, when compared with traditional open resection, robot-assisted surgery was associated with lower average blood loss (87.4 vs 3889 mL),¹⁸ shorter hospitalization (3.4 vs 6 d),¹⁹ and shorter operative time (145 vs 434 min).²⁰ One patient experienced a temporary motor neurological deficit after resection of an obturator schwannoma, and no patients experienced significant permanent neurological deficits.^{5,7,10,17,21-33}

The present review is limited by the small number of reported cases of robotically resected retroperitoneal nerve sheath tumors. All included studies were case reports or series published over the course of more than 20 years, and therefore data such as EBL, operative times, and duration of hospitalization were not available for all patients. Nevertheless, the available data indicate that robot-assisted surgery results in substantially reduced operative times, blood loss, and hospitalization compared with open resection.

Retroperitoneal schwannomas may be asymptomatic and might not be discovered until they have grown relatively large. Because they are insensitive to radiation and chemotherapy, definitive treatment of these tumors is surgical, and complete surgical resection confers a good prognosis.^34,35 Their depth, however, presents a significant challenge. Historically, tumors in this location were resected through prolonged open surgeries with potentially significant risk of intraoperative bleeding or visceral injury. Later, laparoscopic refinements allowed for minimally invasive techniques; however, these suffered from inferior 2-dimensional vision and reduced instrument freedom of movement.³⁶ There is a paucity of neurosurgical data on robotic-assisted surgery for retroperitoneal nerve sheath tumors. However, extensive data in urology show that robotic-assisted surgery leads to diminished postoperative pain, shorter hospital stay, and decreased need for blood transfusions.³⁷ In the general surgery literature, they found no difference in incisional hernia incidence after robotic-assisted cholecystectomy compared with the traditional laparoscopic approach.³⁸

The first robot-assisted retroperitoneal approach to such tumors occurred in 2009, with the transdiaphragmatic retroperitoneal resection of a thoracolumbar neurofibroma.¹⁷ The ability to access these tumors through laparoscopic ports with 3-dimensional visibility and dexterous, precise instrument arms was a significant step forward in the management of these tumors. Multiarm telesurgical systems, such as the da Vinci Surgical System (Intuitive Surgical), are particularly well suited for approaching retroperitoneal schwannomas and neurofibromas, particularly in patients with disorders such as NF1 predisposing to the development, transformation, and recurrence of these tumors. The neurosurgery team was trained in the use of the da Vinci System before each procedure. Robotic-trained urologists maintained close supervision over the neurosurgery team using the dual surgeon console robotic system. With multiple training sessions and close supervision, the neurosurgery team was able to perform fascicle-sparing capsular dissection with a short learning curve.

A fundamental principle for the treatment of many retroperitoneal tumors—in particular sarcomas—has classically been complete surgical resection with large negative margins, which may entail resection of adjacent vessels and viscera (such as kidney or colon).^39,40 In the case of benign or borderline nerve sheath tumors, however, marginal extracapsular resection of tumor with sparing of associated parent motor nerve fascicles and adjacent visceral structures can provide durable tumor control and prevent malignant transformation while preserving motor function.¹⁴ Although there is limited neurosurgical literature on robotic nerve-sparing surgery for abdominopelvic and retroperitoneal approaches, there are significant data in urology. Robotic-assisted surgery has improved nerve-sparing techniques and functional preservation in radical prostatectomies, primarily because of the 3-dimensional visibility and enhanced maneuverability.^41,42

Limitations

At the NIH, we have begun treating patients with retroperitoneal and pelvic schwannomas and neurofibromas with robot-assisted minimally invasive surgery using the da Vinci Surgical System in a collaborative partnership with the NCI Urologic Oncology Branch. The addition of EMG stimulation mapping through a manual laparoscopic port to identify motor nerve fascicles associated with these tumors is a further technical refinement that permits the development of an extracapsular dissection plane for marginal resection and preservation of parent nerve fascicles, potentially reducing the incidence of postoperative motor neurological deficits. A limitation of this technique is that the use of manual ports and existing bipolar stimulators limit the reach within the intraperitoneal space. Ideally, bipolar stimulators would be integrated into the robotic instruments, such as a low-power, pulsed stimulation setting on the bipolar cautery instrument. This would allow rapid and dexterous mapping of the tumor surface.

CONCLUSION

These 3 cases demonstrate that a multidisciplinary approach to robot-assisted surgery with EMG stimulation mapping is a safe and effective treatment approach for retroperitoneal and pelvic nerve sheath tumors, providing excellent surgical visibility and operative precision with minimal morbidity and excellent outcomes.

Funding

This study was supported by the Intramural Research Program of the National Institute of Neurological Disorders and Stroke and the National Cancer Institute in Bethesda, Maryland, USA.

Disclosures

The authors have no personal, financial, or institutional interest in any of the drugs, materials, or devices described in this article. The views expressed are solely those of the authors and do not reflect the official policy or position of the US Navy, the Department of Defense, the National Institutes of Health, or the US Government.