A Q&A with MSK Spine Neurosurgeon Ori Barzilai, MD
Since 2019, MSK has been pioneering the use of robotic surgery for resecting complex, benign paraspinal tumors.
Spine neurosurgeon Ori Barzilai, MD, collaborates with thoracic, urologic, and gynecologic surgeons to perform minimally invasive robotic resections, which provide patients with shorter hospital stays, faster recoveries and less morbidities than traditional, open procedures.
Dr. Barzilai and colleagues recently published a retrospective case series in Operative Neurosurgery (Hagerstown). (1) They describe outcomes for the first 19 patients who underwent minimally invasive robotic surgery between 2019 and 2023 for noncancerous paraspinal tumors requiring thoracic, retroperitoneal, and transperitoneal approaches.
In this Q&A with Dr. Barzilai, we discuss the MSK experience that led to this innovation, the multidisciplinary collaboration in the operating room, highlights of strategies for addressing complex cases, how patient outcomes compared to traditional approaches, and what’s next.
Robotic surgery is common in orthopedics for placing screws and pins but has not traditionally been used for resecting benign spine tumors. How did the idea of using it for this indication arise?
My colleague, spine neurosurgeon Mark Bilsky, MD, Chief of our Multi-Disciplinary Spine Tumor Service, published a case report 10 years ago describing how he and his collaborators used a robotic technique for thoracoscopic resection of an apical paraspinal schwannoma. They used an intraoperative neuromonitoring probe to ensure safe dissection of the tumor from the T1 nerve root. The patient maintained full hand strength postoperatively. (2)
For that operation, only the thoracic surgeon used the robotic-assisted platform. However, the experience led us to explore the benefits of having a robotic-trained neurosurgeon actively perform robotic procedures. Fast forward a few years later, I’m fully convinced our multdisciplinary approach is a meaningful advance for the field of neurosurgery.
How does MSK’s approach differ from the standard of care offered at most other centers?
At other centers, the use of a computer-assisted platform in neurosurgery is typically limited to planning and executing the placement of spinal instrumentation, performing brain biopsies, and placing intra-cranial depth electrodes.
For our innovative approach to resecting paraspinal tumors, I work on one computer-assisted platform, and a robotic-trained colleague from another discipline works on a second platform during the same operation. The platforms provide 3D imaging and wristed instruments with a greater range of motion than the human hand, allowing for much smaller incisions than open procedures.2
To date, I have worked on a range of procedures with urologic surgeon Alvin Goh, MD, thoracic surgeon Bernard Park, MD, gynecologic surgeon Mario Leitao Jr., MD, FACOG, FACS, Director of the Minimal Access and Robotic Surgery Program, and colorectal cancer surgeon Martin Weiser, MD.
These collaborations have led to advances in our capabilities, including intraoperative nerve monitoring, (3) rib osteotomies, resection with intradural extension, and using a combined open-robotic back-front operations.
Has the FDA approved robotic surgery for this indication?
Our approach is still investigational. Since the FDA has not approved the computer-assisted surgical platform we use for neurosurgery, I train with and perform the operations with robotic-trained multidisciplinary colleagues.
Patients provide informed consent after consultations with both surgeons preoperatively, including extensive discussions about risks and the potential need to convert to a traditional open procedure if deemed necessary.
Extensive training and simulations will be required before this approach becomes the standard of care within neurosurgery.
What were the overall findings from your case series?
Only one of 19 cases was converted to open surgery. Gross total resection was achieved in all cases. The median estimated blood loss was low, only 50 cc, and the median length of stay was one day, with 9 patients (almost half) discharged on postoperative day 1 and three discharged on an outpatient basis. (1)
Compared to our previously reported outcomes for open resection of 25 tumors, the robotic approach resulted in a significantly higher gross resection rate (100% versus 60%) and a shorter length of stay of 1 day versus 5 days. (1)
While robotic surgery typically has longer surgical setup and operative times, we have found that the resection portion is comparable with and often shorter than traditional open approaches. However, our primary goal remains achieving total gross resection while optimizing patient outcomes.
An excellent example of our success with this approach is removing a schwannoma the size of a clementine that was pressing a 35-year-old woman’s spine and hip, causing stabbing pains that did not respond to pain medication or physical therapy. The patient was discharged the same day as her surgery and returned to running four months later.
Your patient series included novel workarounds for addressing complex cases. Can you provide highlights of what some of those procedures involved?
Removing a nerve sheath tumor with significant foraminal and spinal canal involvement is challenging, often requiring a facetectcomy and fusion. Drawing on our experience with microscopic resection of cervical nerve sheath tumors, we explored the feasibility of resecting the intradural tumor extension through an extraforaminal approach, the “outside-in” technique. The computer-assisted platform allowed us to separate the tumor from the pseudocapsule, minimizing the risk of encountering cerebrospinal fluid.
In another case, we needed access to small tumors located dorsally to a rib head, but the thoracic approach was obscured. As there are no available instruments for multifaceted robot drilling, we used a spinal endoscopic drill via the robotic assistant port (meaning it was operated manually by another surgeon) to resect the rib head and gain access to the dorsal tumor component. The improved exposure allowed us to resect all of the tumors.
For high-grade spinal cord compression cases, we combined an open-posterior with a lateral robotic approach to allow for a smaller dorsal exposure, rib resection, or thoracotomy as needed, providing a less invasive solution for the residual thoracic component. We started with a dorsal approach, decompressing the spinal cord, and then stabilized and resected nerve roots as needed.
Were there any adverse events in your initial patient series?
Two patients developed postoperative Horner’s syndrome, and one patient experienced an intraoperative low-flow cerebrospinal fluid leak that was closed primarily with sutures with no postoperative sequelae.
How many cases have you performed to date?
We have performed this surgical approach on 30 patients and have more procedures scheduled.
Previously, because recovery from open surgery was so challenging for patients, we usually offered active surveillance, especially for those with smaller tumors. Now that our technique is proving successful and recovery is much easier and faster, we are offering it to a broader range of patients.
What training is involved in using the robotic system?
In general, robotic surgery is very different from open or minimally invasive surgical approaches because it lacks haptic feedback. The transition to visual-only cues for resecting microscopic central nervous system tumors requires a significant learning curve and technical adjustments.
The training was challenging and required ongoing support from MSK, given that this technology had not been used before for neurosurgery. My training included online learning modules, robotic simulation, completing a checklist of standard skill-based tasks, learning the basics of the operating room setup in a dry laboratory, and approach-related training in an animal laboratory.
In addition to publishing our institutional experience, we present at international conferences.
What’s next for advancing robot-assisted surgery for benign spinal tumors?
We have only begun to realize the many potential possibilities with this approach. There will be more opportunities to integrate robotic technology into creating multilevel osteotomies, placing spinal implants, performing deformity corrections, and even resecting cancerous tumors when manufacturers develop enhanced surgical tools in the future.
