Special Topics: Technology and Innovations
Over the past decade, the indications for robotic surgery have expanded to include many surgical subspecialities including urology, gynecology, head and neck, cardiac, thoracic, bariatric, hepatobiliary, and colorectal surgery.1 Because most plastic surgery is not minimally invasive, the role of robots in plastic surgery is not immediately obvious. The title of this article, provided by the editorial staff, presupposes that robotic surgery may not “make sense,” and I imagine that this is what was meant. That said, the superb precision and high-resolution, three-dimensional optics make the robot a valuable instrument in the hands of the plastic surgeon.2,3 In this article, I describe the current clinical applications of robots in plastic surgery and some of the barriers to initiating a robotic practice.
There is currently one commercially available robotic surgical system, the da Vinci (Intuitive Surgical, Inc., Sunnyvale, Calif.). Using this platform, the surgeon sits at a console and remotely controls robotic arms and instruments at the surgical field in real time (Fig. 1). Gross movements of the surgeon’s arms translate into movements of the external robotic arms, and finer movements of the surgeon’s wrist and fingertips correspond to motion at the tips of the instruments. Robotic surgical motion is smooth and precise. Tremor is eliminated completely and there is motion scaling, meaning that the motion paths of the robotic instruments are smaller than the surgeon’s movements. There is no haptic, or sensory input from the surgical field to the console, and thus tissue response must be abstracted from visual information that is interpreted by the surgeon based on experience with conventional surgery. Although a disadvantage, this is easily compensated for after a brief period of cognitive adjustment.
ROBOTIC OPERATING TEAM
Because of the complex nature of the device and its underlying technology, a skilled team must be present to operate it. Sitting at the console and controlling the arms is straightforward. Setting up and troubleshooting a machine requires more experience. It is critical to work with a skilled scrub team that is trained in instrument use and exchange, calibration and cleaning of the endoscope and camera, port management, and vision tower control. Like a pit crew in car racing, having a good surgical team allows the surgeon to make adjustments and transitions quickly and seamlessly. If the team is unsure about the technology, its use will be a burden in both time and effort. Leadership of the operating room team is the surgeon’s responsibility, and thus a comprehensive understanding of both robotic hardware and software is critical.
My own robotic practice focuses on three clinical areas. Each of the following three areas was developed first in the laboratory and transferred to the operating room: (1) transoral robotic reconstructive surgery for head and neck reconstruction, permitting complex oropharyngeal reconstruction without dividing the lip or mandible; (2) robotic muscle harvest, permitting minimally invasive harvest of the latissimus dorsi and rectus abdominis muscle; and (3) robotic microanastomoses, extending the capabilities of the human hand in microvascular and microlymphatic surgery.
Transoral Robotic Surgery
Transoral robotic reconstructive surgery was developed as a minimally invasive approach to oropharyngeal tumors to avoid the morbidity of mandibular swing procedures and/or high-dose chemoradiation therapy.4–7 The technique has been adopted at multiple centers and was approved by the U.S. Food and Drug Administration in 2009. To perform transoral robotic reconstructive surgery, a mouth retractor sets the interdental opening, and two arms and the endoscope are introduced into the oropharynx, converging on the target anatomy (Fig. 2). To facilitate reconstructive access to a severely constrained physical space, transoral inset of a free flap or local flap using robotic assistance has been shown to be both effective and advantageous.8,9 Robotic reconstructive techniques can be used for a pure transoral approach or as a hybrid approach in conjunction with a lateral pharyngotomy, preserving an intact mandible in either scenario. By taking this approach, the plastic surgeon can provide minimally invasive reconstructive support for extirpation of a wide range of tumors that would be difficult to reconstruct through traditional methods.10,11 (See Video, Supplemental Digital Content 1, which demonstrates the transoral robotic reconstructive surgery technique, permitting complex oropharyngeal reconstruction without dividing the lip or mandible, available in the “Related Videos” section of the full-text article on PRSJournal.com or, for Ovid users, at http://links.lww.com/PRS/C78. The video is double speed to save time.)
Although developed for cancer resection, transoral robotic reconstructive surgery is applicable in any case that targets the posterior oropharynx. One such underdeveloped application is robotic cleft and velopharyngeal incompetence surgery. This is a natural extension of the current indications for transoral robotic reconstructive surgery and a fruitful area for further research and development.
Robotic Muscle Harvest
The latissimus dorsi and rectus abdominis muscle flaps are two of the most reliable and frequently used muscle flaps in reconstructive surgery. Harvest of these muscles requires a sufficiently long incision to access the muscle origin, insertion, and vascular supply. Such incisions are cosmetically undesirable and can be the source of donor-site morbidity.12,13 Endoscopic approaches to muscle harvest have largely been abandoned because of technical challenges. The robotic interface supplies the necessary exposure, picture clarity, precision, and control to accomplish both muscle and pedicle dissection with minimal external incisions.
Robotic harvest of the latissimus dorsi muscle can be performed through a skin-sparing or nipple-sparing mastectomy incision alone, or a short axillary incision for a free flap (Fig. 3). Dissection of the pedicle and anterior border of the muscle is performed through the open incision, which is temporarily closed to maintain insufflation (Fig. 4). Three ports are placed along the anterior border of the muscle to introduce the robotic instruments, and the patient-side cart is docked (Fig. 5). Insufflation provides the optical window for the robotic dissection (Fig. 6). (See Video, Supplemental Digital Content 2, which shows the positioning and motion of the patient-side cart during a robotic latissimus dorsi harvest, available in the “Related Videos” section of the full-text article on PRSJournal.com or, for Ovid users, available at http://links.lww.com/PRS/C79. The technique is facilitated by insufflation, and eliminates the need for a back incision. See Video, Supplemental Digital Content 3, which demonstrates the internal view of the harvest of the latissimus dorsi muscle, available in the “Related Videos” section of the full-text article on PRSJournal.com or, for Ovid users, available at http://links.lww.com/PRS/C80. The video is double speed to save time. First, the deep surface of the muscle is dissected off the chest wall, followed by dissection between the superficial surface of the muscle and the subcutaneous adipose tissue.) The incision is inconspicuous and the port sites are used as drain exit sites (Fig. 7).14,15 The technique is very reproducible, and I have not had to abort or convert to open surgery. For me, a robotic harvest requires 2 hours to perform, whereas an open harvest requires 1 hour.
In delayed-immediate breast reconstruction, patients receive a tissue expander at the time of mastectomy. If irradiation is required, the expander is deflated to optimize external beam delivery and then rapidly reexpanded at the conclusion of irradiation. Often, the reexpansion volume is adequate, but the skin is too thin or of poor quality for a lasting implant reconstruction. The robotic latissimus dorsi is a perfect solution. The muscle nourishes and protects the thin skin, and permits the surgeon to be aggressive when removing a thick, radiation-induced capsule (Fig. 8). There is virtually no visible donor site and, with a 1-day hospitalization and minimal pain, there is little downside (Fig. 9). I also occasionally use the robotic latissimus dorsi for upper-outer quadrant lumpectomy defects that are either too large or in an awkward location for local tissue rearrangement, or when volume reduction is undesirable. Finally, in immediate implant-based breast reconstruction, a muscle-only latissimus dorsi flap can be used interchangeably with acellular dermal matrix for lower pole or total implant coverage and support (Fig. 10), and a robotic latissimus harvest can be performed at approximately half the cost of most comparably sized acellular dermal matrices. (The variable cost of using the surgical robot for a latissimus dorsi flap harvest at M. D. Anderson is $1000 per case. This includes increased staff, operating room time, and disposables/reusables. Most pieces of acellular dermal matrix, priced between 25 and 30 cm2, will cost between $2000 and $3000 per breast.)
The rectus abdominis muscle is very useful as either a free or a pedicled flap. The robotic harvest of the muscle is performed intraperitoneally using three ports through the lateral abdominal wall.16 (See Video, Supplemental Digital Content 4, which demonstrates robotic harvest of the rectus abdominis muscle flap, available in the “Related Videos” section of the full-text article on PRSJournal.com or, for Ovid users, available at http://links.lww.com/PRS/C81. The video is double speed to save time. The harvest is performed intracorporeally, allowing easy access to the muscle and the pedicle. The flap can be used inside or outside the body, depending on the clinical scenario.) The ports are placed on the contralateral side of the muscle being harvested, between the iliac crest and the costal margin along a line connecting the anterior superior iliac spine and the midaxillary line (Fig. 11). The entire muscle can be pedicled for use in the pelvis, or extracted as a free flap using a laparoscopic retrieval bag (Fig. 12). The posterior rectus sheath is closed with a running suture (I prefer barbed), and there is no violation of the anterior rectus sheath.17 The robotic rectus harvest requires approximately 1 hour to perform, which is equivalent to open harvest time.
The robotic rectus harvest is very useful as an adjunct to the omentum to provide vascularized tissue for pelvic reconstruction following multiteam robotic resections. It is particularly effective in combination genitourinary/gastrointestinal procedures where significant pelvic dead space is created in an irradiated field, or where colonic and vesicourethral anastomoses are in close proximity to one another, placing them at risk for fistulization. If plastic surgery is the only service that requires a laparotomy in such a multiteam, robotic case, the minimally invasive advantages conferred to the patient by the collective robotic expertise of the other teams is lost.
Robotic Microvascular Anastomosis
The field of microsurgery has evolved very little from a technological standpoint over the past 100 years. Alexis Carrel developed the technique of microvascular anastomosis at the turn of the twentieth century; 60 years later, Jules Jacobsen developed the operating microscope. Since then, most incremental advances have been in flap design. Because robotic technology is capable of superhuman precision, the use of robotic technology for the anastomosis seems natural. In fact, it may be the ideal microsurgery platform by combining the executive functions of the human mind (judgment and decision-making) with the precision of a machine. Realizing this, several companies are designing and building surgical robots exclusively for microsurgery [e.g., MicroSure (Amersterdam, The Netherlands) and Medical MicroInstruments, Inc. (Pisa, Italy)].
Even the da Vinci system, which was designed for laparoscopy and represents a primitive microsurgical interface, has sufficient precision and optics to perform microvascular anastomoses, and provides advantages in certain situations.18 (See Video, Supplemental Digital Content 5, which shows a robotic microvascular anastomosis between a lateral circumflex femoral artery and a facial artery, available in the “Related Videos” section of the full-text article on PRSJournal.com or, for Ovid users, available at http://links.lww.com/PRS/C82. The video is double speed to save time. Advantages include compete tremor elimination and motion scaling.) Where access to recipient vessels is difficult (e.g., some head and neck cases, hepatic artery reconstruction, intraabdominal or intracerebral bypass) or vessels are sufficiently small caliber such that hand-suturing becomes difficult (e.g., left ventricular bypass, digital replantation, and other forms of supermicrosurgery), the robot can augment or overcome the limits of human technical ability.19–21 (See Video, Supplemental Digital Content 6, which demonstrates robotic lymphovenous anastomosis, available in the “Related Videos” section of the full-text article on PRSJournal.com or, for Ovid users, at http://links.lww.com/PRS/C83. The video is double speed to save time. Supermicrosurgery is technically extremely demanding. Robotic precision may allow microsurgeons to exceed the limits of human capabilities once optics and instrumentation improve.) Because of its inherent advantages, it is not unreasonable to suppose that a surgical robot might ultimately become a replacement technology for the microscope. Even at this point, robotic anastomotic times are roughly equivalent to open anastomotic times: between 15 and 45 minutes for a single vessel depending on the situation.18
BARRIERS TO ENTRY
Because it represents such early adoption, starting a robotic practice requires a significant investment in time, effort, and self-education, and surmounting a number of obstacles.
The most commonly cited obstacle to robotic surgery is cost. The robot is expensive to buy, but not to use. If the hospital already owns a robot, it will lose money if use is low and will recuperate cost by attracting patient and surgeon volume. The real determinant of whether a robotic case is profitable is the contribution margin per case. Robotic procedures can have a positive contribution margin compared to open cases if the revenue per procedure is higher than the cost. For this analysis, the price of the machine itself is irrelevant. Case revenue is composed of the professional fee charged by the surgeon, usually based on the Current Procedural Terminology code, and the hospital or technical fees, based on the Diagnosis-Related Group. There are no modifiers for robotic surgical procedures at this time, so the Current Procedural Terminology codes are the same as for the corresponding open procedures, and the diagnoses and thus Diagnosis-Related Groups are the same as well. Increased costs for robotic procedures are attributable to disposables/reusables, additional instrumentation, increased staffing levels, and increased procedural time. In contrast, costs may be decreased if minimally invasive procedures are associated with shorter lengths of stay, lower complication rates, and enhanced recovery, which all come out of a fixed Diagnosis-Related Group reimbursement. The balance of revenue and cost, or contribution margin per procedure, will determine whether it is cost-effective to perform a procedure robotically.
Robotic Training Pathways
As with most complex surgical tasks, graduated responsibility is a critical part of ascending the learning curve in robotic surgery, but there is no formal robotic plastic surgery training curriculum. Simulators are useful for learning to manipulate the instruments at the console, but do not provide experience with management of the patient-side cart (the component that interfaces with the patient). This is the more difficult and less glamorous skill set but one that is absolutely necessary to safely perform robotic surgery. (Urologists and general surgeons are significantly advantaged in patient-side cart management, as they spend considerable time in training providing bedside assistance during robotic cases and have a better overall knowledge of port management and minimally invasive techniques.) Most hospitals require that a surgeon receive the Intuitive certification, which involves a course. The Catch-22 is that Intuitive resists training plastic surgeons because of the lack of U.S. Food and Drug Administration–approved indications for the robot. This makes graduated responsibility difficult to acquire and creates a barrier to developing the necessary skill and experience.
The regulatory pathway for surgical robotics is difficult to navigate. Because there are no U.S. Food and Drug Administration–approved indications in plastic surgery, all use of the robot is “off-label.” The standard approach to surgical innovation outside of an approved indication is through an institutional review board protocol. Most institutional review boards will require an Investigational Device Exemption before approval, which needs to be granted by the U.S. Food and Drug Administration. This is a long road to travel.
Assuming you can overcome these external obstacles, there are a number of “in-house” sources of resistance. The standard operating procedures in the operating room are designed to create regularity and consistency, not promote surgical innovation. The hospital’s medical practice committee will likely have a privileging process that requires certification through Intuitive training, an internal proctoring system, or both (as in my hospital). Access to the robot may be limited because of short supply and high demand. My hospital has four robots, all at 100 percent use. I have block time, but it has been hard fought and, like many scarce resources, requires defense. These internal obstacles conspire against robotic innovation, and the cumulative effect of inertia and multiple hurdles may ultimately thwart a well-intended effort to initiate a practice.
WHY DO IT?
With so few applications and so many barriers, why would a plastic surgeon choose to develop a robotic practice? For me, it was and is about participating in the future of surgery. If you believe, as I do, that robotic technology will ultimately replace much conventional surgery, embracing the technology is a rational choice. Plastic surgery applications are less obvious because we are rooted in open surgery, but the advantages of robotic surgery will likely soon apply to open techniques as well. The precision and visualization of current platforms are only the beginning—coming enhancements include haptic feedback, anatomical navigation, and augmented reality using combinations of tissue elastography, perfusion analysis, and stereotactic guidance. Motion tracking in a virtual environment will replace the clumsy console just as it has in video games. Multiple smaller, sleeker patient interfaces will replace the large, cumbersome device of today. The avalanche of computing speed, networking capability, and robotic technology will ultimately make robotic surgery more convenient, safer, and more easily simulated (and thus taught) than conventional techniques—all things it is not at present.
After the first automobile was built, most people preferred horse and buggy for a decade. When Stephen Sasson invented the digital camera in 1975, the Board of Eastman-Kodak asked him why anybody would want to view a photograph on a computer, and sat on the technology for a decade. In the 1800s, the Luddites (British textile weavers union) nearly burned London to the ground to protest the introduction of the power loom, which automated the labor-intensive weaving process. The history of humanity is dense with resisted adoption of technology that augments human physical and intellectual capability, from the printing press, to the sewing machine, to the plow. In contrast, technology should not be implemented simply because it is available, and thoughtful integration is required. I have tried to steward the process of understanding and assimilating this technology safely and appropriately and I will continue to do so without an excess of passion or prejudice. Being involved with robotic surgery has expanded my conception of what is possible, and altered the way I think about the role of technology in society. It is in small part about the future direction of plastic surgery, but in larger part about the future direction of humanity in general. This newest development in our specialty, as all developments in plastic surgery invariably do, merely reflects the evolving world in which we live.
1. Lee SH, Lim S, Kim JH, Lee KYRobotic versus conventional laparoscopic surgery for rectal cancer: Systematic review and meta-analysis.Ann Surg Treat Res201589190–201
2. Selber JCRobotics in plastic surgery.Semin Plast Surg2014283–4
3. Selber JCRobotic surgery.J Reconstr Microsurg201228433–434
4. Weinstein GS, O’Malley BW Jr, Snyder W, Hockstein NGTransoral robotic surgery: Supraglottic partial laryngectomy.Ann Otol Rhinol Laryngol200711619–23
5. Weinstein GS, O’Malley BW Jr, Snyder W, Sherman E, Quon HTransoral robotic surgery: Radical tonsillectomy.Arch Otolaryngol Head Neck Surg20071331220–1226
6. Park YM, Lee WJ, Lee JG, et alTransoral robotic surgery (TORS) in laryngeal and hypopharyngeal cancer.J Laparoendosc Adv Surg Tech A200919361–368
7. Genden EM, Desai S, Sung CKTransoral robotic surgery for the management of head and neck cancer: A preliminary experience.Head Neck200931283–289
8. Selber JC, Robb G, Serletti JM, Weinstein G, Weber R, Holsinger FCTransoral robotic free flap reconstruction of oropharyngeal defects: A preclinical investigation.Plast Reconstr Surg2010125896–900
9. Selber JCTransoral robotic reconstruction of oropharyngeal defects: A case series.Plast Reconstr Surg20101261978–1987
10. Longfield EA, Holsinger FC, Selber JCReconstruction after robotic head and neck surgery: When and why.J Reconstr Microsurg201228445–450
11. Selber JC, Sarhane KA, Ibrahim AE, Holsinger FCTransoral robotic reconstructive surgery.Semin Plast Surg20142835–38
12. Ascherman JA, Seruya M, Bartsich SAAbdominal wall morbidity following unilateral and bilateral breast reconstruction with pedicled TRAM flaps: An outcomes analysis of 117 consecutive patients.Plast Reconstr Surg20081211–8
13. Kroll SS, Schusterman MA, Reece GP, Miller MJ, Robb G, Evans GAbdominal wall strength, bulging, and hernia after TRAM flap breast reconstruction.Plast Reconstr Surg199596616–619
14. Selber JC, Baumann DP, Holsinger CFRobotic harvest of the latissimus dorsi muscle: Laboratory and clinical experience.J Reconstr Microsurg201228457–464
15. Selber JC, Baumann DP, Holsinger FCRobotic latissimus dorsi muscle harvest: A case series.Plast Reconstr Surg20121291305–1312
16. Ibrahim AE, Sarhane KA, Pederson JC, Selber JCRobotic harvest of the rectus abdominis muscle: Principles and clinical applications.Semin Plast Surg20142826–31
17. Patel NV, Pedersen JCRobotic harvest of the rectus abdominis muscle: A preclinical investigation and case report.J Reconstr Microsurg201228477–480
18. Selber JC, Alrasheed TRobotic microsurgical training and evaluation.Semin Plast Surg2014285–10
19. Parekattil SJ, Gudeloglu A, Brahmbhatt J, Wharton J, Priola KBRobotic assisted versus pure microsurgical vasectomy reversal: Technique and prospective database control trial.J Reconstr Microsurg201228435–444
20. Lee JY, Mattar T, Parisi TJ, Carlsen BT, Bishop AT, Shin AYLearning curve of robotic-assisted microvascular anastomosis in the rat.J Reconstr Microsurg201228451–456
©2017American Society of Plastic Surgeons
21. Willems JI, Shin AM, Shin DM, Bishop AT, Shin AYA comparison of robotically assisted microsurgery versus manual microsurgery in challenging situations.Plast Reconstr Surg20161371317–1324