Journal Logo

Technical Reports

Advancing Surgical Simulation in Gynecologic Oncology

Robotic Dissection of a Novel Pelvic Lymphadenectomy Model

Kiely, Daniel J. MDCM; Gotlieb, Walter H. MD, PhD; Jardon, Kris MD, MSc; Lau, Susie MD; Press, Joshua Z. MD, MSc

Author Information
Simulation in Healthcare: Journal of the Society for Simulation in Healthcare: February 2015 - Volume 10 - Issue 1 - p 38-42
doi: 10.1097/SIH.0000000000000054

Abstract

Pelvic lymphadenectomy is an essential component of many surgical procedures in gynecologic oncology. To perform the procedure safely, the following structures are identified: superior vesical artery, ureter, obturator nerve, as well as external iliac artery and vein. In addition, the relationship of these structures to the genitofemoral nerve, psoas muscle, internal iliac artery, as well as obturator artery and vein must be understood. Pelvic lymphadenectomy involves dissecting lymphatic tissue adjacent to the external and internal iliac vessels, superior to the obtuator nerve, and lateral to the ureter. Major complications can occur as a result of surgical error in the performance of this procedure. Injury to external iliac vessels can lead to substantial bleeding. Injury to the ureter or obturator nerve requires surgical repair. The 3-dimensional (3D) relationship between these structures may be difficult for trainees to learn from the 2-dimensional representations in textbooks. Evidence is emerging that physical models may lead to improved learning as compared with textbooks or even 3D (virtual) computer models.1

One model for surgical training in laparoscopic pelvic lymph node dissection has previously been described, composed of fluid-filled penrose drains to model iliac vessels, placed on a block of wood with cotton balls and beans serving as lymph nodes and a plastic wrap simulating the peritoneum.2 Our aim was to build on this previous effort to create a model with more accurate 3D anatomic detail and thereby provide a more realistic training experience.

Robotic surgery is a fertile ground for simulation research because of the challenges of intraoperative teaching, especially the potential for a single surgeon to control all 3 robotic arms and the robotic camera (using a combination of foot pedals and hand controls), thereby eliminating the need for a surgical assistant.3 Furthermore, simulation of complex procedures such as pelvic lymphadenectomy is in its infancy in robotic surgery: the vast majority of work so far, particularly in virtual-reality simulation, has focused on basic robotic surgical skills such as transferring pegs and placing a suture.3

There is a hierarchy of skills required in surgery, and skills at a basic level are required for accomplishment of skills required at higher levels. We envision this hierarchy with 2 basic levels, technical performance and knowledge of anatomy. The higher level, surgical dissection, is based on the integration of anatomic knowledge and technical skills. Our model aims to challenge the trainee to practice at this level.

In this study, we describe a low-cost model of pelvic lymphadenectomy. We are optimistic that this tool will be beneficial for surgical training given the rapidly accumulating evidence in favor of simulation training in surgery4–7 and in medicine in general.8 Our model builds on a tradition of surgical simulation in gynecologic oncology9–15 and of inanimate models for surgical training.2,9,16–22

METHODS

Ethics approval was obtained from the McGill Faculty of Medicine Institutional Review Board. The model was built using the components shown in Table 1. The frame of the model consisted of a commercially available model of the bony pelvis (Female Pelvic Skeleton Model A61, 3B Scientific). Several novel techniques were used in model design. First, colored rubber tubing was stented with thin, pliable wire to model the blood vessels. By directing the wire out of nicks in the tubing and then covering the wire with additional pieces of tubing, major branches of the blood vessels were modeled. Smaller branches were modeled with primary wire (Fig. 1). Second, visualization of a 3D virtual, computer-based anatomic model was used to guide the positioning of structures in 3D space (Interactive Pelvis and Perineum, Primal Pictures Ltd, London, England). Anatomic structures were positioned according to the principle that by knowing the origin and insertion of structures in the bony pelvis, their course could be defined. Third, the entire model was placed in a stainless steel bowl, and the structures of interest were secured in place with clamps attached to the rim of the bowl, thus preserving their course in 3D space (Fig. 2). Fourth, the model was covered with adherent plastic wrap to simulate the peritoneum after all the structures of interest were positioned and then hot, liquid gelatin solution was poured into the steel bowl to immerse the model. The gelatin solution was then allowed to cool, creating a substance that modeled the loose areolar and adipose tissue surrounding the lymph nodes and blood vessels. Without this technique, we would have great difficulty placing model adipose tissue in the very narrow confines and crevices between the blood vessels, nerves, and lymph nodes. Interfaces between the gelatin matrix, the plastic wrap, the cotton balls, and the model blood vessels simulated the dissection planes. Detailed instructions on how to create the model are available (see Text Document, Supplementary Digital Content 1, https://links.lww.com/SIH/A144, which provides a step-by-step description) (see Video, Supplementary Digital Content 2, https://links.lww.com/SIH/A145, which demonstrates assembly of the model) (see Video, Supplementary Digital Content 3, https://links.lww.com/SIH/A146, which demonstrates assembly of the model) (see Video, Supplementary Digital Content 4, https://links.lww.com/SIH/A147, which demonstrates assembly of the model) (see Video, Supplementary Digital Content 5, https://links.lww.com/SIH/A148, which demonstrates assembly of the model) (see Video, Supplementary Digital Content 6, https://links.lww.com/SIH/A149, which demonstrates assembly of the model) (see Video, Supplementary Digital Content 7, https://links.lww.com/SIH/A150, which demonstrates assembly of the model).

T1-6
TABLE 1:
Approximate Cost of the Robotic Pelvic Lymph Node Dissection Model (US Dollars)
F1-6
FIGURE 1:
Pelvic lymph node model with model vasculature and ureter. 1, external iliac artery; 2, external ilac vein; 3, anterior division of internal iliac artery; 4, ureter; 5, internal iliac vein; 6, obturator nerve; 7, obturator artery; 8, superior vesical artery; 9, uterine artery; 10, iliacus muscle; 11, ischio-coccygeus muscle; 12, bladder; 13, posterior division of internal iliac artery; 14, piriformis muscle; 15, obturator internus muscle.
F2-6
FIGURE 2:
Pelvic lymph node model with clamps in place. 1, ureter; 2, common iliac artery; 3, common iliac vein; 4, rectosigmoid; 5, infundibulopelvic ligament; 6, uterus; 7, round ligament; 8, utero-ovarian ligament; 9 –ovary; 10, lymph node; 11, external iliac vein; 12, external iliac artery; 13, bladder. Note that several additional lymph nodes are added before the dissection.

The cost of our model was approximately $200. The components of the model and their individual costs are shown in Table 1. Building the model for the first time takes approximately 5 hours. To clean the model and rebuild it for a subsequent dissection takes approximately 1 to 2 hours plus overnight in a refrigerator for the gelatin solution to set. The cost of reusing the model is approximately $20, including the cost of the gelatin mix, suture, cotton balls, adherent plastic wrap, and a few silk sutures.

All gynecologic oncologists and gynecologic oncology fellows at the Jewish General Hospital and Royal Victoria Hospital in Montreal, Quebec, Canada (n = 8), were invited to dissect the model using the surgical robot (da Vinci Si Surgical System, Intuitive Surgical, Inc, Sunnyvale, CA) and then complete a 10-item survey assessing the face and content validity of the simulation. Comments were elicited from participating surgeons regarding ways in which the model could be improved. Electrosurgery was not used during the dissection because we have not studied the response of our model to this modality. Figure 3 shows the setup on dissection day, and Figure 4 shows the model just before dissection. Costs of running a dissection session included salary for a nurse specialized in robotics to be present for the setup, operation, and shut down of the surgical robot as required by our institution, and the cost per life used of the robotic training arms. Each da Vinci Si Surgical System training arm costs $3000 (as per our institution’s experience) and is programmed to last 30 lives. One life lasts from the time when the training arm is inserted into the robotic system until the time when the robot is shut down, regardless of how much it is used during that training day. Therefore, the cost for this component was $200 to $300 per training session depending on whether 2 or 3 training arms were used. The training instruments we used were Maryland bipolar forceps (nondominant hand) and curved scissors (dominant hand). For dissections with a third robotic arm, we added the ProGrasp forceps.

F3-6
FIGURE 3:
Setup of the pelvic lymph node model during a trial robotic dissection.
F4-6
FIGURE 4:
Pelvic lymph node model just before dissection with robotic instruments.

Further validation of the model occurred at the Annual Society of Gynecologic Oncologists of Canada (GOC) meeting on June 14, 2013, where participants were invited to complete a short survey (4 items) assessing the model, after viewing a poster presentation accompanied by a short video displaying highlights of the robotic dissections and the model itself.

RESULTS

The pelvic lymph node model was dissected by 3 gynecologic oncologists (of a possible 5) and 2 gynecologic oncology fellows (of a possible 3), each of whom completed the postdissection survey. All the dissections were recorded (see Video, Supplementary Digital Content 8, https://links.lww.com/SIH/A151, which demonstrates highlights) (see Video, Supplementary Digital Content 9, https://links.lww.com/SIH/A152, which demonstrates highlights) (see Video, Supplementary Digital Content 10, https://links.lww.com/SIH/A153, which demonstrates highlights) (see Video, Supplementary Digital Content 11, https://links.lww.com/SIH/A154, which demonstrates highlights) (see Video, Supplementary Digital Content 12, https://links.lww.com/SIH/A155, which demonstrates highlights). Two of the gynecologic oncologists who performed dissections were also involved in the research project in a supervisory role. At the GOC meeting, 5 gynecologic oncologists completed our survey, of approximately 12 who viewed the poster presentation.

Ratings by gynecologic oncologists and fellows who dissected the model were high, with almost all ratings 4 of 5 or higher, including a range from 3 to 5 (Tables 2 and 3). Ratings at the conference were somewhat lower but still strong, with a range between 3 and 5 (Table 2). The model was rated as “useful for training throughout residency and fellowship” unanimously by the 5 participants who dissected the model.

T2-6
TABLE 2:
Median Likert Scale Ratings and Range (in Parentheses) for Questions Answered by Participants Who Dissected the Pelvic Lymph Node Model Using the Surgical Robot (da Vinci Surgical System) and by Participants at the Society of Gynecologic Oncology of Canada Meeting
T3-6
TABLE 3:
Median Likert Scale Ratings and Range (in Parentheses) for Additional Questions Asked Only of Participants Who Dissected the Pelvic Lymph Node Model Using the Surgical Robot (da Vinci Surgical System)

Participants who rated the model made several useful suggestions as to ways the model might be improved. First, position the superior vesical artery more cephalad and the obturator artery deeper in the pelvis. Second, make the lymph nodes smaller and add cotton batting to create connections between the lymph nodes. Third, thicken the gelatin solution so that it better represents fatty tissue. Fourth, develop fluid-filled tubes with fragile walls to simulate blood vessels with a real possibility of vascular injury. These suggestions are being considered for incorporation into future versions of the model.

DISCUSSION

Our study has a number of strengths, which support the utility of this model. The inanimate pelvic lymph node dissection model we have developed incorporates detailed 3D anatomic features, using materials that emulate the sensation of genuine tissues. The model incorporates innovations in 2 key areas, which are challenging to simulate: accurate complex vascular patterns and realistic dissection planes. The high ratings received for face and content validity indicate that this model has potential as a training tool for pelvic lymphadenectomy.

As in most initial validation studies of novel simulations, finding participants to rate the model with no previous research relationship to the lead author and designer of the simulation was difficult. The evaluation at the Society of Gynecologic Oncology of Canada (GOC) meeting was an attempt to compensate for this. Of note, that a video demonstrating assembly and dissection of the model was selected via a peer review process for oral presentation at the Society of Gynecologic Oncology (SGO) Annual Meeting on Women’s Cancer is an additional evidence of the validity of the simulation.

Although the model was assessed by robotic dissection, the structure of the model would also allow it to be used for training in open or laparoscopic surgery. This versatility is one of the advantages of a physical model over a virtual-reality model. It is worth noting that developing a physical model may later inform the development of a virtual-reality model as it is easier for a virtual reality team to study the texture and location of anatomic structures in a physical model than in an animal or human cadaver. Our model may facilitate distributed practice by being lower cost than alternatives such as training on animals3 or human cadavers.23,24 Distributed practice, defined as practice in regular, short sessions designed to allow maximal concentration and effort during training, has been correlated with improved learning in surgery25,26 and other domains.27

We have identified some important limitations with our current model, which we anticipate can be addressed in creating future prototypes. First, there is room to improve the anatomic fidelity of the model in terms of haptic properties, color, and appearance. Second, the evaluation of the model is limited by the fact that our questionnaire has not been previously validated and by the relatively small number of respondents. Third, the 5 participants at the conference based their ratings of the model on a video of the dissection rather than on actual dissection. Fourth, preparing the model takes a significant amount of time, which may be considered a hidden cost. However, 3D printing of model components28–30 or collaboration with industry is an option that may facilitate reproduction of our prototype with greater efficiency in the future. Fifth, the model does not presently allow for incorporation of electrosurgery (monopolar and bipolar energy). Overcoming this important limitation will require further study as to how various inanimate materials respond to electrosurgery and as to the safety of the fumes that may be produced. In the meantime, it is possible for the trainee to state when energy would be applied. Sixth, the model does not presently incorporate fluid-filled blood vessels that realistically simulate vascular injury. This limitation is not insurmountable as other groups have described ways to simulate pulsatile, fluid-filled blood vessels.31,32 Seventh, the inclusion of a simulated body wall designed for robotic surgery simulation, such as the clear plastic prototype described by Marecik et al,17 may help to improve the realism of the model. This may allow more accurate placement of prespecified trocar port sites and further ensure that they remain fixed in space. Of note, the robotic trocars are designed to respect their designated pivot point in space once docked, a feature intended to protect the patient’s abdominal wall from trauma.33 This feature allowed the current version of our model to be functional without a simulated body wall but required us to position the trocars using a gestalt approach.

Despite all these limitations, our model has nevertheless introduced a number of novel techniques that open the door to more complex inanimate models for simulation training in gynecologic oncology.

In conclusion, we have developed a novel inanimate model of pelvic lymph node dissection, particularly suitable for training gynecologic oncologists, and have demonstrated the face and content validity of the model through assessment using the robotic approach. At a conceptual level, we have demonstrated the feasibility of modeling complex anatomy in a low-cost manner to allow the most advanced type of surgical simulation: dissection, which integrates technical skills, and anatomic knowledge. Further work may include developing the model to be used by open and laparoscopic approaches and attempting to demonstrate construct validity of the model. The importance of developing simulations of procedures rather than just basic tasks for robotic surgery simulation has recently been highlighted.3 Our model is a step in this direction, by simulating the procedure of pelvic lymphadenectomy.

ACKNOWLEDGMENT

The authors wish to acknowledge Sonia Brin and Claire Deland from nursing for the support with logistics.

REFERENCES

1. Preece D, Williams SB, Lam R, Weller R. “Let’s get physical”: advantages of a physical model over 3D computer models and textbooks in learning imaging anatomy. Anat Sci Educ 2013; 6: 216–24.
2. Bowring J, Shepherd JH, Ind TE. Assessment of an in vitro model for laparoscopic pelvic lymphadenectomy. BJOG 2007; 114: 964–969.
3. Hung AJ, Jayaratna IS, Teruya K, et al. Comparative assessment of three standardized robotic surgery training methods. BJU Int 2013; 112: 864–871.
4. Sroka G, Feldman LS, Vassiliou MC, et al. Fundamentals of laparoscopic surgery simulator training to proficiency improves laparoscopic performance in the operating room—a randomized controlled trial. Am J Surg 2010; 199: 115–120.
5. Stegemann AP, Ahmed K, Syed JR, et al. Fundamental skills of robotic surgery: a multi-institutional randomized controlled trial for validation of a simulation-based curriculum. Urology 2013; 81: 767–774.
6. Grantcharov TP, Kristiansen VB, Bendix J, et al. Randomized clinical trial of virtual reality simulation for laparoscopic skills training. Br J Surg 2004; 91: 146–150.
7. Palter VN, Grantcharov TP. Development and validation of a comprehensive curriculum to teach an advanced minimally invasive procedure: a randomized controlled trial. Ann Surg 2012; 256: 25–32.
8. Cook DA, Hatala R, Brydges R, et al. Technology-enhanced simulation for health professions education: a systematic review and meta-analysis. JAMA 2011; 306: 978–988.
9. Finan MA, Clark ME, Rocconi RP. A novel method for training residents in robotic hysterectomy. J Robotic Surg 2010; 4: 33–39.
10. Finan MA, Silver S, Otts E, Rocconi RP. A comprehensive method to train residents in robotic hysterectomy techniques. J Robotic Surg 2010; 4: 183–190.
11. Hoffman MS. Simulation of robotic hysterectomy utilizing the porcine model. Am J Obstet Gynecol 2012; 206: 523.e1–523.e2.
12. Hoffman MS, Ondrovic LE, Wenham RM, et al. Evaluation of the porcine model to teach various ancillary procedures to gynecologic oncology fellows. Am J Obstet Gynecol 2009; 201: 116.e1–116.e3.
13. Hammond I, Taylor J, McMenamin P. Anatomy of complications workshop: an educational strategy to improve performance in obstetricians and gynaecologists. Aust N Z J Obstet Gynaecol 2003; 43: 111–114.
14. Hammond I, Taylor J, Obermair A, et al. The anatomy of complications workshop: an educational strategy to improve the training and performance of fellows in gynecologic oncology. Gynecol Oncol 2004; 94: 769–773.
15. Goff BA. Surgical education for gynecologic oncologists. Gynecol Oncol 2009; 114: S45–S6.
16. Barrier BF, Thompson AB, McCullough MW, et al. A novel and inexpensive vaginal hysterectomy simulator. Simul Healthc 2012; 7: 374–379.
17. Marecik SJ, Prasad LM, Park JJ, et al. A lifelike patient simulator for teaching robotic colorectal surgery: how to acquire skills for robotic rectal dissection. Surg Endosc 2008; 22: 1876–1881.
18. Botden SM, Goossens R, Jakimowicz JJ. Developing a realistic model for the training of the laparoscopic Nissen fundoplication. Simul Healthc 2010; 5: 173–178.
19. Hong A, Mullin PM, Al-Marayati L, et al. A low-fidelity total abdominal hysterectomy teaching model for obstetrics and gynecology residents. Simul Healthc 2012; 7: 123–126.
20. Kurashima Y, Feldman L, Al-Sabah S, et al. A novel low-cost simulator for laparoscopic inguinal hernia repair. Surg Innov 2011; 18: 171–175.
21. Siddighi S, Kleeman SD, Baggish MS, et al. Effects of an educational workshop on performance of fourth-degree perineal laceration repair. Obstet Gynecol 2007; 109: 289–294.
22. Lee JY, Mucksavage P, McDougall EM. Simulating laparoscopic renal hilar vessel injuries: preliminary evaluation of a novel surgical training model for residents. J Endourol 2012; 26: 393–397.
23. Levine RL, Kives S, Cathey G, et al. The use of lightly embalmed (fresh tissue) cadavers for resident laparoscopic training. J Minim Invasive Gynecol 2006; 13: 451–456.
24. Cundiff GW, Weidner AC, Visco AG. Effectiveness of laparoscopic cadaveric dissection in enhancing resident comprehension of pelvic anatomy. J Am Coll Surg 2001; 192: 492–497.
25. Moulton CA, Dubrowski A, Macrae H, et al. Teaching surgical skills: what kind of practice makes perfect? A randomized, controlled trial. Ann Surg 2006; 244: 400–409.
26. Schaverien MV. Development of expertise in surgical training. J Surg Educ 2010; 67: 37–43.
27. Anders Ericsson K, Krampe RT, Tesch-Römer C. The role of deliberate practice in the acquisition of expert performance. Psychol Rev 1993; 100: 363–406.
28. Jones N. Science in three dimensions: the print revolution. Nature 2012; 487: 22–23.
29. Lantada AD, Morgado PL. Rapid prototyping for biomedical engineering: current capabilities and challenges. Annu Rev Biomed Eng 2012; 14: 73–96.
30. Abboud M, Orentlicher G. Computer-aided manufacturing in medicine. Atlas Oral Maxillofac Surg Clin North Am 2012; 20: 19–36.
31. Miller SF, Sanz-Guerrero J, Dodde RE, et al. A pulsatile blood vessel system for a femoral arterial access clinical simulation model. Med Eng Phys 2013; 35: 1518–1524.
32. Fann JI, Caffarelli AD, Georgette G, et al. Improvement in coronary anastomosis with cardiac surgery simulation. J Thorac Cardiovasc Surg 2008; 136: 1486–1491.
33. Intuitive Surgical®. The da Vinci Surigcal System Web site. Patient side cart. Available at: http://www.intuitivesurgical.com/products/davinci_surgical_system/. Accessed March 15, 2014.
Keywords:

Anatomic models; Patient simulation; Graduate medical education; Operative surgical procedures; Lymph node excision; Gynecology

Supplemental Digital Content

© 2015 Society for Simulation in Healthcare