Pediatric Robotic and Reconstructive Urology: A Comprehensive Guide

Robotic urological surgery is one of the most significant urological developments in recent years. It allows for greater precision than laparoscopic methods while retaining quicker recovery time and reduced morbidity over classical open surgical techniques. For children, where the room for error is already reduced because of smaller anatomy, it takes on even more importance for urologists.  As a result, robotic surgery is rightly considered one of the most exciting contemporary developments in pediatric urology.

Pediatric Robotic and Reconstructive Urology: A Comprehensive Guide provides specialist and trainees with an innovative text and video guide to this dynamic area, in order to aid mastery of robotic approaches and improve the care of pediatric patients.

Full-color throughout and including over 130 color images, this comprehensive guide covers key areas including:

• Training, instrumentation and physiology of robotic urologic surgery
• Surgical planning and techniques involved
• Adult reconstructive principles applicable to pediatrics
• Management of complications, outcomes and future perspectives for pediatric urologic surgery

Also included are 30 high-quality surgical videos illustrating robotic surgery in action, accessed via a companion website, thus providing the perfect visual tool for the user.

With chapters authored by the leading names in the field, and expertly edited by Mohan Gundeti, this ground-breaking book is essential reading for all pediatric urologists, pediatric surgeons and general urologists, whether experienced or in training.

Of related interest

Smith's Textbook of Endourology, 3E
Smith, ISBN 9781444335545

Pediatric Urology: Surgical Complications and Management
Wilcox, ISBN 9781405162685

• Language: English
• Publisher: Wiley
• Release date: Feb 3, 2012
• ISBN: 9781444345278

Book preview

I

History, Training, Instrumentation, and Physiology

1

The Evolution of Robotic Surgery and Its Clinical Applications

Shyam Sukumar, Mahendra Bhandari and Mani Menon

Vattikuti Urology Institute, Henry Ford Health System, Detroit, MI, USA

Robotics makes the transition to the Information Age complete by looking at information (the monitor) and manipulating information (hand motions send electronic signals which control the tip of the instruments). It is no longer blood and guts, but it is bits and bytes.

Richard M. Satava [1]

Introduction

Surgical robots assist the modern-day surgeon in achieving a technically satisfactory performance by enabling precise surgical actions on a clearly visible surgical target. Robotic surgery preserves the precision of tissue handling even in remote body cavities, which often are visually and tactilely inaccessible to traditional open surgery. A synergy in real-time three-dimensional, magnified vision translates the psychovisual coordination of the surgeon's fingers to fine instruments with an unprecedented seven degrees of freedom, thereby minimizing collateral tissue insult. The speedy return to baseline function following surgery, a unique patient-centric comfort profile, the mitigation of operative complications, and its safe application for intricate surgical procedures even in sick patients have increased its popularity among surgeons and patients alike. In this chapter, we venture off the beaten path, avoiding the usual chronological narrative of various industrial robots to focus instead on the seminal role of robotic surgery in the evolution of minimal access; we also place special emphasis on the clinical applications and interdisciplinary cross-fertilization that has ensued.

Evolution of surgical access

Minimally invasive surgery in general and robotic surgery in particular have challenged the age-old practice of long or multiple incisions and wider exposure to handle complex surgeries. Studies examining the systemic effect of surgical intervention (including the size and location of incisions) support the beneficial effect of minimally invasive surgery on the biological and immunological level [2,3].

Minimal access

John Wickham, a British urologist, coined the term minimally invasive surgery in 1983 [4–6]. A vociferous champion of endoscopic techniques, he classified the history of surgery in three phases, the first being the brutal and ablative medieval epoch, followed by the era of improved resuscitation and carefree incisions, and lastly the modern age beginning from the 1960s onwards. The modern age involved small pockets of enlightened surgeons gradually embracing techniques involving minimal access [4].

The possibility of intra-abdominal insufflations and endoscopic visualization (and hence the laparoscopic concept) was first reported by G. Kelling, who insinuated a cystoscope trans-abdominally and applied pneumoperitoneum for treating blood loss [7]. Contemporaneously, culdoscopic access was described by D.O. Ott in a pregnant woman [7]. A significant milestone was the application of the Veress needle, initially developed for the creation of therapeutic pneumothorax to treat tuberculosis [7].

Kurt Semm further revolutionized the field of laparoscopic surgery by advocating the keyhole approach in gynecologic surgery, and controversially performing the first laparoscopic appendectomy. He is to be credited with developing endoscopic hemostatic suturing and, more than anyone else, was singularly influential in designing the range of innovative instruments indispensable for minimal access [8].

Erich Muhe performed the first human laparoscopic cholecystectomy in 1985. The French triumvirate of Mouret, Dubois, and Perissat then standardized and popularized laparoscopic cholecystectomy as a mainstream procedure [9].

The evolution of tools and techniques for port placement has significantly reduced access-related morbidity such as vascular, bowel, and wound site complications and mitigated oncologic complications related to access – the abdominal metastatic seeding of neoplasms [10]. Access for most urologic procedures involves one of three approaches: open (Hasson), closed (using a Veress needle), or optical ports [11]. Optical ports allow direct visualization of the various tissue layers during access. Trocar type has moved from the use of pyramidal and conical trocars to the use of trocars with bladeless tips, which are less traumatic and hence result in favorable wound parameters.

Evolution of vision systems

The single seminal breakthrough which paved the way for state-of-the-art endoscopic surgery was the development of the fiber-optic system. It was indeed a giant leap for surgeons – from the era of working in natural passages under candlelight brightness through incandescent bulb illumination energized by dry cells to bright vision, which could be carried to any body cavity through fiber-optic carriers.

Taking a radical approach to endoscopic clarity and definition, Harold Hopkins (along with Kapany) published a report in Nature in 1954 describing the fiberscope, in which light passed through bundles of fine fibers of glass increased the quality of the images. Building on Hopkins’ work, Basil Hirshowitz developed the first clinically useful gastroscope [12]. Hopkins was responsible for another seminal innovation, this time in cystoscopy (his initial assignment being to capture images inside the bladder), with his introduction of the rod lens system. Whereas the traditional endoscope consisted of a tube of air with lenses of glass (as the objective) Hopkins’ version consisted of a tube of glass with lenses of air, which greatly increased the amount of light transmitted [12]. Having been denied a conducive commercial response in Britain, Hopkins was fortunate enough to find a collaborator in Karl Storz [12], thereby beginning, arguably, the most important collaboration in the history of endoscopic instrumentation.

Another obstacle with the early systems was that they hampered concerted coordination between the chief surgeon and his or her assistant by limiting the visual plethora of the surgical field to the endoscopist, and the participation of an assistant was dependent on a verbal transliteration of the surgical field. Photodocumentation, so critical for consultation and education, was also severely hampered [13].

The introduction of a miniature electronic camera that transduced the afferent optical image into efferent electronic impulses with the help of a charge-coupled device (CCD) radically transformed the way in which endoscopy was performed [13]. Thus electronic video-endoscopy was born, combining the electronic endoscope and the television. The entire surgical team could now view the magnified images and collaboration between surgeons was taken to a whole new level.

Operating rooms equipped for robotic surgeries with the da Vinci Surgical System provided 3D stereoscopic vision for the operating surgeon but the assistants and allied personnel were only allowed to visualize a 2D version of the unfolding events. The augmented-reality surgical suite, first proposed and implemented by Shrivastava and Menon [14], made tangible stereoscopic 3D vision a reality for everyone in the operating room (Figure 1.1).

Figure 1.1 An augmented-reality operating room at the Vattikuti Urology Institute, Henry Ford Hospital, Detroit, MI.

ch01fig001.eps

While the vision systems improved, the operating endoscopist's ability to manipulate the surgical field themselves was still impaired by the assistant holding the camera in place. This is where robotic technology first made its mark. The first robot to have the US Food and Drug Administration (FDA)'s imprimatur was AESOP (Automated Endoscopic System for Optimal Positioning). Computer Motion (Berkeley, CA, USA), which first introduced AESOP in the mid-1990s, was initially funded by a NASA research grant for the development of a robotic arm for the US space program. AESOP used voice (or, alternatively, foot or hand) control to direct the movements of a robotic arm, which usually held the laparoscope (but could also hold a retractor). The surgeon used a preprogrammed voice card that allowed the device to understand and respond to his or her commands. Allaf et al. [15] examined the optimal interface (voice versus foot control) to manipulate the robot and found that foot control was faster but voice control was more accurate. Kavoussi et al. [16] found the laparoscopic camera positioning to be significantly steadier with less inadvertent movements when under robotic control and concluded that operative times during dissections were not significantly different between robot-assisted and human-assisted procedures.

Telepresence platforms

The da Vinci system

The state-of-the-art platform for the performance of telemanipulative procedures is the da Vinci Surgical System (Intuitive Surgical, Sunnyvale, CA, USA), which is a master–slave system. The system has three separate parts that dovetail to produce an immersive interface.

The console

The surgeon is ensconced at the console, which is designed as an ergonomically comfortable perch with his or her hands fitting into the masters (basically, freely mobile finger controls). In the United States, the FDA has made it mandatory for the console to be in the same room as the patient.

The console also consists of a stereoscopic viewer with sensors. The act of placing the (surgeon’s) head into the console vivifies the system and removing it deactivates and locks the robotic arms. The surgeon's dexterous hand movements are electronically translated to the robotic instruments in direct contact with the operative target. Foot controls for camera positioning and diathermy complete the console.

Stereoscopic vision system

The stereoscopic vision system provides high-resolution 3D images of the operative field to the surgeon at his or her vantage point in the console. The binocular vision system, with images that combine magnification and depth resolution, is juxtaposed with the hand controls to produce an intuitive experience mimicking open surgery.

Surgical cart

The end effectors of the robotic system are in striking contrast to conventional laparoscopy, with the surgeon's controls articulating wristed instruments with seven degrees of freedom (three orientation, three translational, and one for grip) and two degrees of axial rotation. The tools are also steadier than the human hand because it has an inbuilt tremor-filtering functionality. Variable motion scaling allows a versatile degree of motion of the instruments (as, e.g., a 3:1 ratio allows 3 cm of movement of the controls to translate into 1 cm of movement of the instruments in the operative field) and, combined with the magnified vision system, it becomes possible for the surgeon to finesse very nimble dissections in anatomic minefields.

Newer developments include a fourth arm (which gives the console surgeon greater independence and control), an integrated touchpad (for audio and video controls), and superior stereoscopic vision. TilePro allows the projection of intraoperative ultrasound images and preoperative computed tomography (CT) images on to the console screen for precise tumor resection [17]. The dual console systems are a step in the direction of enhanced interspecialty collaboration, providing a platform for the cross-pollination of ideas and sharing of instruments to develop a multidisciplinary robotic program.

Clinical applications

Cardiothoracic surgery

The da Vinci robot (as was the Zeus) was engineered specifically for the performance of minimally invasive cardiac surgery. In 1999, Carpentier and co-workers [18] reported the first totally endoscopic cardiac bypass (TECAB) procedure (LITA–LAD grafting) in Paris. All the surgeons underwent a preliminary evaluation phase during which they optimized the various steps in human cadavers. The rate-limiting step in attaining the holy grail of totally endoscopic closed-chest surgery was felt to be the optimal placement of ports in the rigid chest wall. The excellent kinematic (jointed) dexterity of the anastomotic step of surgery in a 3D vision environment was hailed as the signal feature of the robotic interface.

The FDA approval for robotic coronary revascularization (da Vinci) was the result [19] of a prospective multicentric Investigational Device Exemption (IDE) trial by Argenziano et al. [20]. Patients were enrolled for arrested-heart, single-vessel TECAB. The trial concluded that the TECAB had comparable results pertaining to efficacy (freedom from graft failure and reintervention) and safety (freedom from major adverse cardiac events) endpoints to conventional open surgery.

The new era of collaboration between minimally invasive cardiothoracic surgeons and interventional cardiologists then began in earnest. Katz et al. [21] reported a series of patients treated with hybrid revascularization – TECAB for the LITA–LAD placement combined with percutaneous coronary intervention (PCI) for secondary coronary targets [22].

Urology

Schuessler et al. [23] reported the outcomes for the first transperitoneal laparoscopic prostatectomy series of nine patients and concluded that it offered no relative advantages in surgical outcomes(oncologic cure, potency, continence) and perioperative outcomes (length of stay, convalescence, cosmesis) as compared with the traditional procedure. They also reported other disadvantages – longer operating time, an exceptionally steep learning curve, and increased fixed (and variable) costs. Guilloneau and Vallencien of the Montsouris group improved the technique that scaled down the operating time, duration of catheterization, and length of stay, and proffered equivalent oncologic efficacy and superior functional outcomes, all the while buoyed by a viable cost–benefit model at their institution [24].

Working in one of the earliest centers practicing robotic cardiac surgery, Binder and Kramer in Frankfurt performed the first robotically assisted radical prostatectomy in May 2000 [25].

Figure 1.2 The Veil of Aphrodite after robot-assisted radical prostatectomy.

ch01fig002.eps

Figure 1.3 The first complete intracorporeal robot-assisted laparoscopic augmentation ileocystoplasty and Mitrofanoff appendicovesicostomy. (a) Completed bowel anastamosis; (b) completed cystoplasty; (c) completed Mitrafanoff appendicovesicostomy. Courtesy of Dr. Mohan S. Gundeti.

ch01fig003.eps

Menon and colleagues at the Vattikuti Institute of Urology in Detroit were the first to demonstrate the advantages of robotic prostatectomy over the open procedure [26], and engineered a smooth transition from the Montsouris laparoscopic approach to a replicable robot-specific technique – the Vattikuti Institute prostatectomy [27]. The group also laid the anatomic foundations for the Veil of Aphrodite technique [28] (high anterior release of periprostatic fascia, Figure 1.2). Currently, the group has the largest published series [29] reporting excellent functional and surgical outcomes for the procedure. It was the work at the Vattikuti Urology Institute that laid the foundations for the acceptance of robotics as a viable surgical tool.

Pediatric urology

The assimilation of laparoscopy into pediatric urology has lagged behind its adoption into adult urology [30]. Nevertheless, since its first introduction by Cortesi et al. [31] for the evaluation of impalpable testes, it has become an integral part of many reconstructive procedures in few select hands [30]. Robotic instruments have been tailored to facilitate pediatric surgery (5 mm instruments as opposed to the 8 mm instruments used in adults) and 2D/3D endoscopes have been developed.

A team of pediatric urologists led by Craig Peters and Joseph Borer at the Children's Hospital Boston, who had regularly been performing complex laparoscopic surgery in the 1990s, performed one of the earliest robotic pediatric pyeloplasties in an adolescent with symptomatic ureteropelvic junction (UPJ) obstruction on 1 March 2002 (C. Peters, personal communication).

The most commonly performed procedure with robotic assistance is pyeloplasty for the UPJ obstruction. Other procedures that are performed using the robotic platform are nephrectomies, heminephroureterectomies, ureteral reimplantation, appendicovesicostomy, and orchiopexy, among others [32]. Ever more complex procedures are being performed laparoscopically with the added dexterity of the robot – a case in point being the report published by Gundeti et al. [33] of the first complete intracorporeal robotic-assisted laparoscopic augmentation ileocystoplasty and Mitrofanoff appendicovesicostomy (Figure 1.3).

General surgery

General surgery has proven less permissive to the adoption of robotic technology. Three reasons are commonly cited [34] for this early trend: relatively advanced laparoscopic skill sets in elective general surgeons, equipment limitations (particularly for bowel surgery), and procedural complexity (that would demand a robotic rather than laparoscopic intervention).

On 3 March 1997, Cadiere and colleagues, building on Dubois et al.'s established techniques of laparoscopic access [35], performed the first robot-assisted laparoscopic cholecystectomy on a 72-year-old woman at the St. Blasius hospital in Dendermonde, Belgium. The contemporary role of robotic cholecystectomy is controversial, with some regarding it as the ideal launch pad for getting on to the robotic learning curve [36] whereas others, among them some early pioneers [37] of the robotic approach, have reverted to the standard laparoscopic technique, citing cost-effectiveness.

Marescaux et al. [38] espoused a more sophisticated rationale for the dogged adoption of robotic cholecystecomy (and the robotic approach in general). They envisioned the integration of the different digital interfaces into an augmented-reality operating room, where preoperative images would permit entire simulative surgeries beforehand; the preoperative images and the resultant simulation would then integrate with real-time images to facilitate meticulous identification of anatomic structures (pathology, vasculature, anatomic aberrancy, margins) to a degree heretofore unknown, thereby completely transforming the surgical act into ever more manipulable digital data.

Numerous other procedures have also been performed with robotic assistance. One of the earliest was robotic Nissen fundoplication, which has been shown to have equivalent outcomes in a number of trials comparing it with the laparoscopic approach. At this juncture, robotics does appear to be more costly and surgeons early in the learning curve show longer operative times [34]. In contrast, reports have emerged that the difference disappeared with 10–20 operations for robotic fundoplication (and robotic cholecystectomy) [34]. As in other allied fields, the robotic approach offers special advantages when the need for precise intracorporeal suturing in confined spaces is paramount.

Conclusion

The events of the preceding decade have provided a strong affirmation for the versatility of the robotic interface. The arduous task of actually tailoring this amorphous platform to specific applications behoves a strong commitment by intrepid surgeons and maverick hospitals. Robotic surgery is the beginning and not the end of the journey to minimize further the therapeutic invasion of the human body. Its application to children appears to be intuitive but we need a dedicated robot, supported by finer tools, to handle pediatric surgical procedures. The equipment also needs to be molded to suit infantile needs. The current prohibitive costs of the robotic system have deprived large swathes of underprivileged patients of the benefits of minimally invasive procedures and it is to be hoped that this deplorable scenario will change in the near future.

References

1. Satava RM. Advanced technologies and the future of medicine and surgery. Yonsei Med J 2008; 49(6):873–8.

2. Ueo H, Inoue H, Honda M, et al. Production of interleukin-6 at operative wound sites in surgical patients. J Am Coll Surg 1994;179(3):326–32.

3. Bruce DM, Smith M, Walker CB, et al. Minimal access surgery for cholelithiasis induces an attenuated acute phase response. Am J Surg 1999;178(3):232–4.

4. Wickham JE. The new surgery. Br Med J (Clin Res Ed) 1987;295(6613):1581–2.

5. Wickham JE. Minimally invasive therapy. Health Trends 1991;23(1):6–9.

6. Wickham JE. Minimally invasive surgery. Future developments. Br Med J 1994;308(6922):193–6.

7. Himal HS. Minimally invasive (laparoscopic) surgery. Surg Endosc 2002;16(12):1647–52.

8. Litynski GS. Endoscopic surgery: the history, the pioneers. World J Surg 1999;23(8):745–53.

9. Litynski GS. Profiles in laparoscopy: Mouret, Dubois, and Perissat: the laparoscopic breakthrough in Europe (1987–1988). JSLS 1999;3(2):163–7.

10. Munro MG. Laparoscopic access: complications, technologies, and techniques. Curr Opin Obstet Gynecol 2002;14(4):365–74.

11. Pemberton RJ, Tolley DA, van Velthoven RF. Prevention and management of complications in urological laparoscopic port site placement. Eur Urol 2006;50(5): 958–68.

12. Gow JG. Harold Hopkins and optical systems for urology – an appreciation. Urology 1998;52(1):152–7.

13. Berci G, Paz-Partlow M. Electronic imaging in endoscopy. Surg Endosc 1988;2(4):227–33.

14. Shrivastava A, Menon M. Augmented reality surgical suite for robotic surgery. Presented at the 22nd Engineering and Urology Society Annual Meeting, May 2007 (abstract).

15. Allaf ME, Jackman SV, Schulam PG, et al. Laparoscopic visual field. Voice vs foot pedal interfaces for control of the AESOP robot. Surg Endosc 1998;12(12):1415–8.

16. Kavoussi LR, Moore RG, Adams JB, Partin AW. Comparison of robotic versus human laparoscopic camera control. J Urol 1995;154(6):2134–6.

17. Rogers CG, Laungani R, Bhandari A, et al. Maximizing console surgeon independence during robot-assisted renal surgery by using the Fourth Arm and TilePro. J Endourol 2009;23(1):115–21.

18. Loulmet D, Carpentier A, d’Attellis N, et al. Endoscopic coronary artery bypass grafting with the aid of robotic assisted instruments. J Thorac Cardiovasc Surg 1999;118(1):4–10.

19. Rodriguez E, Chitwood WR. Robotics in cardiac surgery. Scand J Surg 2009;98(2):120–4.

20. Argenziano M, Katz M, Bonatti J, et al. Results of the prospective multicenter trial of robotically assisted totally endoscopic coronary artery bypass grafting. Ann Thorac Surg 2006;81(5):1666–74; discussion 1674–5.

21. Katz MR, Van Praet F, de Canniere D, et al. Integrated coronary revascularization: percutaneous coronary intervention plus robotic totally endoscopic coronary artery bypass. Circulation 2006;114(1 Suppl):I473–6.

22. de Canniere D, Jansens JL, Goldschmidt-Clermont P, et al. Combination of minimally invasive coronary bypass and percutaneous transluminal coronary angioplasty in the treatment of double-vessel coronary disease: two-year follow-up of a new hybrid procedure compared with on-pump double bypass grafting. Am Heart J 2001;142(4):563–70.

23. Schuessler WW, Schulam PG, Clayman RV, Kavoussi LR. Laparoscopic radical prostatectomy: initial short-term experience. Urology 1997;50(6):854–7.

24. Guillonneau B, Vallancien G. Laparoscopic radical prostatectomy: the Montsouris experience. J Urol 2000;163(2):418–22.

25. Binder J, Kramer W. Robotically-assisted laparoscopic radical prostatectomy. BJU Int 2001;87(4):408–10.

26. Menon M, Tewari A, Baize B, et al. Prospective comparison of radical retropubic prostatectomy and robot-assisted anatomic prostatectomy: the Vattikuti Urology Institute experience. Urology 2002;60(5):864–8.

27. Menon M, Tewari A, Peabody J. Vattikuti Institute prostatectomy: technique. J Urol 2003;169(6):2289–92.

28. Kaul S, Savera A, Badani K, et al. Functional outcomes and oncological efficacy of Vattikuti Institute prostatectomy with Veil of Aphrodite nerve-sparing: an analysis of 154 consecutive patients. BJU Int 2006;97(3):467–72.

29. Badani KK, Kaul S, Menon M. Evolution of robotic radical prostatectomy: assessment after 2766 procedures. Cancer 2007;110(9):1951–8.

30. Smaldone MC, Sweeney DD, Ost MC, Docimo SG. Laparoscopy in paediatric urology: present status. BJU Int 2007;100(1):143–50.

31. Cortesi N, Ferrari P, Zambarda E, et al. Diagnosis of bilateral abdominal cryptorchidism by laparoscopy. Endoscopy 1976;8(1):33–4.

32. Casale P, Kojima Y. Robotic-assisted laparoscopic surgery in pediatric urology: an update. Scand J Surg 2009;98(2):110–9.

33. Gundeti MS, Eng MK, Reynolds WS, Zagaja GP. Pediatric robotic-assisted laparoscopic augmentation ileocystoplasty and Mitrofanoff appendicovesicostomy: complete intracorporeal – initial case report. Urology 2008;72(5):1144–7; discussion 1147.

34. Wilson EB. The evolution of robotic general surgery. Scand J Surg 2009;98(2):125–9.

35. Dubois F, Berthelot G, Levard H. Laparoscopic cholecystectomy: historic perspective and personal experience. Surg Laparosc Endosc 1991;1(1):52–7.

36. Ballantyne GH. Telerobotic gastrointestinal surgery: phase 2 – safety and efficacy. Surg Endosc 2007;21(7):1054–62.

37. Ruurda JP, Broeders IA, Simmermacher RP, et al. Feasibility of robot-assisted laparoscopic surgery: an evaluation of 35 robot-assisted laparoscopic cholecystectomies. Surg Laparosc Endosc Percutan Tech 2002;12(1):41–5.

38. Marescaux J, Smith MK, Folscher D, et al. Telerobotic laparoscopic cholecystectomy: initial clinical experience with 25 patients. Ann Surg 2001;234(1):1–7.

2

Stepwise Approach to Training for Robotic Surgery and Credentialing

Jason Y. Lee and Elspeth M. McDougall

Department of Urology, University of California, Irvine, Orange, CA, USA

Introduction

Urologic surgery has come a long way since the first cystoscope was introduced to the world in 1877 by Maximilian Nitze [1]. On waves created by pioneers and champions of innovation, minimally invasive urologic surgery (MIUS) has permeated throughout the entire field, having found applications not only in uro-oncology and endourology but also female and pediatric urology. In particular, the successful completion of the first laparoscopic nephrectomy by Clayman et al. [2] galvanized surgical innovation and, until recently, laparoscopic surgery has been the poster child of the MIUS revolution. Its significant advantages in decreasing patient morbidity have been demonstrated not only in urology but also in many other surgical fields [3–7], all the while demonstrating comparable efficacy, safety, and cost-effectiveness to the open surgical approach. With increased acceptance, adoption, and application, laparoscopic surgery is no longer considered a revolution but a standard treatment modality for many of the common diseases seen in urology today.

Throughout this transformative process, however, laparoscopic surgery has always been plagued by its demanding technical difficulty. So much so that, for some, the transition from traditional open surgery to laparoscopy has been an endeavor too challenging to overcome. With the introduction of robotic-assisted laparoscopic surgery (robotic surgery), however, many of these ergonomic and instrument dexterity issues have been overcome and allowed more surgeons to adopt a minimally invasive approach to their surgical procedures. Originally described as tele-presence surgery by NASA, robotic surgery has enabled surgeons more readily to surmount the often steep learning curve associated with conventional laparoscopic surgery. With its ease of use, improved precision, and marketability, robotic surgery has been embraced by more surgeons than has pure laparoscopy. Despite the lack of long-term data, this recent rapid adoption of robotic surgery is no more evident in any other field than urology [8]. As of 2007, approximately 60% of all radical prostatectomies in the United States were performed using the robotic da Vinci Surgical System [9], and there are estimates that by 2010 that number will have ballooned to over 85%.

Over the last several years, robotic surgery has formally taken the focus away from traditional laparoscopic surgery and now leads the charge in the MIUS movement. Through the vehicle of robotic surgery, MIUS is now being applied to ever more complex procedures, having progressed from simple diagnostics to applications in complex extirpative and reconstructive procedures. As the applicability and incorporation of robotic surgery move forward, it is critical that strategies be developed to provide trainees with an infrastructure for technical training and development of expertise, and simultaneously to ensure proficiency amongst those practicing clinicians.

It is well recognized that surgical skills performance and assessment can be broken down into several categories that are part of the Accredited Council for Graduate Medical Education (ACGME) core competencies. These include cognitive, psychomotor, communication, and affective skills. The ACGME has recently added a seventh core competency directly related to surgical procedure skills [10]. Sweet et al. proposed that validity and curriculum development are interdependent, representing an ongoing process such that concurrent validation occurs throughout the design of the educational curriculum rather than once it is completed [11]. They described an application of the Wiggins and McTighe's backwards design approach to curriculum development for technical skills, starting with outcomes and tailoring research questions as burden of proof towards this purpose [12]. Therefore, validity and curriculum development are interdependent, iterative processes that are never truly complete. Curriculum design for technical skill education involves setting expert-determined goals and objectives at the commencement, designing interventions targeted to these goals, and developing assessment tools that can certify competency in the desired skills [13].

The American Urology Association (AUA) Laparoscopy and Robotic Surgery Committee has embraced these concepts of curriculum development as the platform on which the development of a skills training program, and also assessment devices, will be created for basic robotic surgery training.

It remains incumbent on the profession to develop, endorse, and assist with the implementation of necessary educational programs and credentialing processes to ensure the safe and efficacious application of this new technology. Innovation and technological advancements in science, engineering, and medicine have allowed surgeons to deliver ever more optimized care to their patients. With such advances, it is crucial that the application of such technologies does not come at the expense of patient safety and efficacy. Primum non nocere.

Robotic surgery training for residents and fellows

The healthcare system of the twenty-first century is no longer the paternalistic environment in which physicians once practiced in past decades. With an emphasis on transparency, patient safety, and accountability, and also fiscal responsibility and resource management, surgical residency and fellowship training programs across the country face significant hurdles in providing opportunities for the transfer of surgical expertise to their trainees. This is particularly evident in the transfer of skills associated with new surgical technologies such as robotic surgery.

As such, the training paradigm for robotic surgery in urology today must include a large portion of proficiency development outside of the operating room (OR) in preclinical training. This two-stage approach to the training of surgeons in robotic surgery has been recommended by several different groups [14–16], and attempts to minimize the footprint of surgical education on patient outcomes. Regardless of the number or type of components included in the preclinical and clinical training stages, it is essential that the process be structured and objective or competency based. Unfortunately, to date, there is no validated robotic surgery training curriculum (RSTC) and the optimal format and requisite curriculum components are still in the development stages. However, several national and international organizations have an interest in the development of training guidelines for robotic surgery. The AUA is currently focusing attention on creating basic robotic surgery training guidelines, enlisting input and feedback from various surgical organizations and expert clinicians nationwide.

The initial components of a structured RSTC involve preclinical training (Figure 2.1). First and foremost, it is essential for the trainee to be familiar with the da Vinci Surgical System (henceforth referred to simply as the robot), currently the only commercially available robotic surgery platform. Didactic sessions are an important means of providing the trainee with an opportunity to become familiar with the components and the proper utilization of the robot. Intuitive Surgical (Sunnyvale, CA, USA), vendor of the robot, has created and provides an online tutorial on the fundamentals of robotic surgery which involves a technical overview of the robot, functional aspects of the system, and some troubleshooting tips. This online tutorial also includes a multiple-choice question-based examination which can be used by training programs to evaluate the trainee's knowledge of the basic functional aspects of the robot. Currently there are three different robot models: the standard da Vinci robot, the da Vinci S, and the newest da Vinci Si model. The functional aspects of each model are slightly different and as such the required training should be completed for the system that the trainee will be utilizing clinically at their institution. Informal tutorials with the OR robot nursing team, all of whom are trained on the mechanics and functionality of the robot, should also be an integral part of the educational experience for trainees. This allows for a low-stress, hands-on interactive experience with the robot outside of a live clinical setting. Trainees should practice the docking process and instrument insertion and exchange, and also become familiar with sitting at the surgeon's console and controlling the various aspects of the robotic interface through this platform.

Figure 2.1 Structured robotic surgery training curriculum.

ch02fig001.eps

The second component of the preclinical stage of an RSTC involves inanimate dry-laboratory practice. Through a set of basic skills tasks such as ring-peg transfer, precision cutting, or simple suturing and knot-tying, this dry-laboratory training allows the trainee to become facile in instrument movement and articulation, camera and clutch navigation, and basic cutting and suturing skills, and also permits familiarization with, and proprioceptive training in, the 3D environment, all of which are the building blocks necessary to perform more complex robotic procedures. The limiting factor in dry-laboratory training is access to the robot itself. Unlike traditional laparoscopy, where all that is needed is a pelvic box trainer, laparoscopic instruments, and a camera with monitor, dry-laboratory robotic training requires access to a fully functional robot. This type of educational experience may not be logistically or financially feasible at all training centers. Therefore, alternative teaching strategies for this portion of the curriculum have been developed and are becoming more prevalent in training programs.

Virtual reality (VR) surgical simulators have been shown not only to improve subsequent surgeon performance but also to shorten the learning curve associated with the acquisition of a new technologic skill [17–19]. Unfortunately there are only a few VR robotic simulators currently in development, most of which have not been fully validated. In addition, for many centers, even the cost of such a VR robotic simulator may be significantly prohibitive.

Animate or cadaveric robotic training is another important component of the preclinical training stage [16,20–22]. It provides the next graduated step in the acquisition of basic skills in robotic surgery. These more expensive educational modalities are best utilized in teaching the nuances of high-level skills or procedure-specific training. By using the robot on animal models or cadavers in the laboratory, the trainee has an opportunity for high-fidelity, interactive simulation of situations and potential problems that may be encountered in live clinical experiences. This allows the trainee an opportunity to experience potential clinical scenarios or complications and learn how to avoid or manage these complications in a low-stakes environment away from the patient. The integration of such hands-on laboratory sessions during residency or fellowship training is much more common for traditional laparoscopy as the availability of equipment is not as significant an issue as with robotic surgery. Industry-directed courses are available to trainees in the form of one- or two-day programs during which didactic sessions are combined with a hands-on laboratory session [23].

An immersive 5-day Mini-Fellowship training program developed at the University of California, Irvine (UCI) provides trainees with not only dry-laboratory inanimate training but also procedure-specific, hands-on training with pigs and cadavers [24]. In addition, this program includes didactic tutorial sessions and live case observation in the OR. This program requires the dedication of faculty to provide the tutorial experiences, an extensive team of laboratory personnel, and an overseeing MIS fellow to organize and implement the animal and cadaveric learning experience. In order to cover just the cost of the laboratory expenses for this type of course, a tuition fee of $3500 is required. Unfortunately, this program is currently not duplicated anywhere else in the United States.

Although there is a paucity of data related to the construct validity of a structured training program, for the acquisition of basic robotic skills there is evidence to suggest that formal training for more advanced robotic skills is indeed beneficial [25]. Particularly with the animate and cadaveric robotic experiences, it is important for residents and fellows to have structured, interactive instruction by expert robotic surgeons who can provide real-time feedback [26,27].

Although the preclinical components of an RSTC have been shown to be advantageous [20,28,29], access to such resources may be difficult for some training programs. As such, many programs will incorporate a minimal amount of preclinical training into the curriculum before proceeding directly to the clinical stages of the RSTC. However, this educational approach necessitates that the learning curve associated with robotic surgery be addressed in the clinical setting, a situation that may not be in patients’ best interests. Studies have shown that even three preclinical sessions practicing basic robotic skills can improve preclinical operative times by up to 40% [20].

Prior to advancement from the preclinical stage of an RSTC, the trainee should demonstrate that they have acquired the basic aptitude necessary to begin experiential, clinical training on the robot. This should be accomplished either through informal assessment by an expert robotic surgeon or, ideally, through a formal evaluative process. Regardless, this should be a decision based on demonstrated proficiency, as opposed to a specific volume of cases completed or static length of time spent on the robot. Trainees must have attained the requisite basic skills necessary to proceed to more complex robotic tasks and procedures before moving to the next level of their robotic training.

The clinical training stage of an RSTC is composed of two main components: procedure-specific familiarization with the robot and direct robot console time (Figure 2.1).

It is important to start the clinical stage of an RSTC with a procedure-specific approach. By focusing on one specific procedure, whether it be robotic-assisted laparoscopic radical prostatectomy (RARP), robotic partial nephrectomy, or any other robotic procedure, the trainee will acquire skills in a much more efficient manner. Ideally, the procedure selected should be the type of procedure with the highest volume at that particular training institution. This will afford the trainee the greatest number of opportunities to work effectively through the clinical components of an RSTC. The specific procedure should be clearly defined by the steps required to complete the operation, from the initial positioning of the patient through to the final removal of ports and recovery of the patient. These steps should delineate the degree of complexity of the surgical tasks involved to complete each of the steps and then be ordered from least difficult to most difficult. This provides a graduated surgeon experience with the clinical application of their robotic surgical skills.

The experiential stage of an RSTC begins with clinical observation. Whether prerecorded operative videos or live operative cases in the OR by experts, the trainee is given an opportunity to observe directly the execution of the various steps involved in completing a specific robotic procedure from start to finish [30]. As with traditional open surgery, the trainee will be better able to assist and perform the procedure if they can anticipate the sequence of steps required to complete the procedure. A library of prerecorded operative video footage should be provided to the trainee for reference, and it is best if this footage is broken down into the defined essential key steps of the procedure so that the trainee can review the specific steps as they proceed through their graduated RSTC.

Once the trainee has demonstrated that they have learned the basic knowledge and skill tasks of the steps of a given robotic procedure, they must first begin their operative experience as the bedside assistant to the main console surgeon. The ability to assist effectively for these robotic procedures demonstrates that the trainee has gained the knowledge of the steps of the procedure, a general proficiency in working in the robotic environment, a knowledge of the functionality and limitations of the robot itself, and the strategies and techniques employed by the console surgeon(s) to complete the specific procedure. The importance of beginning the operative experiences as a surgical assistant has been reinforced by several authors [16,30–35], and serves to consolidate the trainee's basic robotic knowledge and skills before commencing their clinical training on the console. The number of cases recommended as the bedside assistant remains without a consensus, although most reports suggest a minimum of 10 cases during robotic surgery training [34,36].

The final component of any RSTC involves time on the surgeon console. Regardless of the amount of training in the preclinical stage of a, RSTC, the steepest learning curve will be encountered once the trainee sits at the console. As such, it is crucial that this component of an RSTC be structured and involve an iterative process of trainee evaluation. In an ideal situation, this would involve a graduated, stepwise progression of defined tasks and steps of the procedure, based on degree of difficulty, under the direct supervision of an expert robotic surgeon who is either at the bedside or at the console with the trainee.

The graduated, stepwise approach allows for the development of proficiency in specific defined tasks prior to advancement to the next, more complex task or step of the procedure. This approach to robotic training has been emphasized by several studies [21,28,32–34,36] and allows the trainee to acquire skills through repetition of specific skill tasks. As discussed earlier, the procedure should be divided into defined steps and should be ordered in a graduated manner based on increasing difficulty. When the trainee has demonstrated proficiency in a predefined step, either through formal evaluation or based on the expert surgeon's judgment, they would be moved on to the next sequentially difficult step of the procedure. Eventually, the trainee will be able to integrate skills learned and practiced during each defined step into a comprehensive ability to complete the entire procedure. Video recording and review of operative performance with the mentor or expert surgeon can also enhance this learning process for the trainee, as it provides formative feedback [37].

Several challenges may be encountered when integrating an RSTC into an existing urology training curriculum. First, the clinical training strategy is somewhat more difficult to integrate in urology than in other specialties such as general surgery. Urology does not have a short, simple, and common procedure such as a laparoscopic cholecystectomy or appendectomy from which to gain early, basic, minimally invasive surgery experiences. As such, the hands-on clinical experience is often an immediate swim in the deep end for the trainee. This supports even further the idea of progressing the hands-on clinical experience using a proficiency-based, graduated stepwise model.

Second, unlike traditional open surgical training, where the mentor can be in close proximity to the trainee and facilitate hands-on teaching, in robotic surgery the mentor and trainee are separated in space and the attending surgeon may not have full control of the operation as they do in open surgery. The fact that only one surgeon can be at the console at any one time is an educational issue that has been documented previously [28,30,33]. In response to this concern, Intuitive Surgical has developed the new da Vinci Si model, which has a dual console available that potentially will allow for expert surgeon direction and supervision for procedural robotic training and collaboration. Although theoretically this new robotic system lends itself well to the educational curriculum concept, the increased cost incurred for the dual console will be a significantly prohibitive factor for most training programs (currently approximately $500 000).

Finally, as with traditional open surgical training, the ideal situation involves an expert surgeon mentoring or teaching a trainee. The current situation in some academic institutions, however, is more often an experienced trainee mentoring a novice trainee, and, until robotic surgery permeates more widely into the current cohort of postgraduate urologists, this will continue to be the case in many institutions, the issue being even more apparent at non-academic institutions.

Postgraduate robotic surgery procedural training

Unlike residents and fellows, practicing urologists wishing to develop skills in robotic surgery do not have protected and dedicated time for educational activities. Logistically, it is much easier to teach new surgical skills to residents than to those in postgraduate clinical practice due to the availability of structured educational opportunities. In addition, postgraduate urologists need to acquire the skill set rapidly in order to be able to integrate robotic surgery effectively into their current practices. As such, they must either work with already experienced colleagues to integrate the training opportunities into their daily clinical practice or take formal time off, finding alternative coverage of their practice, in order to participate in robotic surgery training opportunities. In any event, this will take dedication and patience on the part of the surgeon. Precious elective operative time will need to be committed to the learning curve of this new technological surgery technique and often can result in a backlog of cases. Support from the trainee's institution, manifested by additional elective operative time dedicated to the launch of this technology, can ease this burden.

A similar RSTC is recommended for postgraduate trainees as for residents and fellows, but the time frame during which this curriculum takes place is usually much more condensed and with a longer clinical phase of training. Whether through industry-developed programs, AUA-organized continuing medical education courses, or procedure-specific intensive training programs such as the UCI Robotic Surgery Mini-Fellowship program, postgraduate robotic surgery training ideally involves preclinical components in addition to the mandatory clinical learning curve that all must endure.

Postgraduate trainees can easily become acclimated with the robot itself through both on-line didactic materials and industry-organized training programs. The difficulty will be gaining access to dry-laboratory and animate or cadaveric practice opportunities. This usually requires the trainee to take time away from their clinical duties and attend off-site courses. These courses should include an integrated curriculum that involves not only didactic sessions but also preclinical hands-on experience on the console with either animal or cadaveric models. The 5-day Mini-Fellowship training curriculum previously described has been shown to be effective in enabling postgraduate urologists successfully to incorporate and maintain RARP in clinical practice in both the short and long term. The study of this dedicated RARP curriculum reported that almost 90% of trainees who successfully completed the course were performing RARP at 3 years following the Mini-Fellowship training [24].

Gamboa et al. [24] also demonstrated that postgraduate trainees attending the course with another colleague were more likely to adopt robotic surgery into their practices: 90% versus 78% at 3 years follow-up. In fact, several studies have noted the benefits of partnered training [38–40], which allows for a supportive, synergistic learning environment for the surgeon, both during the training course and in the home institution.

For those postgraduate trainees fortunate enough to have an expert robotic surgeon at their institution, local preceptorship or mentorship by an expert colleague is another training model that has shown to be very effective [39,41]. Early preceptorship by an expert is not only beneficial from a technical standpoint, but also means that the trainee's OR team is familiar with robotic surgery and that the organizational aspect of starting up a robotics program is not an active concern for that trainee. The ideal length of a preceptorship or mentorship has yet to be determined; however, most recommend that this be a competency-based as opposed to a time-based process.

For postgraduate trainees who do not have access to a preceptor or partner-in-training, the learning curve becomes much more of an exercise in independent learning. Although proctoring can aid in dealing with this learning curve, the feasibility of having an on-site proctor throughout the initial phase of the clinical learning curve is unrealistic. In addition, from a medical–legal perspective, a proctor, unlike a preceptor or partner, assumes no liability in the outcomes of the procedure and is not allowed any hands-on participation in the surgical cases.

Credentialing

Although the necessity is ever paramount, currently there are no governing body-mandated credentialing guidelines for robotic surgery. Without a standardized process for surgeon accreditation, the granting of robotic surgery privileges remains institution based; a method that does not ensure patient accountability in a universally consistent manner.

The AUA is currently working towards developing a robotic surgery standard operating practices document to aid institutions and surgeons in this endeavor. However, due to the lack of an established standard of care, credentialing is currently left in the hands of individual healthcare institutions. Hospital credentialing should not be an industry-controlled or -mandated process, as is the case in many hospitals today, but one that is a result of a standardized, competency-based, peer evaluation system. This system should be self-regulated by robotic surgery experts in a transparent and assiduous manner. Some groups, such as the Society of Urologic Robotic Surgeons (SURS), have been vocal in their support for a policy of this nature [42]. As such, it is incumbent on surgeons to change the current landscape so that we can ensure the safe and efficacious application of this new technology to patient care.

The Society of American Gastrointestinal and Endoscopic Surgeons (SAGES) has created guidelines for laparoscopic surgery training and credentialing in the form of the Fundamentals of Laparoscopic Surgery (FLS) curriculum. The FLS curriculum has been validated as a means of training and credentialing trainees [43] and is now endorsed by the American College of Surgeons (ACS). Effective July 2010, all general surgery certification candidates will be required to have successfully completed the FLS training curriculum prior to being eligible for their American Board of Surgery Certification. Following this lead, urologists also need to develop a validated RSTC and credentialing process, initiated and regulated by expert urologic robotic surgery peers.

As a minimum for surgeon credentialing, one should meet the following requirements: (1) demonstrate proficiency in basic laparoscopy, (2) have robotic certification for the use of the specific da Vinci robot model that the surgeon plans to use clinically, (3) provide proof of basic preclinical training in robotic surgery, and (4) receive clinical proficiency status from an approved robotic surgery proctor.

There is currently no consensus as to the number of proctored cases that need to be performed in order for a trainee to be considered safe. Examination of the recent literature regarding robotic surgery training, specifically for RALP, shows that there is a wide range in the recommended number of cases required to move beyond the initial learning curve. Ahlering et al. stated that a laparoscopically naïve surgeon with significant open surgical experience could successfully adopt RALP in 8–12 cases [44], whereas Herrell and Smith determined that a minimum of 250 RALP cases were required to achieve comfort and confidence comparable to open radical prostatectomy [45]. This emphasizes the dichotomous nature of the robotic surgical learning curve. As with any surgical technique, whether open or robotic, there is a difference between the surgeon overcoming the technical learning curve as opposed to the outcomes learning curve for the procedure. Although the technical learning curve can be conquered during a defined training interval, a surgeon's outcomes learning curve is an ongoing, iterative process that can extend well into their senior years of practice. For example, a surgeon may be able to complete a RALP in a timely and safe manner after 10 cases; however, their oncologic and functional outcomes may not be comparable to the generally acceptable established outcomes in the literature. As such, there is no consensus on the number of cases that need to be performed in order to demonstrate proficiency. Likely, this number lies somewhere between 15 and 30 cases for RALP and is influenced by individual surgeon skill and also the interval during which these initial cases are performed. In any event, it is the responsibility of the institution to determine whether the surgeon is proficient and safe, not whether the surgeon is an expert. As with any credentialing or accreditation body, the goal should not be to single out the experts in the group, but to ensure that those treating patients with the new technologic surgical procedure are safe and effective surgeons. Above all, credentialing should be a competency-based process directed at ensuring safety and efficacy for every patient.

Conclusion

The process of expeditiously and effectively training surgeons in robotic surgery is one that has been mandated not only through patient demand for this surgical technology, but also by the principle that patients should not be placed at undue risk for the sake of introducing a new procedure or technology into surgical practice. There currently does not exist a validated RSTC, nor is there a consensus on a robotic surgery credentialing process. Until these important aspects of this new surgical technology are defined, the training and credentialing of robotic surgeons will continue to be an iterative process that requires dedication, commitment, and vigilance. Organizations such as the AUA are directing concerted attention at addressing these two inadequacies of robotic urologic surgery training and credentialing, which will be a benefit for their surgeon members and ultimately for patients nationwide. Structured, proficiency-based advancement, whether for training or credentialing, will ensure that surgeons acquire the necessary skills to provide optimum care to their patients.

References

1. Herr HW. Max Nitze, the cystoscope and urology. J Urol 2006;176:1313–6.

2. Clayman RV, Kavoussi LR, Soper NJ, et al. Laparoscopic nephrectomy: initial case report. J Urol 1991;146(2):278–82.

3. Grace PA, Quereshi A, Coleman J, et al. Reduced postoperative hospitalization after laparoscopic cholecystectomy. Br J Surg 1991;78(2):160–2.

4. Lane BR, Gill IS. 7-year oncological outcomes after laparoscopic and open partial nephrectomy. J Urol 2010;183(2):473–9.

5. Berger A, Brandina R, Atalla MA, et al. Laparoscopic radical nephrectomy for renal cell carcinoma: oncologic outcomes at 10 years or more. J Urol 2009;182(5):2172–6.

6. Bronitsky C, Payne RJ, Stuckey S, Wilkins D. A comparison of laparoscopically assisted vaginal hysterectomy vs traditional total abdominal and vaginal hysterecomties. J Gynecol Surg 1993;9(4):219–25.

7. Seder CW, Hanna K, Lucia V, et al. The safe transition from open to thoracoscopic lobectomy: a 5-year experience. Ann Thorac Surg 2009;88(1):216–25.

8. Ahmed K, Khan MS, Vats A, Nagpal K, et al. Current status of robotic assisted pelvic surgery and future developments. Int J Surg 2009;7:431–40.

9. Wexner SD, Bergamaschi R, Lacy A, et al. The current status of robotic pelvic surgery: results of a multinational interdisciplinary consensus conference. Surg Endosc 2009;23:438–43.

10. Sachdeva AK. Invited commentary: educational interventions to address the core competencies in surgery. Surgery 2004;135(1):43–7.

11. Sweet RM, Hananel D, Lawrenz F. A unified approach to validation, reliability, and education study design for surgical design for surgical technical skills training. Arch Surg 2010;145(2):197–201.

12. Wiggins G, McTighe J. Understanding by Design. Englewood Cliffs, NJ: Prentice Hall, 2001.

13. Wright AS. Validity in educational research: critically important but frequently misunderstood. Arch Surg 2010;145(2):201.

14. Shah G, Haas G. Teaching daVinci Robot Surgical System – a new paradigm. J Urol Urogynakol 2006;13: 22–3.

15. Moles JJ, Connelly PE, Sarti EE, Baredes S. Establishing a training program for residents in robotic surgery. Laryngoscope 2009;119:1927–31.

16. Chitwood RC Jr, Nifong W, Chapman WHH, et al. Robotic surgical training in an academic institution. Ann Surg 2001;234(4):475–86.

17. Seymour NE, Gallagher AG, Roman SA, et al. Virtual reality training improves operating room performance. Ann Surg 2002;236:458–64.

18. Gurusamy K, Aggarwal R, Palanivelu L, Davidson BR. Systematic review of randomized controlled trials on the effectiveness of virtual reality training for laparoscopic surgery. Br J Surg 2008;95:1088–97.

19. Fried GM, Feldman LS, Vassiliou MC, et al.