Surgical Simulation: A Urological PerspectiveGeoffrey R. Wignall, John D. Denstedt,*,† Glenn M. Preminger,‡ Jeffrey A. Cadeddu,Margaret S. Pearle,§ Robert M. Sweet� and Elspeth M. McDougall¶From the Division of Urology, University of Western Ontario, London, Ontario, Canada (GRW, JDD), Division of Urology, DukeUniversity Medical Center, Durham, North Carolina (GMP), Department of Urology, University of Texas, Southwestern Medical Centerat Dallas, Dallas, Texas (JAC, MSP), Department of Urologic Surgery, University of Minnesota, Minneapolis, Minnesota (RMS),and Department of Urology, University of California, Irvine, Irvine, California (EMM)

Purpose: Surgical education is changing rapidly as several factors including budget constraints and medicolegal concernslimit opportunities for urological trainees. New methods of skills training such as low fidelity bench trainers and virtualreality simulators offer new avenues for surgical education. In addition, surgical simulation has the potential to allowpracticing surgeons to develop new skills and maintain those they already possess. We provide a review of the background,current status and future directions of surgical simulators as they pertain to urology.Materials and Methods: We performed a literature review and an overview of surgical simulation in urology.Results: Surgical simulators are in various stages of development and validation. Several simulators have undergoneextensive validation studies and are in use in surgical curricula. While virtual reality simulators offer the potential to moreclosely mimic reality and present entire operations, low fidelity simulators remain useful in skills training, particularly fornovices and junior trainees. Surgical simulation remains in its infancy. However, the potential to shorten learning curves fordifficult techniques and practice surgery without risk to patients continues to drive the development of increasingly moreadvanced and realistic models.Conclusions: Surgical simulation is an exciting area of surgical education. The future is bright as advancements incomputing and graphical capabilities offer new innovations in simulator technology. Simulators must continue to undergorigorous validation studies to ensure that time spent by trainees on bench trainers and virtual reality simulators willtranslate into improved surgical skills in the operating room.

Key Words: computer simulation, education, surgery, endoscopy, laparoscopy

T he face of surgical education is changing, a reality thatis becoming increasingly clear to practicing urologistsand trainees. While the traditional system of resi-

dency training as introduced by Halsted more than a cen-tury ago continues to be used by the majority of urologicaltraining programs, it is becoming apparent that change isnot only probable but also inevitable.1 Some have questionedthe efficacy of the “see one, do one, teach one” model ofsurgical skills training. Others point to the decreasing op-portunities for residents to learn surgical skills throughmentoring in the operating room.2 Several factors may beresponsible for this trend. Residency programs are underincreasing pressure to limit the hours worked by their train-

Submitted for publication July 6, 2007.* Correspondence: Division of Urology, University of Western On-

tario, 268 Grosvenor St., London, Ontario, Canada N6A 4V2 (tele-phone: 519-646-6036; FAX: 519-646-6037; e-mail: denstedt@uwo.ca).

† Financial interest and/or other relationship with Boston Scien-tific, Cook Urological and Olympus.

‡ Financial interest and/or other relationship with Mission Phar-macal and Boston Scientific.

§ Financial interest and/or other relationship with Cook Urologi-cal, Boston Scientific and Percutaneous Systems.

� Financial interest and/or other relationship with Medical Edu-cation Technologies, Inc. and Red Llama Technologies Inc.

¶ Financial interest and/or other relationship with Astellas, Karl

Storz, Intuitive Surgical, Simbionix, Ethicon Endo-Surgical, Endo-Care and METI, Inc.

0022-5347/08/1795-1690/0THE JOURNAL OF UROLOGY®

Copyright © 2008 by AMERICAN UROLOGICAL ASSOCIATION

1690

ees and, thus, decrease the time spent in surgery.3 Financialconstraints stress surgical efficiency and limit teaching op-portunities while medicolegal concerns and fear of litigationmake it increasingly difficult for consultant surgeons to“hand over the scalpel.” New technologies emerge at a stag-gering rate, meaning that practicing surgeons are oftenlearning new skills themselves while simultaneously tryingto provide opportunities to residents. Advancements in en-doscopy and laparoscopy often require exceptional hand/eyecoordination, familiarization with new instruments and theability to work in a 3-dimensional environment using 2-di-mensional images. Simulators that recreate these skills holdthe potential to improve the learning environment for thesurgeon and trainee alike. Finally tertiary care hospitals aretreating increasingly complex cases, minimizing trainee ex-posure to basic surgical skills. As clinical teaching opportu-nities become scarce, the task of providing adequate surgicalteaching to the next generation of urologists becomes para-mount. Urology is an evolving profession and all urologists,from resident to consultant, require opportunities to practiceand expand their surgical skills. With each new technologyor surgical technique, practicing urologists are under pres-sure to expand their skill set and offer the current standardof care.

In the surgical literature a common mistake is to use theterm simulator interchangeably with virtual reality. Simu-

lator is defined as “a device that enables the operator to

Vol. 179, 1690-1699, May 2008Printed in U.S.A.

DOI:10.1016/j.juro.2008.01.014

SURGICAL SIMULATION 1691

reproduce or represent under test conditions phenomenalikely to occur in actual performance” and may apply tophysical and virtual models.4 In this article we apply theterm simulator to any model used to represent surgery in-cluding physical models (eg bench trainers, cadavers) as wellas computer generated virtual reality models.

Simulation has been used in several highly specializedindustries for decades. Trainees in the aerospace industryspend countless hours practicing basic and advanced skillson simulators before hands-on training at the controls of anaircraft. In fact, the Federal Aviation Administration re-quires mandatory annual flight certification including as-sessment through simulation.5 In the military combat pilotsas well as submarine and tank crews are required to dem-onstrate competence in simulated environments before be-ing charged with the operation of billions of dollars worth ofequipment. Conversely simulation is a relatively new phe-nomenon in the field of medicine. In 1990 Delp et al devel-oped the first virtual reality simulator for surgical training.6

Their model consisted of a simulated lower extremity forteaching reconstructive tendon surgery.

Surgery would seem ideally suited to simulated training,a notion that has led to the recent development of numeroussimulators. Certainly the ability to learn and practice vari-ous surgical tasks on a simulator before exposure to patientsin the operating room would seem practical if not essential.Why, then, are not all residents required to undergo trainingon surgical simulators before introduction to the operatingroom in the manner that pilots train before taking the con-trols of an aircraft? There are many potential answers tothis question. The home computer and video game indus-tries push the limits of graphics and processing technology.Computers today are capable of incredible processing speedand realistically rendered virtual environments. While thevast market for home electronics and entertainment soft-ware may support the high costs of software design, therelatively small market for surgical simulation presents anobstacle that will be challenging to overcome. In addition,relatively few simulators have been validated for teaching orproved effective in teaching a specific task. Finally manysimulators are costly, limiting availability primarily toteaching centers. Despite these criticisms simulators arecurrently in use with many more in development or in vary-ing stages of validation.

We provide an overview of the status of surgical simula-tors in urology. We begin with a brief background includingthe various levels (fidelity) of simulation and introduce thestages of validation required before a particular simulationcan be proved useful for education. The current status ofsimulation in urology follows with emphasis on simulatorsthat have undergone validation studies as well as those thatare of particular interest to the urologist. Finally we presentsome thoughts regarding the future of surgical simulation inurological education and clinical practice.

BACKGROUND

In the last 2 decades many open urological procedures havebeen supplanted by minimally invasive techniques. Thesenew surgical methods are increasingly attractive in today’smedical environment, as they minimize patient morbidityand reduce overall costs. Unfortunately the learning curve

for these procedures is often quite steep, requiring many

hours of specialized training to perfect these skills. Severalvariables including the specific procedure, surgeon experi-ence and case volume influence this curve, which may rangefrom as few as 20 to more than 100 procedures. It is fre-quently difficult for urologists in training to acquire ade-quate experience because of limited opportunities in theoperating room. While anatomical differences and social is-sues have limited interest in animal models for surgicaltraining, these models are still commonly used for proce-dural training such as for laparoscopic nephrectomy. De-spite the hope that animal models will eventually be re-placed by simulation, skills training in animal laboratoriesremains a necessary staple of many surgical education pro-grams. In addition, while human cadavers most closely dem-onstrate human anatomy, poor tissue compliance and highcost limit their use.

The integration of simulators into urological curriculaoffers exciting opportunities for trainees to perfect theirskills in an inanimate but dynamic model in which anatom-ical structures are accurately reproduced and the feel of theactual procedure is captured. Through practice and repeti-tion the ability to accurately simulate the skills requiredduring a procedure may ultimately shorten an otherwisedaunting learning curve. Increased surgical efficiency trans-lates into reduced operating room time needed to completeprocedures and greater overall patient safety.

A way to categorize the various types of simulation is tofocus on the concept of fidelity (see Appendix). Low fidelitysimulators are those that are not lifelike such as laparo-scopic box trainers or animal tissue on which to practicesuturing. That is not to say that these simulators do nothave their uses. In fact, they are some of the most usefulsimulators currently available. Advantages of low fidelitysimulators include low cost and portability. The main disad-vantages are the lack of realism and the inability to teachentire operations. High fidelity simulators are those that aremore lifelike, often with the ability to move beyond simpleskill or task training and simulate partial or whole opera-tions. Traditional high fidelity simulators include animalmodels and cadavers. Animal models such as the pig modelused for training in laparoscopic surgery allow for realistictissue handling and blood flow. Unfortunately animal mod-els are costly, require veterinary assistance, raise social andethical questions, and differ anatomically from humans. Al-though anatomically ideal, cadaveric models are not alwaysreadily available and the tissues do not truly mimic thecompliance of live surgery. Commercially available high fi-delity simulators such as the Uro-Scopic Trainer (Limbs andThings, Bristol, United Kingdom) offer the advantages ofreusability, realistic anatomy and the ability to use realsurgical instruments such as rigid and flexible endoscopes.However, these simulators are significantly more expensivethan low fidelity models and require maintenance. Matsu-moto et al performed a randomized controlled trial compar-ing low and high fidelity bench models for teaching endouro-logical skills to novices.7 Using a validated global ratingscale the authors concluded that students trained on benchmodels performed significantly better than those who under-went didactic teaching alone. Interestingly there was nosignificant difference between students who trained on thelow fidelity model and those who trained on the high fidelitymodel. The authors also noted that the low fidelity model

was produced at less than 1% of the purchase cost of the high

SURGICAL SIMULATION1692

fidelity model. Other studies have reported similar results,leading to the conclusion that low fidelity simulators arebest used for the teaching of basic tasks to junior traineesdue to the lower cost and equivalent performance comparedto high fidelity models.8,9

A relatively new category of simulation, virtual reality,has arisen as a result of significant improvements in com-puting and graphics capabilities. Virtual reality is defined as“an artificial environment which is experienced through sen-sory stimuli (as sights and sounds) provided by a computerand in which one’s actions partially determine what hap-pens in the environment.”4 While expensive (prohibitively sofor most institutions) and requiring maintenance, VR simu-lators offer the opportunity to practice basic skills or entiresurgical procedures in virtually rendered environments. Fora VR model to be realistic and useful it must correctlyreproduce anatomy, and preserve anatomical characteristicssuch as weight and deformability.10 Ideally the simulationmust also provide forms of tactile feedback to inform thesurgeon when surgical instruments are in direct contactwith virtual structures. Another visual factor essential torealism is a 3-dimensional portrayal of the environment.Actual endoscopic instruments can also be incorporated intothe simulation and the sensation of pressure conveyed di-rectly through the instruments.

VR simulators have the potential to provide models tai-lored to specific pathological conditions, allowing the sur-geon to practice various operative approaches before theactual case and determine the optimal surgical approach. Inaddition, through computer dissection and analysis of skillperformance VR simulators are beginning to provide virtualmentoring and objective assessment of skill performance. Asthe simulator can create a standard scenario, the operatorcan program specific factors that must be performed cor-rectly or within a certain amount of time. Mistakes such asapplying too much pressure with instruments in delicateareas or accidentally severing vascular structures can bedetected, virtual instructional cues can be provided andsurgeon learning can be enhanced. In the early days ofvirtual reality surgical simulation endoscopic, laparoscopicor angiographic scenarios were initially more realisticallyproduced because the simulation avoided the complexities ofrecreating the tactile sensation and more varied approachesneeded for open cases. Since real instruments such as endo-scopes, catheters or wires could be incorporated into mini-mally invasive simulations, it was much easier to generaterealistic sensations of pressure and force feedback throughendoscopic or angiographic tools. These capabilities wereespecially advantageous for the field of urology, which incor-porated many procedures amenable to VR simulation suchas laparoscopic techniques, cystoscopy, TURP and ureteros-copy.

SIMULATOR VALIDATION

Before simulation can be incorporated into surgical curric-ula or used as a tool for assessment of competence it isessential that the validity and reliability of the assessmenttools be established. Reliability of a training instrumentrefers to the consistency or reproducibility of the test. Sub-ject measurement should be consistent not only when as-

sessed by a single observer on different occasions, but also

when assessed by various observers. Establishment of testreliability is necessary before the tool can be proved valid.

Validity broadly implies that the instrument appropri-ately measures what it was intended to measure. To thatend a variety of subjective and objective benchmarks havebeen developed to assess validity. Subjective assessment iscomprised of face validity and content validity, which de-scribe the appropriateness of the test in the opinion of ex-perts. Face validity establishes that the test seems reason-able and appropriate, while content validity assures that thecontents of the test cover the relevant areas of the subjectbeing assessed.

Correlation of the results of a new assessment tool withthose of an established tool constitutes criterion validity.This is comprised of concurrent validity, when the new testand an established test are performed simultaneously, andpredictive validity, the extent to which scores on the newtest predict future clinical performance. In the event that noclear standard exists with which the new test can be com-pared, construct validity is sought, in which the skill to bemeasured is correlated with another characteristic. Typi-cally construct validity is established by demonstrating dif-ferences in test performance between experts and novices inthe skill to be measured.

To validate a surgical simulator performance on the sim-ulator must accurately predict performance in the operatingroom. Unfortunately there are few reliable measures of sur-gical performance. At present the best measure of operativeperformance is the Objective Structured Assessment ofTechnical Skills examination, which relies on a combinationof checklists and global rating scales for the evaluation ofparticular tasks and characteristics.11,12 Despite the avail-ability of a variety of physical and virtual reality simulators,relatively few have undergone rigorous testing that wouldfully validate their use for training or proficiency assess-ment.

CURRENT STATUS OFUROLOGY SPECIFIC SIMULATORS

TURP SimulationTURP remains the gold standard surgical procedure forsuccessfully treating medically refractory lower urinarytract symptoms associated with benign prostatic hyperpla-sia.13,14 Newer technologies such as holmium laser enucle-ation and GreenLight™ laser therapy are still in their in-fancy yet offer alternatives to TURP that many urologistsare beginning to use. These new, laser based proceduresrequire unique skills that must be acquired by surgeons.However, there are no currently available simulators toteach these skills. As these techniques gain in popularity,appropriate simulators would be beneficial to facilitatetraining in these procedures.

TURP training remains challenging for several reasons.TURP involves working in a small 3-dimensional space, with2-dimensional visual feedback requiring the operator tohave or develop unique visual-spatial and psychomotor abil-ities. The field is often visually obscured by tissue and blood,which can be disorienting. There are many potential ana-tomical hazards. An error in judgment or skill could resultin total urinary incontinence, bladder, urethral, rectal orureteral injury, profuse blood loss, erectile dysfunction and

life threatening levels of hyponatremia. Much like any pro-

SURGICAL SIMULATION 1693

cedural skill in medicine, with board certification the abilityto perform TURP is not distinguished from the judgmentand cognitive skills that are measured via oral and writtenexaminations.

Historically this training problem was addressed withcase volume. A decade ago residents performed 120 TURPson average before graduating.15 Since the new millennium,the mean number of TURPs a graduating resident performshas plateaued at approximately 62.16 Sweet et al surveyed72 board certified urologists at the AUA annual meeting in2002.17 The mean number of TURPs they believed necessarybefore entering independent practice was 66.8, which isgreater than the mean and applies to the majority of grad-uating residents in the United States. Given such issues,simulation training for TURP has been pursued. There arerelatively few low fidelity models for the simulation ofTURP, the majority of which involve standard equipmentwith various surrogate tissues meant to represent the pros-tate, eg chicken breast or vegetable matter. These modelsoffer little more than allowing the trainee to practice assem-bling the equipment and feeling the mechanism of action ofthe resectoscope.

The first reported VR TURP simulator was described byLardennois et al in 1990 after having seen a colonoscopysimulator.18 Kumar et al reported on work on their TURPsimulator since 1999.19 This unit used an optical trackingdevice, and a hybrid/computer generated and physical pros-tate model. The loop is tracked with a potentiometer and thescope tracked in space. This simulator was constructed as avirtual 3-dimensional training aid to eventually be used inthe operating room but it lacked features such as bleeding.Two urologists evaluated the content, and believed that thetrainer had a poor frame refresh rate, lack of bleeding, modelmovement and inaccurate deformation, contributing to in-accuracy in vitro. Validation studies are currently lack-ing.19,20 Ballaro et al created a VR TURP simulator thatunfortunately lacks real-time interactivity and tactile feed-back.21 Manyak et al described a simulator for lower urinarytract procedures which adds haptics force feedback.22 KarlStorz (Tuttlingen, Germany) has been developing and dem-onstrating a transurethral bladder tumor resection/TURPsimulator which also provides force feedback. Their URO-Trainer has demonstrated face, content and construct valid-ity as a simulator of basic lower urinary tract procedures.23

While it has not yet been validated as a TURP simulator, thedevelopers promise that a future module for the unit willsimulate TURP. Kallstrom et al describe their real-time, VRTURP simulator comprised of a computer, a robotic armwith 6 df connected to a modified resectoscope and a dummycamera.24 Although it is still in the rudimentary designphase, the authors established face and content validitythrough questionnaires sent to 17 practicing urologists. Thelevel of anatomical detail is quite limited and the designerspoint out that the simulator is meant to be conceptual ratherthan anatomically realistic.

In partnership with Gyrus/ACMI (Reading, Berkshire,United Kingdom) Oppenheimer et al began development ofthe University of Washington virtual reality TURP simula-tor in 2000.25 Recognizing that management of hemostasiswas a critical skill set to learn during TURP, they created ableeding movie texture map library, creating realistic bleed-ing. Fluid flow detection by potentiometers attached to the

stopcocks triggers the changing of the blood flow movies. A

force feedback device was integrated using a Mantis (MimicTechnologies, Inc., Seattle, Washington). Didactics and agraphical user interface as well as subtask training moduleshave been added to train orientation, cutting and coagula-tion skills independently. The simulator logs motion andforce data, operative errors, and resection safety and effi-ciency metrics. Version 1.0 was validated by 72 board certi-fied urologists and 19 novices at the AUA annual meeting in2002. All measured aspects of the simulator content weredeemed “more than acceptable” with experts committingfewer errors and outperforming novices in all of the keymetrics.17 Correlations of key metrics differed among thenovice, trainee and expert groups.26 The authors endorsedthe simulator for training but cautioned its use for assess-ment and accreditation. A multi-institutional predictive va-lidity study examining simulator ability to improve perfor-mance in the operating room with residents is under way.The University of Washington TURP simulator was com-mercially released at the AUA annual meeting in 2007 onthe SurgicalSIM® platform (fig. 1).

Cystoscopy and Ureteroscopy SimulationUrological endoscopic training has experienced several sig-nificant changes through the years. Historically residentswould alternate looking through the cystoscope with theconsulting urologist. Aids such as separate optical viewingarms allowed students to observe surgeons while theyworked. With the advent of videoendoscopy, trainees arenow able to simultaneously observe the same field of visionas the surgeon. It would seem a logical step to take theseimage based procedures and render them in a simulated

FIG. 1. SurgicalSIM TURP simulator (photo courtesy of METI®)

SURGICAL SIMULATION1694

environment. Indeed, models of basic endourological proce-dures such as cystoscopy and ureteroscopy are among themost commonly used and studied types of simulation.

Relatively few low fidelity simulators exist for basic lowerand upper urinary tract endourological procedures. Thedearth of these simulators may be due in part to the vari-ability of anatomy as well as the numerous componentsinvolved in even the most basic endourological procedure.Matsumoto et al hypothesized that the key to the design ofbench models is the identification of essential constructsrather than the accurate representation of anatomy.7 In atrial analyzing the effect of bench model fidelity on endouro-logical skills, the authors constructed a low fidelity benchmodel consisting of a Penrose drain (urethra), an invertedStyrofoam™ cup (bladder), molded latex and 2 embeddeddrinking straws (ureters) (fig. 2). Small holes were cut in thestraws to allow the insertion of mid ureteral stones. Stan-dard endoscopic instruments were used for low and highfidelity models. The entire cost for the construction of thismodel was $20, which is much less than the $3,700 requiredto purchase the high fidelity bench model. Most importantly,the novices trained on the low fidelity model experienced asimilar level of skill improvement as those trained on thehigh fidelity simulator.

Several high fidelity bench trainers are currently avail-able including the Uro-Scopic Trainer and the Scope Trainer(Mediskills Ltd., Edinburgh, United Kingdom). The Uro-Scopic Trainer consists of a mannequin of the male genito-urinary tract through which standard instruments may bepassed. This is an obvious advantage as trainees may practicewith the same instruments they will use in the operating room.The Uro-Scopic Trainer allows the user to simulate severalendourological techniques including examination of the uri-nary tract, stent and guidewire insertion, lithotripsy andstone retrieval. Chou et al examined the ability of 2 simu-lators, a high fidelity bench model (Uro-Scopic Trainer) anda VR simulator (URO Mentor™), to teach basic uretero-scopic skills to inexperienced medical students.27 A total of16 medical students received didactic training and videoinstruction by a single faculty member. Participants weretrained on the bench model or the VR simulator until able toperform the ureteroscopy independently. After 2 months

FIG. 2. Low fidelity ureteroscopy simulator (reprinted with author’spermission).7

participants were graded on the ability to perform basicureteroscopy on a pig model. The authors found no signifi-cant difference between the 2 groups in their ability toperform the steps of the procedure and concluded that eitherof these training modalities may improve the initial clinicalperformance of urological trainees. The Scope Trainer offerssimilar features including a distensible bladder, ureteralorifices and ureters that follow the natural course for anadult male (fig. 3). Standard procedures such as ureteral orrenal intracorporeal lithotripsy and urothelial tumor biopsymay be simulated with the Scope Trainer.

Currently there are several commercially available cys-toscopic and ureteroscopic virtual reality simulators. Thebest studied urological simulator is the URO Mentor, aWindows based virtual reality ureteroscopy simulator. Wil-helm28 and Watterson29 et al independently tested groups ofinexperienced medical students who were randomized totraining or no training on the URO Mentor, and showed thatthe trained group outperformed the untrained group by ob-jective and subjective measures, thereby establishing face,content and concurrent validity. Construct validity was sub-sequently proven by Jacomides et al who compared theperformance of 16 VR trained medical students with 16 VRnaïve urology residents, and showed that trained studentsperformed comparably to first year urology residents whentested on a VR clinical scenario and that task completiontimes for the residents correlated inversely with year oftraining.30

Ogan et al further demonstrated the concurrent andpredictive validity of the URO Mentor by comparing theperformance of VR trained medical students with that ofuntrained residents during ureteroscopy with a cadavermodel.31 Due to the inherent variability of anatomy in hu-mans a cadaver model was chosen to provide a standardizedmodel to simulate clinical ureteroscopy. Indeed student per-formance but not resident performance on the simulatorcorrelated strongly with performance on the cadaver bysome objective and subjective measures. Resident cadaverperformance correlated strongly with level of training. Theauthors concluded that performance on the URO Mentor canpredict performance in the operating room for novices, but

FIG. 3. Scope Trainer high fidelity simulator (photo courtesy ofMediskills Ltd.).

for more experienced endoscopists clinical training and ex-

SURGICAL SIMULATION 1695

perience outweigh the contribution of VR training to overalloperative proficiency.

Matsumoto et al compared VR simulation with the UROMentor to a previously validated high fidelity bench trainer(Uro-Scopic Trainer) for the assessment endourologicalskills.32 Urology residents of various postgraduate yearswere assessed on the ability to perform routine endourologi-cal tasks using a global rating scale. Senior residents scoredsignificantly higher than junior residents and required lesstime to complete the task on the VR simulator. The authorsconcluded that performance on the VR simulator was com-parable to performance on the high fidelity bench model andthat the URO Mentor ureteroscopy simulator is a useful toolfor the assessment of resident performance.

Recently Reich et al reported on a high level VR simulatorspecifically designed for endourological procedures in thelower urinary tract.23 The URO-Trainer features virtualreality images derived from digital video footage of hundredsof endourological procedures (fig. 4). The authors note thepotential benefit of this simulator for the teaching of lowerurinary tract procedures compared to the URO Mentorwhich focuses primarily on upper urinary tract procedures.The user is able to choose from 4 optical endoscopes (0, 30, 70and 120 degrees) with realistic force feedback. Bleeding isresponsive to fluid flow and blood loss can be recorded.Multiple procedures may be performed using a variety oftools including resection loops and laser fibers. During train-ing objective data such as percentage of bladder mucosainspected, percentage of lesions treated and blood loss arerecorded. In this study 24 medical students and 12 residentsperformed specific tasks during 2 separate 60-minute ses-sions using the URO-Trainer. Exposure to the URO-Trainerimproved the performance of all groups with more experi-enced participants showing greater aptitude, thus demon-strating construct validity. Further validation studies usingthe URO-Trainer have been proposed to achieve clinicalvalidity.

Percutaneous Renal Access SimulationWhile urologist acquired renal access is becoming more com-mon, at present only 11% of urologists performing percuta-

FIG. 4. URO-Trainer VR simulator (photo courtesy of Karl Storz)

neous nephrolithotomy obtain their own renal access.33 Thismay be due to a lack of training, a low level of comfort oreven perceived time constraints. Watterson et al recentlyreviewed their experience with percutaneous access gainedby radiologists or a single, fellowship trained urologist.34

Access related complication rates were significantly higherin the radiology access group with significantly better stone-free rates in the urology access group (86% vs 61%). The factremains that most urologists who perform percutaneousnephrolithotomy do not obtain their own access and, thus,many residents may never learn this important skill. Sur-gical simulation is a possible solution for practicing andtraining urologists to gain the necessary skills to obtainpercutaneous renal access.

Ex vivo biological bench models simulating percutaneousrenal surgery using porcine kidneys have been described.Hammond et al described the use of porcine kidneys containingpreimplanted artificial calculi placed inside intact chicken car-casses to simulate percutaneous nephrolithotomy.35 Residenttrainees submitted anonymous questionnaires and all par-ticipants rated the exercise as worthwhile for improvingcomfort level with the procedures. Similarly Strohmaier andGiese used porcine kidneys with intact ureters embedded insilicone to allow ultrasound guided percutaneous renal ac-cess.36 The authors noted that this model allows simulationof all flexible and rigid endourological techniques, andprovides more realistic tissue handling than nonbiologicalmodels. These models have yet to be rigorously validatedand comparisons to nonbiological models remain to beperformed.

Two previously discussed high fidelity nonbiological mod-els offer modules for training percutaneous renal proce-dures. The Percutaneous Nephrolithotomy Trainer (Limbsand Things) allows for simulated needle insertion, passageof guidewires, dilation of the nephrostomy tract, nephros-copy and stone removal. The model consists of a slab thatresponds realistically to insertion of wires, dilators andsheaths. The slab contains multiple calices where stonesmay be placed for procedural variability. X-ray imaging issimulated by the translucent slab and a light box. The PercTrainer (Mediskills Ltd.) is a high fidelity bench model thatsimulates ultrasound or fluoroscopy guided renal access.This model also allows for percutaneous nephrostomy tractdilation, stone extraction or fragmentation and nephrostomytube placement.

Virtual reality simulation of percutaneous renal accesshas been demonstrated with the PERC Mentor™, a recentcomplement to the URO Mentor (fig. 5). This percutaneousrenal access simulator is comprised of a mannequin repre-senting the human flank including simulated skin and pal-pable ribs. Access needles may be passed through the ab-dominal wall under simulated fluoroscopic guidance withreal-time feedback including fluoroscopy time used, the abil-ity to perform retrograde or antegrade pyelography, needleaspiration and rotation of the C arm. A computer interfaceprovides multiple learning tasks designed to guide the userthrough components such as accessing various calices, pass-ing guidewires and using ureteral catheters to assist withwire passage. Face, content and concurrent validity of thesimulator were established by Knudsen et al who showedthat inexperienced medical students and residents random-ized to training on the PERC Mentor outperformed controls

in a VR case scenario by subjective and objective criteria.37

SURGICAL SIMULATION1696

In addition, trained students and residents obtained percu-taneous access in a porcine model faster and more accu-rately than their untrained counterparts, thereby demon-strating concurrent and predictive validity.38 Knudsen et alfurther established construct validity using a global valida-tion scale for 63 subjects (medical students and residents)randomized to training on the PERC Mentor vs no train-ing.39 The group randomized to training demonstrated sig-nificantly better performance than the group that did notreceive training.

Laparoscopic SimulatorsLaparoscopic surgery is certainly one of the more difficultsurgical techniques to learn, let alone simulate. Shortly af-ter the introduction of laparoscopic cholecystectomy it be-came evident that the skills required to perform laparoscopyare different than those to perform open surgery. In the1990s the surgical community realized that the early phaseof the learning curve, familiarity with instrumentation andbasic tasks, could be taught in a systematic and logicalfashion outside the operating room. The tools available forlaparoscopic procedural training and simulation range fromthe original low fidelity pelvic trainers to virtual realitysimulators of varying complexity. The best studied of thesesimulators is the MISTELS, which is a low fidelity physicallaparoscopy simulator comprised of 5 exercises.40 Validitywas established in a series of studies comprised of more than250 subjects (surgeons and trainees).41 Dividing subjects into 3groups based on laparoscopic experience, investigators showedthat MISTELS scores differed significantly among the groupswith successively higher scores achieved with increasing levelof laparoscopic experience. Concurrent and predictive validitywere established by comparing MISTELS scores with clinicaloperative performance as judged by trained observers in theoperating room using a validated global assessment tool. TotalMISTELS scores correlated strongly with global assessmentscores. Furthermore, level of training and MISTELS score

FIG. 5. Combined URO Mentor/PERC Mentor virtual reality simu-lator (photo courtesy of Simbionix, Cleveland, Ohio).

proved to be independent predictors of intraoperative technical

skill. After rigorous testing and validation of the MISTELS, theSociety of American Gastrointestinal and Endoscopic Surgeonsincorporated the MISTELS into their Fundamentals of Lapa-roscopic Surgery program, which provides a standardized cur-riculum comprising cognitive and technical aspects of laparos-copy.

Low fidelity pelvic (box) trainers permit the practice ofmultiple basic tasks such as transferring, cutting, suturingand knot tying. Task practice in the box trainer has beenshown to improve similar in vivo task performance.42 Assuch, these trainers are routinely used in laparoscopycourses and for resident education in combination with di-dactic teaching. However, a criticism of box trainers is thatthey are too simplistic and do not reflect the complexity ofthe real procedure. Scott et al were the first to address thisissue by randomizing 22 surgery residents to box training orno training and then assessing performance of laparoscopiccholecystectomy in a patient.43 The trained group demon-strated improved skill in actual surgery indicating thatlaparoscopic box training was transferable to the operatingroom. Conversely in the only urology specific study Traxeret al randomized 12 residents in the same fashion and thenassessed their performance of laparoscopic nephrectomy in aporcine model.44 The authors did not find any benefit toskills training on the performance of laparoscopic nephrec-tomy, questioning the value of inanimate simulation formore complex laparoscopic procedures.

Intracorporeal suturing is one of the most difficult basiclaparoscopic skills to acquire. Laparoscopic radical prosta-tectomy and pyeloplasty are examples of urological proce-dures that rely on adept suturing skills to ensure watertightclosure. Several studies have demonstrated the value ofbox trainers in combination with simulated tissues, egchicken skin, for the training of intracorporeal anastomo-sis and knot tying.45,46 Box trainers clearly permit theassimilation of essential basic tasks such as suturing, butare no substitute for in vivo training where the surgeonmust marry anatomical knowledge, procedural knowledgeand laparoscopic skills to successfully complete the surgi-cal procedure.

While the inability of low fidelity box trainers to simulatecomplete procedures quickly led to the development of VRsimulators, researchers at Johns Hopkins have developed ahigh fidelity trainer for the training and evaluation of uro-logical skills.47 Lapman is a synthetic torso based on theVisible Human Project® of the National Library of Medicine.This trainer is a replica of the torso including the abdominalcavity and thorax. The developers enrolled 25 medical stu-dents, residents and urologists who were randomized to theLapman or a standard box trainer. Users reported thatLapman was significantly easier to use, provided a betterview and was a superior approximation of human anatomy.

The need for a dynamic model with accurate reproduc-ibility of anatomical structures and tactile feedback led tothe development of virtual reality laparoscopy simulators.The first simulators, eg MIST-VR™, essentially mimickedbox trainers in that they simulated abstract tasks enablingthe acquisition of similar psychomotor skills rather thanprocedural/anatomical knowledge. Nevertheless, the promisethat skills learned in the virtual reality environment are trans-ferable to real surgery was demonstrated by Grantcharovet al.48 Randomizing 16 laparoscopic novices to MIST-VR

training or no training, the trained group performed lapa-

SURGICAL SIMULATION 1697

roscopic cholecystectomy significantly faster, with fewer er-rors and better economy of motion. Subsequent simulatorshave augmented basic skills training modules to includesimulated segments of actual procedures such as tissue ma-nipulation and suturing skills. The goal of reproducing anentire laparoscopic procedure in the virtual reality environ-ment has recently been accomplished with the introductionof laparoscopic cholecystectomy simulators, eg LAP Men-tor™ (fig. 6). This device enables trainees to perform acomplete virtual laparoscopic cholecystectomy with the ben-efit of some tissue interaction (force) feedback. Althoughthese systems clearly demonstrate the potential of virtualreality laparoscopy simulation, they simulate a clinicallyeasy procedure where the clinical learning experience isreadily available. The challenge and clear benefit will be themodeling of more complex procedures such as laparoscopicprostatectomy. Unlike endourological simulation there is novirtual reality software currently available simulating alaparoscopic urological procedure. As no clinical data yetexists evaluating the transferability and validity of the cur-rent laparoscopic cholecystectomy simulator, the develop-ment and investment in virtual reality simulators for moreadvanced urological procedures should be appropriatelymeasured. Nevertheless, the potential of laparoscopic simu-lators to reduce the apprenticeship learning of basic, inter-mediate and even advanced skills, and to permit the practiceof difficult procedures before actual surgery promises toimprove the quality of urological care.

FIG. 6. LAP Mentor virtual reality simulator (photo courtesy ofSimbionix).

An additional note should be made regarding the rapidlygrowing field of robotic assisted laparoscopic surgery. Pri-marily used in the United States for robot assisted laparo-scopic radical prostatectomy, the da Vinci® surgical systemhas gained widespread use among urologists as an adjunctto traditional laparoscopy. Despite the obvious advantagesof enhanced surgeon comfort, 3-dimensional vision andgreater degrees of freedom, the da Vinci system presentsunique challenges such as the absence of tactile feedback.While there is little doubt that robotic assisted procedureswould benefit greatly from simulation, at present there is nocommercially available robot assisted laparoscopic VR sim-ulator. Basic skills such as suturing and object transfer maybe simulated using low fidelity models in the same fashionas traditional laparoscopy. Surgeons using the da Vinci sys-tem presently rely on visual cues to determine whethertissue is under appropriate tension or that surgical knotsare properly secured. Clearly robot assisted VR simulatorswill need to provide users with superior graphical renderingto accurately simulate these precise skills. Prototype simu-lators such as those being developed by Mimic Technologies,Inc. show great promise and it seems to be only a matter oftime before surgeons are able to benefit from these trainingtools.49

THE FUTURE OF SURGICAL SIMULATION

The random opportunities of our current apprenticeship sys-tem for surgical training need to be replaced by a curriculumor learning system that meets the needs of residents andtheir future. Surgical simulation, whether model or com-puter based, provides a unique opportunity for repetitiveskills training with the exploration of all possible outcomesin a risk-free environment that can maximize the educa-tional experience and reduce the training time for surgeonsin complex surgical techniques.48

For practicing urologists technical advances are happen-ing so quickly that it is not uncommon for surgeons 5 to 10years from residency training to lack the skills for many ofthe minimally invasive surgical techniques. As such, “learn-ing how to learn” surgery may well be the most importantlesson we can impart to residents during training, for thismore than anything else will equip them for the future. Tothis end model and VR based simulator training will allowsurgeons to learn a new set of skills in a postgraduateeducational format that can then be combined to complete atask, which in turn can be combined to allow for the masteringof a new procedure. Low fidelity trainers have proven to beexcellent, cost-effective resources for basic skill and task train-ing, particularly for novice trainees, and should remain animportant component of surgical education. Through repetitivepractice surgeons can acquire the confidence and competencenecessary to safely introduce a new procedure into their prac-tice.50 In addition, surgeons who perform specific proceduressporadically will be able to practice and maintain proficiencybetween cases.

The development of centers of excellence in surgical ed-ucation will provide surgeons with the opportunity to prac-tice and maintain surgical skills. With proper validationand integration into educational programs, these virtualreality and other training modalities will eventually tran-sition from training tools to testing modalities. In this

regard they will become invaluable in board examina-

SURGICAL SIMULATION1698

tions, certification processes and re-certification. At longlast chirugie, or the “hand work” that we all do, will beobjectively evaluated for the cognitive and manual aspectsthat are essential to the successful completion of all sur-gical procedures.

Surgical simulation will not replace the need for welldesigned, comprehensive educational curricula or reduce theimportance of dedicated and committed educators, but will

enhance and complement these efforts. The integration of

ED, Sidhu RS et al: The educational impact of bench model

simulation into the surgical training curriculum will allowthe trainee to acquire the basic surgical skills foundation.This will allow the surgeon educator to then concentrateon teaching judgment and professionalism, and strengthen-ing the knowledge and interpretation of what is observed inthe clinical setting to create the competent surgeon at acognitive and manual level. Simulation must be used withinan effective learning environment, underpinned by knowl-

edge and professional attitudes.

APPENDIX

Categories of Surgical Simulation

Category Examples Advantages DisadvantagesRelative

Cost Uses

Low fidelity Peg boards, syntheticsuturing mats

Widely available, reusable,may use actual instruments

No procedural training � Basic task training for novices

High fidelity:Biological Live animals, dissected

animal tissues,human cadavers

Whole procedure simulation,dynamic anatomy in liveanimals (ie bleeding),realistic haptics

Anatomical differences (animals),poor tissure compliance (humancadavers), ethicalconsiderations, not reusable

���� Whole procedural training (ielaparoscopicnephrectomy/cystectomy)

Nonbiological High fidelity benchmodels (Uro-ScopicTrainer, ScopeTrainer)

Reusable, allow use of actualinstruments, some elementsof procedural trainingbeyond basic tasks

Not proven to be more effectivethan low fidelity models forbasic skills training, lessrealistic tissue haptics thanbiological models

�� Basic skills training, often goodoption for procedural training,especially for endourologicalprocedures (ie ureteroscopy,percutaneous renal surgery)

Virtual reality LapSim®, PercMentor, SurgicalSim

Reusable, instantperformance feedback, maysimulate basic tasks orwhole procedures

Maintenance, costly, unreliablehaptics

��� Basic skills or proceduraltraining, simulators availablefor endourological,laparoscopic and percutaneous

renal procedures

Abbreviations and Acronyms

AUA � American Urological AssociationMISTELS � McGill Inanimate System for Training

and Evaluation of Laparoscopic SkillsTURP � transurethral prostate resection

VR � virtual reality

REFERENCES

1. Carter BN: The fruition of Halsted’s concept of surgical train-ing. Surgery 1952; 32: 518.

2. Anastakis DJ, Wanzel KR, Brown MH, McIlroy JH, HamstraSJ, Ali J et al: Evaluating the effectiveness of a 2-yearcurriculum in a surgical skills center. Am J Surg 2003;185: 378.

3. Leach DC: A model for GME: shifting from process to out-comes. A progress report from the Accreditation Council forGraduate Medical Education. Med Educ 2004; 38: 12.

4. Merrium-Webster Online Dictionary. Available at www.merriam-webster.com. Accessed July 6, 2007.

5. Satava RM: Accomplishments and challenges of surgical sim-ulation. Surg Endosc 2001; 15: 232.

6. Delp SL, Loan JP, Hoy MG, Zajac FE, Topp EL and Rosen JM:An interactive graphics-based model of the lower extremityto study orthopaedic surgical procedures. IEEE TransBiomed Eng 1990; 37: 757.

7. Matsumoto ED, Hamstra SJ, Radomski SB and Cusimano MD:The effect of bench model fidelity on endourological skills: arandomized controlled study. J Urol 2002; 167: 1243.

8. Anastakis DJ, Regehr G, Reznick RK, Cusimano M, Mur-naghan J, Brown M et al: Assessment of technical skillstransfer from the bench training model to the humanmodel. Am J Surg 1999; 177: 167.

9. Grober ED, Hamstra SJ, Wanzel KR, Reznick RK, Matsumoto

fidelity on the acquisition of technical skill: the use ofclinically relevant outcome measures. Ann Surg 2004; 240:374.

10. Satava RM: Virtual reality, telesurgery, and the new worldorder of medicine. J Image Guid Surg 1995; 1: 12.

11. Wanzel KR, Ward M and Reznick RK: Teaching the surgicalcraft: from selection to certification. Curr Probl Surg 2002;39: 573.

12. Feldman LS, Sherman V and Fried GM: Using simulators toassess laparoscopic competence: ready for widespread use?Surgery 2004; 135: 28.

13. Holtgrewe HL: Surgical management of benign prostatic hy-perplasia in 2001—a pause for thought. J Urol 2001; 166:177.

14. AUA Guideline on management of benign prostatic hyperpla-sia: (2003). Chapter 1: diagnosis and treatment recommen-dations. J Urol 2003; 170: 530.

15. Sweet R, Porter J, Oppenheimer P, Hendrickson D, Gupta Aand Weghorst S: Simulation of bleeding in endoscopic pro-cedures using virtual reality. J Endourol 2002; 16: 451.

16. Urology Residency Case Log Report. Chicago: AccreditationCouncil for Graduate Medical Education 2004.

17. Sweet R, Kowalewski T, Oppenheimer P, Weghorst S andSatava R: Face, content and construct validity of the Uni-versity of Washington virtual reality transurethral pros-tate resection trainer. J Urol 2004; 172: 1953.

18. Lardennois B, Clement T, Ziade A and Brandt B: Computerstimulation of endoscopic resection of the prostate. AnnUrol (Paris) 1990; 24: 519.

19. Kumar PV, Gomes MP, Davies BL and Timoney AG: A com-puter assisted surgical trainer for transurethral resectionof the prostate. J Urol 2002; 168: 2111.

20. Gomes MP, Barrett AR, Timoney AG and Davies BL: A com-puter-assisted training/monitoring system for TURP struc-ture and design. IEEE Trans Inf Technol Biomed 1999;

3: 242.

SURGICAL SIMULATION 1699

21. Ballaro A, Briggs T, Garcia-Montes F, MacDonald D, Ember-ton M and Mundy AR: A computer generated interactivetransurethral prostatic resection simulator. J Urol 1999;162: 1633.

22. Manyak MJ, Santangelo K, Hahn J, Kaufman R, Carleton T,Hua XC et al: Virtual reality surgical simulation for lowerurinary tract endoscopy and procedures. J Endourol 2002;16: 185.

23. Reich O, Noll M, Gratzke C, Bachmann A, Waidelich R, SeitzM et al: High-level virtual reality simulator for endouro-logic procedures of lower urinary tract. Urology 2006; 67:1144.

24. Kallstrom R, Hjertberg H, Kjolhede H and Svanvik J: Use of avirtual reality, real-time, simulation model for the trainingof urologists in transurethral resection of the prostate.Scand J Urol Nephrol 2005; 39: 313.

25. Oppenheimer P, Gupta A, Weghorst S, Sweet R and Porter J:The representation of blood flow in endourologic surgicalsimulations. Stud Health Technol Inform 2001; 81: 365.

26. Rashid HH, Kowalewski T, Oppenheimer P, Ooms A, KriegerJN and Sweet RM: The virtual reality transurethral pros-tatic resection trainer: evaluation of discriminate validity.J Urol 2007; 177: 2283.

27. Chou DS, Abdelshehid C, Clayman RV and McDougall EM:Comparison of results of virtual-reality simulator andtraining model for basic ureteroscopy training. J Endourol2006; 20: 266.

28. Wilhelm DM, Ogan K, Roehrborn CG, Cadeddu JA and PearleMS: Assessment of basic endoscopic performance using avirtual reality simulator. J Am Coll Surg 2002; 195: 675.

29. Watterson JD, Beiko DT, Kuan JK and Denstedt JD: Random-ized prospective blinded study validating acquistion of ure-teroscopy skills using computer based virtual reality en-dourological simulator. J Urol 2002; 168: 1928.

30. Jacomides L, Ogan K, Cadeddu JA and Pearle MS: Use of avirtual reality simulator for ureteroscopy training. J Urol2004; 171: 320.

31. Ogan K, Jacomides L, Shulman MJ, Roehrborn CG, CadedduJA and Pearle MS: Virtual ureteroscopy predicts uretero-scopic proficiency of medical students on a cadaver. J Urol2004; 172: 667.

32. Matsumoto ED, Pace KT and D’A Honey RJ: Virtual realityureteroscopy simulator as a valid tool for assessing endo-urological skills. Int J Urol 2006; 13: 896.

33. Bird VG, Fallon B and Winfield HN: Practice patterns in thetreatment of large renal stones. J Endourol 2003; 17: 355.

34. Watterson JD, Soon S and Jana K: Access related complica-tions during percutaneous nephrolithotomy: urology versusradiology at a single academic institution. J Urol 2006; 176:142.

35. Hammond L, Ketchum J and Schwartz BF: A new approach tourology training: a laboratory model for percutaneous

nephrolithotomy. J Urol 2004; 172: 1950.

36. Strohmaier WL and Giese A: Ex vivo training model for per-cutaneous renal surgery. Urol Res 2005; 33: 91.

37. Knudsen BE, Matsumoto ED, Chew BH, Margulis V, CadedduJA and Denstedt JD: Impact of virtual reality (VR) trainingon percutaneous renal collecting system access. J Endourol,suppl., 2004; 18: A20.

38. Margulis V, Matsumoto ED, Knudsen BE, Chew BH, PautlerSE, Cadeddu JA et al: Percutaneous renal collecting systemaccess: can virtual reality training shorten the learningcurve? J Urol, suppl., 2005; 173: 315, abstract 1162.

39. Knudsen BE, Matsumoto ED, Chew BH, Johnson B, MargulisV, Cadeddu JA et al: A randomized, controlled, prospectivestudy validating the acquisition of percutaneous renal col-lecting system access skills using a computer based hybridvirtual reality surgical simulator: phase I. J Urol 2006; 176:2173.

40. Derossis AM, Fried GM, Abrahamowicz M, Sigman HH,Barkun JS and Meakins JL: Development of a model fortraining and evaluation of laparoscopic skills. Am J Surg1998; 175: 482.

41. Fried GM, Feldman LS, Vassiliou MC, Fraser SA, StanbridgeD, Ghitulescu G et al: Proving the value of simulation inlaparoscopic surgery. Ann Surg 2004; 240: 518.

42. Fried GM, Derossis AM, Bothwell J and Sigman HH: Compar-ison of laparoscopic performance in vivo with performancemeasured in a laparoscopic simulator. Surg Endosc 1999;13: 1077.

43. Scott DJ, Bergen PC, Rege RV, Laycock R, Tesfay ST, Valen-tine RJ et al: Laparoscopic training on bench models: betterand more cost effective than operating room experience?J Am Coll Surg 2000; 191: 272.

44. Traxer O, Gettman MT, Napper CA, Scott DJ, Jones DB,Roehrborn CG et al: The impact of intense laparoscopicskills training on the operative performance of urology res-idents. J Urol 2001; 166: 1658.

45. Laguna MP, Arce-Alcazar A, Mochtar CA, Van Velthoven R,Peltier A and de la Rosette JJ: Construct validity of thechicken model in the simulation of laparoscopic radicalprostatectomy suture. J Endourol 2006; 20: 69.

46. Teber D, Dekel Y, Frede T, Klein J and Rassweiler J: TheHeilbronn laparoscopic training program for laparoscopicsuturing: concept and validation. J Endourol 2005; 19: 230.

47. Mazilu D, Patriciu A, Gruionu L, McAllister M, Ong A, EllisonL et al: Synthetic torso for training in and evaluation ofurologic laparoscopic skills. J Endourol 2006; 20: 340.

48. Grantcharov TP, Kristiansen VB, Bendix J, Bardram L,Rosenberg J and Funch-Jensen P: Randomized clinical trialof virtual reality simulation for laparoscopic skills training.Br J Surg 2004; 91: 146.

49. Albani JM and Lee DI: Virtual reality-assisted robotic surgerysimulation. J Endourol 2007; 21: 285.

50. Shalhav AL, Dabagia MD, Wagner TT, Koch MO and Linge-man JE: Training postgraduate urologists in laparoscopic

surgery: the current challenge. J Urol 2002; 167: 2135.