Task
1
3
5
Identify anatomy
No knowledge
Moderate identification
100% identification
Endoscopic maneuvers
Significant trauma, trouble with luminal view
Rare trauma, reasonable luminal view
No trauma, excellent luminal view
Employment of adjunct instrumentation
Required multiple attempts with guidewire, basket, or laser
Relatively good use guidewire, basket, or laser
Smooth use of guidewire, basket, or laser
Procedural autonomy
Needed significant assistance, frequent pauses
Minimal assistance required
No assistance
Overall proficiency
Poor
Average
Excellent
Simulators
Simulators are commonly categorized as low or high fidelity. Low-fidelity simulators are defined as those inanimate models that have decreased similarity to real-life situations, such as suture tying boards and laparoscopic box trainers [6]. These trainers allow the improvement of certain surgical skills, but not the ability to practice or evaluate the completion of an entire procedure [6]. Low-fidelity simulators also require a mentor to instruct in and observe for appropriate technique. One benefit of low-fidelity simulators is that they tend to be low cost and easily accessible. High-fidelity simulators are those that more closely mimic reality, traditionally with the use of cadaveric or animal models, but more recently with the use of accurate inanimate constructs, computer-aided technology, and the creation of simulators which may also include virtual instruction. The highest fidelity simulation is achieved by maximizing representation of visual, physical, psychological, and tactile reality [21]. Changes in the model and environment are designed to elicit trainee responses that would mimic those in real-life operative situations [13].
Early reports of endoscopic simulation were made by Trindade and colleagues in 1981, when they described the use of a cadaveric dog model for cystoscopic and ureteroscopic procedures [18]. Subsequently, the development of low fidelity of bench-top models and virtual reality simulators has provided an accessible training alternative, avoiding ethical issues and high costs of using cadaveric models. Preminger and colleagues first described the use of a VR simulator for ureteroscopy in 1996, and over the past 10–15 years, other high-fidelity simulators have become available for ureteroscopic instruction [22]. However, low fidelity models have also been studied and found to be efficacious in resident training, with similar improvement in surgical technique as what is observed with high-fidelity simulators [6]. Schout and colleagues performed a systematic review of the literature on endourologic simulators in 2008 [23]. Of the 26 validation studies reviewed for this publication, they found that high fidelity models (as described below) generally had construct validity. However, they note that relatively few studies have been performed to look that the transfer of skills, from simulator to real patient.
Low-Fidelity Simulators
Matsumoto and colleagues advocate the use of a cost-effective, low fidelity bench model for ureteroscopy, with a 2002 study demonstrating that a model constructed from readily available materials provided equitable improvement in performance, provided that the model is constructed thoughtfully, with specific maneuvers required to successfully complete the task at hand [4]. Their model was comprised of a Styrofoam cup to mimic the bladder, two straws to replicate ureters, and a Penrose drain to act as the urethra, and cost $20 to build. Importantly, in designing the low-fidelity model, expert consensus of the key steps of ureteroscopy was obtained, with the construction of the simulator intended to replicate these steps [4]. The authors were able to confirm construct validity, with low-fidelity and high-fidelity trainees with statistically significant superior performance as compared to didactic-only trainees [4]. However, predictive validity of this model was not established, as the medical student subjects could not be ethically allowed to operate autonomously on a real patient [4, 18].
Although some low cost simulators have been described for other endoscopic procedures, often times incorporating the use of standard surgical equipment and substitute tissue (i.e., chicken breast for a prostate tissue), few low-fidelity ureteroscopic simulators have been described [6]. This is likely due to the significant variability in procedural steps encountered in upper tract interrogation, due to both patient and operative factors [6]. The primary advantages of low-fidelity simulators include low cost and easy portability [6]. However, low-fidelity simulators cannot generally replicate entire procedures and will have a decreased relationship to real-life situations encountered in the operating room [6].
High-Fidelity, Non-virtual Reality Simulators
High-fidelity, non-VR simulators used in surgical training are currently constructed using materials such as latex and silicone to replicate human tissues and often employ the use of standard endoscopic equipment [8, 24]. Significant advances in available materials have allowed more accurate representation, allowing for a more realistic simulation experience. These models can be manipulated to introduce procedural variability that may be encountered in the operating room, in real-life situations. Some authors argue that use of the high-fidelity bench models is preferable to virtual reality models, as they use real surgical equipment and can be placed in the operating theater to most closely approximate a true surgical case [8]. Several simulators have been described and validated, incorporating the use of objective testing, including OSATS criteria, and subjective measures with global rating scores.
Uro-Scopic Trainer
The Uro-Scopic Trainer™ (Limbs and Things, Bristol, UK) is one high-fidelity bench model that incorporates a mannequin replica of the adult male genitourinary tract, with model urethra, bladder, ureters, and intrarenal collecting systems. It is designed to be used with standard endoscopic flexible and rigid equipment. Multiple simulation activities can be carried out with this model, including urethroscopy and cystoscopy, in addition to therapeutic and diagnostic ureteroscopy. Irrigation can be used with this system. The simulator incorporates built-in ports that can be used to place calculi into the urinary tract, and realism can be reinforced with simulated blood [18]. The simulated ureters are transparent to allow for direct observation of instrument placement during the training activity.
Matsumoto and colleagues evaluated this simulator, with validation studies including 11 junior and 6 senior urology residents [4]. All subjects underwent a pretest, a didactic training session, a post-didactic test, a simulation hands-on training session, and a final posttest. Hands-on skills were assessed using a global rating score and checklist. Senior residents outperformed juniors on the global rating score, establishing construct validity and simulation training appeared to have a direct association with overall performance [18].
Scope Trainers
The Scope Trainer™ and Advanced Scope Trainer™ (Mediskills Ltd, Edinburgh, UK), are high-fidelity bench models, with a distensible bladder and anatomic representation of the location of the ureteral orifices, the course of ureters, and orientation of the collecting systems of an adult male [18]. The Advanced Scope Trainer also has a see-through dome to allow for direct visualization of endoscopic instruments. Irrigation can be used with these models, and is emptied through the bladder. These models can also be used with a percutaneous access trainer, to teach percutaneous nephrolithotomy. In addition, the simulators contain both an upper tract calculus and an upper tract tumor, to teach both lithotripsy and endoscopic tumor management, and can be used with fluoroscopy [18].
In Brehmer’s initial validation studies of the Scope Trainer, performed in 2002, the model was prepared in the operating room, and draped as a live patient, to maximally replicate a real-life operative case [25]. Fourteen urologists were tested on rigid ureteroscopy, including nine consultants and five residents. All participants were studied on the model once and on live patients twice, to account for intraoperative variability. The mean score on a task-specific checklist was identical between the model and live patients, loosely suggesting predictive validity, and did not vary between trainees and attendings; however, subspecialists in endourology scored higher, demonstrating construct validity. Content validity was demonstrated as well, as individuals felt that the activity accurately replicated real surgery [18].
Brehmer’s group expanded the initial study of the Scope Trainer to include residents with little endourologic experience [16]. The performance of all resident groups with rigid ureteroscopy for a distal ureteral calculus, as assessed by a task-specific and global checklist, including those who had never performed ureteroscopy before, improved following instruction and 2 days of simulation training [16]. Residents who had never before performed ureteroscopy improved somewhat more than those with some endoscopic experience. All residents assessed had improved confidence with the technique following the educational program [16].
Adult Ureteroscopy Trainer
The Adult Ureteroscopy Trainer™ (Ideal Anatomic Modeling, Holt, Michigan) is a novel, high-fidelity model that was developed by a Urology resident as an affordable alternative to commercially produced simulators [24]. A computerized tomography study from a patient who had difficulty with spontaneous stone passage was used to recreate the urinary collecting system in silicone using rapid prototyping technology. This model can be used with standard ureteroscopic and fluoroscopic technologies.
White and colleagues performed validation studies of the Adult Ureteroscopy Trainer, the first with 19 consultant urologists, 11 Urology residents, 10 medical students, and 6 biotechnology industry representatives performing both ureteroscopy and nephroscopy, with questionnaires completed before and after the procedures [24]. The groups’ second study assessed construct validity with two, ten-person groups, one of which had never performed ureteroscopy and the second who performed more than 30 cases per year [24]. The first study established face and content validity, with all participants rating the simulator as a realistic training tool that was easy to use, while the second demonstrated construct validity, with consultant urologists outperforming trainees, as evaluated by a task-based scoring system [24].
High Fidelity Virtual Reality Simulators
The UroMentor™ (Simbionix, Lod, Israel) is a high fidelity, commercially available virtual reality cystoscopic and ureteroscopic simulator. It employs the use of standard semirigid and flexible endoscopic instruments and a mannequin shell, with computer-generated graphics and haptic technology to simulate anatomical structures and intraoperative situations [8, 18, 26, 27]. A database of real-life cases is available, allowing the trainee exposure to a variety of patient and surgical factors, including situations with decreased visualization and difficult anatomic configurations. This high fidelity, virtual reality model is linked to teaching modules and the simulator itself has the ability to evaluate the practitioner and record the performance history for a particular individual [18, 27].
The UroMentor has undergone numerous validation studies, looking at content, realism, and the predictive ability of simulation activities [1, 8, 20, 27–30]. Watterson and colleagues assert that content and face validity are established in this model, as it incorporates high-fidelity graphics, the use of real endoscopic instruments, and has multiple real surgical cases available for training [8]. Two studies, one by Wilhelm and colleagues and the other by Watterson and colleagues, evaluated medical students with randomization to control groups or instruction with UroMentor simulation [20, 28]. Both studies demonstrated significant improvement in a subjective global rating score following UroMentor training. Further studies by Jacomides and associates evaluated the performance of simulation-trained medical students as compared to nonsimulation-trained residents [1]. The performance of the simulation-trained medical students paralleled that of first-year Urology residents, who had performed 14 ureteroscopic procedures on average, while more senior residents performed better overall than the aforementioned groups, establishing construct validity. The medical students demonstrated significant reduction in operative time following UroMentor training as compared to their preoperative assessments. Ogan and colleagues also performed initial studies establishing predictive validity of this model, with medical students who performed well in simulation activities demonstrating higher proficiency in cadaveric models [29]. Interestingly, resident scores between the simulation activities and cadaveric procedures were not well correlated, perhaps indicating that simulation activities are most useful in instructing and evaluating the clinically inexperienced surgeon [29].
High Fidelity, Biological Models
A number of various live and cadaveric animal models have been used historically to study the upper urinary tract, namely dog, rabbit, and more recently, swine [11, 31–33]. The use of a live porcine model is favored, due to the similarity of the anatomical structure to that of humans, including multi-papillary kidney architecture [31]. Sampaio and associates conducted morphologic studies of the porcine kidney and found that, despite some minor differences, they are similar to human kidneys with regard to length, caliceal orientation, and variability and urinary drainage, making them ideal representative models. This model was also advocated as an accurate model for the study of urological procedures by Paterson and colleagues, although they noted the lack of perinephric fat in swine (causing increased kidney mobility), tortuosity of the ureters, and location of the ureteral orifices at the bladder neck, all of which may make ureteroscopic procedures more difficult and decrease the content validity [32]. The main concerns regarding the use of live animals include ethical considerations, the high cost required to maintain the research subjects as well as issues related to animal–human disease transmission, including bovine encephalopathy [15].