Several training modalities can be used when constructing a training programme for robotic surgery. In general a programme starts with knowledge development, followed by skills training using a combination of simulation modalities (Table S1), followed by real-life case observation in the operating room. When starting with actual robotic surgery there is a role for bedside assisting, proctoring and the mentoring console. Ten training courses for robotic surgery in a skills laboratory were identified (Table 1).
Table 1. Training courses in a laboratory setting
|Hanly et al.54||2004||23 surgeons||Standardised hand on dVSS training |
Self-guided learning in porcine model
|Setup time, operating time, complications||No|
|Hernandez et al.67||2004||13 surgeons||Synthetic bowel anastomosis||OSATS score, motion analysis||No|
|Ro et al.21||2005||4 experts, 17 novices||7 designed drills||Calculated performance score||Construct|
|Narazaki et al.19||2006||7 students||3 designed tasks||Time, motion analysis and muscular activation pattern||No|
|Mehrabi et al.53||2006||4 trainees with varying experience||Porcine and rat model||Operating time, quality of operation, complications||No|
|Vlaovic et al.55||2008||35 urologists||4 exercises including suturing and cutting||OSATS score||No|
|Marecik et al.24||2008||11 residents||Porcine intestine for anastomosis||Time and leak pressure||No|
|Moles et al.22||2009||7 residents||Self-designed teaching model 5 exercises||Time and errors||No|
|Finan et al.89||2010||16 students||5 designed advanced drills for hysterectomy||Time||No|
|Chandra et al.20||2010||20 novices, 9 experts||ProMIS™ (Haptica Ltd, Dublin, Ireland) Hybrid Surgical Simulator||Time, pathlength and smoothness||Construct|
When starting with robotic surgery the trainee/surgeon needs to gain knowledge and understand the robot technology, device functions, basic troubleshooting, device parameters and the limitations of the system. The next step will be the development of knowledge for specific surgical procedures. This includes patient selection and indications, preoperative preparation, patient and system positioning, port placement, procedural steps, complications and their management. To make sure every surgeon starting with robotic surgery has a basic level of theoretical knowledge, a theoretical examination on these items could be helpful.
Just as in laparoscopy, training for robotic surgery can be scheduled in a skills laboratory. In such facilities exercises on pelvic trainers and other exercises can be performed. A skills laboratory usually has the advantage of high accessibility, but a disadvantage is the need for an expensive robot for dedicated use in the training facility. With this in mind, most hospitals could probably not afford a separate robot for use in a skills laboratory only. In these cases the available robot at the operating room could be used for training after working hours or at scheduled times when no surgery is performed.
Several authors compared conventional laparoscopy with robotic assisted laparoscopy in a skills laboratory. Conventional exercises for laparoscopy can be used and can actually be performed faster and more accurately with robotic surgery.12–15 The exercises have a shorter learning curve and are performed more accurately with robot assistance.16,17 Residents without any laparoscopic experience demonstrated the capacity to rapidly learn basic surgical manoeuvres.18,19
It is important that training exercises are validated and have a proper goal. Several levels of validation can be distinguished (Table S2) Exercises should at least have face validity (the simulation resembles the real task) and construct validity (the ability to differentiate between groups with different levels of competence), and translate well to the clinical setting, before they are used in a robotic training programme. Unfortunately, there are only a few reports of validated exercises.20,21 A curriculum, consisting of five tasks for training basic robotic skills was developed by Moles et al.22 They were not able to demonstrate a significant learning curve, although there was a trend suggesting that learning had taken place. A portable, reusable, relatively inexpensive pelvic model was developed to simulate the dissection phase of the rectum. This model could be used for mounting additional exercises and be prepared for other specific surgical procedures.23 Another model that can be used is animal intestine, which is suitable for learning the suturing skills for an intestinal anastomosis.24 Training for the difficult parts of an operation in such specifically designed models can have a positive impact on the learning curve for complex procedures.
Another aspect of skills laboratory training is the transferability of the basic skills acquisition to real surgical performance.25 Surgeons tend to move slower, make more curved movements and use more grip force during human surgery. During robotic training it is possible to record objective measures of the robotic instruments. These parameters can be used to describe aspects of robotic surgical performance.26 In addition, using real-time augmented visual feedback during training can enhance the actual surgical performance.27–29
Suzuki et al.30,31 developed a tele-surgery simulation training system for cholecystectomy, which consists of a soft tissue model that reflects a person’s anatomy and an operation console using an internet connection. The authors aim to apply the system to other surgical procedures in the future.
Virtual reality (VR) training could play an important role in training and learning robotic surgery.32 Since 2006, several, mainly small, studies with respect to VR systems for robotic surgery were published. Studies that address the validation or the learning capacity of the different simulators are summarised in Table 2. Depending on budget and training purpose, several simulators are commercially available, all yet to be validated.
Table 2. Training using virtual reality
|Halvorsen et al.41||2006||SEP-robot||26 medical students||Basic suturing skills (inanimate model)||Difference in number of stitches placed between pre-test and post-test||No||Yes|
|Fiedler et al.36||2007||VR environ vs dVSS||5 experienced users||2 standard tasks: bimanual carrying, needle passing||Time, instrument tip distance and speed, range of motion (elbow)||No||Yes|
|Balasundaram et al.42||2008||SEP-robot||10 surgical novices |
|5 suturing tasks with increasing difficulty||Time, pathlength, errors||Construct||No|
|Brown-Clerk et al.33||2008||VR environ vs dVSS||9 subjects with no or limited prior dVSS experience||1 standard advanced surgical task: mesh alignment||Task completion time, distance travelled, instrument tip speed, range of motion (wrist and elbow), EMG||No||Yes|
|Kenney et al.47||2008||dVT||7 experts |
|2 endowrist modules and 2 needle-driving modules||Post-task survey, time, maximal force, maximal strain, instrument motion, instrument collisions, time instrument out of view, number of targets reached or missed, total score percentage||Face, content, construct||No|
|Lendvay et al.48||2008||dVT vs dVSS||15 course enrollees in urology with varying experience||1 ring transfer module||Pre-task and post-task survey, time for placing first three rings on floor peg, economy of motion, peak ring strain, instrument collisions, time instruments out of view, time master tele-manipulators out of centre||Face, content, construct||No|
|Katsavelis et al.37||2009||VR-environ vs dVSS||8 novices||2 standard tasks: bimanual carrying, needle passing||Task completion time, travelling distance and speed of instrument tips, range of motion (wrist and elbow), EMG||No||Yes|
|Lin et al.40||2009||SEP-robot vs (ProMIS or SurgicalSIM)||14 experts |
|Suture-knot tying||Post-task survey, time, path length, smoothness, errors||Face, construct||No|
|Mukherjee et al.34||2009||VR-environ||10 students||3 standard exercises||Instrument collisions, time instruments out of view, time master tele-manipulators out of centre||No||Yes|
|Seixas-Mikelus et al.45||2009||RoSS||15 experts |
|2 standard exercises||Post-task survey||Face||No|
|Sethi et al.50||2009||dVT||5 experts |
|3 standard exercises||Post-task survey, time, time instruments out of view, TLX||Face, content, construct||No|
|Lerner et al.51||2010||dVT vs dVSS||11 minimal robotic experienced subjects |
|5 standard exercises||Time, economy of motion, instrument collision, time instruments out of view, peak instrument force, number of targets reached or missed||No||Yes|
|van der Meijden et al.43||2010||SEP-robot||7 MIS experts |
9 MIS novices
|1 suturing task||Pre-task and post-task survey, time, path length||Face*, construct*||No|
Surgical skills training in a virtual environment had a significant learning effect and the learned skills are consistent with and transferable to actual robot-assisted procedures.33–35 However, further research is needed to develop this as an effective and reliable VR environment.36,37 Sun et al.38 developed a prototype of a computer-based simulator to practice simple surgical skills and port placement.
The SEP Robot™ simulator (SimSurgery AS, Oslo, Norway) is part of a conventional VR trainer for laparoscopy, which can be converted into a simulator for robotic surgery.39 Direct comparison of both trainer modalities showed no significant difference for a standardised suturing task.40 Training of a robotic suturing skill on this simulator equalled training on a mechanical simulator41 and practice sessions improved the technical performance of novices.42 Concepts of face validity and construct validity for this simulator seem to be present.43
Recently two more advanced simulators were introduced. The Robotic Surgical Simulator (RoSS) (Figure S2)44 demonstrated face validity.45 The development of procedural tasks for the robotic prostatectomy and robotic hysterectomy are ongoing. The dV-Trainer™ (dVT) is a simulator that uses the same kinematics as the da Vinci® Surgical System (dVSS) (Figure S3).46 During the development phase of this system several validation studies demonstrated face, content and construct validity.47–50 Training on the dVT improved performance on the robot system equal to training with the robot itself.51 The software of the dVT is suitable to use within the actual robotic console, allowing virtual tasks to be performed in a real-life environment.
Animal and human cadaver training
Animal and cadaver simulation models have the advantage of simulating the human anatomy and these models can be used for procedural training. This kind of training was considered one of the most important components of a robotic training programme52 and has hence been incorporated in several courses.53–55 These courses seem to enable participants to successfully incorporate robotic assisted surgery and maintain this technique in clinical practice in the short term and long term.52,56 Although operating on animal models is almost similar to operating on people, it is expensive and there are ethical concerns (in some countries it is banned).57 However, compared with performance assessment on human cadaver models, assessment in the animal laboratory is usual easier to schedule, cheaper and more reproducible. An animal laboratory requires a separate robotic system, which will raise costs substantially.
Live case observation
Evaluation of a robotic training programme demonstrated that operating room observation is an important component.52 There are several ways to implement operating room observation in a training programme. Watching live surgery and actually being present in the operating room or to watch the surgery in another room with the possibility to communicate with the surgeon gives a real-life experience. To watch a video registration of an operation together with a teacher is another option. The video recording has the advantage that illustrative surgeries are selected in advance and the educational moments can be planned ahead.58
Proctoring, that is providing direct supervision of an expert, takes place in the initial phase of a learning curve and the proctor is responsible for the assessment of skills and knowledge of the trainee. A review regarding proctoring, underlines the importance for robotic surgery and institutional credentialling, and addresses the medico-legal aspects. Although extended proctorship is an expensive way of training, it provides a relatively safe way to introduce a new technique and prevents surgeons from beginning to perform procedures before they have mastered the technique.59
There are different ways of proctoring. Usually the proctor will visit the hospital of the trainee, and surgery is performed together, giving the trainee more responsibilities depending on his or her skills. Sometimes the trainee visits the proctor first to view a number of cases. Proctoring is a very time-consuming and expensive way of teaching, so it is interesting to look at alternatives. Modern communication technology, tele-mentoring and tele-procotoring will save time and travelling. Alternatively, a trainee can make a video recording of the performed procedure and send it to a proctor; the evaluation can then be carried out by watching the video online together. Surprisingly, after a 5-day intensive robotic course only 37.5% of the attendees used the possibility of proctoring, even at no extra cost. This could be because most of the trainees attended as a team and on returning to their hospital they performed surgery together.52
Mentoring during actual performance of a robotic operation can be carried out in several ways. First, the mentor can observe the trainee closely while performing an operation and give verbal instruction and take over the operation when necessary. Another possibility is using the availability of the mentoring console (Figure S4). This is a second console, which facilitates the surgeon to collaborate with the trainee during surgery. The mentoring console has two collaborative modes: the ‘swap’ mode, which allows the mentor and the trainee to operate simultaneously and actively swap control of the robot arms, and the ‘nudge’ mode, which allows them both to have control over two robot arms. The ‘nudge mode’ seems to be particularly useful for guiding the trainee’s hands during some steps of an operation.60 There is also the possibility for the trainee to sit at the mentoring console and passively follow the motions of the telemanipulators of the instructor (haptic learning).61
Preliminary studies show that the teaching possibility of drawing lines over a moving picture (telestration) also seems possible in robotic surgery and does not negatively impact performance.62
Previous video game experience shortens the time to learn laparoscopic skills on a simulator.63 This association was, however, not found for robotic surgery and previous extensive video game experience was even inversely correlated with the ability to learn robotic suturing. This could be explained by the fact that the robotic system transforms intuitive three-dimensional (3D) hand motions into 3D hand movements on a 3D screen. Students who had significant experience in activities requiring intuitive hand movements (athletics and musical instruments) performed better in the robotic tasks.64 Hagen et al.65 confirmed this and showed that there is no correlation between robotic performance and logical thinking, 3D understanding or general dexterity.
Assessment of skills training
There are several options to assess performance of the trainee during skills training. Objective assessments can be carried out while using a validated rating scale. Most commonly used is the Objective Structured Assessment of Technical Skills (OSATS).66 Several authors used this method to assess robotic training.55,67 There is no robotic specific rating scale available yet. The Application Programming Interface (API) included in the robotic system can provide time and motion analysis of the robot.19 These parameters showed an objective and accurate correlation between surgical performance and OSATS score.67 This feature could be used for further development and validation of training exercises. A nonvalidated grading system for different steps of a robotic procedure was used to objectively demonstrate progression of the trainee, allowing them to proceed to the next step of the procedure.68
The ‘learning curve’ refers to the amount of surgical procedures performed before a surgeon reaches an accepted plateau in outcome parameters (operating time, blood loss, complication rate, quality of surgery). More complex procedures have a relatively long learning curve. The length of a learning curve may also vary as a result of surgeon-related factors (surgical experience with a similar technology, familiarity with the procedure) or hospital-related factors (availability of theatre time, available case load). Many series of robot assisted laparoscopic procedures have been reported upon, but only a minority of them addresses the aspect of the learning curve.
With respect to the outcome parameter ‘operating time’, there are the different phases of the operation. First, there is the aspect of time needed for the operating team to prepare and activate the robot system (‘setup time’). Second, there is the time phase relating to positioning and installing the robot (‘docking time’). Third, one can differentiate the actual time needed to complete the robotic surgery procedure (‘console time’). Fourth, there is the whole time span in which the person is in the theatre (‘theatre time’). Setup time and docking time can be reduced quickly, when working in a high-volume setting with a dedicated team.69,70 Intraoperatively outcome parameters are blood loss, complication rate, and the conversion rate to open surgery. For the quality of oncological surgery, parameters like ‘number of lymph nodes’, ‘tumour-free margins’ and ‘recurrence rate’ are known to be used. Instead of the learning curve, Sammon et al.69 suggest using the ‘learning rate’, which is defined as the percentage decrease in operative time (minutes) per doubling of cumulative procedure number.
The learning curve for benign gynaecological procedures (mainly hysterectomy) in robotic surgery is considered to be around approximately 50 cases.71 A significant improvement in operating time after 20 cases was found for the next 20 cases of hysterectomy and myomectomy.72 For sacrocolpopexy only small series were published. A 25% reduction of operative time was found after ten cases.73 Several authors address the learning curve for gynaecological oncology procedures. Proficiency for performing the robotic hysterectomy with combined pelvic–aortic lymph nodes dissection can be achieved in approximately 20 cases, but there is a continued gradual improvement in operative time from 50 to 70 cases.70 Similar results were obtained for the radical hysterectomy including pelvic lymphadenectomy.2,74,75 However, these series do not address all parameters regarding the quality of oncological surgery, such as recurrence rate. To achieve proficiency in these parameters the learning curve will probably be longer (like in urology) and still has to be established.
Until recently, it was thought that surgeons become proficient for robot-assisted laparoscopic radical prostatectomy (RALP) within 40 cases.76 Samadi et al.77 introduced an expert level after achieving proficiency, which is considered to be reached when operating outcome parameters no longer improve. The learning curve to attain this level is expected to be longer and will be in the order of 70 cases. The full procedural learning curve also includes patient outcome parameters for quality of surgery (positive margins and recurrence rate). When using these quality parameters much longer learning curves are recently described.78 For positive margins, proficiency was not reached within 100 cases79 and to achieve oncological outcomes comparable to that of an experienced open surgeon, the learning curve is expected to be 250–400 cases.80,81 In a given hospital setting, the transition from open radical prostatectomy to RALP can be made without affecting oncological patient outcomes.82 Doumerc et al.83 found a flattening of the learning curve after 140 cases and a flattening for larger tumours after 170 cases. It is most likely that for other procedures the same principles can be taken into account. Surgeons should consider whether they can build enough experience to minimise suboptimal oncological outcomes, before embarking on or continuing a robotic programme.78
Until today, most studies in general surgery have been case series and only a few studies address the issue of a learning curve.3 These studies present similar results before achieving proficiency in operating parameters.84 Larger series are necessary to address a proper learning curve and the oncological quality parameters.
Training future surgeons
With the increasing popularity of robotic surgery there is a growing need for sophisticated training programmes for residents, fellows and surgeons. Ideally these training programmes should be competence based. Courses are commonly used to share new information and/or learn new skills. Some of them are pure didactic and other mainly consist of skills training, but many of them combine the two aspects. Several authors addressed the issue of training residents and fellows, and described their training programme (Table 3). In contrast to open surgery, robotic skills can improve significantly in a relatively short time.53,55
Table 3. Training approaches for robotic surgery
|Badani et al.98||2006||Urol||Trainee||NA||RALP||NA||Yes||Skills lab||15–20||Stepwise + proctoring|
|Rashid et al.91||2006||Urol||Resident||2||RALP||83||Intuitive course||Skills lab + animal lab||12||Stepwise (5)|
|Schroeck et al.96||2007||Urol||Resident + fellow||5||RALP||383||Yes||ND||10||Stepwise (3)|
|Thiel et al.90||2008||Urol||Resident||?||RALP||50||No||Yes||ND||Stepwise (9)|
|Yoshioka et al.118||2008||Urol||Urologist||2||RALP||14||Intuitive course + image training||Skills lab||ND||Mentorship after 12 cases|
|Hoekstra et al.93||2009||Gyn||Fellow||?||Gyn onc||75||No course, during bedside assistance||Yes||ND||Stepwise (5)|
|Lee et al.94||2009||Gyn||Fellow||2||Gyn onc||21||Yes||Skills lab + animal lab||ND||Stepwise (6)|
|Mirheydar et al.95||2009||Urol||Urologist + fellow||6 + 3||RALP||Phase 2. 5–6* (total each) |
Phase 3. 32** (mean each)
|Intuitive course||Animal lab||5–6||Stepwise (8) extended mentorship|
|Davis et al.68||2009||Urol||Resident + fellow||4 + 3||RALP||124||ND||ND||ND||Stepwise (11)|
|Schachner et al.109||2009||Cardiac surg||Fellow||2||TECEB||44||ND||Skills lab + animal lab||ND||Stepwise (6)|
|Link et al.97||2009||Urol||Fellow||?||RALP||1833||Two separate intuitive courses||Skills lab||25||Stepwise (6)|
|Finan et al.88||2010||Gyn||Resident||16||Hyst||190||Yes||Skills lab||ND||Stepwise (5)***|
With the approval of the dVSS by the Food and Drug Administration (FDA), the manufacturer was told to provide a comprehensive robotic training for all surgeons and their teams. The manufacturer has currently engaged 24 training centres located all over the world.85 This training comprises two parts: on-site training, which highlights the key features of the system, preparation and management in your own hospital, and off-site training, which consists of a course to learn and practice procedural skills. From thereon, support of surgical proctoring in the first cases is provided. It is important to start quickly with regular scheduled cases after completion of a course, because otherwise the newly learned skills can fade away. At least one or two cases a week is recommended to overcome the first part of the learning curve. In addition to the registered training centres there are centres that developed their own training programme and in this way function as a training centre. Mostly these centres focus on specific procedures.
With the expansion of robotic surgery, training for residents during their specialty training has become more and more an issue. In 2006 only 34% of the urology residents thought they would perform robotic surgery after residency.86 A recent survey among residents in obstetrics and gynaecology showed that 79% of the residents feel robotic training should be included in their residency programme and 67% feel their training is not adequate.87
A preliminary programme for residents to train basic robotic skills consisted of a tutorial on the use of the robot and was followed by structured skills laboratory training.22 A structured curriculum, to acquire the knowledge and skills for robot-assisted hysterectomy is described in detail by Finan et al.88,89 The curriculum is divided into familiarisation with the robot itself and console components. The actual procedure is broken down into multiple segments that can be mimicked in a skills laboratory. Five exercises were developed: 1) dexterity, 2) bladder flap development, 3) seal and cut the ligaments, 4) skeletonising the vessels, and 5) suturing the vaginal cuff. During the training period, the residents were bedside assistant and observed a number of robotic surgeries. After completing the program, the residents started stepwise with surgery and no training-related patient complications were noted. Some residents could not complete the programme because of poor eye–hand coordination, these were not allowed to progress to surgery on people.
When using a systematic approach, urology residents can safely and effectively learn a complex procedure such as RALP.90 It is possible to divide a procedure into steps with increasing difficulty and only proceed to the next step when proficiency has been reached. Performance can be rated using an analogue scale and every step can be recorded and reviewed with the trainee. This system based on appropriate supervision, graduate responsibility, real time feedback and objective measurement could also be used for other procedures.91
An increasing number of gynaecological oncology units incorporate robotic surgery in their practice. A survey among fellows and fellowship directors in gynaecological oncology showed that 95% of the responded training centres own a robot and are using it.92 However, there is no standardised training curriculum for fellows and official guidelines for education of this new technology are lacking.
Hoekstra et al.93 describe the transition period of a gynaecological oncology fellowship programme into a robotic programme. They underline the importance of commitment of the whole department during such a transition. The fellows started with limited observation, animal training, video observation and port-side assistant training, followed by operating at the console steps of the procedure with increasing difficulty. For the steps vaginal cuff closure, hysterectomy and pelvic lymph node dissection proficiency can be reached after each five to ten cases. For the para-aortic lymph node dissection approximately ten more cases were required to reach proficiency. A systematic approach was described by Lee et al.;94 their fellow-training programme included 1) didactic and hands-on training with the robotic system, 2) instructional videos, 3) assistance at the operating table, and 4) performance of segments of several gynaecological procedures. The conclusion from this study was that the introduction with a systematic approach for training in robotic surgery is feasible.
In urology, an extended proctorship programme reported a high take-rate and training satisfaction.95 A guideline of what to look for in a trainee’s progression during the separate steps of the operation can be helpful. Although, training requires more operating time, it does not appear to diminish patient outcome and seems possible at a high-volume centre.68 In contrast to these findings, other authors state that the implementation of a training programme will not cost extra operative time.96 All authors reported that fellows can be trained in a complex procedure with no significant adverse impact on patient clinical outcome.68,96,97 However, a structured and systematic approach to learn robotics in a safe and effective way is paramount.98
Costs of training
Robotic surgery is still expensive and several authors addressed the aspect of costs1,2,99–101 or compared the costs of the robotic procedure with laparoscopic or open surgical procedures.102–104 Most studies mainly focus on the costs of the robotic system (±$1,800,000) with the additional 10% per year of fixed service costs and instrument costs (±$700 to $1000 per case).
Rarely addressed are the costs of the learning curve of the surgeon and the surgical team. These are substantial costs that are often underestimated. Steinberg et al.101 constructed a theoretical model to describe the cost associated with the learning curve of a single surgeon for RALP. In this model fixed costs for operating room time and anaesthesia services were used. The model was then applied to several learning curves described in the literature. The learning curves ranged from 13 to 200 cases. Costs associated with the least expensive learning curve were $49 613 and with the most expensive learning curve were $554 966. The average learning curve was 77 cases at a price of $217 034. Because of the diversity of the studies it is difficult to determine whether their conclusions are best applied to a single surgeon, a group practice or a hospital setting. This study illustrates the high costs that are involved with the learning curve for complex robotic procedures and underlines the need for sophisticated training programmes together with a high case load to overcome the learning curve.
Curriculum design and implementation
Pioneers in robotic surgery provided an overview of robotic surgical training, identifying objective-based curriculum levels for system training and specific training for advanced procedures.105,106 The importance of team training has been stressed and it was noted that training of the whole surgical team could be one of the highest barriers.106 It is recommended to start with a cohesive and dedicated operating room team and two surgeons, which need to work closely together during the initial learning curve.52 This team-based approach can reduce the learning curve of robotic assisted surgery.107,108 Only when the initial team has achieved proficiency can new surgeons be introduced and a general strategy should be to avoid multiple learning curves running parallel.109 The team should start with relatively easy cases, to be able to familiarise themselves with patient positioning, the setup of the robot, trocar and robot positioning, and the robotic instruments.110,111 When these issues are mastered, the team can move on to more complex cases. A large case volume is required to maintain skills, not only for the console surgeon but for bedside assistant as well.98 A well-trained bedside assistant is just as important as the console surgeon and requires a combination of open, laparoscopic and robotic skills.112 Slow adoption of robotics and prolonged operative times may result from insufficient trained bedside assistants.113
The operating room nurses, anaesthesiology nurses and the anaesthetist also play an important role in robotic surgery and should be well trained before starting a robotic programme. There are many specific perioperative competencies that a scrub nurse or a circulating nurse must know.114 Training too many nurses at once usually results in inadequate exposure to robot procedures and will delay the success of the programme.
Guzzo et al.115 identified three essential phases in a structured robotic training programme: first the preclinical phase, second the bedside assistant phase, and last the operative console phase. In the preclinical phase, trainees should become familiar with the robotic system and learn basic skills through inanimate simulation models. This can be achieved in the course of several days consisting of didactic and skills training, which should be organised in a structured and competence-based way.116 In the bedside assistant phase, the trainee functions as a co-surgeon, learning trocar and robot placement, instrumentation, troubleshooting and will progressively learn the different steps of the operation. In the console phase, the trainee starts performing parts of the robotic operation. In the ideal situation the operation is divided into multiple steps with increasing difficulty. Optimal proficiency of each step is objectively graded.68 This structured approach allows optimal feedback from mentor to trainee. It is recommended to videotape the procedures, which allows proper reviewing of procedural steps afterwards.
From the urological literature we know a structured training programme for RALP will take 55–80 cases to train a future robotic surgeon. First, observing 10–20 cases before starting to participate as a bedside assistant.98 Participation as a bedside assistant in 10–25 cases before starting with segments as a console surgeon,77,91,97,98 As a console surgeon it will take 20–30 cases before a whole procedure can be performed.98 From there, approximately ten cases should be performed under the direct supervision of an experienced robotic surgeon.98 To maintain credentialling, a minimum of 20 cases a year is recommended.117 Regular self-assessment of performed cases and complications will help to shorten the learning curve and will provide insight into the quality of surgery. It is most likely that these issues will be mandatory for future credentialling.