The time to achieve mastery in basic and advanced robotic tasks in a simulator setting has been estimated at 10 h for junior and senior surgery residents, respectively . However, residents from institutions where structured curricula are not mandatory may fail to complete robotic surgery training, citing barriers such as lack of time, lack of access to robot or simulators, and lack of motivation, in contrary to overwhelming initial interest to gain fundamental robotic skills . Laparoscopy education, notably the Fundamentals of Laparoscopy Surgery (FLS) curriculum, is more accessible to surgery residents, but evidence is lacking as to whether these skills transfer to robot-assisted surgery.
Learning a basic laparoscopy task, such as the FLS peg transfer, follows an inverse learning curve, with 90% of gains occurring in the first 5.8 ± 2.3 trials for a group of 16 medical students . In another study of 16 new surgical residents, peg transfer task times of 201 ± 84 s were measured prior to FLS training . If 90% of gains occur in the first 5.8 ± 2.3 repetitions for this task, then a large effect size should be measurable after a training time of 20 min, which should allow for six repetitions of this task.
We selected the ball placement and rope passing tasks for our study because they satisfied the same psychomotor criteria as the FLS peg transfer task, namely depth perception, visual-spatial perception in a 2D setting, and the coordination of dominant and non-dominant hands, which are important skills for suturing and needle positioning . The maximum time limit allowed for this foundation level task during FLS curriculum testing is 300 s, with a proficiency benchmark of 48 s . For comparison, the robot-assisted version of the ball placement and rope passing tasks used in our study was 600 s, with measured proficiency benchmarks of 43.3 ± 3.1 and 40.0 ± 7.5 s, respectively.
We hypothesized that an effect size would be measurable if laparoscopy learning had occurred, and if these skills then transferred to the robot-assisted digital laparoscopy platform. However, we found no statistically significant differences in objective measures, such as time spent on tasks, completion rate, clutch use, out of view instruments, errors, and manual adjustments.
We also measured subjective differences related to the performance of a task which may not be reflected in objective measures, such as when two or more groups achieve similar performance times despite one group that finds the task to be significantly more difficult than the other groups. NASA-TLX workload was greater in laparoscopy-exposed groups in our study, but this result was not statistically significant. This observation could be due to ergonomic differences between laparoscopy and robotic surgery .
Altogether, our study results suggest that laparoscopy exposure, in the form of limited psychomotor skills training, does not affect initial robot-assisted surgery performance among learners. This study supports the idea that training in robotic surgery ought to take place in a robot-assisted simulation environment.
With respect to study limitations, there was self-selection bias in our study because participants self-enrolled and self-assigned into their own groups. The number of participants in our study was also small, affecting the precision of our results. Practical limitations with respect to operating room time, student availability, surgeon availability, robot availability, and experiment costs make it difficult to organize larger and more comprehensive studies.
Although randomization and blinding could be achieved by performing this study on a virtual reality simulator, simulators are not yet available for the Senhance Surgical System. Simulation is also not the same as real-world surgery in an operating room setting, where technical difficulties and instrument errors can occur, as demonstrated by our study. Additionally, the presence of an attending surgeon could have influenced our experiment, but trainees at most institutions (including ours) are not allowed to operate without direct supervision. Interruptions are stressful and can influence the subjective experience of users, which our study aimed to capture. These arguments support and strengthen the validity of our study.
Several studies have previously reported faster learning curves and improved retention of skills with robotic assistance as compared to laparoscopy [22,23,24,25]. With respect to basic manipulation tasks, improved task speeds with robot assistance have been measured as compared to laparoscopy, but with minimal transfer effects . Among studies looking at skills transfer, one study compared novices completing a ball drop task with laparoscopy or the RoSS simulator, and found that while both groups improved after training, the degree of improvement was equal, indicating there is no skills transfer . Other studies have argued that skills transfer effects from laparoscopy to robotic surgery may be more pronounced with difficult tasks, such as suturing [22, 23]. In our view, laparoscopy and robotic surgery are different domains, perhaps requiring different skills. Previously, novice users have demonstrated rapid adaptation to the Senhance device, regardless of experience level . Validated robotic surgery curricula, such as the Fundamentals of Robotic Surgery (FRS), offer a direct and efficient path for novices to become proficient in robotic skills without embarking on the intermediate step of laparoscopy .
Individual performance was highly variable, and it would have been worthwhile to ask participants about their backgrounds and career goals to investigate factors differentiating high and low performers. One study has found evidence to suggest that objective innate ability may distinguish students interested in surgical careers from others . However, it must be remembered that innate or initial ability may not correlate with the rate of improvement of robotic skills, which requires practice.