Laparoscopic surgery (LS) is the standard technique for an increasing number of operations. Some of the major benefits compared to open surgery include: minimized risk of infection, reduced postoperative pain, shorter rehabilitation time, and better cosmetic results. In practice, LS requires a very different set of both technical and non-technical skills when compared to open surgery, designed to address the absence of 3D vision, the use of long and thin instruments, the fulcrum effect, and limited force feedback [1-3].

Realistic anatomical models and human cadavers have been employed by surgeons to practice the various psychomotor and visuospatial skills required to perform LS procedures. However, these models need frequent replacement and have high maintenance costs [4]. Over the last decade, simulation-based surgical education has received significant attention, and a number of laparoscopic simulators have been employed [5]. Box trainers (BTs) and virtual reality (VR) simulators are the two main training modalities employed in the hospitals and clinical training centers for laparoscopic skills acquisition and development. A number of studies have demonstrated that training with such devices improves the required laparoscopic skills and that these skills are transferable to the operating room [6-8]. High-fidelity models with life-like patient anatomy are employed for the development of special psychomotor skills outside the operating theater [9]. Trainees are able to develop surgical skills and become familiar with a particular procedure in a surgical laboratory before operating on a patient [10-11]. However, there remain some questions about the superiority of laparoscopic VR simulators over BTs with regard to the level of training and the skills achieved, mainly due to the different sense of force-feedback, the difference in the image quality, the behavior of the tissue models, as well as the dissimilarity of the tasks performed with the two modalities [12-13].

A significant amount of research has been published evaluating simulation-based training for general surgery procedures and gynecology. Gor et al. suggested that the MIST-VR simulator provides objective assessment of the laparoscopic skills of gynecologists [14]. Aggarwal et al. demonstrated that an ectopic pregnancy scenario in a VR simulator can differentiate Ob/Gyn residents with different level of surgical experience [15]. Similarly, Larsen et al. showed that expert gynecologists outperformed intermediates and novices in time, instrument pathlength, and total errors [16]. Schreuder et al. reported that gynecologists with minimal laparoscopic experience can improve their skills during short-phase training on a VR procedural simulator [17].

Investigations of the training value of BTs is more focused on the development of appropriate anatomic models for gynecologic surgeries. For example, Scott et al. investigated the laparoscopic cost effectiveness of bench models in relation to operating room experience [18]. Zhen et al. highlighted the use of the fundamentals of laparoscopic surgery (FLS) program to assess the laparoscopic competence of gynecologists [19]. Tang et al. described the design of a phantom that allows trainees to practice key skills required for laparoscopic salpingotomy and salpingectomy [20]. A porcine small bowel was used to simulate the fallopian tube, while porcine liver and red food dye were used to simulate ectopic pregnancies inside the fallopian tube. Levine et al. described the use of a lightly embalmed human cadaver model for practicing laparoscopic techniques for various operations, including laparoscopic hysterectomy [21]. A live porcine model for teaching advanced laparoscopic skills to gynecologic oncology fellows was developed by Hoffman et al. [22]. The model was found to be suitable for laparoscopic training in various procedures such as lymphadenectomy, uretero-neocystostomy, and bowel anastomoses.

The aim of our study was to compare the training efficacy of two LS simulators and also to assess the transferability of skills acquired, during training on basic laparoscopic tasks, to the performance of more advanced gynecologic operations on a VR simulator. To achieve this goal we used two separate groups of residents and studied the impact of basic skills training using a VR simulator or a box trainer, using custom training models that represented female gynecological anatomy. Following this training, the two groups were assessed on the performance of two laparoscopic gynecologic operations on a VR simulator (salpingotomy and salpingectomy). The post-training performance was compared to that before training and also compared between the two groups. Our analysis also included investigation of the effect of gender, video gaming skills and musical instrument ability in the performance of laparoscopic operations.

The increasing popularity of minimally invasive surgery highlights the necessity to develop training programs for improvement of key psychomotor skills such as hand eye coordination. Clinical experience has shown that these skills require a significant training period in the operating theater (10 to 30 patients, depending on the operation), resulting in longer operating times, higher complication rates, higher conversion rate to open laparotomy, and subsequently higher hospital costs [26-27]. Therefore, training using laparoscopic simulators has significant advantages over training on patients, providing a safe and controlled environment without the risk of harm to patients and avoiding the stressful operating conditions with patients.

In this study we aimed to investigate the differential impact of training and skills acquisition in high-fidelity VR simulator vs. low-fidelity BT. The evaluation was conducted on the VR simulator, since there are controversies about the transferability of skills between different laparoscopic training modalities. It has been suggested that the VR simulators are able to assess the skill level of laparoscopic surgeons [28-29]. It also seems that VR simulators, when used appropriately, more closely simulate real laparoscopic procedures [30]. However, VR simulators and BTs have some fundamental inadequacies. For example, VR simulators provide worse depth perception compared to BTs, whereas in some tasks the properties of the graphical models are not so realistic. On the other hand, BTs provide no automated means for performance assessment, whereas the training models require replacement after a task performance, and maintenance.

In this study the training tasks were not identical for the two groups, as there is no consensus on which tasks to include in a basic laparoscopic training program in order to achieve the most effective skills training. A previous study has shown that training on the same tasks, but using different modalities (VR simulator and BT), resulted in almost equivalent performance [23]. In the present study the tasks on the VR simulator were common basic laparoscopic exercises, whereas the ones in the BT were chosen as more advanced: ‘cystectomy’ for ovarian cyst, and ‘salpingotomy’ for ectopic pregnancy. Our main purpose was to investigate whether practice of more advanced tasks on the BT is more effective than VR basic skills training, given that the assessment tasks are simulated surgical operations.

Our results indicated that VR training improves certain laparoscopic skills, in agreement with other studies which have shown that training on basic tasks has a significant impact in the improvement of more complex tasks [29]. In another study it was shown that VR training contributes to the enhancement of key laparoscopic skills, and that experienced surgeons scored better than residents in terms of time and instrument pathlength [25]. Iwata et al. found that basic laparoscopic tasks such as peg transfer and cutting were strong discriminators of laparoscopic experience [24]. Furthermore, Mansour et al. assessed technical and dexterity skills of surgeons by measuring various parameters during peg transfer and cutting, and they found improvement in some aspects of the laparoscopic surgical skills of the trainees [31].

Also in our study, the performance of the subjects improved significantly as a result of the BT training on the two advanced tasks. These exercises were designed to incorporate laparoscopic grasping and cutting skills, which are fundamental for performing laparoscopic ectopic pregnancy surgery. Although the primary goal of training was to increase technical performance (i.e. shorter time and fewer errors), we also observed a decrease in the dexterity variability, as demonstrated by the greater economy of movements performed (i.e. shorter instrument pathlength). This finding was also valid for the other group trained on the VR simulator.

With regard to the pre- and post-training assessments performed on the VR simulator, there was no statistically significant improvement for either of the groups in any performance metric with the laparoscopic operations. Also there was no statistically significant difference between the two groups for any metric. This result may seem surprising, since one would expect the skills acquired during training would be transferred to the performance of the (more advanced) assessment tasks, and/or that one of the two types of training provided would be superior. There are a number of potential reasons for this discrepancy. First, the assessment tasks are actual laparoscopic operations which require the combinative demonstration of various complex skills. Second, in the training sessions, each group practiced a limited number of basic skills separately (such as grasping, cutting, etc.). Thus, when the subjects attempted to perform the laparoscopic operations on the VR simulator, they could not combine the separate skills learned during training. Third, laparoscopic operations require management of additional skills beyond those practiced during basic skills training.

The identical post-training performance of the two groups suggests that VR simulators and BTs are equivalent modalities for laparoscopic basic skill training. One of the limitations of our study could be the fact that both groups were assessed before training and after training in VR simulators and that this might provide a positive bias for residents that trained with VR simulators over those using the BTs. However, we chose VR operations for pre- and post-training assessment as we were not able to assess the groups in real surgical operations. It does appear that the bias for the residents who were trained in VR simulators was small since the skills required to perform VR operations are much more advanced than those required for the VR basic tasks training. In a related study Loukas et al. performed a head-to-head comparison study which showed that both modalities provided significant enhancement in novices’ performance and that the skills learned on one device were transferable to the other [23]. Other studies have also attempted to clarify which training modality is the most effective by comparing the transferability of skills between VR simulators and BTs [2, 32]. However, the reported results are contradictory mainly because of the dissimilarity of the tasks performed on each device.

Finally, with respect to the subjects’ video gaming and music expertise, only video gaming seemed to have a positive effect in the performance of the laparoscopic operations on the VR simulator. Grantcharov et al. suggests that trainees who regularly play computer games make fewer errors and have shorter learning times [33]. Also, it has been suggested that video gamers acquire laparoscopic techniques faster, and training on video games appears to improve laparoscopic performance with VR trainers [34].

In a review article, we provided an overview of the evaluation and capacity of the different types of laparoscopic simulators including laparoscopic box trainers, laparoscopic VR simulators, animal models, human cadavers and lightly embalmed human cadavers in the development and assessment of training in laparoscopic gynecologic procedures. We suggested that the clinical training curriculum of obstetrician-gynecologists should include laparoscopic VR simulators through a simulation-based education program outside the live operative environment under appropriated supervision in order to ensure the acquisition of maximal surgical psychomotor skills and the achievement of high-quality laparoscopic training [35]. Moreover, Papanikolaou et al. suggested that teaching hospitals should introduce training programs using laparoscopic simulators with standardized and reproducible tasks in order to achieve better patient care with safety, efficiency and lower cost [36]. Furthermore, Torricelli et al. suggested that the best way for the dissemination of laparoscopic surgery in obstetrics-gynecology residents is the induction of laparoscopic simulators for a short period of training [37].

The present randomized-prospective study showed that basic skills are improved by training with either VR simulators or BTs. However, groups trained with these devices show no statistically significant improvement in laparoscopic operations in a VR simulator. We propose laparoscopic training laboratories to include VR simulators as a reasonable alternative to BTs for laparoscopic training of inexperienced residents. Compared to BTs, VR simulators provide various means for automated assessment of training performance via a rich set of assessment metrics. These parameters may then be used either by the trainees to evaluate their training progress or by the tutor to adapt the training procedure according to the needs of each individual trainee.

This study was a part of the Thesis for a Master Degree (M.Sc.) in “Minimally Invasive Surgery, Robotic Surgery and Telesurgery”, Medical School, National and Kapodistrian University of Athens from Mr. Michail Varras with tutors the Professor Mr. Nikolaos Nikiteas, and the Assistant Professor Dr. Constantinos Loukas, Medical School, National and Kapodistrian University of Athens.

The authors declare no conflict of interest.