Development of an ex vivo simulated training model for laparoscopic liver resection

Abstract

Background: The number of patients who have undergone laparoscopic liver surgery has increased in the last 15 years. It is technically challenging surgery, requiring both advanced laparoscopic and liver resection skills. Surgeons often require familiarisation with much of the equipment and techniques used in this type of surgery. No ex vivo model currently exists for laparoscopic liver resection (LLR). The aim of this study was to develop a model for acquiring the technical skills involved in LLR that was also able to assess and measure surgical performance.

Methods: The ProMIS augmented reality surgical simulator was selected because performance data other than time could be obtained, and the simulator was adapted to create the laparoscopic trainer. Twenty candidates with differing laparoscopic surgical experience tested the model. Three groups were identified, novice, intermediate, and expert, according to previous exposure to the laparoscopic tasks. Candidates were required to identify a tumour ultrasonographically, mark and transect ex vivo liver, and perform two laparoscopic stitches with intracorporeal knots. The ProMIS recorded the performance data, including instrument path lengths and time.

Results: Measurements taken from the ProMIS simulator were analysed for statistical differences between the groups. Expert surgeons showed a statistically significant difference in the time taken to identify the liver lesion and transect the organ. The results also demonstrate that the more difficult tasks such as laparoscopic suturing are completed by the expert surgeons with statistically significant shorter times and path lengths compared to the less experienced surgeons.

Conclusion: The adapted ProMIS augmented reality simulator provided junior surgeons with a realistic learning environment in which to familiarise themselves with the equipment and techniques required for LLR. The model also allows assessment of the performance of individuals over time and within a peer group. Construct validity is proven for the suturing component of the model.