RESUMEN
BACKGROUND: This pioneering study is aimed to design training models for robotic pancreaticojejunostomy (PJ) and to assess their usefulness using quality improvement exercise in the dry lab. METHODS: Three dry lab models were developed including the anastomosis model of a transected silicon pancreatic stent (model 1), a rough model (model 2) simulating PJ, and an advanced three-dimensional printed model (model 3) more vividly simulating PJ. Three surgeons (A, B, C) with same specialty and levels of expertise in surgery were enrolled in the training which was divided into three rounds of tasks. In the first round, all three surgeons (A, B, C) participated in the training on basic technical tasks before moved on to the next rounds. While surgeons A, B participated in the second round on model 1, only surgeon A worked on model 2 in the third round. Their proficiency of performance was evaluated on model 3. RESULTS: The results of the first and second rounds between surgeons are similar. Surgeon A practiced with model 2 for 6 h, completing 10 cases. In model 3, the times of attempts before achieving a consecutively three times of satisfactory anastomosis procedures were compared, for surgeon A, six cases, 20 for B and 25 for C. CONCLUSIONS: The specifically designed series of dry lab training models may be a potential training tool for advancing the robotic PJ through quality improvement exercise in dry lab. Further larger and well-designed studies are warranted to validate this issue.