Features of laparoscopic liver resection training after mastering of robot-assisted technology

   Aim. To estimate the learning curve for laparoscopic liver resections performed by a surgeon experienced in robot-assisted liver resections using the CUSUM method.   Materials and methods. The study involved a retrospective analysis of the results of laparoscopic liver resections for malignant and benign neoplasms performed from 2015 to December 2020 and robot-assisted liver resections from 2010 to 2020. The author evaluated the learning curve for laparoscopic liver resections of a surgeon who had mastered robot-assisted resections of high difficulty. Selecting the boundaries between training periods according to the obtained CUSUM graphs was determined by critical changes in the resection difficulty score (IWATE and IMM), duration of surgery, blood loss, and incidence of postoperative complications. Major perioperative events were compared between the laparoscopic and robot-assisted resection groups in each of the training periods.   Results. 174 laparoscopic and 57 robot-assisted liver resections were performed. The duration of the first training period comprised 11 robot-assisted resections and 20 laparoscopic resections, the second period – 16 and 20, the third period – 30 and 134, accordingly. In the second period, the resection difficulty score increased significantly for both groups, while the amount of blood loss, the incidence of postoperative complications, and the duration of hospital treatment did not differ significantly. In the second training period, the duration of surgery was significantly longer in both groups.   Conclusion. Studying the dynamics of surgical difficulty using the CUSUM method is considered to be a reliable, controlled way to estimate the learning curve for liver resection. Completing the learning curve for robot-assisted liver resections reduces the duration of the learning curve for laparoscopic resections compared to published data of other authors.