Robotic major hepatectomy: Is there a learning curve?

Background. Robotic hepatectomy has been suggested as a safe and effective management of liver disease. However, no large case series have documented the learning curve for robotic major hepatectomy. Method. We conducted a retrospective study for robotic major hepatectomy performed by the same operative team between January 2012 and October 2015 and evaluated the learning curve for operation time using the cumulative sum method, presented as cumulative sum operation time. Results. Overall, there were 183 robotic hepatectomies, 92 of which were performed in patients who underwent robotic major hepatectomy: left hemihepatectomy was performed in 32 (34.8 %) patients, right hemihepatectomy in 41 (44.6%), left trisectionectomy in 3 (3.3 %), right trisectionectomy in 6 (6.5 %), and 8-5-4 trisegmentectomy in 10 (10.8%). The median duration of surgery was 434 minutes (142-805 minutes) and the median blood loss was 195 mL (50-2,000 mL). Fifty-nine percent of patients had malignancies, and those with advanced stages of cancer had more blood loss during an operation. The cumulative sum operation time model of robotic major hepatectomy suggested that the learning curve comprised 3 characteristic phases: initial (phase 1, 15 patients), intermediate (phase 2, 25 patients), and mature (phase 3, 52 patients). The learning effects were underlined by shorter operation time and hospital stay after phase 1 and less blood loss after phase 2. Conclusion. This is the largest series regarding robotic major hepatectomy. Our findings suggest that a solid training program based on the learning curve should be considered for beginners of robotic hepatectomy. Participants should evaluate the evolution of our minimally invasive hepatectomy before considering our robotic experience.