Initial Experience With Robotic Liver Resection in the United States

Abstract

Background This study’s aim was to show the feasibility and safety of robotic liver resection (RLR) even without extensive experience in major laparoscopic liver resection (LLR). Methods A single center, retrospective analysis was performed for consecutive liver resections for solid liver tumors from 2014 to 2022. Results The analysis included 226 liver resections, comprising 127 (56.2%) open surgeries, 28 (12.4%) LLR, and 71 (31.4%) RLR. The rate of RLR increased and that of LLR decreased over time. In a comparison between propensity score matching-selected open liver resection and RLR (41:41), RLR had significantly less blood loss (384 ± 413 vs 649 ± 646 mL, P = .030) and shorter hospital stay (4.4 ± 3.0 vs 6.4 ± 3.7 days, P = .010), as well as comparable operative time (289 ± 123 vs 290 ± 132 mins, P = .954). A comparison between LLR and RLR showed comparable perioperative outcomes, even with more surgeries with higher difficulty score included in RLR (5.2 ± 2.7 vs 4.3 ± 2.5, P = .147). The analysis of the learning curve in RLR demonstrated that blood loss, conversion rate, and complication rate consistently improved over time, with the case number required to achieve the learning curve appearing to be 60 cases. Conclusions The findings suggest that RLR is a feasible, safe, and acceptable platform for liver resection, and that the safe implementation and dissemination of RLR can be achieved without solid experience of LLR.