Preliminary translational assessment of robotic surgery skills for vascular dissection: from simulator to in vivo porcine model

Robotic-assisted surgery (RAS) offers enhanced visualization, precision, and dexterity, but the absence of haptic feedback poses challenges during delicate dissection tasks such as vascular dissection. Simulation-based training has been proposed as a strategy to mitigate these limitations, yet evidence of translational effectiveness into in vivo surgical performance remains limited. We conducted a prospective, controlled feasibility study to evaluate the impact of a structured, simulator-based training program on robotic vascular dissection. Twelve novice surgeons were included in a prospective, controlled, non-randomized feasibility study. Six underwent structured dry-lab training with a sensorized high-fidelity vascular simulator, while six served as untrained controls, no baseline robotic performance assessment was performed before the intervention. Surgical performance was assessed during robotic vascular dissections in anesthetized porcine models using the da Vinci Xi platform. Performance was assessed by a single expert evaluator who was blinded to group allocation using the Global Evaluative Assessment of Robotic Skills (GEARS) and qualitative parameters including tissue handling, vessel exposure, and stapler placement. The trained group achieved significantly higher overall GEARS scores than the control group (25.7 ± 2.9 vs. 21.2 ± 2.4; p = 0.026). Depth perception was significantly improved in trained participants (4.33 ± 0.81 vs. 2.83 ± 0.75; p = 0.028). Trends toward enhanced bimanual dexterity and efficiency were observed but did not reach statistical significance. Qualitative analysis highlighted safer tissue handling, more consistent vessel exposure, and improved stapler positioning in the trained group compared with the controlgroup. Structured training with a sensorized high-fidelity vascular simulator was associated with better performance in selected components of robotic vascular dissection performance in an in vivo porcine model. These preliminary findings support the feasibility of this translational training pathway but require confirmation in larger randomized studies.