Presentation Authors: Saum Ghodoussipour*, Jian Chen, Jessica Nguyen, Sharath Reddy, Aliasger Shakir, Rut Patel, Inderbir Gill, Andrew Hung, Los Angeles, CA
Introduction: Automated performance metrics (APMs) provide a novel approach to the assessment of surgical skills. Herein, we present an initial construct validation of APMs during robotic partial nephrectomy (RPN). We also seek to determine if APMs correlate to clinical outcomes after RPN.
Methods: We recorded APMs (instrument motion tracking and system events data) and synchronized surgical videos from da Vinci Si systems during RPN using a system data recorder. Each case was segmented into 7 steps: colon mobilization (CM), ureteral identification/dissection (UD), hilar dissection (HD), exposure of tumor within Gerotas fascia (GF), intraoperative US/scoring of tumor (US), excision of tumor (EX) and rhenorrhaphy (R). APMs from each step were compared between expert (E) surgeons (>100 cases) and novice (N) ( < 100 cases). Clinical outcomes were also collected prospectively and correlated to APMs.
Results: We evaluated 39 RPN cases performed by 7 E and 10 N surgeons.During CM, E had a shorter task duration and more efficient robotic instrument usage (shorter duration of movement, shorter path lengths). During UD, E had shorter task duration, greater efficiency and more wrist articulation. During HD, E had a shorter task duration, used the clutch less and had a greater ratio of dominant to non-dominant instrument use vs N. E had greater wrist articulation than N during GF and were more efficient during US, (all p < 0.05). _x000D_
We did not compare APMs amongst groups for EX and R as these steps were largely performed by E. However, we did find significant correlation between APMs and warm ischemia time (WIT) and estimated blood loss (EBL) during these two steps (Table 1): EBL was correlated to task completion time (p < 0.001, r=0.55), instrument efficiency and dominant to non-dominant instrument usage ratio (p=0.01, r=0.56). WIT was correlated to several APMS, including instrument efficiency and task duration (p < 0.001, r=0.75). Nephrometry score did not correlate with either WIT or EBL.
Conclusions: Experts are more efficient and directed in their movement during RPN. APMs during key steps may serve as better predictors of clinical outcomes than other clinical measures such as tumor complexity. These data help establish a standardized metric for surgeon assessment and training during RPN.
Source of Funding: This study was funded in part by an Intuitive Surgical Clinical Grant; Intuitive Surgical provided the systems data recorder.