Related Experiment Video
Updated: Sep 27, 2026
Surgical Robot-Assisted Transanal Specimen Extraction Radical Sigmoidectomy Without an Auxiliary Abdominal Incision
Published on: June 13, 2025
Surgeon- and procedure-specific variability in operative-time learning curves during parallel implementation of
Zsolt Madarasz1, Kira Baginski2, Annika Hoyer2
1Department of Surgery, Bielefeld University, Medical School and University Medical Center OWL, Detmold, Germany.
Background:
Robotic colorectal surgery is associated with learning trajectories that may vary according to procedural complexity and individual surgeon experience. Although robotic colorectal learning curves have been extensively investigated in single-surgeon series, limited data exist regarding inter-surgeon variability during the implementation of multiple robotic colorectal procedures within the same institution. This study evaluated surgeon- and procedure-specific learning curves during the progressive institutional implementation of a robotic colorectal surgery program.
Methods:
We reviewed 369 consecutive patients who underwent elective robotic colorectal resections between 2018 and 2025. Operative performance of three primary console surgeons was analyzed for robotic right colectomy (RRC) with complete mesocolic excision (CME), robotic anterior resection (RAR), and robotic low anterior resection (RLAR). Learning curves were assessed using cumulative sum (CUSUM) analysis of skin-to-skin operative times. Risk-adjusted CUSUM (RA-CUSUM) analysis was used to account for selected patient-related factors potentially influencing operative time.
Results:
CUSUM analysis demonstrated substantial surgeon- and procedure-specific variability in operative learning trajectories. RRC generally demonstrated earlier operative-time stabilization, with CUSUM turning points ranging from 8 to 18 procedures. RLAR demonstrated the greatest variability in operative learning trajectories and the highest operative times, with CUSUM turning points ranging from 16 to 27 procedures, whereas RAR showed intermediate learning trajectories. Despite heterogeneous CUSUM profiles between surgeons and procedures, overall perioperative and oncological outcomes were acceptable across the study cohort.
Conclusion:
Operative learning trajectories in robotic colorectal surgery demonstrate considerable variability according to both surgeon and procedure, even within a standardized institutional robotic program. While RRC generally showed earlier operative-time stabilization, RLAR was characterized by higher operative times and greater variability in operative learning trajectories. These findings suggest that operative learning trajectories vary substantially between surgeons and procedures and should not be interpreted using fixed case-number thresholds alone.
