Technical proficiency assessment of robotic intracorporeal single-stapling colorectal anastomosis using video-based...

This study aimed to evaluate the learning curve of robotic intracorporeal single-stapling anastomosis (RiSSA) using risk-adjusted cumulative sum (RA-CUSUM) analysis based on standardized procedural intervals.

We retrospectively analyzed 36 consecutive patients who underwent robotic left-sided colorectal resection with RiSSA by a single surgeon. Eight intraoperative timepoints were annotated from surgical videos to define two composite metrics: pure RiSSA interval and total purse-string suture time. RA-CUSUM analysis was applied to assess technical proficiency over time.

The RA-CUSUM curve showed an inflection point at case 17 for the pure RiSSA interval and at case 11 for purse-string suture time, indicating earlier acquisition of suture skills compared to overall procedural fluency. Two Clavien-Dindo grade ≥ III complications occurred in the late phase, including one anastomotic leak (5.3%, 1/19), whereas no major complications were observed during the early phase, although the limited sample size precludes any definitive safety interpretation. Technical metrics, including console time, pure RiSSA interval, and total purse-string suture duration, significantly improved after the inflection point.

In this single-surgeon cohort, RiSSA demonstrated a definable learning trajectory, with technical efficiency stabilizing after approximately 17 cases. The occurrence of major complications after the learning phase highlights that technical proficiency does not eliminate procedural risks. The pure RiSSA interval offers a reproducible metric to evaluate anastomotic proficiency and could support skill assessment frameworks in robotic colorectal procedures. Studies involving multiple surgeons and institutions are warranted to determine the generalizability of these findings.