To analyze the learning curve of transanal minimally invasive surgery (TAMIS).

A retrospective cohort study was conducted to analyze relevant clinical data of 45 patients who had undergo TAMIS by the same group of surgeons in the Sixth Affiliated Hospital of Sun Yat-sen University from July 2016 to October 2024. The learning curve of TAMIS was plotted using the cumulative sum (CUSUM) analysis method and the optimal learning cases was determined according to the peak value of the learning curve. Clinical indicators such as duration of surgery, intraoperative blood loss, positive rate of circumferential margin, length of postoperative hospital stay, and incidence of postoperative complications were compared at different stages.

Forty-five patients had been successfully undergo TAMIS. The optimum curve equation was y=0.030x³-2.856x²+71.555x-138.134, R²=0.954, P<0.05. According to the peak value of the curve, seventeen cases were determined as the minimum cumulative required cases of TAMIS for surgeons to cross the learning curve. Forty-five cases were divided into two groups: the learning improvement group (Group A) of the former 17 cases, and the proficiency group (Group B) of the latter 28 cases. Compared with Group A, Group B had shorter duration of surgery(49.6 min vs. 88.4 min, t=7.170, P<0.001), less intraoperative blood loss(5.0 mL vs. 10.0 mL, Z=-2.384, P=0.017), and shorter length of postoperative hospital stay (3.5 d vs. 6.0 d, Z=-3.006, P=0.003). There was no statistically significant difference in the observation indicators including positive rate of circumferential margin and incidence of postoperative complications between the two groups (P>0.05).

The learning curve of TAMIS can be divided into the two stages: learning improvement stage and mastery stage. Seventeen cases may be the optimal required cases to cross the learning curve of TAMIS for surgeons.