Objective To investigate the learning curve of laparoscopic transanal total mesorectal excision (taTME) for rectal cancer operated by one or two surgery teams.

Methods The retrospective cross‑sectional study was conducted. Based on the concept of real-world research, the clinical data of 1 458 patients undergoing laparoscopic rectal cancer taTME from 44 medical centers who were registered in the Chinese taTME registry collaborative (CTRC) database from May 2010 to May 2020 were collected. The 1 458 patients were divided into cohorts with one surgery team or two surgery teams according to the operation method. Patients with one surgery team underwent taTME by transabdominal operation and then by transanal operation. Patients with two surgery teams underwent taTME by transabdominal and transanal operation simultaneously with duration of the simutaneous operation time ≥30 minutes. The entire surgical process of patients with two surgery teams is not required to be performed by two surgery teams simutaneously. The clinical data were collected from the medical centers with similar operation amount according to the operation time sequence to analyze the difference between different operation stages and explore the learning curve. The operation time was taken as the parameter to carry out cumulative sum analysis and draw the learning curve of laparoscopic rectal cancer taTME in each medical center. The clinicopathological characteristics of patients from two medical centers with the largest difference in learning curves were analyzed. Observation indicators: (1) screening results of clinical data; (2) clinical data collection of patients with one surgery team; (3) surgical situations of laparoscopic rectal cancer taTME from the one surgery team in different operation stages; (4) learning curve of the one surgery team; (5) clinical data collection of patients with two surgery teams; (6) surgical situations of laparoscopic rectal cancer taTME from the two surgery teams; (7) learning curve of the two surgery teams. The cumulative sum was calculated by the $CUSUM={\sum }_{i=\mathrm{1}}^n\left(Xi-U\right)$, where Xi represented the operation time of each taTME, U represented the average operation time of all cases, and n represented the operation number. Fitting process was conducted on scatter plot of learning curves. Taking the apex of learning curve as the boundary, the learning curve was divided into two stages. The abscissa corresponding to the apex of learning curve was the number of operations that needed to be performed to cross the learning curve. Measurement data with normal distribution were represented as Mean±SD. Comparison between two groups was conducted using the t test and comparison between multiple groups was conducted using the ANOVA. Measurement data with skewed distribution were represented as M(P25,P75), and comparison between groups was conducted using the Mann‑Whitney U test. Comparison of ordinal data was analyzed using the rank sum test. Count data were analyzed using the chi‑square test or Fisher exact probability.

Results (1) Screening results of clinical data:the clinical data of 661 patients from 7 medical centers with one surgery team and two surgery teams were collected. (2) Clinical data collection of patients with one surgery team: the clinical data of 312 patients undergoing laparoscopic rectal cancer taTME from 5 medical centers were collected including 42 cases in the number 2 medical center, 97 cases in the number 20 medical center, 82 cases in the number 33 medical center, 35 cases in the number 37 medical center and 56 cases in the number 39 medical center, respectively. (3) Surgical situations of laparoscopic rectal cancer taTME from the one surgery team in different operation stages: three medical centers including the number 2, number 37 and number 39 medical center with close operation volume provided the clinical data of cases distributed in five operation stages. Among the five operation stages, the proportion of high‑quality operation of total mesorectal excision (TME) was ≥17/18, the incidence of postoperative complications was ≤13.3%(4/30) and the incidence of anastomotic leakage was ≤10.0%(3/30). There was no significant difference in the TME quality, postoperative complications or anastomotic leakage among the five operation stages (P>0.05). There was no significant difference in the operation time among the five operation stages (χ²=6.950, P>0.05). (4) Learning curve of the one surgery team: the number of operations corresponding to the turning point of learning curve in number 2 and number 20 medical center was 22 and 39, respectively. The number of operations corresponding to the turning points of learning curve in number 33 and number 37 medical center was 15, 66 and 10, 28, respectively. The number of operations corresponding to the turning point of learning curve in number 39 medical center was 20. The overall curve of number 20 medical center was in line with the trend of learning curve and 39 cases of operations was the minimum number needed to cross the learning curve. The biggest difference in learning curve was shown between the number 20 and number 33 medical center. Cases with the gender of male or female, age, body mass index, cases classified as stage 1, stage 2, stage 3 or stage 4 of the American Society of Anesthesiologists (ASA) Classification, cases with neoadjuvant therapy, duration of postoperative hospital stay of the number 20 medical center were 77, 20, (60±10)years, 24 kg/m² (22 kg/m², 26 kg/m²), 1, 88, 8, 0, 8, 8, 11 days (9 days, 13 days), respectively, versus 51, 31, (64±11)years, 23 kg/m²(21 kg/m², 26 kg/m²), 0, 35, 43, 1, 31, 16 days (13 day, 21 day) of number 33 medical center, showing significant differences in the above indicators between the two medical centers (χ²=6.442, t=-2.265, Z=-2.032, -6.870, χ²=22.120, Z=-8.408, P<0.05). (5) Clinical data collection of the two surgery teams: the clinical data of 259 patients undergoing laparoscopic rectal cancer taTME from 5 medical centers were collected, including 46 cases in the number 2 medical center, 47 cases in the number 8 medical center, 78 cases in the number 18 medical center, 43 cases in the number 33 medical center and 45 cases in the number 44 medical center, respectively. (6) Surgical situations of laparoscopic rectal cancer taTME from the two surgery teams: four medical centers including the number 2, number 8, number 33 and number 44 medical center with close operation volume provided the clinical data of cases distributed in four operation stages. Among the four operation stages, the proportion of high‑quality operation of TME was ≥50.0%(13/26), the incidence of postoperative complications was ≤35.0%(14/40) and the incidence of anastomotic leakage was ≤22.5%(9/40). There was no significant difference in the TME quality, postoperative complications or operation time among the four operation stages (χ²=3.252, 4.733, 8.848, P>0.05). There was a significant difference in the incidence of anastomotic leakage among the four operation stages (P<0.05). (7) Learning curve of the two surgery teams: the number of operations corresponding to the turning point of learning curve in number 2 and number 8 medical center was 28 and 16, respectively. The number of operations corresponding to the turning points of learning curve in number 18, number 33 and number 44 medical center was 12 and 58, 10 and 36, 14 and 36, respectively. The overall curve of number 2 medical center was in line with the trend of learning curve and 28 cases of operations was the minimum number needed to cross the learning curve. The biggest difference in learning curve was shown between the number 2 and number 33 medical center. The age and cases with tumor in stage T0 and (or) Tis, stage T1, stage T2, stage T3 or stage T4 of the T staging of the number 2 and number 33 medical center were (60±12)years, 3, 1, 9, 11, 20 and (65±10)years, 2, 3, 22, 15, 0, respectively, showing significant differences in the above indicators between the two medical centers (t=-2.280, Z=-4.033, P<0.05).

Conclusion Thirty‑nine cases of operations was the minimum number for the one surgery team to cross the learning curve of laparoscopic rectal cancer taTME and 28 cases of operations was the minimum number for the two surgery teams to cross the learning curve of laparoscopic rectal cancer taTME.