[Incidence and risk factors of anastomotic leak after transanal total mesorectal excision in China: a retrospective analysis based on national database].

Objective: Transanal total mesorectal excision (taTME) was a very hot topic in the first few years since its appearance, but now more introspections and controversies on this procedure have emerged. One of the reasons why the Norwegian Ministry of Health stopped taTME was the high incidence of postoperative anastomotic leak. In current study, the incidence and risk factors of anastomotic leak after taTME were analyzed based on the data registered in the Chinese taTME Registry Collaborative (CTRC). Methods: A case-control study was carried out. Between November 15, 2017 and December 31, 2020, clinical data of 1668 patients undergoing taTME procedure registered in the CTRC database from 43 domestic centers were collected retrospectively. After excluding 98 cases without anastomosis and 109 cases without complete postoperative complication data, 1461 patients were finally enrolled for analysis. There were 1036 males (70.9%) and 425 females (29.1%) with mean age of (58.2±15.6) years and mean body mass index of (23.6±3.8) kg/m(2). Anastomotic leak was diagnosed and classified according to the International Study Group of Rectal Cancer (ISREC) criteria. The risk factors associated with postoperative anastomotic leak cases were analyzed. The impact of the cumulative number of taTME surgeries in a single center on the incidence of anastomotic leak was evaluated. As for those centers with the number of taTME surgery ≥ 40 cases, incidence of anastomic leak between 20 cases of taTME surgery in the early and later phases was compared. Results: Of 1461 patients undergoing taTME, 103(7.0%) developed anastomotic leak, including 71 (68.9%) males and 32 (31.1%) females with mean age of (59.0±13.9) years and mean body mass index of (24.5±5.7) kg/m(2). The mean distance between anastomosis site and anal verge was (2.6±1.4) cm. Thirty-nine cases (37.9%) were classified as ISREC grade A, 30 cases (29.1%) as grade B and 34 cases (33.0%) as grade C. Anastomotic leak occurred in 89 cases (7.0%,89/1263) in the laparoscopic taTME group and 14 cases (7.1%, 14/198) in the pure taTME group. Multivariate analysis showed that hand-sewn anastomosis (P=0.004) and the absence of defunctioning stoma (P=0.013) were independently associated with anastomotic leak after taTME. In the 16 centers (37.2%) which performed ≥ 30 taTME surgeries with cumulative number of 1317 taTME surgeries, 86 cases developed anastomotic leak (6.5%, 86/1317). And in the 27 centers which performed less than 30 taTME surgeries with cumulative number of 144 taTME surgeries, 17 cases developed anastomotic leak (11.8%, 17/144). There was significant difference between two kinds of center (χ(2)=5.513, P=0.019). Thirteen centers performed ≥ 40 taTME surgeries. In the early phase (the first 20 cases in each center), 29 cases (11.2%, 29/260) developed anastomotic leak, and in the later phase, 12 cases (4.6%, 12/260) developed anastomotic leak. The difference between the early phase and the later phase was statistically significant (χ(2)=7.652, P=0.006). Conclusion: The incidence of anastomotic leak after taTME may be reduced by using stapler and defunctioning stoma, or by accumulating experience.