Objective To investigate the influencing factors of refractory anastomotic stenosis after laparoscopic intersphincteric resection (Ls-ISR) for rectal cancer and construction of nomogram prediction model.

Methods The retrospective case-control study was conducted. The clinicopatho-logical data of 495 patients who underwent Ls-ISR for rectal cancer in two medical centers, including 448 patients in Peking University First Hospital and 47 patients in Cancer Hospital Chinese Academy of Medical Sciences, from June 2012 to December 2021 were collected. There were 311 males and 184 females, aged 61 (range, 20-84)years. Observation indicators: (1) incidence of anastomotic stenosis; (2) influencing factors of refractory anastomotic stenosis after Ls-ISR; (3) construction and evaluation of nomogram prediction model for refractory anastomotic stenosis after Ls-ISR. Follow-up was conducted using outpatient examination and telephone interview to detect the incidence of postoperative anastomotic leakage and anastomotic stenosis up to August 2022. Measurement data with normal distribution were represented as Mean±SD, and comparison between groups was conducted using the t test. Measurement data with skewed distribution were represented as M(range). Count data were described as absolute numbers, and comparison between groups was conducted using the chi-square test. Univariate and multivariate analyses were conducted using the Logistic regression model. Factors with P<0.10 in univariate analysis were included in multivariate analysis. The R software (3.6.3 version) was used to construct nomogram prediction model. The receiver operating characteristic (ROC) curve was drawn and the area under curve (AUC) was used to evaluate the efficacy of nomogram prediction model.

Results (1) Incidence of anastomotic stenosis. All 495 patients underwent Ls-ISR successfully, without conversion to laparotomy, and all patients were followed up for 47(range, 8-116)months. During the follow-up period, there were 458 patients without anas-tomotic stenosis, and 37 patients with anastomotic stenosis. Of the 37 patients, there were 15 cases with grade A anastomotic stenosis, 3 cases with grade B anastomotic stenosis and 19 cases with grade C anastomotic stenosis, including 22 cases being identified as the refractory anastomotic stenosis. Fifteen patients with grade A anastomotic stenosis were relieved after anal dilation treat-ment. Three patients with grade B anastomotic stenosis were improved after balloon dilation and endoscopic treatment. Nineteen patients with grade C anastomotic stenosis underwent permanent stoma. During the follow-up period, there were 42 cases with anastomotic leakage including 17 cases combined with refractory anastomotic stenosis, and 453 cases without anastomotic leakage including 5 cases with refractory anastomotic stenosis. There was a significant difference in the refractory anastomotic stenosis between patients with and without anastomotic leakage (χ²=131.181, P<0.05). (2) Influencing factors of refractory anastomotic stenosis after Ls-ISR. Results of multivariate analysis showed that neoadjuvant therapy, distance from tumor to anal margin ≤4 cm, clinic N+ stage were independent risk factors of refractory anastomotic stenosis after Ls-ISR (hazard ratio=7.297, 3.898, 2.672, 95% confidence interval as 2.870-18.550, 1.050-14.465, 1.064-6.712, P<0.05). (3) Construction and evaluation of nomogram prediction model for refractory anastomotic stenosis after Ls-ISR. Based on the results of multivariate analysis, neoadjuvant therapy, distance from tumor to anal margin and clinic N staging were included to constructed the nomogram prediction model for refractory anastomotic stenosis after Ls-ISR. Results of ROC curve showed the AUC of nomogram prediction model for refractory anastomotic stenosis after Ls-ISR was 0.739 (95% confidence interval as 0.646-0.833).

Conclusions Neoadjuvant therapy, distance from tumor to anal margin ≤4 cm, clinic N+ stage are independent risk factors of refractory anastomotic stenosis after Ls-ISR. Nomogram prediction model based on these factors can predict the incidence of refractory anastomotic stenosis after Ls-ISR.