Identification of risk factors for urethrovesical anastomotic leakage following laparoscopic radical prostatectomy / 中华泌尿外科杂志

ABSTRACT

Objective:To investigate the risk factors of urethrovesical anastomotic leakage after laparoscopic radical prostatectomy.Methods:The clinical data of 292 patients who underwent laparoscopic radical prostatectomy in the Tenth People's Hospital Affiliated to Tongji University from January to December 2021 were retrospectively analyzed. According to whether there was anastomotic leakage, the patients were divided into leakage group (27 cases) and non-leakage group (265 cases). There were no significant differences in age [(71.5±6.5) years vs. (70.2±6.4) years], body mass index [(24.5±3.6) kg/m 2 vs. (24.2±3.0) kg/m 2], prostate volume[40(27.3, 63.2)ml vs. 38(28.1, 56.2)ml], Gleason score, clinical stage, and risk classification between the leakage group and the non-leakage group ( P>0.05), but the total prostate-specific antigen in the leakage group was significantly higher than that in the non-leakage group[20.0 (9.6, 79.0) ng/ml vs. 13.7 (8.5, 25.0) ng/ml, P=0.049]. Propensity score matching (PSM) was used to match the above indicators between the leakage group and the non-leakage group as 1∶1, so that the baseline of the two groups was balanced. The perioperative indicators of the matched two groups of patients were compared and analyzed. Statistically significant indicators were selected and included in univariate and multivariate logistic regression to analyze the risk factors of anastomotic leakage after radical prostatectomy. Finally, the receiver operating characteristic (ROC) curve was drawn, and the area under the curve (AUC) was calculated. The accuracy of each factor in predicting urine leakage was obtained. Results:After PSM, 24 cases were successfully matched. The leakage group had shorter membranous urethral length (MUL) [(15.5±2.2)mm vs. (17.5±1.5)mm, P<0.001], thinner membranous urethral wall thickness (UWT) [(9.5±1.9)mm vs. (10.6±1.5)mm, P=0.024], longer anastomotic time of urethrovesical neck[(21.6±4.1)min vs. (16.9±2.9)min, P<0.001] and higher failure rate of water injection test [16.7% (4/24) vs. 4.2% (1/24), P=0.045] than the non-leakage group. There was no significant difference in other indicators between the two groups. The results of multivariate logistic regression analysis showed that short MUL ( OR=0.544, 95% CI 0.335-0.884, P=0.014), narrow UWT ( OR=0.538, 95% CI 0.313-0.924, P=0.025) and long anastomotic time of urethrovesical neck ( OR=1.519, 95% CI 1.122-2.110, P=0.009) were independent risk factors for anastomotic urine leakage. ROC curve analysis showed that the AUC of MUL, UWT, and anastomotic time were 0.789 (95% CI 0.651-0.927), 0.715 (95% CI 0.562-0.868), and 0.842 (95% CI 0.731-0.953), respectively. Conclusions:Narrow and short membranous urethra and long anastomosis time in patients with laparoscopic radical prostatectomy may be independent risk factors for postoperative anastomotic leakage, which may predict the occurrence of anastomotic leakage.