Comparison Of Late Complications Of Different Urinary Diversion Methods In Patients Undergoing Radical Cystectomy And Construction Of Overall Survival Prediction Model

Objective:The purpose of this paper is to study whether there are differences in late complications in patients with different urinary diversion,and to screen out independent risk factors for the overall survival of muscle invasive bladder cancer patients after radical cystectomy by Cox analysis.According to these prognostic factors,a prognostic model presenting the overall survival probability in the form of nomogram was established,and the model was evaluated by various methods.Methods:In this study,the clinical data of patients underwent radical cystectomy and urinary diversion in our hospital from November 1,2010 to November 1,2019,and patients with muscle invasive bladder cancer who changed fistulas in our ostomy clinic were collected.Through telephone follow-up,the current survival conditions of patients and the late complications of patients after 3 months of surgery were acquired.In our study,chi-square test or Fisher exact test was used to compare the late complications of patients with different urinary diversion.The collected clinical information was included in univariate Cox analysis,and the statistically significant variables in univariate analysis were included in multivariate Cox analysis to screen out the independent risk factors for the overall survival of patients after radical cystectomy.In this study,Kaplan-Meier method was used to draw the survival curve of the overall survival of MIBC patients and the survival curve of a single prognostic factor controlling other covariates.A nomogram covering multiple predictors was established to visually show the risk of 3-year and 5-year overall survival of different patients.Calibration curve,C-index,time-dependent ROC curve and the area under the curve are applied to appraise the accuracy and discrimination of this model,and the decision curve is applied to appraise the clinical effectiveness of this model.Results:1.A total of 225 patients were followed up in our study,and 111 patients’ information was collected excluding those with incomplete follow-up information.Among them,there were 65 patients with cutaneous ureterostomy,24 patients with ileal channel operation and 22 patients with ileal orthotopic neo-bladder.2.The incidence of urinary tract infection was very different in these three kinds of urinary diversion(P<0.001)The cutaneous ureterostomy patients.have the highest incidence of urinary tract infection,followed by ileal conduit,and the lowest in patients undergoing ileal orthotopic neo-bladder.The incidence of intestinal obstruction after ileal orthotopic neo-bladder was higher than that of other urinary diversion methods(P<0.001).Hydronephrosis is more likely to occur after ileal orthotopic neo-bladder surgery than other urinary diversion methods(P=0.003).There was no significant difference in the diversion methods of urine flow for other late complications.3.Through univariate and multivariate Cox analysis,age(P = 0.01;HR= 1.017(1.001～1.031)),diameter of tumor(P = 0.028;HR=1.003(1.000～1.005),histological grade(P < 0.001;HR=3.597(1.887～6.849)),tumor N stage(P <0.001)and urinary diversion mode(P <0.001)are selected as independent risk factors for the overall survival of muscle invasive bladder cancer patients after radical cystectomy.According to the screened prognostic factors,the Kaplan-Meier curves of every single prognostic factor was drafted to correct other confounding factors.4.Five independent risk factors were enrolled in the model,and a nomogram was constructed to visually show the 3-year and 5-year overall survival.The C-index of this model is 0.772(0.735～0.809).The 3-year and 5-year AUC of time-dependent ROC curves are 0.856(0.806-0.906)and 0.816(0.753-0.878)respectively.The calibration curve agreed with its standard line well,and the prediction probability and observation probability of the display model were in good agreement.The model curve of the decision-making curve is obviously far from the two extreme lines,which shows that the clinical net benefit of the model is high.Conclusion:1.The incidence of urinary tract infection after cutaneous ureterostomy is higher than that of patients with other urinary diversion.The incidences of hydronephrosis and intestinal obstruction after ileal orthotopic neo-bladder is higher than that of other urinary diversion operations.2.Age,tumor diameter,high-grade urothelial carcinoma,different urinary diversion methods and the increase of tumor N stage are independent risk factors for the overall survival of muscle invasive bladder cancer patients after radical cystectomy.The overall survival of patients with ileal orthotopic neo-bladder is longer than that of other patients with other urinary diversion.3.The prediction model of total survival of muscle invasive bladder cancer patients after radical cystectomy was established.C-index,calibration curve,time-dependent ROC curve and its AUC show that the model has high accuracy and discrimination ability.The decision curves of 3 years and 5 years show that the model can benefit patients in clinical practice.