Development and validation of a risk prediction model for repeated indwelling urinary catheterization in patients with cervical cancer after surgery

Objective:To study the factors influencing secondary indwelling catheterisation after cervical cancer surgery and to develop a predictive risk model.Methods:A total of 260 patients in a tertiary hospital in Chongqing were selected from January 2020 to December 2021 via convenience sampling.Relevant information of patients was recorded,including age;body mass index;history of hypertension and diabetes,bladder dysfunction,postoperative urinary retention,and postoperative urinary tract infection;Histology;staging;surgical approach;Operation time;Time of first remove of catheter;indwelling catheter days;Hospitalization days.Least absolute shrinkage and selection operator was used to reduce dimensionality and select patient characteristics,and multivariate analysis was performed based on the selected variables.Based on the outcome of analysis,a line chart model was developed for predicting the risk of secondary catheterization in patients with indwelling catheterization after radical cervical cancer surgery.The coefficient of conformity index(C-index)and calibration curves were used to evaluate the accuracy and fit.The model was internally validated via bootstrapping(1000 random samples),and the clinical utility of the model was assessed via decision curve analysis(DCA).Results:Four characteristic variables were selected,including preoperative bladder function,postoperative urinary tract infection,surgical approach,and Time of first remove of catheter.They are independent risk factors affecting urinary tract.The risk prediction model exhibited good discrimination performance with a C-index of 0.722(95%CI,0.661-0.783)and was well calibrated.The C-index was 0.708 in internal validation analysis.DCA showed that the risk model was clinically useful for predicting secondary catheterization,and clinical benefits were observed at the decision threshold of≥11%.Conclusion:A novel model was developed to predict the risk of secondary catheterization.The model was based on preoperative bladder dysfunction,postoperative urinary tract infection,surgical approach,and number of days since the removal of the primary catheter.