Establishment and validation of a predictive model for neurogenic urinary tract injury in children

Abstract

Objective: To establish a predictive model for upper urinary tract damage in children with neurogenic bladder and verify its efficacy. Methods: From January 2011 to December 2021, 143 children with NB in the Children's Hospital of Chongqing Medical University and 84 children with NB in the First Affiliated Hospital of Zhengzhou University were selected as the research objects. The former is set as the training set and the latter is set as the validation set, and the general parameters of the two are compared. The independent risk factors of upper urinary tract damage in children with NB were screened out by Lasso regression, and multivariate logistic regression analysis and a nomogram prediction model was established. The models were validated internally and externally on the training set and validation set, respectively, and the area under the receiver operating curve (ROC) was used to verify the accuracy of the model. Results: A total of 227 children with NB were included in this study, including 121 males and 106 females, aged (10.2±3.8) years. There was no significant difference in other parameters except age between the training set and validation set (all P>0.05); Lasso regression and multivariate logistic regression analysis showed that detrusor leakage point pressure (DLPP) ≥ 40 cmH2O (OR=4.76, 95%CI: 2.01-11.26, 1 cmH2O=0.098 kPa), overactive bladder (OAB) (OR=3.08, 95%CI: 1.34-7.04), bladder compliance (BC)<20 ml/cm H2O (OR=3.65, 95%CI: 1.41-9.47), history of previous urinary tract infection (OR=2.73, 95%CI: 1.09-6.81), and abdominal pressure/other voiding patterns (OR=2.86, 95%CI: 1.20-6.82) were risk factors for upper urinary tract damage in children with NB (all P<0.05). The above parameters were used to establish a nomogram model of upper urinary tract damage in children with NB. The internal and external validation results show that the AUC values for the training and validation sets were 0.84 (95%CI: 0.77-0.91) and 0.86 (95%CI: 0.79-0.94), respectively. Conclusion: The prediction model of upper urinary tract damage in children with NB constructed in this study has high discrimination, accuracy and clinical applicability, which can help clinicians identify high-risk patients and make individualized treatment design for these patients.