Nomogram for soiling prediction in post-surgery hirschsprung children: a retrospective study

Abstract

Purpose: The aim of this study was to develop a nomogram for predicting the probability of postoperative soiling in patients aged >1 year operated for HSCR. Materials and Methods: We retrospectively analyzed HSCR patients with surgical therapy over one year of age from January 2000 and December 2019 at our department. Eligible patients were randomly categorized into the training and validation set at a ratio of 7:3. By integrating the least absolute shrinkage and selection operator [LASSO] and multivariable logistic regression analysis, crucial variables were determined for establishment of the nomogram. And, the performance of nomogram was evaluated by C-index，area under the receiver operating characteristic curve [ROC], calibration curves, and decision curve analysis [DCA]. Meanwhile, a validation set was used to further assess the model. Results: This study enrolled 601 cases, and 97 patients suffered from soiling. Three risk factors, including surgical history, length of removed bowel, and surgical procedures were identified as predictive factors for soiling occurrence. The C-index was 0.871 [95% CI=0.821–0.921] in the training set and 0.878 [95% CI=0.811–0.945] in the validation set, respectively. And, the AUC was found to be 0.896 [95% CI=0.855 − 0.929] in the training set and 0.866 [95% CI=0.767 − 0.920] in the validation set. Additionally, the calibration curves displayed a favorable agreement between the nomogram model and actual observations. The DCA revealed that employing the nomogram to predict the risk of soiling occurrence would be advantageous if the threshold was between 1-73% in the training set and 3-69% in the validation set. Conclusion: This study represents the first efforts to develop and validate a model capable of predicting the post-operative risk of soiling in patients aged >1 year operated for HSCR. This model may assist clinicians in determining the individual risk of soiling subsequent to HSCR surgery, aiding in personalized patient care and management.