Development of a predictive model for surgical intervention following air enema reduction of pediatric intussusception

TL;DR

This study creates a model to predict the need for surgery after air enema treatment for pediatric intussusception, helping reduce serious complications.

Contribution

A novel predictive model using measurable indicators to guide post-enema treatment decisions in pediatric intussusception.

Findings

Age, symptom duration, bloody stools, and blood parameters were independent risk factors for surgical intervention.

The model achieved an AUC of 0.879 and showed good calibration and net benefit via DCA.

The model offers a reliable tool for clinicians to decide on post-enema treatment strategies.

Abstract

Surgical intervention after air enema for paediatric intussusception is very common, and prompt surgical treatment after failure of air enema therapy is the key to reducing serious complications, such as intestinal perforation and intestinal necrosis caused by intussusception. The aim of this study was to develop and validate a prediction model for surgical intervention after air enema in paediatric intussusception to reduce the incidence of serious complications.

A retrospective study was performed on 843 children who were successfully reduced by air enema and 120 children who underwent surgical intervention after air enema in our hospital from January 2011 to December 2021. Baseline information, clinical presentation and test results of the children on admission were recorded. Univariate and multivariate logistic regression analyses were used to identify independent risk factors for surgical intervention after air enema for paediatric intussusception. Meanwhile, we developed a predictive model to predict surgical intervention after air enema for paediatric intussusception based on independent risk factors and validated the model using receiver operating characteristic (ROC) curves, calibration curves, and decision curve analysis (DCA).

Age, duration of symptoms, bloody stools, body temperature, lymphocyte percentage and basophil percentage were independently associated with the composite endpoint (p < 0.05). The patients were randomly divided into a training set and a validation set at a ratio of 7:3 for model construction and validation, respectively. A logistic regression model was constructed based on the above six factors and integrated into the nomogram. The area under the ROC curve of the nomogram constructed by 6 independent risk factors reached 0.879, and the calibration curve was close to the ideal diagonal. In addition, DCA analysis revealed significant net benefits of the model.

Our predictive model for surgical intervention after air enema in pediatric intussusception, developed using objectively measurable indicators, demonstrates reliable predictive capability. It provides clinicians with an effective and dependable tool for early decision-making regarding post-enema treatment strategies—whether to continue with enema or proceed to surgery.

Figures