# Risk factors for complications after infantile enterostomy: development of a clinical prediction model

## TL;DR

This study identifies risk factors for complications after infantile enterostomy and creates a prediction model to help doctors assess and manage risk.

## Contribution

A new clinical prediction model for short-term complications after infantile enterostomy is developed and validated.

## Key findings

- Smaller weight at surgery, small intestine stoma, and prolonged high CRP levels are independent risk factors for complications.
- The prediction model has an AUC of 0.784, indicating good accuracy in predicting complications.
- Calibration curves and DCA confirm the model's clinical utility and predictive value.

## Abstract

Enterostomy is a common surgical procedure for treating acute abdomen in infants. However, the associated complication incidence is high, which significantly impacts infants’ recovery. This study aimed to identify risk factors of short-term complications and develop a prediction model in infants with enterostomy.

We retrospectively analyzed the clinical data of infants who underwent enterostomy at Sichuan University West China Second Hospital from November 2021 to June 2024. Multifactorial logistic regression analysis was used to screen the risk factors for postoperative complications related to enterostomy in infants, and R software was applied to develop a nomogram prediction model. The accuracy and clinical utility of the prediction model were verified by the receiver operating characteristic (ROC) curve, calibration curve, and decision curve analysis (DCA).

A total of 155 infants were included in this study, with 61 cases (39.35%) in the complication group and 94 cases (60.65%) in the non-complication group. Multifactorial logistic regression analysis showed that smaller weight at surgery (OR = 0.999, 95% CI: 0.999 ~ 1.000, p = 0.044), small intestine stoma (OR = 6.405, 95% CI: 1.647 ~ 24.916, p = 0.007), and prolonged duration of postoperative high-level C-reactive protein (CRP) (OR = 1.081, 95% CI: 1.001 ~ 1.167, p = 0.048) were independent risk factors for complications related to enterostomy in infants. The area under the curve (AUC) of the risk prediction model was 0.784 (95% CI: 0.712 ~ 0.857), and the goodness-of-fit test value of the Hosmer-Lemeshow was 0.604, higher than 0.05, indicating that the regression model had a significant fitting effect. The calibration curves and DCA demonstrated high predictive value and clinical efficiency.

The smaller weight at surgery, small intestine stoma, and longer duration of postoperative high-level CRP may be used to identify the risk of short-term complications after enterostomy. This prediction model is provided for medical staff to evaluate complication-associated risk and take measures for those infants at risk.