Early Identification of Pediatric Inflammatory Bowel Disease Based on a Noninvasive Multivariable Predictive Model

Abstract

Background: Early identification of pediatric inflammatory bowel disease (IBD) improves long-term outcomes; yet, significant diagnostic delays persist. This study aimed to establish and validate the optimal model of noninvasive evaluation tests to help clinicians with the early identification of pediatric IBD.

Methods: The study adopted a retrospective development and prospective temporal validation design within the same clinical center. A cohort of 314 pediatric patients (IBD, 103; non-IBD, 211) was used to develop a logistic regression model. The model based on noninvasive features, including IBD-related symptoms, routine laboratory tests, and transabdominal ultrasound findings. Ultrasound parameters included Limberg score >1 (bowel wall thickening with blood flow), increased mesenteric fat, disrupted wall layering, and enlarged lymph nodes. The ultrasound operator was blinded to laboratory and endoscopic results. Feature selection was performed using logistic regression and random forest methods. Model performance was assessed via bootstrapped internal validation (1000 resamples), and temporally validated in a prospective cohort of 66 children (IBD, 19; non-IBD, 47).

Results: In the importance assessment, the ultrasound feature of Limberg level >1 was identified as the most valuable feature, followed by the erythrocyte sedimentation rate, fecal calprotectin, C-reactive protein and hypoalbuminemia. The most valuable clinical symptom identified was active perianal abscess or fistula. The model, constructed from these features, demonstrated high accuracy and robustness in both internal validation (area under the curve, 0.97 [95% confidence interval: 0.95-0.98]) and temporal external validation (area under the curve, 0.94 [95% confidence interval: 0.86-1.00]). In the external validation set, the model showed good calibration, with a calibration slope of 0.86, and a Brier score of 0.08.

Conclusion: The nomogram, based on noninvasive factors, can identify children with IBD at early stages using accessible noninvasive testing.

Keywords: inflammatory bowel disease; intestinal ultrasound; nomograms; noninvasive; pediatric.

Figure 1

Flow diagram of the research participants. (a) Diagram of the flow of the training set. (b) Diagram of the flow of the validation set. (c) Diagnostic protocol for pediatric inflammatory bowel disease. Red-colored text indicates timing: (1) laboratory tests, endoscopy, and intestinal ultrasound were performed within one week; (2) the final reference standard diagnosis of IBD was made within one month following the completion of all assessments.

Figure 2

Visualization of the diagnostic prediction models. (a) Evolutionary selection in LASSO regression. (b) Feature importance for the random forest model based on datasets with ultrasound characteristics. (c) Nomogram for diagnosing pediatric IBD.

Figure 3

Evaluation of model stability and diagnostic performance in the training and validation datasets. (a) Receiver operating characteristic (ROC) curves of the nomogram model in the training dataset. (b) ROC curves of the nomogram model in the validation dataset. (c) Validation of the nomogram model for pediatric inflammatory bowel disease (IBD) in the training dataset. (d) Validation of the nomogram model for pediatric IBD in the validation dataset.