The development and application of pediatric complicated appendicitis prediction model

Area of Science:

• Pediatric Surgery
• Diagnostic Imaging
• Inflammatory Markers

Background:

• Acute appendicitis is a common cause of acute abdomen in pediatric surgery, accounting for 20-30% of cases.
• Accurate diagnosis is crucial for timely intervention and preventing complications.

Purpose of the Study:

• To develop and evaluate a decision tree model for diagnosing complicated appendicitis in children.
• The model integrates appendiceal ultrasound findings with inflammatory indices.

Main Methods:

• Retrospective analysis of 395 children diagnosed with appendicitis.
• Collected data on inflammatory markers (CRP, NLR) and ultrasound features (appendix diameter, echogenicity, fecaliths).
• Developed and validated binary logistic regression and decision tree models using ROC curves.

Main Results:

• Independent risk factors for complicated appendicitis included CRP, NLR, appendicolith, and enhanced mesenteric echogenicity.
• The decision tree model achieved 79% accuracy and an AUC of 0.809.
• The logistic regression model showed 74.9% accuracy and an AUC of 0.823.

Conclusions:

• Appendiceal ultrasound combined with inflammatory markers offers a valuable tool for diagnosing childhood appendicitis.
• The decision tree model is simple, intuitive, and effective in assisting pediatric emergency physicians.