[Establishment and validation of nomogram prediction model for complicated acute appendicitis]

Area of Science:

• Surgical Gastroenterology
• Medical Informatics
• Predictive Analytics

Context:

• Acute appendicitis is a common surgical emergency.
• Distinguishing complicated appendicitis (CA) from uncomplicated cases preoperatively is clinically significant.
• Accurate prediction of CA can guide timely surgical intervention and resource allocation.

Purpose:

• To develop and validate a predictive nomogram model for complicated appendicitis (CA).
• To identify key clinical factors associated with the progression to CA.
• To provide a tool for preoperative risk stratification of appendicitis patients.

Summary:

• A retrospective analysis of 663 appendicitis patients identified elderly status (age ≥ 60), prolonged abdominal pain duration, high fever (≥ 39°C), and elevated total bilirubin as independent predictors of CA.
• A nomogram model incorporating these factors demonstrated high predictive accuracy with an Area Under the Curve (AUC) of 0.935.
• Internal validation using the Bootstrap method confirmed the model's robust discrimination ability (AUC=0.933).

Impact:

• The developed nomogram offers a valuable, non-invasive tool for clinicians to predict the likelihood of complicated appendicitis preoperatively.
• Early identification of CA can lead to optimized treatment strategies, potentially reducing complications and improving patient prognosis.
• This model contributes to the field of surgical decision support systems by integrating clinical data for improved diagnostic accuracy.