Risk Prediction Model for Crohn's Disease Based on Hematological Indicators

Area of Science:

• Gastroenterology
• Internal Medicine
• Diagnostic Medicine

Background:

• Diagnostic errors are common when distinguishing Crohn's disease (CD) from other conditions like ulcerative colitis (UC) and intestinal tuberculosis (ITB).
• An efficient, rapid, and simple predictive model is crucial for clinical application in CD diagnosis.
• Current diagnostic methods can be invasive and time-consuming, necessitating simpler alternatives.

Purpose of the Study:

• To develop a risk prediction model for CD using five routine laboratory tests.
• To create an early warning system and a visual nomograph for CD risk assessment.
• To provide clinicians with a convenient tool for CD risk determination and differential diagnosis.

Main Methods:

• Retrospective analysis of 310 cases (2020-2022) including CD, UC, non-IBD diseases, and healthy controls.
• Risk prediction models were built using logistic regression based on hematological indices: ESR, Hb, WBC, ALb, and CH.
• Models were evaluated for diagnostic accuracy and visualized using a nomograph.

Main Results:

• Significant differences in ESR, WBC, and WBC/CH ratios were observed between CD and non-CD groups (p < 0.05).
• The WBC/CH ratio showed a strong correlation with CD occurrence (correlation coefficient > 0.4).
• A logistic regression model demonstrated high diagnostic accuracy (AUC=0.86) for CD prediction and (AUC=0.88) for differentiating CD from ITB.

Conclusions:

• A novel CD risk prediction model was successfully established using five conventional hematological indices.
• The model, visualized via a nomograph, offers high diagnostic accuracy for CD and its differentiation from ITB.
• This tool can assist clinicians in better managing CD and reducing patient suffering.