Building and Verifying a Prediction Model for Deep Vein Thrombosis Among Spinal Cord Injury Patients Undergoing...

Area of Science:

• Medical Rehabilitation
• Vascular Surgery
• Epidemiology

Background:

• Deep vein thrombosis (DVT) is a significant complication in spinal cord injury (SCI) patients during inpatient rehabilitation.
• Effective risk prediction is crucial for timely intervention and improved patient outcomes.

Purpose of the Study:

• To identify variables associated with DVT in SCI patients undergoing rehabilitation.
• To develop and validate a nomogram model for predicting DVT risk in this population.

Main Methods:

• A cohort of 558 SCI patients was divided into training (n=446) and validation (n=112) groups.
• Logistic regression analysis identified independent risk factors for DVT.
• A nomogram model was constructed and validated using ROC curves and calibration plots.

Main Results:

• Prothrombin time, D-dimer, age, and Caprini score were identified as independent predictors of DVT.
• The nomogram model demonstrated strong predictive accuracy (AUC 0.793 in training, 0.905 in validation).
• The model showed good calibration, with predicted probabilities aligning well with observed DVT frequencies.

Conclusions:

• Prothrombin time, D-dimer, age, and Caprini score are key predictors of DVT in SCI patients.
• The developed nomogram provides an accurate and accessible tool for DVT risk assessment in this cohort.
• This tool aids clinicians in early identification of high-risk patients for targeted prevention and management strategies.