Nomogram for Predicting Deep Venous Thrombosis in Lower Extremity Fractures

Area of Science:

• Orthopedics
• Vascular Surgery
• Clinical Prediction Modeling

Background:

• Deep venous thrombosis (DVT) is a frequent complication in lower extremity fractures, impacting recovery and long-term quality of life.
• Existing research lacks specific clinical predictive models for DVT in this patient population.
• Early DVT prediction and prevention are crucial for reducing patient morbidity and enhancing treatment effectiveness.

Purpose of the Study:

• To develop and validate a novel clinical prediction model for deep venous thrombosis (DVT) in patients with lower limb fractures.
• To identify key risk factors associated with DVT development in this specific patient group.
• To provide a practical tool for clinicians to assess DVT risk and guide preventative strategies.

Main Methods:

• A retrospective study involving 3300 patients with lower limb fractures from two Chinese hospitals.
• Multivariate logistic regression analysis was employed to identify significant predictors of DVT.
• Model performance was assessed using discrimination and calibration curves, including the C-statistic.

Main Results:

• The final analysis included 2662 patients (1666 with DVT, 996 without).
• Significant predictive factors for DVT were identified as age, BMI, fracture-fixation type, injury impact energy, blood transfusion, and anticoagulant use.
• The developed model demonstrated a C-statistic of 0.676, indicating moderate discriminative ability.

Conclusions:

• A new, validated thrombus prediction model specifically for lower limb fracture patients has been established.
• The model incorporates easily accessible clinical variables, facilitating its application in routine practice.
• This tool can aid in the early identification and management of DVT risk in orthopedic trauma patients.