Prevalence and risk factors for proximal deep vein thrombosis at admission in patients with traumatic fractures: a multicenter retrospective study

Objective This study aimed to determine the associated risk factors for proximal deep vein thrombosis (DVT) in patients with lower extremity and pelvic-acetabular fractures. Methods The medical records of 4,056 patients with lower extremity and pelvic-acetabular fractures were retrospectively reviewed. The patients were classified into proximal or non-proximal DVT groups. Logistic regression models were used to determine the independent risk variables for proximal DVT. The predictive value of the related risk factors was further analyzed using receiver operating characteristic curves. Results The prevalence of proximal DVT was 3.16%. Sex, body mass index (BMI), fracture site, injury mechanism, diabetes, coronary heart disease (CHD), injury-to-admission interval, hematocrit, platelet counts, and D-dimer levels differed significantly between the two groups. BMI >= 24.0 kg/m(2), femoral shaft fractures, high-energy injury, diabetes, injury-to-admission interval >24 h were independent risk factors for proximal DVT. CHD decreased the risk of proximal DVT. The platelet and D-dimer had high negative predictive value for predicting proximal DVT formation, with cut-off values of 174 x 10(9)/L and 2.18 mg/L, respectively. Conclusion BMI >= 24.0 kg/m(2), femoral shaft fractures, high-energy injury, diabetes, injury-to-admission interval >24 h were independent risk factors for proximal DVT in patients with lower extremity and pelvic-acetabular fractures. Platelet count and D-dimer level were effective indicators for excluding proximal DVT occurrence. CHD decreased the risk of proximal DVT.