Variable Rate of D-dimer Predicts Venous Thromboembolism in Postoperative Fracture Patients: A Single-Center Retrospective Cohort Study

Background The correlation between plasma D-dimer levels and deep vein thrombosis (DVT) remains inconsistent across studies. This study aimed to investigate whether the variable rate of D-dimer is a more accurate predictor of venous thromboembolism (VTE) compared to D-dimer alone. Methods Demographic data, fracture classification, and D-dimer levels were collected. The variable rate of D-dimer was defined as the ratio of postoperative D-dimer to preoperative D-dimer. Duplex ultrasonography was used to evaluate DVT in the lower extremities, while computed tomography pulmonary angiography was conducted for patients suspected of having pulmonary thromboembolism (PTE). Results A total of 77,609 traumatic fracture patients without VTE (mean age: 51.36 ± 28.72 years) and 2,813 patients with VTE (mean age: 68.13 ± 26.65 years) were included. Among the VTE patients, 36 had upper limb fractures, 522 had spine fractures, 269 had pelvic fractures, 1,265 had lower extremity fractures, and 721 had compound or other types of fractures. Of these, 2,743 patients developed lower extremity venous thromboembolism, and 70 suffered from PTE. The variable rate of D-dimer was significantly higher in postoperative traumatic fracture patients with VTE (2.79 ± 1.77) compared to preoperative patients without VTE (1.35 ± 0.70). Using a cutoff point of 2.0 for the variable rate of D-dimer, sensitivity and specificity were 83.44% and 88.60%, respectively. Multivariate analysis revealed that age > 60 years, female sex, and lower extremity fractures were independent risk factors for VTE. Conclusions The variable rate of D-dimer is strongly correlated with VTE in postoperative traumatic fracture patients. It serves as a useful predictor of VTE, particularly in elderly patients, females, and those with lower extremity fractures.