Abstract
Introduction: Laboratory testing plays a minor role in screening for fractures in blunt trauma patients. Its main value is in the assesssment of co-morbidity. We investigated the association between elevated D-dimer levels and fractures.
Hypothesis: Elevated D-dimer levels are associated with fractures in blunt trauma patients not including polytrauma.
Methods: A retrospective cohort study was performed on blunt trauma patients referred to our emergency department (ED) over two years (2008-2010). Patients were eligible for inclusion if they were referred within 24 hours after injury. Patients with polytrauma, head injury, aortic injury/disease, malignancy, and thromboembolic tendency were excluded. A D-dimer assay was obtained up on ED arrival. Receiver-operator characteristic (ROC) curve analysis was performed to evaluate the cut-off value of D-dimer.
Results: Of 341 eligible patients, 210 (61.6%) had fractures (Fx group) and 131 (38.4%) did not (non-Fx group). Compared to the non-Fx group, the Fx group was significantly older (61±23 vs. 49±23 yr, p<0.01) and had higher D-dimer levels and ISS (DD, median (IQR): 8.4 (3.0-24.9) vs. 1.4 (0-3.1), p<0.01, ISS, mean±SD: 7.4±3.9 vs. 2.8±2.6, p<0.01). ROC analysis revealed the area under the curve was 0.81 (95% CI 0.77-0.86) and identified the D-dimer level of ≥4.5μg/ml as an optimal cut-off value, with a sensitivity of 69.0%, specificity of 84.7%, and a positive predictive value of 90.6% for detecting fractures.
Conclusions: Our findings suggest that an initial D-dimer assay may be efficient in screening for fractures in blunt trauma patients. Further study in the prospective setting is warranted to establish its efficacy.
