Causes of elevated D-dimer in patients admitted to a large urban emergency department

•D-dimer was examined in ED patients with suspected venous thromboembolism (VTE).•A significant correlation was found between age and D-dimer.•Infection was the most frequent diagnosis, followed by VTE and syncope.•As compared with values < 1000 ng/mL, the OR for VTE was 8.5 for D-dimer > 3000 ng/mL.•D-dimer lacks specificity for diagnosing VTE in elderly patients with comorbidities.

BackgroundAlthough the request for D-dimer is widespread in emergency departments (EDs), the causes of elevation and their relationship with D-dimer levels in patients with diagnostic values are uncertain.MethodsIn this retrospective investigation, the study population consisted of all patients who visited our large urban ED in the year 2012, for whom a D-dimer test was requested for excluding or diagnosing venous thromboembolism (VTE). Only patients with D-dimer values > 243 ng/mL were included, regardless of their pre-test clinical probability for VTE.ResultsThe final study population consisted of 1647 patients. A significant positive correlation was found between age and D-dimer. Infection was the most frequent diagnosis (15.6%), followed by VTE (12.1%), syncope (9.4%), heart failure (8.9%), trauma (8.2%) and cancer (5.8%). D-dimer was higher in patients with VTE than in those with other diagnoses (2541 ng/mL vs 1030 ng/mL; p < 0.001). The frequency of VTE gradually increased from patients with values < 1000 ng/mL to those with D-dimer > 3000 ng/mL (4.1 vs 26.7%; p < 0.001). As compared with D-dimer values < 1000 ng/mL, the Odds Ratio for VTE was 8.5 for values > 3000 ng/mL.ConclusionsThese results show that D-dimer lacks specificity for diagnosing VTE, especially in elderly patients admitted to the ED with significant co-morbidities. In older patients, elevated values (> 1000 ng/mL) are more frequently associated with VTE, so the use of higher cut-offs may be advantageous.