Diagnostic accuracy of the Triage® D-dimer test for exclusion of venous thromboembolism in outpatients☆
Received 5 March 2007; received in revised form 5 July 2007; accepted 13 July 2007.
Abstract
Background
We evaluated the diagnostic performance of the Triage D-dimer test, a new fast quantitative point-of-care whole blood D-dimer assay and compared it with the Vidas D-dimer assay.
Materials and methods
The study population comprised 319 outpatients for whom D-dimer testing was requested in order to rule out venous thromboembolism (VTE). Routine testing consisted of a plasma ELISA D-dimer analysis (Vidas). For all included patients, an additional EDTA whole blood D-dimer test (Triage) was performed. Patients were classified by reference imaging or by follow-up of the medical record. Accuracy indices, receiver operating characteristics and the kappa coefficient for agreement were calculated using the cutoff values recommended by the manufacturer.
Results
Prevalence of VTE was 14%. Sensitivity and specificity for VTE were 98% (95%CI: 88–100) and 34% (95%CI: 28–40) for Vidas and 91% (95%CI: 78–97) and 42% (95%CI: 36–48) for Triage, respectively. The differences in sensitivity and specificity between both D-dimer assays were statistically significant (McNemar, p<0.0001). ROC-curve analysis yielded an area under the curve of 0.83 (95%CI: 0.76–0.89) for the Vidas and 0.81 (95%CI: 0.74–0.88) for the Triage (p=0.396). The kappa coefficient for agreement between Vidas and Triage was 0.75 (95%CI: 0.68–0.79).
Conclusions
The Triage and Vidas D-dimer tests show comparable diagnostic accuracy. Vidas showed a significant higher sensitivity. Our findings strongly suggest lowering the cutoff for the Triage D-dimer test from 400 to 350 ng/mL. In this way specificity lowers from 42 to 38%, but, more importantly, sensitivity increases from 91 to 95%.
ABSTRACT