D-dimer for the exclusion of venous thromboembolism: comparison of a new automated latex particle immunoassay (MDA D-dimer) with an established enzyme-linked fluorescent assay (VIDAS D-dimer)

Abstract

The use of D-dimer tests to exclude venous thromboembolism is an important advance in clinical practice and also has economic benefits. Ideally the test should be objective and a test that could be run on the routine coagulometer would obviate the need for additional investment in alternative hardware. A new automated latex particle immunoassay (MDA D-dimer) that can be run on a routine coagulometer was compared with a well established enzyme linked fluorescent assay (VIDAS D-dimer) on the basis of their ability to exclude venous thromboembolism. The assays were compared in 49 patients presenting to the emergency department with clinically suspected deep vein thrombosis or pulmonary embolism. After objective diagnostic imaging, 20 patients were confirmed to have venous thromboembolism. There was strong agreement between the assays in individual patients. Using a cut-off of 500 micrograms/l, both tests had a sensitivity of 100% and therefore a negative predictive value of 100%, however the MDA test would have spared more patients (20% vs. 12%) from further testing if a negative D-dimer was used to rule-out the diagnosis. It was concluded that a rapid, objective latex D-dimer test run on a routine coagulometer (MDA D-dimer) can be used to exclude the diagnosis of venous thromboembolism.

Comparison of an immuno‐turbidometric method (STalia®d‐DI) with an established enzyme linked fluorescent assay (VIDAS®) d‐dimer for the exclusion of venous thromboembolism

“…Using cutoff's of 500 μ g/l FEU on the Vidas (Keeling et al ., 1999), and 500 μ g/l FEU on the STalia (manufacturer's data and confirmed by Receiver Operator Characteristics (ROC) analysis) 42% (42/100) patients were negative (<500) and 46% (46/100) were positive (>500) on both systems. Six per cent (6/100) were positive on the Vidas but negative on the STalia and another 6% (6/100) were positive on the STalia but negative on the Vidas (Figure 1).…”

“…In addition to a high sensitivity, it needs a rapid turn‐around time and a reasonable specificity so that a significant number of patients have the diagnosis excluded. d ‐dimer tests using latex immuno‐turbidimetric methods appear to have test characteristics comparable with those for enzyme‐linked immunosorbent assay (ELISA) methods (Keeling et al ., 1999; Brown et al ., 2003).…”

“…Across a wide range of data (Vidas 83–5656) and (STalia ® <200–>4000), there was good agreement between the two methods. Using cutoff's of 500 μ g/l fibrinogen equivalent units (Keeling et al ., 1999), 42% (42/100) patients were negative (<500) and 46% (46/100) were positive (>500) on both systems. Six per cent (6/100) were positive on the Vidas but negative on the STalia ® and another 6% (6/100) were positive on the STalia ® but negative on the Vidas.…”

Evidence-based management of deep vein thrombosis and pulmonary embolus

“…D-dimer tests should be used in conjunction with pre-test clinical probability scoring because the negative predictive value depends not only on the sensitivity and specificity of the test but also on the prevalence of VTE. Thus, a patient with a negative D-dimer result but a moderate pre-test clinical probability score (see Table 1) still has about a one in 50 risk of VTE, whereas a patient with a high pretest clinical probability score has a one in five risk of VTE even with a negative Ddimer result 3 .…”

Utility of D‐dimer in the diagnosis of cerebral venous sinus thrombosis

“…D-dimer levels were measured using two different assays: an automated enzyme-linked immunosorbent assay (ELISA) (VIDAS; BioMerieux, Durham, NC, USA), and an automated quantatative latex particle immunoassay (MDA; BioMerieux). Cut-off values of > 500 ng mL )1 for the VIDAS assay and 0.5 lg fibrinogen equivalent units (FEU) mL )1 for the MDA assay were used to distinguish positive/negative D-dimer results, as established in the literature for patients with peripheral venous thromboembolism [4,5] There were nine patients with confirmed CVT. Mean age was 38 ± 15 years; six were female and three male.…”