• thrombo-embolic disease
• pulmonary embolism
• Diagnostic Tests

The D-dimer is a stable termination product of fibrin degradation. It was introduced as a biomarker of coagulation activation in the early 1970s and first evaluated as a test to exclude venous thromboembolism (VTE) in the 1980s.1 It has since been widely adopted into diagnostic care pathways across the world, despite well-documented issues with variation across laboratory assays and reporting units.2 D-dimer is often viewed as a dichotomous test (positive or negative) based on the standard threshold of 500 ng/mL fibrinogen equivalent units (FEUs) used in original derivation studies, despite being a continuous variable.3 Like so many tests in medicine, there is potential for more effective use.

Several research groups have worked to this end and evaluated the potential impact of adjusting the D-dimer threshold according to age, clinical pre-test probability or gestalt. This approach has clear merit; the specificity of the test is low at the standard threshold (500 ng/mL FEU); continuous/ordinal biomarkers in other clinical models have been shown to enhance risk prediction4; and patients and clinicians have different levels of risk tolerance. In their EMJ paper, Morris et al5 present a feasibility study evaluating the ‘adjust unlikely’ rule for exclusion of pulmonary embolism (PE) in the ED. This rule asks the treating clinician to determine whether they believe PE to be the most likely diagnosis, using only gestalt. PE can be excluded if it is not considered the most likely diagnosis and the D-dimer value is below an age-adjusted threshold. If PE is considered most likely, a standard threshold is …