Background The diagnostic algorithm for venous thromboembolism (VTE) currently involves a composite of pre-test probability, D-dimer and imaging. Other laboratory tests, however, may assist in the identification of patients with VTE.

Aim To assess the accuracy of different coagulation tests (D-dimer, thrombin generation, phospholipid-dependent (PPL) clotting time, soluble P-selectin (sP-selectin)) as biomarkers of acute VTE.

Methods Random samples arriving at the Coagulation Laboratory at Mater Dei Hospital (Msida, Malta) from the Accident and Emergency Department with a request for D-dimer measurement were collected between August 2015 and February 2016. The following tests were performed: Innovance D-dimer (Siemens Healthcare Diagnostics), HemosIL D-dimer HS (Instrumentation Laboratory), thrombin generation (using the calibrated automated thrombogram), STA Procoag PPL (Diagnostica Stago) and sP-selectin (Affymetrix; eBioscience). VTE was objectively confirmed by compression ultrasonography, CT pulmonary angiography or ventilation/perfusion lung scan.

Results 100 samples were collected (33 with VTE). A strong positive linear correlation was found between the two D-dimer tests (r=0.97, p<0.001). Patients with VTE showed significantly higher sP-selectin concentrations compared with patients without VTE (75.7 ng/mL vs 53.0 ng/mL, p<0.001). In the random forest plot, the two D-dimer assays showed the highest variable importance, followed by sP-selectin. A sP-selectin cut-off of 74.8 ng/mL was associated with 72.7% sensitivity and 78.2% specificity for acute VTE in our cohort.

Conclusion Our results confirmed D-dimer as the main biomarker of VTE and speculated a role for sP-selectin. The impact of thrombin generation was limited and no role emerged for the PPL clotting time. These observations need to be confirmed in large management studies.