Diagnostic Accuracy in Acute Venous Thromboembolism: Comparing D-Dimer, Thrombin Generation, Overall Hemostatic Potential, and Fibrin Monomers

Abstract

Introduction For acute venous thromboembolism (VTE), a biomarker with higher specificity than D-dimer would be of great clinical use. Thrombin generation and overall hemostatic potential (OHP) reflect the hemostatic balance by globally assessing multiple coagulation factors and inhibitors. These tests discriminate between healthy controls and patients with a prothrombotic tendency but have yet to be established as clinical biomarkers of VTE. Objective This study compares endogenous thrombin potential (ETP) and OHP to D-dimer and fibrin monomers (FM) in outpatients with suspected VTE. Methods A cross-sectional diagnostic study where 954 patients with suspected pulmonary embolism or deep venous thrombosis were recruited consecutively from the medical emergency department at Karolinska University Hospital. D-dimer, FM, OHP, and ETP were analyzed in a subpopulation of 60 patients with VTE and 98 matched controls without VTE. VTE was verified either by ultrasonography or computed tomography and clinical data were collected from medical records. Results Compared with healthy controls, both VTE and non-VTE patients displayed prothrombotic profiles in OHP and ETP. D-dimer, FM, ETP area under the curve (AUC), and ETP T _lag were significantly different between patients with VTE and non-VTE. The largest receiver-operating characteristic AUCs for discrimination between VTE and non-VTE, were found in D-dimer with 0.94, FM 0.77, and ETP AUC 0.65. No useful cutoff could be identified for the ETP or the OHP assay. Conclusion Compared with D-dimer, neither ETP nor OHP were clinically viable biomarkers of acute venous thrombosis. The data indicated that a large portion of the emergency patients with suspected VTE were in a prothrombotic state.

Keywords: clinical studies; deep vein thromboses; global hemostatic assays; predictive value of tests; pulmonary embolism.

Fig. 1

Flowchart of study samples. Matched controls were selected for all patients with venous thromboembolism (VTE). Only a subset of the 940 samples could be stored in the acute clinical chemistry laboratory. All included samples ( n  = 174) were analyzed by overall hemostatic potential (OHP) and endogenous thrombin potential (ETP). Nine patients were excluded due to prior anticoagulant treatment and seven due to low technical quality. Primary analysis in this study was performed on 60 samples with VTE and 98 without VTE.

Fig. 2

(A) Overall hemostatic potential (OHP), (B) OCP, (C) OFP, (D) fibrinogen, (E) endogenous thrombin potential area under the curve (ETP AUC), (F) ETP C _max , (G) ETP T _lag , and (H) ETP T _max . Difference in global hemostatic parameters between patients with no venous thromboembolism (VTE) [white], VTE [gray], and healthy controls [light gray]. Box and whisker plots displaying medians [mid-line], interquartile range (IQR) [box], 1.5 × IQR [whisker], outliers > 1.5 × IQR [ring], and extreme outliers > 3 × IQR [asterisk].

Fig. 3

(A) Age, (B) gender, (C) previous venous thromboembolism (VTE), and (D) platelet inhibitors. Box and whisker plots of endogenous thrombin potential area under the curve (ETP AUC) and overall hemostatic potential (OHP) related to age, gender, and previous VTE (samples in analysis = 125). No VTE [white], VTE [gray], medians [mid-line], interquartile range (IQR) [box], 1.5 × IQR [whisker], outliers > 1.5 × IQR [ring], and extreme outliers > 3 × IQR [asterisk].

Fig. 4

Receiver-operating characteristic (ROC) curves for global hemostatic assays, D-dimer, and fibrin monomers. D-dimer and fibrin monomers are plotted in green, overall hemostatic potential (OHP) parameters in blue, and endogenous thrombin potential (ETP) parameters in red-orange. ETP T _max (light orange) is negative because it is the only parameter where a smaller test result indicates a more “positive” test.