Retrospective review of D-dimer testing for venous thrombosis recurrence risk stratification: is this a useful test...

Predicting the risk of recurrence after venous thromboembolism (VTE) remains an important clinical challenge. Post-anticoagulation cessation D-dimer has previously been shown to be associated with increased VTE recurrence in unprovoked major VTE, however this is not routinely used clinically and has not been validated in provoked VTE and isolated distal DVT (IDDVT). We aimed to retrospectively evaluate this practice in the real-world setting including examining its use in provoked VTE and IDDVT. Consecutive patients diagnosed with DVT or PE between January 2013 and December 2016 were retrospectively evaluated. Clinical features, VTE risk factors, recurrence and bleeding rates were evaluated for patients with normal and abnormal post-anticoagulation D-dimer, as well as those patients who did not undergo D-dimer testing. Patients with active malignancy, superficial vein thrombosis and inadequate follow-up were excluded. Of the 1033 patients with a diagnosis of VTE in the study period, 173 were included in the "D-dimer tested" group, and 254 in the "D-dimer un-tested" comparison group. Abnormal post-anticoagulation D-dimer was significantly associated with VTE recurrence (HR 5.96, 95% CI 2.15-14.57, p < 0.001). Abnormal D-dimer was also associated with high risk of VTE recurrence in travel-provoked VTE (67.61 events per 100 patient-years), and unprovoked IDDVT (HR 14.37, 95% CI 1.75-117.83, p = 0.013). Males with abnormal post-anticoagulation D-dimer were associated with the highest risk of VTE recurrence (HR 12.95, 95% CI 2.78-60.20, p = 0.001). Patients with unprovoked proximal DVT and/or PE who underwent D-dimer testing had a lower VTE recurrence rate compared to those who did not have D-dimer testing (HR 0.28, 95% CI 0.10-0.80, p = 0.017). We confirm the utility of post-anticoagulation cessation D-dimer testing to stratify VTE recurrence risk in the real-world setting, including potentially a role of this assay for predicting subsequent VTE in travel-provoked VTE and unprovoked IDDVT.