D‐dimer at venous thrombosis diagnosis is associated with risk of recurrence

Essentials Whether D‐dimer at incident venous thromboembolism (VTE) can predict recurrence‐risk is unknown. We explored this association in 454 cancer‐free patients with a first lifetime VTE. A low D‐dimer at first VTE diagnosis was associated with a low recurrence risk. The association was predominant in patients with deep vein thrombosis and unprovoked VTE. Click to hear Dr Cannegieter's presentation on venous thrombosis: prediction of recurrence Summary Background Venous thromboembolism (VTE) is a common disease with a high recurrence rate. D‐dimer measured after cessation of anticoagulant therapy predicts recurrence, and is used to decide on treatment prolongation. However, whether D‐dimer measured at first VTE diagnosis can be used to assess recurrence‐risk is unknown. Aims To investigate the association between D‐dimer, measured at first VTE diagnosis and risk of recurrent VTE. Methods Information on clinical risk factors and laboratory markers were collected in 454 cancer‐free patients with a first VTE. Recurrent VTEs and deaths during follow‐up (1994–2012) were recorded. Results During a median follow‐up of 3.9 years, 84 patients experienced a recurrent VTE. The crude recurrence rate was 1.7 (95% confidence interval [CI], 1.0–2.9) per 100 person‐years in the lower quartile of D‐dimer (≤ 1500 ng mL−1), and 4.9 (95% CI, 3.9–6.1) per 100 person‐years in the upper three quartiles combined, yielding an absolute risk difference of 3.2 per 100 person‐years. Patients with D‐dimer ≤ 1500 ng mL−1 had 54% lower recurrence‐risk than patients with D‐dimer > 1500 ng mL−1 (HR, 0.46; 95% CI, 0.25–0.82). The association was particularly pronounced among patients with unprovoked events and deep vein thrombosis, showing a 66% (HR, 0.34; 95% CI, 0.15–0.74) and 68% (HR, 0.32; 95% CI, 0.14–0.71) lower recurrence risk among patients with D‐dimer ≤ 1500 ng mL−1, respectively. Conclusions A low D‐dimer (≤ 1500 ng mL−1) measured at first VTE diagnosis was associated with a low recurrence risk, particularly among patients with DVT and unprovoked events. Our findings suggest that a clinical decision to avoid prolonged anticoagulant treatment could be considered based on low D‐dimer at the time of VTE diagnosis.