Serial D-Dimer Measurements Are Useful in Predicting the Recurrence of Venous Thromboembolism after Discontinuation of Anticoagulation

The optimal duration of anticoagulation following a venous thromboembolism (VTE) is influenced by the site of the event and the presence of risk factors. After anticoagulation is discontinued, approximately 10% of patients will have a recurrent event. At present, there is no accurate method of identifying this group of patients, in whom the benefits of reanticoagulation may outweigh the bleeding risks. In 2006, Palaretti et al reported the use of a single qualitative d-dimer measurement, 4 weeks after stopping anticoagulation for a spontaneous VTE, to identify patients at increased risk of a recurrent VTE. Our observational study investigates whether serial quantitative d-dimer measurements, after the discontinuation of anticoagulation for VTE, are of use in identifying patients at higher risk of VTE recurrence. We followed an unselected group of patients attending a hospital based Thrombosis Clinic in whom anticoagulation for VTE was stopped. Over a 2 year period from July 2005 to July 2007, anticoagulation was discontinued in 216 patients (112 females, 104 males) after a median period of 6 months’ treatment (range 2–324 months). The patients ranged in age from 16–88 years, with a mean age of 54 years. Of the group, 146 had been anticoagulated for a deep vein thrombosis (DVT), 59 for pulmonary embolism (PE) +/− DVT, and 11 had been anticoagulated for VTE at other sites. Major risk factors for VTE (recent surgical procedure, malignancy, pregnancy) were present at diagnosis in 80 patients. Minor risk factors (minor trauma, prolonged travel or immobility, hormone therapy) were present in 61 patients, while no risk factors were identified in 69 patients. Presence of risk factors was unknown in the remaining 6 patients. After discontinuation of their anticoagulation, patients were followed up in the Thrombosis Clinic for a median of 14.5 months (range 0–41 months). D-dimer measurements were recorded at the point of stopping anticoagulation, 4 weeks later, and then on subsequent clinic visits. Quantitative d-dimer results were obtained using an automated latex immunoassay (Instrumentation Laboratories, d-dimers HS) on an ACL TOP CTS analyser. Recurrence of VTE occurred in 23 of the 216 (10.7%) patients. D-dimer results off anticoagulation were available in 207 patients. Forty six patients had repeatedly high d-dimer measurements (> 300ng/ml) off anticoagulation. D-dimers remained within the normal range in 112 patients. In 33 patients, d-dimers were initially normal, but subsequently became high, while the opposite was true in 16 patients, where initially high d-dimers later fell into the normal range. Of the recurrences, 8 occurred in the group who had repeatedly high d-dimers after anticoagulation was stopped (17.4%), whilst only 3 recurrences occurred in patients whose d-dimers were consistently normal (2.7%). In the group whose d-dimers were initially normal, but subsequently became high during follow up, there were 6 recurrences (18.2%). There was 1 recurrence in the group whose d-dimers were initially high, but then became normal. In the 9 patients in whom serial d-dimer results were not available, 5 recurrent VTE were observed: 4 of these occurred within 4 weeks of the patient stopping anticoagulation. After statistical analysis of the data, taking into account the nature of the original event, age, gender and d-dimer measurements, a high d-dimer off anticoagulation was the only significant independent variable predicting for recurrence of VTE (Chi-square 13.1 p<0.00001). This data supports a role for monitoring d-dimer measurements in identifying patients at increased risk of VTE recurrence. Further investigation is needed to establish the benefits of reanticoagulation in these patients.