TORONTO – The risk of recurrence following an initial episode of venous thromboembolism is highest in the first 3 months, but remains high for up to 3 years, according to findings from a population-based study involving 2,989 adults.

Over a mean of 23 months (median, 30 months), there were 329 VTE recurrences in the study subjects. Cumulative incidence rates were 5.1% at 3 months, and 14.5% at 3 years. The corresponding rates were 8.7% and 24.8% among those with active cancer, 5.2% and 13.0% among those with provoked VTE, and 3.8% and 13.1% among those with unprovoked VTE, Dr. Wei Huang reported at the International Society on Thrombosis and Haemostasis congress.

Courtesy Yale Rosen/Wikimedia Commons

Independent predictors of recurrence within 3 years after the index event were active cancer with chemotherapy (hazard ratio, 2.59), active cancer without chemotherapy (HR, 1.59), hypercoagulable state (HR, 2.53) superficial thrombophlebitis (HR, 1.62), varicose vein stripping (HR, 1.75), and inferior vena cava (IVC) filter placement (HR, 2.04), said Dr. Huang of the University of Massachusetts, Worcester.

Individuals included in the study were all residents of the Worcester Metropolitan Statistical Area (WMSA) who had a validated diagnosis of acute first-time deep vein thrombosis and/or pulmonary embolism in a hospital or ambulatory care center that provided short-term care for WMSA residents between 1999 and 2009. Medical records and national and local death registry data were reviewed to examine outcomes up to 3 years after the index event.

Subjects were adults with a mean age of 64 years; 44% were men, and 94% where white. Pulmonary embolism with or without deep vein thrombosis occurred in 42%, and 17% of cases were associated with cancer, 43% involved provoked VTE, and 40% involved unprovoked VTE.

Provoked VTE was defined as VTE occurring within 3 months of a prior surgery, pregnancy, trauma, fracture, or hospitalization in patients without presence of active cancer.

Though limited by the lack of information about variations in physician practices across regions, and by the high proportion of white resident in the WMSA, which both raise questions about whether the findings are generalizable to the U.S. population, the identification of these predictors could allow for improved estimation of risk for individual patients, and may aid in the design of new interventional studies, Dr. Huang concluded.

This study was supported by the National Institutes of Health.