• J D Douketis, C Gu, A Piccioli, A Ghirarduzzi, V Pengo, and P Prandoni.
• Department of Medicine, McMaster University, Hamilton, ON, Canada. jdouket@mcmaster.ca
• J. Thromb. Haemost. 2009 Apr 1; 7 (4): 546-51.

BackgroundIn patients with venous thromboembolism (VTE), 15-20% will have prevalent cancer when VTE is diagnosed but little is known about such patients' long-term risk, time course and predictors of new cancer.Patients And MethodsWe studied an inception cohort of patients with a first VTE who were not diagnosed with cancer within 3 months after VTE and who had follow-up for up to 120 months. We determined the annual risk for a new cancer [number of events and 95% confidence interval (CI)] per 100 person-years in all patients and in those with unprovoked VTE and identified predictors for new cancer.ResultsWe studied 1852 patients with VTE who received anticoagulant therapy for 12 months (mean) and were followed for 4.2 years (mean). During follow-up, there were 105 (5.7%) patients diagnosed with new cancer during the period after the initial 3 months from diagnosis, for an annual risk of 1.32 (CI, 1.09-1.60) per 100 person-years. The risk for new cancer appeared uniform over time. The annual risk for new cancer was more than 2-fold higher in patients presenting with unprovoked compared with those with provoked VTE [1.76 (CI, 1.39-2.20) vs. 0.83 (CI, 0.58-1.16) per 100 person-years; P<0.001]. Clinical predictors for new cancer were increasing age [hazard ratio (HR), 1.23; CI, 1.05-1.44] and unprovoked VTE (HR, 1.86; CI, 1.21-2.87).ConclusionIn patients with a first VTE and without prevalent cancer, the risk for new cancer is about 1-2% per year, appears to be uniform over time, and is higher in patients with unprovoked VTE and those with advanced age.

22 keywords...

hide…