The predictive value of D-dimer measurement for cancer in patients with deep vein thrombosis.

BACKGROUND AND OBJECTIVES Venous thromboembolism can be related to malignancy, but routine screening for cancer in patients with deep vein thrombosis (DVT) is not a recommended practice. The aim of this study was to evaluate the value of D-dimer concentration in predicting cancer in patients with DVT. DESIGN AND METHODS D-dimer levels were measured in outpatients presenting with DVT. In a proportion of patients, D-dimer levels were measured daily for 4 days. The occurrence of malignancy was documented. RESULTS Patients were followed for a median of 34 months. Fifty (23%) of 218 patients with thrombosis had cancer in the study period including 14 who developed cancer during the follow-up. High initial D-dimer levels (levels > 4000 mg/L) were associated with more cancer during follow-up than were lower D-dimer levels: 13% versus 4% (p=0.048). High D-dimer levels after 4 days of treatment were associated with a 15% prevalence of cancer whereas the prevalence in patients with lower D-dimer levels was 5% (p=0.1). The total cancer prevalence (including cancer diagnosed before thrombosis) in patients with initial D-dimer levels < 4000 mg/L was 16% compared to 32% in patients with higher levels (p=0.009, RR=2.0). After 4 days of treatment, total cancer prevalences were 14% and 46%, respectively (p=0.02, RR=3.4). In patients aged < 60 years, initial D-dimer levels of < 4000 mg/L were associated with a cancer prevalence of 3% whereas higher levels were associated with a prevalence of 23% (p=0.001, RR=8.6). After 4 days of treatment, the prevalences associated with lower and higher levels of D-dimer were 0% and 100%, respectively (p=0.003). There was no difference in older patients. INTERPRETATION AND CONCLUSIONS High D-dimer concentrations at presentation or during the first days of treatment are indicators of an increased probability of overt or occult forms of cancer, especially in patients under 60 years old.