The D-dimer levels are considered to be useful for the diagnosis of thrombosis, and they can be clinically used as a negative predictive value (NPV). However, evidence for the efficacy of diagnosing thrombosis based on the D-dimer levels is still not well established. The present study was designed to evaluate the cut-off values of D-dimer levels as a negative predictor for thrombosis. The plasma concentrations of D-dimer were measured in inpatients suspected of having thrombosis, and then the findings were evaluated to assess the correlation with the diagnosis of thrombosis. In healthy volunteers, the median value of VIDAS-D-dimer was 0.12 microg/ml, and the 95% confidence interval was from 0.05 to 0.38 microg/ml. However, the plasma D-dimer levels were significantly higher in patients with thrombosis than in those without thrombosis; there was no significant difference in D-dimer levels among various thromboses such as pulmonary embolism (PE), deep vein thrombosis (DVT), and disseminated intravascular coagulation (DIC). The NPV for venous thromboembolism was 100% in patients with 0.5 microg/ml VIDAS-D-dimer and 1.2 microg/ml LPIA-D-dimer levels. Elevated D-dimer levels might indicate a high risk of thrombosis, especially DVT/PE, and they are thus considered to be useful as a negative predictor for thrombosis.