Incidence of Patients with Deep Vein Thrombosis and Pulmonary Embolism in Elevated D-Dimer Levels

Abstract

Elevated levels of D-dimers with unknown significance often irritate in clinical practice. No detailed information in medical literature exists about the percentage of deep vein thrombosis (dvt) and pulmonary embolism (pe). 236 patients with elevated levels of D-dimers were divided by pretest-probability for deep vein thrombosis and pulmonary embolism (Wells' Score). Subsequently every patient was evaluated by compression ultrasound of deep vein thrombosis of the legs. In cases of medium or high probability of pulmonary embolism an additional lung perfusion scintigraphy was performed. The statistic evaluation of the medians of D-dimers was performed by Wilcoxon-Mann-Whitney-Test for asymptotic significance. 39% of the patients proved to have deep vein thrombosis, 23% proved to have pulmonary embolism. The median of D-dimers in patients without dvt (1356 ng/ml), p = 0.01) was significantly lower in comparison to patients with dvt of pelvic veins (3541 ng/ml), upper legs (2694 ng/ml) and lower legs (1796 ng/ml). In patients with pulmonary embolism the D-dimers median was significantly higher in comparison to patients without pe (1518 ng/ml, p = 0.01). In clinical practice elevated D-dimers correlate in 39% with dvt and 23% with pe. Elevated D-dimer levels should always be correlated with pretest-probability in order to avoid unnecessary compression ultrasound.