Diagnostic value of D-dimer in patients with a low pretest probability of deep venous thrombosis of lower extremities

We aimed to assess the accuracy of a diagnostic strategy including broth clinical assessment and determination of D-dimer (DD) in patients with clinically suspicion of low pretest probability of deep venous thrombosis (DVT). 149 outpatients (mean age 69; SD 16) with clinically suspected proximal DVT attending our Emergency Department and classified as low pretest probability were included in an observational prospective study. In patients with a DD (STA Liatest D-Di, Diagnostica Stago, Asnières sur Seine, France) concentration below the cut-off value (0.4 ng/ml) the diagnosis of DVT was readily ruled out, whereas those individuals with a positive DD result underwent compression Doppler venous ultrasound. A 3-month clinical follow-up was carried out in those patients in whom a diagnosis of DVT was initially excluded. Only 2 cases of DVT were confirmed (prevalence 1.3%; CI 95%, 0.2-5.3). In 47.6% of cases, a DD negativity ruled out the diagnosis of DVT. The rate of negative DD results was significantly lower in patients below 70 years of age as compared to older patients (73.6 versus 33.3%) (p < 0.001). Sensitivity, specificity, positive predictive value and negative predictive value of DD in our series were 100% (CI 95%,19.7-95.1), 48.3% (CI 95%, 40.0-56.7), 2.6% (CI 95%, 0.4-9.8) and 100% (CI 95%, 93.6-99.8) respectively, the latter being similar in the two age groups. No case of DVT was diagnosed during the follow-up period. In patients with a low pretest probability of DVT a negative DD result reliably and safely rules out such diagnosis. However, the diagnostic value of DD is lower in elderly patients (>= 70 years of age) due to a lower rate of negative results in these individuals.