This study was to determine the clinical diagnostic value of D-dimer for DVT in patients with ischemic stroke.
During July 2013 to December 2014, a cohort study of ischemic stroke patients who presented with symptoms of DVT in
upper or lower extremities was performed, with a total of 255 patients at baseline. D-dimer levels were measured from
each patient using Colour Doppler Ultrasonography (CDUS), and all patients underwent venous duplex examinations. In
ours study, 56 patients were diagnosed as DVT (22.0%). When compared to the patients without-DVT, a significantly increased
trend of plasma D-dimer levels was found in stroke patients with DVT [3.07 (IQR, 2.26-4.05)mg/L VS. 0.54
(IQR, 0.27-1.14) mg/L; P<0.0001]. From the analysis results of the ROC curve, optimal cutoff value was 1.51 mg/L for
diagnosing of DVT (sensitivity: 91.1 %; specificity: 85.4%; the AUC: 0.914 [95%CI, 0.878—0.950; P<0.001]). If cut-off
value of 0.5 mg/L, the diagnosis sensitivity was 100%, the specificity was 46.2%, and the positive predictive value was
34.3%. In addition, 36.1% (92/255) stroke patients who suspected with DVT did not need perform CDUS, and those patients
could be excluded by plasma D-dimer tested. Collectively, plasma D-dimer level may have a guiding meaning for
diagnosing DVT in ischemic stroke patients, and the D-dimer assay is a reliable method for ruling out DVT.