Evaluation of the usefulness of a D dimer test in combination with clinical pretest probability score in the prediction and exclusion of Venous Thromboembolism by medical residents.

INTRODUCTION

Venous thromboembolism (VTE) requires urgent diagnosis and treatment to avoid related complications. Clinical presentations of VTE are nonspecific and require definitive confirmation by imaging techniques. A clinical pretest probability (PTP) score system helps predict VTE and reduces the need for costly imaging studies. d-dimer (DD) assay has been used to screen patients for VTE and has shown to be specific for VTE. The combined use of PTP and DD assay may improve exclusion of VTE and safely avoid imaging studies.

MATERIALS AND METHODS

We prospectively used the Wells PTP score and a DD test to evaluate 230 consecutive patients who presented with VTE symptoms. The receiver operating characteristic curve was used to identify a new DD cutoff value, which was applied to VTE diagnosis and compared with the upper limit of locally established reference range for prediction of thrombosis alone and in combination with the clinical PTP score.

RESULTS

We evaluated 118 patients with VTE symptoms fulfilling the inclusion criteria, 64 (54.2%) with clinically suspected deep vein thrombosis (DVT) and 54 (45.8%) with symptoms of pulmonary embolism (PE). The PTP was low in 28 (43.8%) and moderate/high in 36 (56.25%) of the suspected DVT patients, and low in 29 (53.7%) and moderate/high in 25 (46.3%) of the suspected PE patients. Eighteen cases were confirmed by imaging studies: 9 DVT and 9 PE. The agreement between confirmed cases and PTP was significant with PE but not DVT. The negative predictive value for both DVT and PE with current DD cutoff value of <250 μg/L DDU was 100%, whereas with the calculated cutoff the NPV was 88%.

CONCLUSIONS

We confirm that PTP score is valuable tool for medical residents to improve the detection accuracy of VTE, especially for PE. The DD cutoff value of 250 μg/L FEU is ideal for excluding most cases of low PTP; however, the calculated cutoff was less specific for the exclusion of VTE.