A volumetric index for the quantification of deep venous thrombosis

The evaluation of treatment strategies for deep venous thrombosis (DVT) is assessed through the use of a reliable method of quantifying the extent of the thrombotic process. Previous indices of thrombus burden have suffered from various limitations, including lack of clinical relevance, poor correlation with actual thrombus mass, and failure to include important venous segments in the methodology. The use of a novel scheme of quantifying venous thrombus was evaluated as an alternative method that would avoid some of the drawbacks of existing indices.

The volumes of 14 venous segments (infrarenal inferior vena cava, common iliac, hypogastric, external iliac, common femoral, profunda, superficial femoral, and popliteal and six tibial veins) were calculated from computed tomography (pelvic vein diameter), duplex ultrasound scan (infrainguinal vein diameter), and contrast venography (length of all segments) measurements. A venous volumetric index (VVI) was assigned with the normalization of the values to the volume of a single calf vein. The VVI was validated with the assessment of the ability to discriminate between asymptomatic and symptomatic DVT and between those DVT that were associated with pulmonary emboli and those that were not.

With the imaging data, the VVI ranged from 1 for a single calf vein thrombus to 26 for the infrarenal inferior vena cava. Each VVI unit represented 2.3 mL of thrombus, with a maximum possible score of 63 representing a thrombus burden of 145 mL for a single extremity, including the infrarenal inferior vena cava. In 885 patients with DVT, the VVI ranged from 1 to 56, averaging 3.9 +/- 0.2 in patients who were asymptomatic and 8.7 +/- 0.3 in patients who were symptomatic (P <.001). The VVI was similar in the patients with pulmonary emboli as compared with those without (9.6 +/- 1.2 vs 7.7 +/- 0.2, respectively). In comparison with the three existing methods of quantifying venous thrombus burden, the receiver operating characteristic curve analysis results suggested that the VVI and the Venous Registry index were better than the other two indices in the discrimination of patients with symptomatic and asymptomatic DVT (P <.001).

A novel index of venous thrombus burden, on the basis of actual venous volume measurements, was more accurate than present indices in the differentiation between clinical categories of patients with DVT. As such, it offers an acceptable alternative to current scoring systems.