Freehand 3D Ultrasound to Measure Thrombus Volume in Patients with Acute Deep-Vein Thrombosis

Abstract

Objective —Current imaging techniques are limited in their ability to quantify thrombus burden, progression, or resolution in patients with acute deep-vein thrombosis (DVT). These assessments are critical measures of therapeutic success when thrombolytic or thrombectomy treatment protocols are used for DVT. We have developed a novel freehand three-dimensional (3D) ultrasound method to measure thrombus volume. In this study, we evaluated the reliability of this new technology. Methods —We studied consecutive hospital inpatients with a first episode of acute DVT. Treatment decisions were not influenced by the study protocol. A combination of routine imaging in grayscale, color-flow, and power-Doppler modes along-with freehand 3D volumetric imaging with a linear transducer attached to an external tracking sensor was performed using a personal computer–based ultrasound scanner. Image-processing software loaded on the personal computer was used to process and reconstruct the serial cross-sectional images of the affected veins and outlined thrombus into a composite 3D image. Imaging and processing was performed twice by one sonographer, and repeated once by another sonographer. Results —The mean age of patients was 56.7 ± 15 years; 40% were women. A total of 70% of DVTs were in the lower extremity and 60% received anticoagulation. Thrombus volume was reliably determined by our protocol. The mean inter- and intraobserver differences in measurements were −0.09 ± 0.59 cm3 and 0.23 ± 0.43 cm3 (mean ± SD), which were well within 2 SD (Bland Altman statistics) of the mean volumes. Measurements of the second sonographer correlated well with those of the first (regression slope = 0.95 ± 0.07, p = 0.001) and those of the first observation correlated with the second observation performed by the same sonographer (regression slope = 0.91 ± 0.03, p = 0.001). Interrater agreement was good between the two sonographers (kappa = 0.71) and between two measurements by the same sonographer (kappa = 0.82). Conclusions —We have developed a novel freehand 3D-ultrasound imaging technique and protocol that reliably measures venous thrombus volume. There is good agreement in measurements between two observers and repeated measurements by one observer. The technique can be readily adapted for routine clinical practice. This protocol will be of increasing value as the appreciation for the deleterious effects of residual thrombus after DVT increases; and the use of aggressive thrombus removal treatments for acute DVT increases.