Three-dimensional contrast-enhanced ultrasound improves endoleak detection and classification after endovascular aneurysm repair

2017 
Abstract Background Three-dimensional contrast-enhanced ultrasound (3D-CEUS) is a novel technology allowing surgeons to view duplex ultrasound images in three dimensions with ultrasound contrast highlighting blood flow in endoleaks after endovascular aneurysm repair (EVAR). It potentially reduces the need for computed tomography angiography (CTA) and catheter angiography. This study compares 3D-CEUS with both CTA and the final vascular multidisciplinary team (MDT) diagnosis using all available imaging. Interoperator variability for detection of endoleak and the influence of 3D-CEUS on patient management were studied. Methods A consecutive 100 patients undergoing CTA for EVAR surveillance were invited to undergo standard CEUS and 3D-CEUS on the same day, with 3D-CEUS reported independently by two blinded vascular scientists. Presence and type of endoleak were compared between CTA, standard CEUS, 3D-CEUS, and the final diagnostic decision made in the vascular MDT meeting. Interoperator reliability of 3D-CEUS was analyzed using the κ statistic. Results The 100 paired CTA, CEUS, and 3D-CEUS studies were analyzed. Compared with CTA, the sensitivity, specificity, positive predictive value, and negative predictive value of 3D-CEUS to endoleak were 96%, 91%, 90%, and 96%, respectively. Compared with the MDT decision with access to all imaging modalities, the sensitivity, specificity, positive predictive value, and negative predictive value of 3D-CEUS were 96%, 100%, 100%, and 96%. The κ statistic for interoperator agreement was 0.89. Conclusions 3D-CEUS was more sensitive and accurate than CTA for endoleak detection and classification after EVAR. 3D-CEUS is now our initial investigation of choice in cases of sac expansion during duplex ultrasound follow-up or if there is diagnostic uncertainty on standard duplex ultrasound or CTA.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    20
    References
    21
    Citations
    NaN
    KQI
    []