Surveillance of patients post endovascular aneurysm repair: How does duplex ultrasound compare with computed tomography (CT)?.

Abstract

Endovascular aneurysm repair (EVAR) has become increasingly popular over the last few years, but is associated with complications such as endoleaks, graft fracture and migration (Ouriel, Clair et al. 2003), which prompt urgent intervention. This has necessitated lifelong follow up of patients with CT scans at regular intervals, which has significant cost implications and exposes patients to a large radiation dose. Duplex ultrasound has been proposed as an alternative (Henao, Hodge et al. 2006, Sandford, Bown et al. 2006), but substantial variation has been reported in the literature in terms of its sensitivity and specificity for detecting complications. The aim of this research project was to assess the limitations of duplex ultrasound and to establish whether its sensitivity and specificity in routine clinical imaging for the surveillance of patients post EVAR could be improved to the same level as CT. Novel ultrasound blood and tissue mimicking phantoms were used to assess the limits of ultrasound, and the effects of operator experience, in detecting complications (Ramnarine, Nassiri et al. 1998, Teirlinck, Bezemer et al. 1998, Madsen, Dong et al. 1999, Surry, Austin et al. 2004). The study was divided into a clinical arm and a laboratory-based arm. The clinical arm compared results from CT and Duplex ultrasound for patients post EVAR. In the laboratory, novel tissue and blood mimicking ultrasound phantoms with known geometry, material and blood flow parameters were used to assess the limitations of duplex ultrasound, independent of the operator. The effect of operator experience on the reliable detection and classification of complications was also assessed. In 12 addition, machine dependent parameters were studied to optimise the sensitivity of duplex ultrasound in EVAR evaluations. The results of our laboratory experiments demonstrated that endoleak detection with duplex ultrasound proved more difficult with increasing depth of the endoleak from the viewing position and when the endoleak was in a plane distal to the aortic flow. Our results also demonstrated an increasing trend toward flow detection with increasing flow rates of the endoleak. Our clinical arm corroborated the findings of the laboratory arm with endoleaks that were missed on DUS being identified as slow flow endoleaks located posteriorly and in direct apposition with the stent graft. Although our results did not unequivocally demonstrate the superiority of DUS compared to CT, there was a clear trend towards diagnosis of all endoleaks that required intervention. Our experiments support the use of DUS for the surveillance of patients post EVAR in a complimentary role to CT, thus reducing the substantial radiation exposure for these patients and the associated cost burden to health providers.