Dynamic Computed Tomography Angiography as Imaging Method for Endoleak Classification after Endovascular Aneurysm Repair: A Case Series and Systematic Review of the Literature

Abstract

Introduction: This study evaluated our experience with dynamic computed tomography angiography (dCTA) as a diagnostic tool after endovascular aortic aneurysm repair (EVAR) with respect to the endoleak classification and the available literature. Methods: We reviewed all patients who underwent dCTA because of suspected endoleaks after the EVAR and classified the endoleaks in these patients based on standard CTA (sCTA) and dCTA. We systematically reviewed all available publications that investigated the diagnostic accuracy of dCTA compared with other imaging techniques. Results: In our single-center series, 16 dCTAs were performed in 16 patients. In 11 patients, the undefined endoleaks that appeared on sCTA scans were successfully classified using dCTA. In three patients with a type II endoleak and aneurysm sac growth, inflow arteries were successfully identified using dCTA, and in two patients, aneurysm sac growth was observed without a visible endoleak on both sCTA and dCTA scans. The dCTA revealed four occult endoleaks, all of which were type II endoleaks. The systematic review identified six series comparing dCTA with other imaging methods. All articles reported an excellent outcome regarding the endoleak classification. In published dCTA protocols, the number and timing of phases varied greatly, affecting radiation exposure. Time attenuation curves of the current series show that some phases do not contribute to the endoleak classification and that the use of a test bolus improves the timing of the dCTA. Conclusions: The dCTA is a valuable additional tool that can identify and classify endoleaks more accurately than the sCTA. Published dCTA protocols vary greatly and should be optimized to decrease radiation exposure as long as accuracy can be maintained. The use of a test bolus to improve the timing of the dCTA is recommended, but the optimal number of scanning phases is yet to be determined.