Recurrent endoleak detection and measurement of aneurysm size with CTA after coil embolization of endoleaks

PURPOSE: The optimal modality for following aneurysm size and detecting endoleaks after endovascular aneurysm repair (EVAR) remains controversial. Computed tomographic angiography (CTA) has been widely employed but can be limited by metal artifact from stents, which is exacerbated by embolization coils placed during the treatment of type 2 endoleaks. The authors assessed interobserver agreement of CTA for measuring aneurysm size and presence of recurrent endoleak in patients with prior coil embolization of type 2 endoleaks. MATERIALS AND METHODS: A total of 65 CTAs were retrospectively reviewed in a cohort of 27 patients (25 men; two women; mean age, 77.4 years) who had prior endoleak embolization after EVAR. Endoleak embolizations included transarterial (n = 8) and translumbar (n = 19) approaches. In each patient, maximal aneurysm diameter and presence/absence of recurrent endoleak was measured independently by two observers. Cohen's Kappa statistic was used to assess interobserver agreement, as well as paired two-tailed Student t tests for aneurysm diameter. RESULTS: Recurrent type 2 endoleaks were detected with CTA in eight of 27 patients (30%) and on 13 of 65 CTAs (20%). A high degree of correlation (98.5%) was also seen between the two observers for presence of endoleak (Kappa = 0.95). Mean aneurysm diameter for the entire cohort correlated closely between both observers: 54.8 mm +/- 1.1 for observer A and 54.9 mm +/- 1.1 for observer B (P =.66). There was a disagreement between the readers of greater than 2 mm regarding aneurysm size in 13.8% of the CTAs (nine of 65 CTAs). CONCLUSION: Despite the presence of streak artifact on CTA following coil embolization of type 2 endoleaks, CTA remains a useful study for following patients. The presence of embolization coils does not prevent CTA measurement of aneurysm diameter and detection of recurrent endoleak with a high degree of interobserver agreement.