Risk factors for target vessel endoleaks after physician-modified fenestrated or branched endovascular aortic repair for postdissection thoracoabdominal aortic aneurysms.

OBJECTIVE: Patients with postdissection thoracoabdominal aortic aneurysms (TAAAs) have been more likely to develop endoleaks than those with degenerative TAAAs after fenestrated or branched endovascular aortic repair (F/BEVAR). In the present study, we aimed to determine the risk factors for target vessel (TV)-related endoleaks after visceral segment F/BEVAR for postdissection TAAAs. METHODS: We performed a retrospective analysis of all patients with degenerative and postdissection TAAAs treated with F/BEVAR between 2017 and 2021. All the patients had undergone computed tomography angiography before and 3 months, 6 months, and annually after discharge. Two experienced vascular surgeons had used data from computed tomography angiography and vascular angiography to judge the presence of endoleaks. The study end points were mortality, aneurysm rupture, and the emergence of and reintervention for TV-related endoleaks. RESULTS: A total of 195 patients (mean age, 66 ± 10 years; 69% men) had undergone F/BEVAR for 99 postdissection TAAAs and 96 degenerative TAAAs. During a mean follow-up of 16 ± 12 months, we found that the patients with postdissection TAAAs were younger (age, 64 ± 10 years vs 69 ± 9 years; P = .001), had required more prior aortic repairs (58% vs 40%; P = .012), and had had a higher body mass index (26.1 ± 3.4 kg/m2 vs 24.8 ± 3 kg/m2; P = .008), a larger visceral segment aortic diameter (47.1 ± 7.5 mm vs 44.5 ± 7.5 mm; P = .016), and more TV-related endoleaks (18% vs 7%; P = .023) compared with those with degenerative TAAAs. Of the 99 patients with postdissection TAAAs, 327 renal-mesenteric arteries were revascularized using 12 scallops, 141 fenestrations, and 174 inner or outer branch stents. A total of 25 TV-related endoleaks were identified among 18 patients during follow-up, including 6 type Ic (retrograde from the distal end of the branch), 3 type IIIb (bridging stent fabric tear), and 16 type IIIc endoleaks (detachment or loose connection of the bridging stent). The patients with an endoleak had had a larger visceral aortic diameter (52.7 ± 6.4 mm vs 45.8 ± 7.2 mm; P < .001) and had undergone revascularization of more TVs (3.7 ± 0.7 vs 3.2 ± 0.9; P = .032). In contrast, true lumen compression did not seem to affect the occurrence of TV endoleaks (39% vs 27%; P = .323). The use of presewn branch stents in the fenestration position was associated with a lower risk of TV-related endoleaks (5% vs 11%; P = .025). In addition, TVs derived entirely or partially from the false lumen were more prone to the development of endoleaks after reconstruction (19% vs 4% [P < .001]; and 15% vs 4% [P = .047], respectively). CONCLUSIONS: We found that patients with postdissection TAAAs were more likely to have TV-related endoleaks after F/BEVAR in the visceral region than those with degenerative TAAAs. Additionally, patients with a larger aortic diameter and a greater number of fenestrations in the visceral region were more likely to have experienced TV-related endoleaks. Branch vessels deriving from the false lumen were also more likely to develop endoleaks after reconstruction, and prefabricated branch stents were related to a lower possibility of TV-related endoleaks.