Geometric and hemodynamic analysis of fenestrated and multibranched aortic endografts

The objective of this study was to determine the influence of hemodynamic force on the development of type III endoleak and branch thrombosis after complex endovascular thoracoabdominal aortic aneurysm repair.

Patients with thoracoabdominal aortic aneurysm, within surgical range, treated with a fenestrated or branched endovascular aneurysm repair from 2014 to 2018 and with 3-month control computed tomography angiography were selected. Demographic variables, aneurysm anatomy, and endograft conformation were analyzed retrospectively from a prospective registry. The hemodynamic force was calculated using the mass and momentum conservation equations.

Twenty-eight patients were included; the mean follow-up period was 24.7 ± 19.3 months. There were 102 abdominal vessels successfully catheterized (19 celiac arteries, 29 superior mesenteric arteries, 27 right renal arteries, 26 left renal arteries, and 1 polar renal artery). The rate of type III endoleak was 11.5% (n = 12); six cases were associated with branches that received two stents (P < .001). A higher rate of endoleak was observed with wider stents (8.50 ± 1.0 mm vs 7.17 ± 1.3 mm; P = .001) but not with longer stents (P = .530). All cases of type III endoleak affected visceral arteries (eight celiac arteries and four superior mesenteric arteries). The freedom from type III endoleak at 24 months was 86%. The rate of thrombosis was 5.9% (n = 6). A higher rate of thrombosis was observed in smaller vessels (5.00 ± 1.3 mm vs 7.16 ± 1.8 mm; P = .001), with higher stent oversizing (36.87% ± 23.6% vs 5.52% ± 15.0%; P < .001), and with a higher angle of curvature (124.33 ± 86.1 degrees vs 57.71 ± 27.9 degrees; P < .001). All cases of thrombosis were related to renal arteries (two left renal arteries, two right renal arteries, and two polar renal arteries). The freedom from thrombosis at 24 months was 92%. The area under the curve for the angle of curvature was 0.802 (95% confidence interval, 0.661-0.943; P = .013), and the cutoff point was established at 59.5 degrees (sensitivity, 100%; specificity, 60.4%). The receiver operating characteristic curve for the stent oversize showed an area under the curve of 0.903 (95% confidence interval, 0.821-0.984; P = .001), and the cutoff point was 14.5% (sensitivity, 100%; specificity, 77.1%). A higher hemodynamic force was associated with thrombosis (23.35 × 10^-3 N ± 18.7 × 10^-3 N vs 12.31 × 10^-3 N ± 6.8 × 10^-3 N; P = .001) but not with endoleak (P = .796). The freedom from endoleak and thrombosis at 24 months was 86% and 90%, respectively.

Longer stents should be preferred to avoid type III endoleak. A higher angle of curvature leads to a higher hemodynamic force that results in a higher rate of thrombosis. Accordingly, we recommend maintaining the angle of curvature under 59.9 degrees. Small vessels and excessive stent oversizing entail a higher risk of thrombosis; as such, we advise a maximum stent oversize of 14.5%. Renal arteries are more susceptible to thrombosis, whereas visceral arteries are more prone to endoleak.