Assessment of changes in stent graft geometry after chimney endovascular aneurysm sealing

Background: Chimney endovascular aneurysm sealing (ch-EVAS) could potentially minimize gutter-associated endoleaks in patients with juxtarenal abdominal aortic aneurysms resulting from the use of the conformable endobags surrounding the chimney stent grafts (ch-SGs). The aim of the present study was to quantify the (non)apposition of the endobags in the proximal aortic neck, migration of the endograft stent frames, and changes in geometry of the ch-SGs during the follow-up period.

Methods: The prospective data from 20 patients undergoing elective ch-EVAS were retrospectively reviewed. The aortic anatomy was analyzed on preoperative and postoperative computed tomography scans. The (non)apposition of the endobags in the aortic neck, Nellix (Endologix, Irvine, Calif) stent frame migration, and chimney graft geometry and migration were assessed.

Results: The median preoperative infrarenal neck length was 4.0 mm (interquartile range [IQR], 0-6.0 mm). The median seal length in the juxtarenal aortic neck at the first follow-up was 23.0 mm (IQR 18.0-30.8 mm). Five type IA endoleaks were identified on postoperative imaging; one at 1 month and four newly diagnosed at 1 year. Of these five type IA endoleaks, two were type Is1 (not extending into the aneurysm sac) and did not need reintervention and other three were type Is2 (extending into the aneurysm sac). One of these patients died of malignancy before reintervention could be performed. Bilateral ch-SG occlusions in one patient were documented at the 1-month follow-up (patient needed hemodialysis) and two patients with a new single ch-SG occlusion were found at the 1-year follow-up. No reinterventions were performed for the ch-SG occlusions. An occluded Nellix stent frame in one patient was treated with femorofemoral crossover bypass. Kaplan-Meier estimate of reintervention-free survival was 85.0% after 1 year. Migration ≥5 mm of the proximal end of the Nellix stent frames was observed in 20.0% of the patients, but no reintervention was performed at the 1-year follow-up. Imaging showed 20.1% of the available sealing surface was not used, and the nonapposition surface increased to 30.6% of the preoperative aortic neck surface at 1 year. Median migration was 3.5 mm (IQR, 2.4-5.0 mm) and 3.1 mm (IQR, 2.0-4.8 mm) for the left and right proximal end of the Nellix stent frames, respectively, and was 3.0 mm (IQR, 2.2-4.8 mm) for the proximal end of the ch-SGs at 1 year of follow-up.

Conclusions: Substantial distal migration of the Nellix endograft and positional changes of the ch-SGs in the juxtarenal aortic neck were observed at 1 year of follow-up, resulting in a 25.0% type IA endoleak rate, with three of these type IA endoleaks extending into the aneurysm sac. The reintervention-free survival rate was 85.0% at 1 year in this cohort of 20 patients. Careful follow-up after ch-EVAS is advised because changes are often subtle. The authors have stopped the ch-EVAS procedure so far. Long-term follow-up data on the stability of the Nellix endograft and the consequences of migration on ch-SGs is required before this technique should be used in clinical practice.