Management of failed endovascular aortic aneurysm repair with explantation or fenestrated-branched endovascular aortic aneurysm repair

• Published in: Journal of vascular surgery 2018-12, Vol.68 (6), p.1676-1687.e3
• Main Authors: Dias, Agenor P., Farivar, Behzad S., Steenberge, Sean P., Brier, Corey, Kuramochi, Yuki, Lyden, Sean P., Eagleton, Matthew J.
• Format: Article
• Language: English
• Subjects: Abdominal aortic aneurysm; Endoleak; EVAR; Explantation; Failed EVAR; Fenestrated-branched endograft; Open conversion; Thoracoabdominal aortic aneurysm
• ISSN: 0741-5214; 1097-6809
• DOI: 10.1016/j.jvs.2018.03.418

The incidence of failed endovascular aneurysm repair (EVAR) is increasing, and understanding the different methods of management and repair is paramount. The objective of this study was to evaluate the clinical management and rescue of failed EVAR by either explantation or fenestrated-branched EVAR (F/B-EVAR). A retrospective analysis (1999-2016) of 247 patients who underwent either explantation (n = 162) or F/B-EVAR (n = 85) for failed EVAR was performed. F/B-EVAR was performed under a physician-sponsored investigational device exemption. Demographics of the patients, clinical presentation and failure etiology, perioperative management, rate of reinterventions, morbidity, and mortality were analyzed. Those undergoing surgical explantation were compared with those undergoing F/B-EVAR conversion. Statistical analysis included multivariable logistic regressions, Fisher exact test, and χ2 test. The majority of patients were male (n = 216 [87%]), with a mean age of 75 years (range, 50-93 years). The mean time from primary EVAR was higher in F/B-EVAR (46 ± 7 months vs 69 ± 41 months; P < .001). Graft manufacturer did not differ between those requiring explantation and those having endovascular rescue (P = .170). All emergencies (n = 24 [10%]) and infections (n = 28 [11%]) were treated with open conversion. Endoleak was the most common reason for failure in both explantation and F/B-EVAR groups (75% vs 64%, respectively; P = .052). Type I endoleak was the most common endoleak reported in both groups, occurring more frequently in F/B-EVAR (64% vs 40%; P < .001); type II endoleak was more common in those undergoing open repair (28% vs 2%; P < .001). Graft migration (12% vs 26%; P = .005) and neck degeneration/disease progression (14% vs 59%; P < .001) were more prevalent in F/B-EVAR, but aneurysm enlargement was more common in explantation (68% vs 33%; P < .001). Thirty-day reintervention rates did not differ between F/B-EVAR and explantation (odds ratio, 0.6258; 95% confidence interval, 0.2-1.86; P = .4115); however, 30-day mortality was lower in the F/B-EVAR group (5% vs 10%; P = .0192). Similarly, aneurysm-related mortality was also lower in the F/B-EVAR group (hazard ratio, 0.0683; 95% confidence interval, 0.01-0.44; P = .0048). A subset analysis excluding emergencies and infections did not alter the lack of difference in terms of freedom from reinterventions (P = .1175), 30-day mortality (P = .6329), or aneurysm-related mortality (P = .7849). Explantation and