Impact of secondary interventions on mortality after fenestrated branched endovascular aortic aneurysm repair
Kristina A Giles,Salvatore T Scali,Benjamin J Pearce,Thomas S Huber,Scott A Berceli,Dean J Arnaoutakis,Martin R Back,Javairiah Fatima,Gilbert R Upchurch Jr,Adam W Beck
DOI: https://doi.org/10.1016/j.jvs.2019.02.029
Abstract:Background: Fenestrated and branched endovascular aortic repair (F/BEVAR) is increasingly used to manage pararenal and thoracoabdominal aortic disease (TAAA). Device-related reintervention after F/BEVAR is common, but little is known about its impact on postoperative mortality. The purpose of this analysis was to describe secondary intervention (SI) after F/BEVAR and determine the impact of these procedures on patient survival. Methods: A single-center review was done on all consecutive F/BEVARs performed from 2010 to 2016. Primary end points were incidence of secondary aortic, branch, and/or access vessel‒related SI, and survival. SI was categorized as minor endovascular (branch restenting, access vessel treatment, or percutaneous coil embolization), major endovascular (new aortic graft placement), or open (bleeding, access vessel, and/or aortic). Kaplan-Meier methodology was used to estimate freedom from SI and survival. Multivariable analysis was used to identify predictors of SI. Results: A total of 308 F/BEVAR procedures were performed (75% physician-modified, 18% custom, 7% Zfen), with 1022 vessels revascularized (celiac, 228; superior mesenteric artery [SMA], 263; renal, 525). There were 117 (39%) extent I-III TAAA, 132 (44%) extent IV TAAA/4-vessel pararenal, and 54 (18%) <4-vessel pararenal repairs performed. Any type of SI occurred in 24% (74) of patients during the mean follow-up of 20 ± 21 months. The majority of reinterventions were endovascular (minor, 53% [n = 39]; major, 32% [n = 24]), whereas 12% (n = 9) were open and 3% (n = 2) hybrid. Primary indication for SI included: 22 (29%) with branch-related endoleaks (1C or III); 15 (22%) with proximal or distal aortic degeneration; 8 (12%) with branch vessel thrombosis/stenosis; 10 (11%) with aortic device type III endoleak/loss of overlap; 4 (6%) with postoperative mesenteric or renal bleeding events; 5 (5%) with type II endoleak; 3 (5%) with access vessel complication; and 2 (3%) with graft infection. Most SIs were elective (65%; n = 48) with the remainder occurring emergently (24%; n = 18) or for symptoms/urgently (11%; n = 8). Compared with endovascular remediation, open SI was more likely to be emergent (89%, 8 of 9; P = .001). Freedom from SI was 80 ± 3% and 64 ± 4% at 1 and 3 years, respectively. One- and 5-year survival with or without SI was: 1 year, 88 ± 4% vs 81 ± 3%; 5 years, 76 ± 5% vs 59 ± 4% (log rank test, P = .06). There was no survival difference based on type of SI (log rank test, P = .3). Extent I-III TAAA (HR, 1.6; 95% CI, 0.98-3.3; P = .06) and history of cerebrovascular disease (HR, 1.8; 95% CI, 0.97-2.6; P = .07) were predictive of SI. Conclusions: SIs after F/BEVAR most frequently involve branch vessel or aortic device remediation procedures; however, they do not negatively impact out-of-hospital survival. These results further highlight the crucial role of imaging surveillance after F/BEVAR to maintain durability. Discussions with patients, periprocedural planning, and the next generation of device design must focus on issues surrounding the risk of device-related SI events.