Long-term results of total endovascular repair of arch-involving aortic pathologies using parallel grafts for supra-aortic debranching

Objective: We evaluated the long-term morphologic and clinical outcomes after thoracic endovascular aortic repair combined with parallel grafts (PG-TEVAR) for arch-involving aortic pathologies. Methods: We performed a retrospective analysis of perioperative and follow-up data of patients who had undergone PGTEVAR at a single vascular surgery center from November 2010 to April 2018. Patients with prior or simultaneous open chest or cervical debranching procedures or arch repair were excluded. The primary endpoint was freedom from overall PG-TEVARerelated reintervention. The secondary endpoints were parallel graft sealing zone failure (presence of gutter-related type I or Ic endoleak), PG failure (occlusion or reintervention), stroke, and 30-day and overall PG-TEVARerelated and all-cause mortality. Kaplan-Meier curves were used to estimate the freedom from reintervention and survival. Receiver operating characteristics curves were used to find the optimal cutoff to prevent type Ia endoleak-related reintervention. Results: A total of 33 patients, including 8 women, with a median age of 74 years (interquartile range, 67-79 years) had undergone PG-TEVAR (chimney, periscope, and sandwich in 20, 15, and 13 patients, respectively) with proximal landing in Ishimaru zone 0, 1, or 2 in 4, 5, and 24 patients, respectively. The aortic pathologies included type B aortic dissection (acute and chronic, eight and six, respectively), degenerative aneurysm (n = 10), type Ia endoleak (n = 3), para-anastomotic/patch aneurysm (n = 4), left subclavian artery aneurysm (n = 1), and traumatic rupture (n = 1). The perioperative stroke rate and 30-day mortality was 6% and 9%, respectively. Direct postoperative computed tomography revealed 28 endoleaks (gutter-related type Ia, 12; gutter-related type Ib, 9; type Ia, 2; type Ic, 2; type III, 1; undetermined, 2) in 27 patients. The technical and clinical success rate was 37% and 30%, respectively. The mean follow-up for survival was 48 +/- 31 months. The latest radiologic follow-up demonstrated 12 remaining and 1 new endoleak. The early and overall PG sealing zone failure and PG failure was 73% and 36% and 9% and 18%, respectively. The overall PG-TEVARerelated reintervention rate was 33% (n = 11). The estimated freedom from overall PG-TEVARerelated reintervention was 68% at 60 months. The main graft oversizing and length oversizing rates were not significantly associated statistically with the type Ia endoleak-related reintervention rate. The PG-TEVARerelated and all-cause mortality were 18% and 34%, respectively. Conclusions: PG-TEVAR for total endovascular repair of arch-involving aortic pathologies resulted in a high rate of type I endoleaks and the need for long-term reintervention. Gutter-related endoleaks might be more frequent than reported and should not be underestimated because they can lead to sac enlargement and reintervention. Frequent radiologic surveillance is mandatory. Further studies comparing PG-TEVAR to other total endovascular alternatives are required to confirm these findings.