Assessment of risk factors for mortality, endoleak and late reintervention after EVAR

Introduction. Endoleak (EL) and late reintervention represent the Achilles heel of endovascular repair of abdominal aortic diseases (EVAR) if compared to open surgical repair (OSR), as they countermand the early advantage of the former over the latter in terms of postoperative mortality and morbidity. Moreover, differently from OSR, they mandate a long-term postoperative surveillance, which increases the overall costs of EVAR. Aim of our study was to review our experience with EVAR, analyzing any preoperative and intraoperative factor which could predict a late reintervention or occurrence of EL and help in a proper selection of patients before the procedure. Materials and methods. Data of all consecutive patients who underwent EVAR from January 2003 and May 2012 at our Institution were retrospectively collected. Outcomes were analyzed to evaluate any factor which could affect survival, the occurrence of EL or the need for reintervention. The following items were specifically assessed: history of smoke, assumption of antiplatelet/anticoagulant drugs, sac diameter, proximal neck (diameter, length and angle), percentage of circumference of the sac covered by thrombus, number of patent lumbar arteries arising from the sac, patency of inferior mesenteric artery (IMA), graft oversizing > 25%, hypogastric arteries coverage, IMA or lumbar preoperative embolization. P value < .05 was considered statistically significant. Results. One-hundred and sixty patients (143 men, 89.4%; median age 77 years, range 45-92) underwent EVAR, most of them for degenerative infrarenal AAA (139 patients, 86.8%). Seven procedures were performed in an emergent setting for acute contained rupture; 8 patients were symptomatic for lumbar pain (5), anemia and melena (2) and blue toe syndrome (1). An aorto-bisiliac endograft was deployed in most cases (124, 77.6%), a chimney technique was used in 2 cases. Median duration of the procedure was 190 minutes (IQR 155-210 min) and median in-hospital stay was 5 days (IQR 4-9 days). Perioperative mortality was 4.4% and 12 patients (7.5%) were discharged having a type II EL under close surveillance. Long-term follow-up was available for 146 patients (median 16.6 months, IQR 6.2-37.6 months). Survival was 69.7%+4.3% at 3 years and 53.8%+5.5% at 5 years. There were 2 fatal ruptures. Survival was significantly affected by age (P=.03), preoperative rupture and symptoms (P<.001, RR 0.3, 95%CI 0.2-0.5 and P=.01, RR 0.45, 95%CI 0.28-0.82 respectively). Freedom from EL at 3 and 5 years was 67.8%+4.3% and 60.1%+5.8% respectively. We observed 13 type Ia EL (all of them treated with placement of a proximal aortic cuff) and 31 type II EL (2 required selective embolization of the guilty vessel, 12 spontaneously regressed, the remaining are still under surveillance). Furthermore there were 3 graft thromboses which contributed to an overall reintervention rate of 12.3%. None of the assessed factors affected significantly the occurrence of EL or the need for reintervention. Conclusions. In our study, survival after EVAR was significantly affected by age, preoperative rupture and symptoms. None of the analyzed factors has been shown to be predictive of the occurrence of EL or complications that required reintervention. These results further justify the need for close follow-up after EVAR.