Endovascular aneurysm repair (EVAR) will fail over time in a percentage of patients. Mechanical failure of the device, progression of aortic disease, or interface complications between the device and the native vasculature may contribute. Our aim was to evaluate the role of fenestrated and branched endografts as treatment options for failed devices.
Methods
Between January 2001 and June 2013, 970 patients were enrolled into a physician-sponsored investigational device exemption (PSIDE) study and treated with a fenestrated/branched endograft. All patients treated for nonurgent proximal neck failure of an infrarenal endoprosthesis previously implanted during EVAR comprised the study group. Patients treated for a primary aneurysm within the PSIDE were evaluated as a comparison group to identify preoperative risk factors for failure. A retrospective review was undertaken to determine the details of the initial EVAR, whereas the prospective PSIDE database was used to assess outcomes of secondary treatment. Three-dimensional imaging techniques were used to define all morphologic measurements. Statistical analysis included comparisons between categoric variables with the 蠂2 test and between continuous variables with the Wilcoxon rank sum test between patients with late failures and those with native aortic repair. Kaplan-Meier curves were used to analyze overall survival.
Results
Of 970 patients enrolled in the PSIDE, 54 (5.6%) had late failure of a prior endograft. Fenestrated/branched devices were used to address the failure in each patient. The etiology of failure was related to a proximal neck issue in all patients: type Ia endoleak in 38, stent migration in 18, neck degeneration in 28, or some combination of these factors. The endovascular rescue procedure took place a mean of 61 months after the primary procedure. The mean aneurysm diameter at reintervention was 67 mm. Patients requiring a secondary fenestrated procedure were younger at the time of their primary intervention (P = .039) and were more likely to have a history of chronic renal insufficiency (P = .05) compared with other patients in the PSIDE. Technical success rate in the study group was 85% (44 of 52). Successful stenting was achieved in 71 of 77 (92%) target vessels. Thirty-day mortality was 3.8% (two of 52). Fluoroscopy dose and operating time were longer in the rescue group (P = .07) than in the control group (P = .008). Secondary interventions were required in 36.5% (19 of 52) of patients.
Conclusions
Our series demonstrates the risk for late failure after EVAR is greater in patients who are younger and have chronic renal impairment at the time of implantation. Branched and fenestrated repair after failed EVAR is more complex than repair in the native aorta. More research is needed to identify patients at higher risk of failure after EVAR to prevent the need for rescue in the future.
