Variability in aneurysm sac regression after endovascular aneurysm repair based on a comprehensive registry of patients in Eastern Ontario

Background: Although the absence of aneurysm-related mortality, postimplantation rupture, and reintervention after endovascular aneurysm repair (EVAR) is desirable, it may not necessarily reflect successful aneurysm sac exclusion. Sac regression may be a more sensitive marker for EVAR success and may be influenced by factors beyond the presence or absence of an endoleak. The objective of this study is to determine the rate of overall long-term sac regression after EVAR and the influence of nonanatomic factors, and endograft devices used at our center. Methods: This retrospective cohort study included all EVARs performed for intact and ruptured abdominal aortic aneurysms (AAAs) at a university teaching hospital. Preoperative, operative, and follow-up data were collected using clinical and radiologic institutional databases. Preoperative and post-EVAR sac diameters were determined by a blinded observer in accordance with Society for Vascular Surgery guidelines. Absolute and relative sac regression was determined at the following intervals: 0 to 6 months, 6 to 12 months, 12 to 18 months, 18 months to 2 years, 2 to 5 years, 5 to 10 years, and more than 10 years. Results: From 1999 to 2015, 1060 patients underwent EVAR for an AAA at the Ottawa Hospital. Procedures were performed using a total of nine unique endograft devices, with five devices (Cook Zenith, n = 398; Medtronic Endurant, n = 375; Medtronic Talent, n = 183; Cook Zenith LP, n = 52; and Terumo Anaconda, n = 23) used in 97% of the procedures. The mean preoperative AAA diameter was 61.2 mm, with no detectable differences between endograft devices with respect to age, preoperative AAA diameter, or rupture diagnosis. Overall mean sac regression increased from -1.3 mm at 6 months, to -14.9 mm beyond 10 years. The majority of sac regression was achieved within 2 years. Only 90 of the 1060 patients (8.5%) experienced sac expansion of greater than 5 mm at some point during their follow-up period. Kaplan-Meier analyses revealed statistically significant device-specific variability in sac regression rates, even in the absence of an endoleak. Cox proportional hazard modeling demonstrated that age less than 75 years (hazard ratio [HR], 1.4; P = .001), female sex (HR, 1.4; P = .003), absence of type I endoleak (HR, 4.6; P < .0001), AAA greater than 70mm (HR, 1.6; P < .0001), and both the Zenith (HR, 2.0; P < .0001) and Endurant (HR, 1.7; P = .001) devices were associated with shorter time to more than 5 mm sac regression. Conclusions: This study demonstrated a pattern of sac diameter change after EVAR, with the majority of sac regression occurring within the first 2 years. Variability in sac regression was influenced by nonanatomic variables including age, sex, original AAA diameter, and specific endograft device, even after controlling for the presence or absence of an endoleak. The biophysical relationship between specific endograft design and materials, and sac regression is yet to be determined.