AneuRx Device Migration: Incidence, Risk Factors, and Consequences

Success after endovascular abdominal aortic aneurysm repair (EVAR) is dependent on device positional stability. The quest for such stability has motivated different endograft designs, and the risk factors entailed remain the subject of debate. This study aims at defining the incidence, risk factors, and clinical implications of device migration after EVAR with the AneuRx® endograft. In this study we included all consecutive 109 patients submitted to primary AneuRx placement for infrarenal aortic or aortoiliac aneurysms. Preoperative computed tomography (CT) scans were reviewed for the following anatomic characteristics: neck length, diameter, angulation, calcification, and thrombus load; and sac diameter and thrombus load. Percentage of device oversizing relative to the proximal neck diameter was determined. All postoperative CT scans were reviewed, and the distance between the lowest renal artery and the craniad end of the device was measured. A ≥5-mm increase in such distance was considered indicative of device migration. Migration cumulative incidence was estimated by the Kaplan-Meier method, and its association with any of the preoperative anatomical characteristics was tested using Cox proportional hazards models. Median follow-up time was 9 (range, 1-31) months. Migration occurred in nine patients, corresponding to a 15.6% estimated probability of migration at 30 months (SE = 5.1%). Migration was associated with the risk of proximal type I endoleak (hazard ratio = 3.39, 95% confidence interval = 1.46-7.87; p = 0.007). This type of endoleak occurred in three of the migration-affected patients (33.3%); all of them were resolved by additional cuff placement at the proximal landing zone. No other migration-related reinterventions were performed. The only significant associations between anatomic factors and device migration probability were the protective effects of longer necks (odds ratio [OR] = 0.71 for each additional 5 mm, p = 0.045) and longer overlapped portions of neck and device (OR = 0.56 for each additional 5 mm, p = 0.003). There was a trend toward higher probability of migration among reverse-tapered necks (OR = 1.75, p = 0.109). Percentage of device oversizing correlated with early neck dilation (between preoperative and first postoperative diameters, correlation coefficient = 0.4, p < 0.0001), but not with late neck dilatation (between first postoperative and 1.5-year scan diameters, correlation coefficient = 0.29, p = 0.112). There was a trend toward higher mean percentage of late dilation among migrators (11.4%, standard error of the mean [SEM] 2.6) than nonmigrators (5.7%, SEM = 1) (p = 0.08), but both groups had similar mean percentages of early dilation (3%, SEM = 1.6%, vs. 5.5%, SEM = 0.6%; p = 0.365). This result indicates that device migration is not a rare event after AneuRx implantation. This phenomenon is associated with proximal type I endoleaks. Deployment of the endograft immediately below the renal arteries might help to prevent migration, since use of greater lengths of overlapped device relative to the proximal neck has a protective effect. Migration seems to be independent of the degree of device oversizing.