Device migration after endoluminal abdominal aortic aneurysm repair: analysis of 113 cases with a minimum follow-up period of 2 years.

PURPOSE: Device migration (DM) has been shown to cause late failure after endoluminal abdominal aortic aneurysm (AAA) repair. To establish the incidence rate and the predictive factors of distal migration of the proximal portion of the endograft, computed tomographic (CT) scans performed at different time intervals during follow-up examination of 113 patients were reviewed. PATIENTS AND METHODS: Between April 1997 and March 1999, 148 patients underwent endoluminal AAA repair with a modular endograft with infrarenal fixation (Medtronic-AVE AneuRx, Santa Rosa, Calif) at our unit. CT scans performed at 1, 6, and 12 months after surgery and yearly thereafter were prospectively stored in a computer imaging database. Patient demographics, risk factors, operative details, and follow-up events were prospectively collected. No patients were lost to follow-up examination. Twelve patients died within 2 years of surgery, four patients underwent immediate conversion to open repair, and adequate CT measurements were not feasible in 19 cases, which left 113 patients available for a minimum 2-year assessment and 418 CT scan results reviewed. Two vascular surgeons, blinded to patient identity and history with tested interobserver agreement (kappa = 0.64), separately reviewed axial reconstructions of CT scans. DM was defined as changes of 10 mm or more in the distance between the lower renal artery and the first visible portion of the endograft at follow-up examination. Ten possible independent predictors of DM were analyzed with multivariate Cox proportional hazards regression model.
RESULTS: One AAA rupture, which was successfully treated, occurred at a mean follow-up period of 28 months (range, 24 to 46 months). Seventeen patients (15%) showed DM. Eight patients (47%) with DM underwent reintervention: a proximal cuff was positioned in six patients and late conversion to open repair was performed in two patients. Of the 10 variables analyzed with Cox proportional hazards regression model, AAA neck enlargement of more than 10% after endoluminal repair (hazard ratio, 7.3; confidence interval, 1.8 to 29.2; P =.004) and preoperative AAA diameter of 55 mm or more (hazard ratio, 4.5; confidence interval, 1.2 to 16.7; P =.02) were positive independent predictors of DM. The probability of DM at 36 months was 27% according to life table analysis.
CONCLUSION: DM occurred in a significant portion of our patients, yet aggressive follow-up examination and a high reintervention rate prevented aneurysm-related death. According to our data, dilatation of the infrarenal aortic neck is an important factor that contributes to the distal migration of stent grafts, and patients with large aneurysms are at high risk for DM.