Objective: To validate a means by which migration of thoracic stent grafts can be reliably detected and quantified.
Methods: Patients treated for thoracic aneurysms (without dissections) with either the Cook Zenith TX1/TX2 or the Gore TAG device were retrospectively reviewed. Patients with digital imaging data at a baseline study (discharge or 1-month computed tomographic scan) and a minimum of 6 months' follow-up were evaluated on a three-dimensional workstation. Centerline of flow (CLF) calculations were used to determine length measurements to establish distances from native vascular landmarks (left common carotid artery, left common carotid artery, and celiac artery) to the proximal and distal aspects of the fixation systems of stent grafts. Patients with evidence of fixation system migration (>10 mm of movement) or increasing thoracic aortic lengths (left common carotid artery to celiac artery distance) were subjected to more detailed reviews.
Results: Of 194 patients evaluated (133 Zenith and 61 TAG), 46 were treated for dissections and excluded. Fifty-seven patients did not have a digital baseline study and available DICOM data for follow-up imaging at 6 months or later or had died before such follow-up imaging. The remaining 91 patients underwent assessment for device migration. Analyses were conducted on 19 patients at 6 months, on 42 at 12 months, on 12 at 24 months, on 13 at 36 months, and on 5 at 48 months. CLF analysis noted more than 10 mm of caudal movement of the proximal device in 10 patients and cranial movement of the distal device in 3 patients. When this subset was further scrutinized with regard to morphologic changes remote from the prosthesis and in the context of the overall aortic repair (such as elephant trunk grafts), only four patients had movement of the proximal or distal fixation system with respect to the initially deployed location. Two-dimensional axial image analysis identified migration in only one of the four patients with CLF-detected fixation system movement.
Conclusions: The importance of early migration detection cannot be overstated given the potential to avert consequences as evidenced by analyses of counterpart abdominal aortic aneurysm devices. In vivo thoracic device analysis is more complex than that for devices used for infrarenal aneurysms. Distance calculations based on CLF measurements may overestimate the frequency of true migration, yet they serve as a reasonable initial screening tool. The resultant subset of patients then must undergo a more detailed evaluation of device position in the context of the aortic morphology to differentiate true migration from devices that maintain stable fixation system positions.