A novel measurement technique for the design of fenestrated stent grafts: Comparison with three-dimensional aorta...

Stent graft placement is an acceptable treatment option for aortic disease, particularly for abdominal aortic aneurysm. At present, the use of stent grafts is expanding beyond current indications for use. Fenestrated stent grafts are used in patients with abdominal aortic aneurysms whose aortic anatomy is unsuitable for repair using standard devices. The success of fenestrated stent graft placement is largely dependent on planning, including obtaining measurements and designing the stent.

To demonstrate a measurement technique that may be used for the design of fenestrated stent grafts to repair endovascular aneurysms, and to compare these measurements, obtained using archived two-dimensional patient data, with measurements obtained using a three-dimensional (3-D) computer-assisted design model.

Fenestrated stent grafts were designed and fabricated based on computed tomographic angiography images. 3-D models were constructed using modelling software and rapid prototyping technology incorporated with fused deposition modelling. The stent grafts were trunk-type, with four holes for the visceral branches (celiac axis, superior mesenteric artery, right renal artery and left renal artery). Computed tomography scans of 10 patients with abdominal aortic aneurysms were reviewed. Axial, multiplanar reconstruction and curved multiplanar reconstruction images were used to measure 11 parameters. Sizing of the fenestrated aortic stent grafts was performed independently by an experienced interventional radiologist, and the results were compared with the same measurements calculated using the 3-D aorta model (generated using Materialise Interactive Medical Image Control System software [Materialise NV, Belgium]). Data were reported as the mean of the measurements. Measurements were evaluated using Bland-Altman analysis and concordance correlation coefficients (CCCs).

A total of 10 fenestrated stent grafts were fabricated. The proximal landing section above the celiac axis (one point of the wall being defined as the standard point) was 3 cm, and the distal flared section was 3 cm below the lowest renal artery. Ten computer-assisted design aorta models were successfully constructed. Measurements of the aortic diameter showed high agreement between those obtained using the archived patient computer system stent graft and those obtained using the 3-D aorta model. The CCC for variability was 0.9974. The distance from the standard point to the branch vessels also demonstrated good agreement. The CCC for variability was 0.9999.

A direct measurement technique using a standard point was simple to perform and was easily applied to the fabrication process. Preparation time will likely be shortened and the versatility of stent grafts will be improved using this method. It will be possible to produce standardized fenestrated stent grafts once patients' measurements are recorded and analyzed.

A fenestrated stent graft design technique using measurements of distance from a standard point generally showed a high level of agreement with a 3-D aorta model.