The use of inflated balloons to improve fenestration alignment during the deployment of fenestrated grafts.
Noah T Scherrer, Zachary J Minion, Eleftherios S Xenos, Joseph L Bobadilla, David J Minion
University of Kentucky Medical Center, Lexington, KY
Target vessel occlusion after Fenestrated Endovascular Aneurysm Repair (FEVAR) occurs in approximately 2-4% of stented vessels. Poor alignment of fenestrations and scallops with the target vessels may contribute to this complication. Traditionally, sheaths have been used to align fenestrations and scallops to target vessels during deployment, relying ultimately on stents in the target vessels to later correct any mal-alignment that may have occurred during the deployment. However, sheaths have a cross-sectional area approximately one-sixth of a normal renal or visceral lumen, providing significant leeway for mal-alignment. In light of this, we routinely use inflated balloons in the target vessels instead of sheaths during FEVAR deployment for the theoretical advantage of improved alignment. We also feel this simple maneuver may protect from atheroembolism during graft opening. The purpose of this study was to compare sheaths to inflated balloons for fenestration to target vessel alignment during deployment of a FEVAR bench top model with varying aortic configurations.
A 32 mm diameter Z-Fen proximal body (Cook Medical, Bloomington, IN) was used for all test deployments. The device was configured with a single scallop (height 10 mm and width 10 mm) placed at the 1200 position and two small fenestrations (height 8 mm and width 6 mm) placed 15 mm from the edge of the fabric and an arc length of 21.2 mm from the center of scallop on each side. Vinyl tubing with holes drilled to represent the superior mesenteric and renal arteries was used to create models of varying aortic anatomy. The vinyl tubing had an inner diameter of 25.4 mm. The superior mesenteric artery hole was 6.4 mm in diameter and the renal artery holes were 5.6 mm in diameter. The position of the Z-Fen device in the vinyl tubing was controlled using string threaded through the struts of the proximal free-flow component and the most distal z-stent. A Newton wire was used to simulate the constraining wire and a short piece of plastic cylindrical tubing was used to simulate the top cap during the deployments.
Three different aortic models were used. Model A was straight with holes drilled in perfect alignment with the fenestrated graft. Model B was designed to represent tortuous anatomy, the holes were kept aligned, but a 45 degree angle was placed in the vinyl tubing 4 mm distal to the renal arteries. Model C was designed to represent an ill-measured fenestrated graft and was straight but the right renal hole was intentionally misplaced 5.6 mm ventral to the appropriate position and the left renal hole was misplaced 5.6 mm cranial to the appropriate position.
For each model, the fenestrated graft was deployed 6 times using 6 french Ansel sheaths (Cook Medical, Bloomington, IN) for alignment and 6 times using Mustang balloons (Boston Scientific, Marlborough, MA) inflated to nominal pressure for alignment. A 7 mm balloon was used in the superior mesenteric artery hole and 6 mm balloons were used in the renal artery holes. Based on preliminary findings, all deployments were performed with upward traction placed on our deployment model during removal of the constraining wire. Percent area encroachment of graft material on target vessel ostium was calculated using image analysis software and compared with Student’s t-test.
In Model A, average percent area encroachment when using sheaths for alignment was 6.4% ± 1.5% compared to 5.8% ± 1.0% for balloons (p=0.73). In Model B, average percent area encroachment was 45.9% ± 7.8% for sheaths compared to 20.6% ± 3.8% for balloons (p<0.01). The renal fenestration on the outer curve was responsible for the majority of the mal-alignment, with an average encroachment of 79% due to the fenestration opening distal to its intended site. In Model C, average percent area encroachment was 44.1% ± 3.4% for sheaths compared to 13.3% ± 2.4% for balloons (p<0.01).
For completeness, our preliminary findings using sheath alignment in Model A with no upward traction during removal of the constraining wire resulted in an average percent area encroachment of 25.3 ± 6.6%, leading us to alter our protocol to include upward traction during all deployments.
Our results provide some insight into the behavior of fenestrated grafts during the actual mechanism of deployment. While sheaths appear adequate for alignment of fenestrations to target vessels in straightforward anatomy with precisely measured fenestrations and scallops, inflated balloons in our bench top model provide significant improvement in targeted vessel alignment in tortuous anatomy and in fenestrated grafts with inaccurately measured fenestrations and scallops. Since it is standard practice that balloons are inserted in the target vessels to aid tracking of the sheaths into the vessels, it would add little to the case to use these same balloons to improve alignment of the fenestrations.
It should also be noted that adequate alignment using sheaths required upward pressure during removal of the constraining wire in our model. While this finding may in part be due to the decreased axial support of the graft inherent in our deployment model, it also demonstrates the advantage of using the natural fulcrum of the femoral-based sheaths (or balloons) in the target vessels. Therefore, there is intuitive logic that this other simple maneuver can help to further improve alignment.
Admittedly, while mal-alignment can be at least partially overcome using stents, it does require biomechanical forces to maintain proper alignment. This can lead to an acute or chronic stresses on the targeted vessel tangential to the fenestrated stent graft which, in theory, may cause stent fracture or accelerated stent intimal hyperplasia that causes early or late stent occlusion. Certainly, there can be little disadvantage to improved initial alignment of the fenestrations and decreased chronic forces on the branch vessel stents.
We believe that another very interesting finding of our study was that balloon alignment during deployment can also overcome intentional mis-placed fenestrations. This is because there is excess material in the circumference of the graft compared to the smaller lumen of the aorta. By manipulating the excess material so that it is unevenly distributed throughout its circumference, one can essentially move fenestration in multiple directions. The implications is that balloon alignment during deployment may extend the applicability of off-the-shelf designs to greater variations in anatomy.
Finally, our results also suggest that renal fenestrations should likely be placed more proximal than expected during planning for vessels arising from the outer curve of angulated anatomy to compensate for the poor tilt that occurs because of the tortuosity.
Back to 2016 Annual Meeting Program