Comparison of sacrificed healthy aorta during thoracoabdominal aortic aneurysm repair using off-the-shelf endovascular branched devices and open surgery

Background: Off-the-shelf devices for branched endovascular aortic repair of thoracoabdominal aortic aneurysm (TAAA) have been developed to overcome the manufacturing- and logistics-related delays characteristic of device customization. Nonetheless, the structural requirements of branched endovascular aortic repair, together with the need for additional thoracic components to suit different anatomies, might lead to a large sacrifice of healthy aorta. Methods: We enrolled 18 consecutive TAAA patients treated between 2010 and 2016 in two centers employing the off-the-shelf t-Branch (Cook Medical, Bloomington, Ind) device (TB group). We matched these 18 patients one-to-one with another 18 patients of the 464 consecutive TAAAs treated in the same period by means of open repair (O group) in one of the two centers. The patients were matched on the basis of t-Branch feasibility, distance from the left subclavian artery to the beginning of the aneurysm, and diameter of proximal landing zone evaluated at preoperative computed tomography angiography. We compared the overall length of sacrificed healthy native aorta (the sealing zone in the TB group) in the two groups according to postoperative computed tomography angiography, and we investigated the number of intercostal arteries lost from that healthy aortic segment. We calculated also the real unnecessary sacrificed aorta by subtracting the compulsory minimum neck coverage requested to obtain the proximal sealing (25 mm). Results: We matched one-to-one six extent II, seven extent III, and five extent IV TAAAs. Overall, the length of sacrificed healthy aorta was longer (P < .001) in the TB group (median length: TB group, 49 mm [interquartile range (IQR), 31-60 mm]; O group, 14 mm [IQR, 10-20 mm]), and consequently the number of intercostal arteries lost (median number of pairs lost: TB group, four [IQR, three-four]; O group, two [IQR, one-three]) was higher (P = .004). Stratifying the results according to TAAA extent and after having subtracted from the TB group the length of aorta required for proximal sealing (25 mm), the length of sacrificed healthy aorta was longer (P = .008) only in the extent IV TAAA, 125 mm (IQR, 56-155 mm). The number of intercostal artery pairs sacrificed was significantly higher only in extent III TAAA (P = .043) and extent IV TAAA (P = .024). No significant 30-day clinical outcomes difference was observed between the matched groups. Conclusions: The use of the t-Branch device compared with open repair is associated with a greater sacrifice of healthy aorta in extent IV TAAA and with an increased loss of intercostal arteries arising from healthy aortic proximal neck. The theoretical possible increased risk of spinal cord ischemia secondary to the routine use of this off-the-shelf device should thus be analyzed further.