Endovascular repair of traumatic thoracic aortic disruptions with “stacked” abdominal endograft extension cuffs
Article Outline
Objective
Endovascular stent graft repair of a traumatic thoracic aortic disruption (TTAD) is rapidly becoming an accepted alternative to open surgical repair. The use of currently approved thoracic stent grafts especially in younger patients with small, “steep,” tapered aortas, remains a concern due to the acute thoracic endograft collapse and enfolding. The objective of this study, the largest report to date, was to evaluate the mid-term results of TTAD treated with abdominal aortic “stacked” extension cuffs, with follow-up extending to 41 months.
Methods
Thirty-one patients with multi-system trauma (age range, 15 to 61; mean 31.4 years) were seen after motor vehicle accidents between January 1, 2003 and July 1, 2007. Chest x-ray findings warranted thoracic CT scans, which revealed disruptions of the thoracic aorta. Intra-operative arteriograms in all patients and intravascular ultrasound (IVUS) (n = 17) delineated the extent of the aortic injuries. The aortic length from the subclavian artery to the injury averaged 2.5 cm (range, 1.5 to 4.0 cm). The repairs were performed with Gore (W.L. Gore & Associates, Flagstaff, Ariz) (n = 15), Aneuryx (Medtronic, Santa Rosa, Calif) (n = 15), and Zenith (Cook, Inc., Bloomington, Ind) (n = 1) Aortic Extension Cuffs. A femoral artery approach was used in 27 patients and a supra-inguinal retroperitoneal iliac approach in four. All patients underwent thoracic CT scans during follow-up.
Results
In all patients, the stent-graft cuffs successfully excluded the TTAD: 21 patients had 2 cuffs, 9 had 3 cuffs, and 1 had 4 cuffs. The aorta adjacent to the injury mean diameter was 18.5 mm (range, 17-24 mm). No subclavian arteries were covered. Two patients required an additional cuff for exclusion of the Type I endoleaks at the distal attachment site within 6 weeks of initial endograft repair. There were no procedure-related deaths; 2 patients died of other injuries. At follow-up, extending to 41 months (range, 3 to 41 months), two pseudo-aneurysms occurred which required open operative repair: 1 due to infection (4 months) and a leaking pseudoaneurysm (14 months). A CT scan in all other survivors demonstrated no device-related complications, endoleaks, or cuff migrations.
Conclusion
Stent-graft repair of TTAD is technically feasible. The technique of “stacked” aortic cuffs provides an acceptable option when urgent therapy is needed, when patients are deemed high-risk for open operative repair, or until thoracic endografts are designed which can safely treat focal, smaller aortic diameter injuries.
Blunt trauma to the thoracic aorta is a life-threatening, often fatal condition, in patients who frequently have devastating multiple system injuries. Indeed, operative repair of thoracic aortic injuries are associated with mortality rates approaching 30%1, 2 and paraplegia rates reported as high as 18%.3
The classic injury mechanism due to blunt thoracic aortic trauma is related to the combination of deceleration and traction at the immobile aortic isthmus, which represents the junction between the mobile aortic arch and the fixed descending thoracic aorta. The isthmus is the most common site of rupture (70%), followed by the ascending aorta or aortic arch (18%) and the distal thoracic aorta (14%).3, 4
The endovascular repair of traumatic thoracic aortic disruption (TTAD) is rapidly emerging as a favorable alternative to open surgical repair, and several studies have reported lower peri-operative mortality, paraplegia, operative time, operative blood loss, and hospital length of stay.5, 6, 7 The use of currently approved thoracic stent-grafts, however, may not be possible, especially in younger patients with small, “steep,” tapered aortas. The objective of this study, the largest reported to date, was to evaluate the results of the use of commercially available “stacked” abdominal endograft extension cuffs to treat TTAD caused by blunt trauma.
Methods
A retrospective review of 31 patients who underwent endoluminal repair of blunt TTAD between January 1, 2003, and July 1, 2007, was performed. During the same time interval, 2 other patients with TTADs underwent repairs with the Gore TAG device (W.L. Gore & Associates, Flagstaff, Ariz). These 2 patients' thoracic aortas measured 28 and 30 mm in diameter and they were not included in the study. No open operative procedures were performed during this interval. All patients had been in motor vehicle accidents and sustained severe blunt trauma. Patients underwent chest x-rays and suspected findings for TTAD mandated computed tomography angiography (CTA) scan (Fig 1). Measurements of the proximal and distal thoracic aortic diameter and the length between the left subclavian artery and the aortic disruption were obtained from the CTA scan.
Fig 1.
Computed tomography angiography demonstrates blunt aortic disruption with peri-aortic hematoma (arrow).
Repairs were performed in the operating room with the patient under general anesthesia with standby cardiovascular surgery support available. Intravenous heparin was administered judiciously (ie, 50u IV/kg) as the majority of the patients had severe concomitant injuries (mean ISS of 40). Cutaneous femoral artery access was used for aortography. A 5 French sheath was placed in the femoral artery and a marker pigtail catheter was passed under fluoroscopic guidance to the aortic arch for aortography. The contralateral femoral artery was exposed in standard fashion for stent-graft access, and an 8 French sheath was placed. After aortography, intravascular ultrasound (IVUS) (n = 17) (Volcano Corporation, Rancho Cordova, Calif) was performed of the aortic arch and descending thoracic aorta to further delineate the anatomy and to visualize the location of the disruption in relation to the left subclavian artery. After the IVUS catheter was removed, a wire exchange was performed and an Amplatz (Meditech/Boston Scientific, Natick, Mass) Super-Stiff Wire was placed in the ascending arch for stent graft deployment. When the femoral arteries were too small to accept the device, or the device was not long enough to reach the thoracic aorta, a supra-inguinal incision was made for proximal access to the common iliac artery. In 4 patients, a supra-inguinal retroperitoneal iliac artery approach allowed access to the iliac. The endograft extension cuffs were oversized approximately 10-15%, and a distance of 1.5-2.0 cm between subclavian artery and injury site was considered adequate for endograft cuff placement. The endograft cuffs were passed “sheathless” (without a separate introducer sheath) as vessels were free of occlusive disease and this allowed for the maximal length of device to reach the thoracic aorta. The endograft extension cuffs were deployed, distal to proximal whenever possible, (Fig 2). Ballooning was reserved for malposition or Type I endoleaks in order to minimize the potential for further aortic injury and retrograde dissection. CT scans were obtained within 72 hours after implantation, and follow-up CT scans were obtained per endograft surveillance protocols.
Fig 2.
Technique of “distal to proximal” deployment of endograft extension cuffs to improve seal and wall apposition (arrows indicated distal to proximal deployment).
Results
The mechanism of the TTAD injury was motor vehicle crash in 26 patients and motorcycle crash in 5. Twenty-four patients underwent endovascular repair less than 36 hours after injury. The delay from injury to repair in the other 7 patients, all of whom were in an unstable condition from multiple injuries, averaged 2.5 days (range, 2 to 5 days). The mean injury severity score (ISS) was 40 and concomitant injuries included coexisting head injuries (21), extremity fractures (19), pelvic fractures (9), sternal/clavicle fractures (6), and solid organ injuries (5).
There were no procedure-related deaths, paraplegia, renal failure, or other cardiovascular complications. Two patients died from causes unrelated to the endograft procedures: 1 from multi-system organ failure and the other when life support measures were withdrawn. The mean age was 31.4 years (range, 15 to 61 years) and 19 patients were men. Twenty-seven procedures were successfully done through a femoral artery approach. Iliac artery Dacron graft conduits for access were necessary in 2 patients due to small femoral arteries. In 2 other patients the device was too short to read the site of injury and these patients had direct puncture of the iliac artery after the stent-graft device was tunneled under the inguinal ligament (similar to an aorto femoral bypass) for easy access to the common iliac artery; this obviated the need for a conduit and provided a less angled approach to the iliac artery. No intra-operative conversions to an open procedure were necessary.
In all patients, the endograft extension cuffs successfully excluded the traumatic disruptions based on intra-operative aortography. Gore Excluder Aortic Extension Cuffs (W.L. Gore & Associates, Flagstaff, Ariz) were used in 15 patients, AneuRx Extension Cuffs (Medtronic, Santa Rosa, Calif) were used in 15, and a modified Zenith (Cook, Inc., Bloomington, Ind) Cuff was used in 1 patient, per surgeon's preference. Mean aortic diameter was 18.5 mm (range, 17 to 24 mm). No subclavian arteries were covered. Twenty-one patients had 2 cuffs, 9 had 3 cuffs, and 1 had 4 cuffs. The aortic length from the subclavian artery to the aortic disruption, the “landing zone”, averaged 2.5 cm (range, l.5 to 4.0 cm). The average estimated blood loss was 150 mL (range, 75 to 300 mL).
Postoperative CT imaging was performed in all patients prior to discharge. Two patients required an additional cuff for exclusion of Type I endoleaks at the distal attachment site within 6 weeks of initial endograft repair. These devices were deployed uneventfully. CT scan follow-up extending to 41 months (range, 3 to 41 months) was performed in 27 patients; 2 were lost to follow-up. Two pseudo-aneurysms occurred which required open operative repair: 1 was due to a pulmonary abscess which infected the endograft 4 months after endovascular repair, and the other a “leaking” pseudoaneurysm which occurred 14 months after endovascular repair. CT scans in the other 25 survivors demonstrated no device-related complications, endoleaks, or cuff migrations compared to immediate postoperative images.
Discussion
Since the initial report by Dake et al,8 of the successful repair of a traumatic aortic disruption with stent-grafts, many case reports and several institutional reviews have reproduced their results.9 However, traumatic aortic injuries tend to affect younger populations, in contrast to thoracic aneurysms, and commercially available devices are not designed for use in young thoracic aortas. The young aorta tends to be more narrow and the turn radius of the aortic arch much tighter and “steeper.”10
The small size of the aorta in young patients creates new challenges that one rarely sees in patients with aneurysm disease and access vessels are often smaller as well. In our series of 31 patients, however, the mean aortic diameter was 18.5 mm (range, 17 to 24 mm), but only 4 patients required iliac artery access, as the access vessels were free of occlusive disease or tortuosity and readily dilated to accommodate the endograft extension cuff delivery systems which were passed “sheathless” and were an approximate 18F diameter.
The smallest available thoracic endograft in the United States is the Gore TAG device (26 mm diameter) which was designed for patients with thoracic aortic aneurysm who have a larger diameter thoracic aorta, and “flatter” aortic arch curvature due to aortic elongation associated with aging. The 26 mm TAG device should be used in treating aortas equal to or larger than 23 mm in diameter. If, for example, a 26 mm TAG device is placed in 18.5 mm aorta, it represents an approximate 40% oversize. It appears that the adverse events reported to date with the use of the Gore TAG device to treat TTAD may be due to device oversizing beyond the recommended Instructions for Use (IFU) approved by the FDA, which may result in crimping, fracture, endoleak, migration, and device collapse.11, 12 Adverse events, such as acute endoprosthesis collapse or enfolding, however, are not unique to the TAG endograft and have been reported with other endografts, when such grafts were oversized as well.11 The unique packaging and release mechanism of the TAG device may exacerbate the problem of enfolding as the device is packaged in a furled configuration, and a smaller aorta may prevent it from unfolding completely.13 Although small aortic diameter may promote collapse, other variables such as poor apposition of the endograft to the aortic wall, acute aortic arch angulation, and other anatomic factors may also play a causative role in endoprosthesis collapse, although this remains to be proven.13, 14 Because of these concerns, many surgeons are reluctant to use the TAG device in patients with aortas measuring <23 mm in diameter.
Another concern with the use of thoracic aneurysm endografts for treating TTAD is the potential need to cover the left subclavian artery. Coverage of the left subclavian artery is innocuous in most cases, but may be fraught with hazard when a left internal mammary-coronary bypass or dominant left vertebral artery is present, or a history of previous abdominal aortic surgery, or internal iliac artery occlusions, which may increase the risk of postoperative spinal cord ischemia is noted.15 In this setting, carotid-subclavian bypass is indicated. Coverage of the left subclavian artery may also place the device in the transverse portion of the aortic arch which may lead to poor apposition against the aortic wall, device collapse, and the creation of an inferior “lip” which may create a proximal endoleak.10
With stacked cuffs, both of these problems are readily avoided and the left subclavian artery is not covered. A limitation of abdominal endograft cuffs, however, is their short length, which necessitates the use of multiple cuffs and creates the potential for device separation and a potential Type III endoleak. For this reason, the cuffs are deployed distal for proximal to improve seal and wall apposition. In this series, 21 patients had 2 cuffs, 9 had 3 cuffs, 1 had 4 cuffs, and only 2 endoleaks occurred. Another limitation with the abdominal endograft cuffs is the relatively short delivery system. However, in only 2 patients were we unable to reach the desired deployment location by femoral access; these patients underwent uneventful iliac artery access.
Although the use of aortic extension cuffs to treat TTAD is “off label,” no enfolding or device collapse has been reported with this technique. Two patients, however, had Type I endoleaks within 6 weeks of initial endograft repair which were treated with an additional extension cuff early in our experience (case 4 and 9). The wisest course of action appears to be to add an extra cuff if any doubt of treatment length is present. Two late pseudo-aneurysms also occurred (one due to infection from a pulmonary abscess at 4 months and one at the distal attachment site at 14), for an overall complication rate of 13%.
Nevertheless, we believe that the use of abdominal extension cuffs are best to treat TTAD when the aortic diameter is less than 23 mm or the radius of curvature is small, which it inherently tends to be in the younger patient population commonly affected by TTAD. The long-term performance of abdominal aortic cuffs in the thoracic aorta is unknown and will only be answered by other reports with long-term follow-up. Patients with TTAD tend to be young and have decades of life expectancy within which the aorta may enlarge resulting in loss of fixation and material fatigue (ie, stent fractures, fabric fatigue) may occur. If these problems occur, it will likely be years later, but patients must be made aware of the necessity for life long follow-up. Regardless, until smaller diameter, shorter length thoracic endografts are made with increased radial force, better attachment systems, and increased flexibility to better oppose the inner wall of the aortic arch, the technique of “stacked” abdominal aortic cuffs provides an acceptable option when urgent therapy to treat TTAD is needed, especially in younger patients and when patients are deemed high-risk for operative repair.
Author contributions
References
- . Rupture of thoracic aorta caused by blunt trauma: a fifteen-year experience. J Thorac Cardiovas Surg. 1990;100:652–661
- . Traumatic aortic rupture: twenty-year meta analysis of mortality and risk for paraplegia. Ann Thorac Surg. 1994;58:585–593
- Prospective study of blunt aortic injury: multicenter trial of the American Association for the Surgery of Trauma. J Trauma. 1997;42:374–380discussion 380-3
- . Traumatic rupture of the thoracic aorta third decade of experience. Am J Surg. 2002;183:571–575
- . Outcomes of surgical and endovascular treatment of acute traumatic thoracic aortic injury. J Vasc Surg. 2007;46:934–940
- . Management of blunt thoracic aortic injuries: endovascular stents versus open repair. J Trauma. 2004;56:565–570
- . Acute thoracic aortic trauma: a comparison of endoluminal stent grafts with open repair and nonoperative management. Ann Vasc Surg. 2003;17:589–595
- . Transluminal placement of endovascular stent grafts for the treatment of descending thoracic aortic aneurysms. N Eng J Med. 1994;331:1729–1734
- . Traumatic thoracic aortic injury. Endovasc Today Feb. 2006;58–66
- . Twenty consecutive cases of endograft repair of traumatic aortic disruptions: lessons learned. J Vasc Surg. 2007;45:487–492
- . Collapse of a stent-graft following treatment of a traumatic thoracic aortic rupture. J Endovasc Ther. 2005;12:503–507
- Endovascular repair of proximal endograft collapse after treatment for thoracic aortic disease. J Vasc Surg. 2006;43:609–612
- . Anatomic factors associated with acute endograft collapse after Gore TAG treatment of thoracic aortic dissection or traumatic rupture. J Vasc Surg. 2007;45:655–661
- . Thoracic endovascular aortic repair of traumatic injuries involving the descending thoracic aorta. J Cardiovasc Surg. 2007;48:741–750
- . Thoracic endovascular aortic repair for traumatic aortic transaction. J Vasc Surg. 2007;46:928–933
Competition of interest: none.
PII: S0741-5214(08)00829-X
doi:10.1016/j.jvs.2008.05.055
© 2008 The Society for Vascular Surgery. Published by Elsevier Inc. All rights reserved.
