The introduction of endovascular stent graft technology has ushered in a new era in therapy for diseases of the descending thoracic aorta. This ‘radical and revolutionary’ stent graft technology has been accompanied by simultaneous advances in non-invasive imaging and materials technology. The technical challenges of stent graft deployment in the descending thoracic aorta—such as proximity to the great vessels and arch tortuosity— have been and remain a device engineering focus. The promise of these new tools is the long-term repair of thoracic aortic disease processes such as aneurysm or dissection with minimal associated morbidity and mortality.
However, it must be remembered that the pathophysiology of vascular disease is fundamental. The lessons learned from decades of open surgical management of thoracic aortic disease are applicable to the implementation of this endovascular technology. Therefore, we believe clinicians should not deviate from the established indications for operative intervention. In addition, the anatomical limitations of various stent devices should be respected. One should not hesitate to proceed with open repair in the face of unfavorable endovascular anatomy and a surgical indication.
The proper implementation of stent graft technology is dependent on high-quality imaging in the pre-, intra-, and post-operative periods. Specifically, multidetector computed tomography (MDCT) with 3D reconstructions has made device selection and stent deployment strategy (i.e. access route for large sheaths) more exact and safer. The authors generally prefer computed tomography angiography (CTA) to magnetic resonance imaging (MRI) unless confronted with renal insufficiency. Also, intravascular ultrasound is a useful tool when attempting to minimize contrast use and in the analysis of aortic wall pathology, e.g. in acute and chronic aortic dissection. In addition, the introduction of improved non-invasive imaging has brought disease processes to the attention of physicians that previously were encountered relatively rarely, i.e. intramural hematoma and penetrating aortic ulcer. The natural history and management of these entities is not well defined in the literature, and represent another area for investigation and broadening the scope of thoracic aortic intervention.
Currently, there are 12 thoracic endografts commercially available in Europe. However, in the US there is only one Food and Drug Administation (FDA)-approved device: the Gore Thoracic Aortic Graft (TAG) device (WL Gore and Associates, Flagstaff, Arizona), which is approved for the treatment of descending thoracic aortic aneurysms. Two other devices— the TX2 device (Cook, Bloomington, Indiana) and the Talent graft (Medtronic Endovascular, Minneapolis, Minnesota)—have completed multicenter trials and are awaiting FDA approval. Other new devices, such as the Relay Thoracic Stent-Graft (Bolton Medical, Sunrise, Florida) and the Valiant Thoracic Stent-Graft (Medtronic Endovascular, Minneapolis, Minnesota), are currently recruiting patients for FDA trials.
Finally, even though thoracic endografting holds the promise of decreased morbidity and mortality, problems such as spinal cord injury and stroke, while possibly less common than with open aortic surgery, remain a challenge.1–3 Careful attention to intra-operative arterial perfusion pressure and spinal cord function with somatosensory-evoked potentials are often used during our stent graft deployments and are routine in high-risk cases for neurological complications. In addition, the liberal use of spinal cerebrospinal fluid (CSF) drains is indicated when extensive lengths of descending thoracic aorta are to be covered. This vigilance should be continued into the early post-operative period.
Aneurysms
Recently, the incidence of degenerative thoracic aortic aneurysms encountered by clinicians has increased.4 The natural history of thoracic aneurysms is progressive expansion, subsequent increased aneurysm wall stress, and eventual rupture.5 Once ruptured, emergent repair is extremely challenging, with an associated mortality in the mid-90% range.6 Great strides have been made in reducing morbidity and mortality associated with elective open repair of descending thoracic aneurysms; however, even at high-volume centers operative mortality and paraplegia rates generally run in the 10% range.7,8
Since FDA approval in 2005 of the Gore TAG Thoracic Endoprosthesis, the treatment paradigm for thoracic aneurysmal disease has quickly moved toward stent grafting. This has occurred even though late results and level 1 evidence in the form of randomized trials are non-existent. However, the early results are quite compelling. In phase II multicenter data used by the FDA, the Gore TAG device was demonstrated to be a safe alternative with relatively low morbidity—4% stroke and 3% temporary or permanent paraplegia—and excellent freedom from aneurysm-related death of 97% at two years.9
These results compared favorably with an open-repair cohort compiled for the FDA analysis of the Gore TAG. Paraplegia, operative mortality, intensive care unit (ICU) stay, and time to return to normal activity all measured favorably for endovascular repair versus open repair. Areas in which the endovascular group fared worse were vascular injuries secondary to femoral or iliac access problems and the presence of endoleaks post-repair. Three re-interventions occurred in the endograft cohort (n=140) and none in the open surgical cohort (n=94) in two years of follow-up.10
Mid-term data have also been encouraging. A single-center report from Boston comparing stent graft repair with an open cohort demonstrated a reduction in operative mortality by half, with stent graft repair and similar late survival for both cohorts. Re-intervention rates and spinal cord ischemic complications were similar.7 These findings have been supported by additional reports from the institution of the authors and others.11–13 As the long-term fate of the aorta following stent graft repair is unknown, scheduled long-term follow-up imaging is required. Type I endoleaks should be addressed when discovered. This can usually be accomplished endovascularly. The significance of type II endoleaks in the setting of a stable aneurysm is uncertain, but should be followed carefully.14
The authors, along with others, are pushing this technology into areas such as aortic arch and thoracoabdominal aneurysms with a hybrid approach (open surgical debranching followed by stent deployment). While only limited short-term data exist, this approach seems promising. The avoidance of deep hypothermic circulatory arrest and cardiopulmonary bypass may reduce peri-operative morbidity and mortality.15,16 Soon there will be other devices available for the treatment of thoracic aortic aneurysms. The next generation of devices promises to allow for greater flexibility in conforming to difficult anatomy and the ability to preserve important side branches while still excluding the aneurysm, i.e. in the aortic arch and thoracoabdominal anatomy.
Aortic Dissection and Acute Aortic Syndromes
Acute aortic syndromes requiring surgical intervention such as ruptured descending thoracic aortic aneurysms, symptomatic acute type B aortic dissections, intramural aortic hematoma, aortopulmonary fistulae, penetrating aortic ulcers, and traumatic aortic transections have been associated with as much as a six-fold increase in morbidity and mortality compared with elective scenarios.17
Here, in the acute setting, stent grafts, if indicated, offer the potential of an expeditious life-saving procedure in critically ill patients. At a minimum, stent grafts can be seen as a salvage strategy, which can allow for elective surgical repair under controlled circumstances. Optimally and potentially, stent grafts can be seen as a definitive therapeutic modality, although long-term data are lacking to support the latter statement.18
According to the International Registry of Acute Aortic Dissection (IRAD),19 uncomplicated acute type B dissection treated with medical therapy is associated with a mortality rate of 10.7%. However, up to 20% of patients with type B dissection develop complications, i.e. rupture and malperfusion requiring surgical intervention with a subsequent mortality of 31.4%.19 However, the outcomes in both complicated and uncomplicated type B dissection seem to be improving at high-volume centers. In 2007, Estrera et al. reported a 17% mortality associated with complicated type B dissection and only a 1.2% mortality associated with uncomplicated type B dissection.20 The endovascular treatment of uncomplicated and complicated type B aortic dissections is an area of active investigation. A recent meta-analysis of outcomes following endovascular treatment of type B aortic dissections revealed a greater than 95% technical success rate and low rates of paraplegia (0.8%) and stroke (1.9%). However, they did note a 14–18% overall major complication rate. The acute and mid-term results of an endovascular approach to type B dissection appeared to compare favorably with open repair.21
Multiple series with limited patient numbers and limited follow-up indicate that stent graft repair is a viable alternative to open repair for traumatic aortic injuries. In fact, peri-operative mortality and paraplegia rates appear to compare favorably with traditional open repair. These results are apparent despite increased intercostal artery occlusion and frequent coverage of the left subclavian artery compared with open repair. The reason for this is unclear, but may be due to the fact that mean arterial pressures can be run higher after stent repair versus open repair. Leurs et al. in an analysis of the European Collaborators on Stent/Graft Techniques for Aortic Aneurysm Repair (EUROSTAR) database reported on 50 patients treated with traumatic aortic transections: the operative mortality and paraplegia rates were both 6%.22
In summary, endovascular stent grafting of disease processes of the descending aorta is feasible and relatively safe. Exquisite judgment is essential for good results. Generally, these results rest on a broad knowledge base of thoracic aortic disease processes and experience in both open and endovascular surgery. Careful attention to patient anatomy and device specifications must be maintained. The key to the successful implementation of this technology lies in careful pre-operative planning (pre-operative imaging, device selection, and access matters), intra-operative execution with safe device delivery, and prevention of central nervous system injury.