Jon S. Matsumura
The traveling fellowship was conducted between June, 2003, and June, 2004. The purpose of the fellowship was to visit centers of excellence in thoracic endografts and meet personally with expert physicians around the world. An additional benefit has been the compilation of the clinical experience of the physicians/centers that were visited. I am indebted to the expert physicians and their colleagues who gave freely of their time and graciously opened their labs and procedure rooms to the fellow.
Worldwide Clinical Review
Peer-reviewed scientific publications are the gold standard of evidence for medical decision making. However, in rapidly emerging technologies, early experiences are dominated by prototype devices, single site small experiences, and initial learning periods. Data from US clinical trials is useful because of the well-defined protocol, regulatory oversight, and generally more diligent follow-up and corelab review. However, practical clinical use often does not match the rigid entry criteria of these studies, and industry-sponsored trials focus on the sponsor’s contemporary device version. In order to get a grasp of the general clinical practice in thoracic endografts, a survey was conducted during personal visits with eight physicians from leading thoracic endovascular centers of excellence around the world. Table 1 lists the primary investigator who was interviewed and their respective medical center. This pooled data of 1180 patients provides reflections on a large series of patients from the perspectives of experts in the field. Weaknesses of this compilation are the lack of uniformity in surveillance protocols, definitions, follow-up intervals, and compliance, and overlap with existing publications.
Table 2 summarizes the types of thoracic pathology that has been treated and the 30 day mortality in each of these categories. It is the majority impression that endovascular treatment is an equivalent or better therapy compared to conventional treatment in each of these categories except for thoracoabdominal aneurysm, chronic dissection, and stenosis where there was disagreement or it was felt to be unknown. This is conditional upon the patient having suitable anatomy for endorepair. The majority of patients (64.2%) have been treated for degenerative descending thoracic aortic aneurysms. A wide variety of endografts have been used for treatment.(Table 3.)
Most patients (70.3%) have been treated in an operating room environment, and 49.2% used fixed unit fluoroscopy units. Intravascular ultrasound was used in 12%, and transesophageal echocardiography in 38.1%. Primary access was in the femoral artery in 84.4% (percutaneous in 1.1%), iliac in 7.5%, conduit or aortofemoral graft limb in 6.5%, infrarenal aorta in 1.1%, and other sites in 0.4%.
The left subclavian artery was covered in 29.8%, and the left common carotid in 8.0%, the innominate in 3.7%, celiac artery in 3.1%, and SMA in 0.8%. The option to revascularize these major branches is exercised at different rates, and seems to be undergoing evolution. Most centers revascularize the carotid, innominate and visceral arteries when they are occasionally covered. In contrast, revascularization is performed before or concurrently with endorepair in only 36.7% of patients who have coverage of the left subclavian artery. This is an area of increasing consensus—that is, fewer patients are having routine subclavian revascularization except when there is diseased or hypoplastic right vertebral arteries or fistulas/reconstructions based on left subclavian branches. The majority of patients are managed expectantly, and in the infrequent scenario when symptoms develop after left subclavian coverage, delayed revascularization can be performed.
Spinal cord drainage was used routinely in 1.0%, and selectively in 6.8% of patients with extensive descending aortic coverage or history of AAA. Adenosine (2.1%) and high dose beta blocker (0.1%) were used rarely to arrest or slow the heart for deployment. The 30-day morbidity rates are stroke in 2.8%, renal failure in 1.6%, and paraplegia in 2.5%. 43% of the paraplegia complications were delayed in onset.
Later outcomes are best described by time-dependent estimates, but are not precisely combinable because of unequal or unknown individual lengths of follow up. Nevertheless, general estimates are useful to gauge the emerging technology. Endoleak was noted in 10.5% of patients, sac expansion in 4.0%, proximal neck dilation or dissection in 2.7%, distal neck dilation in 2.1%, intercomponent migration in 1.7%, proximal migration in 1.3%, distal migration in 0.4%, asymptomatic device failure such as fracture in 6.3%, and symptomatic device failure in 0.1%, and aneurysm rupture in 0.9%.
Recommendations by the experts on future device design were ranked, and the top desired attributes are: 1) more flexible delivery systems, 2) more flexible devices with better conformation to irregular wall contour, and 3) branched devices.
The treatment of thoracic aortic disease is rapidly evolving. Alternatives to the traditional, open surgical approach are being developed and evaluated in clinical trials. While mortality rates for open surgical repair of the thoracic aorta have decreased with experience, early reports of thoracic aortic endografts suggest that endovascular repair may offer a less invasive option with lower morbidity and mortality rates.
Several engineering advancements are being pursued to address shortcomings in earlier devices for thoracic endorepair. These include endografts that conform to the individual aortic anatomy, more flexible and accurate delivery systems, more robust construction to accommodate higher thoracic aortic forces, and mechanisms to treat pathology close to or involving the aortic arch and visceral sidebranches. As these technologies mature, it is possible that endovascular therapy may become the preferred initial therapy for many thoracic aortic diseases.
Table I. Visited physicians and institutions by reported total thoracic endograft volume.
Shin Ishimaru, MD
Michael Dake, MD
Roy Greenberg, MD
Ludger Sunder-Plassman, MD
Rodney White, MD
Geoffrey White, MD
Takao Ohki, MD
Michael Laurence-Brown, MD
Tokyo Medical University
Stanford University Medical School
University of Ulm
University of California, Los Angeles-Harbor
University of Sydney
Montefiore Medical Center
Table II. Various pathology treated by endografts and periprocedural mortality.
Primary aortic pathology
Descending thoracic aneurysm
Acute traumatic disruption
Intramural hematoma with ulcer
Giant penetrating ulcer
Percent of total cases
30 day Mortality (%)
* Often due to concomitant neurologic injury.
Table III. Distribution of endovascular grafts in this compilation.
Posted June 2010