|Home | About | Journals | Submit | Contact Us | Français|
Type III aortic dissection is associated with high morbidity and mortality. There is a shifting paradigm in the treatment of complicated and uncomplicated acute Type III aortic dissection towards earlier endovascular repair. In this review, the authors present the current perspective on the endovascular management of acute complicated and uncomplicated Type III aortic dissection.
The first report of an aortic dissection was in the 1800s by Shekelton when he noted an obliterated false lumen of a healed dissection.1 The misnomer aneurysme dissequant or ‘dissecting aneurysm’ was subsequently coined by Laennec in 1826.2 Thereafter, the term acute aortic dissection has undergone numerous refinements in definition and is a commonly recognized catastrophe of the aorta today. The pathognomonic lesion is an intimal tear resulting in blood dissecting into the medial layer of the aortic wall. Aortic dissections are thought to begin where shear force (ΔP/ΔTmax) and pressure fluctuations along the aortic wall are the greatest. The original description of a distal (Type III) aortic dissection according to DeBakey originated at level of the left subclavian artery.3 Type IIIa aortic dissections are confined to the descending thoracic aorta, while Type IIIb aortic dissections extend into the abdominal aorta. This classification was further refined by Daily with the introduction of the Stanford schema in 1970 (Figure 1).4 Distal aortic dissections that limit blood flow to the viscera and the lower extremities are frequently fatal. According to the International Registry of Acute Aortic Dissection (IRAD), the mortality from distal (Type III) dissections is 27.4%.5 Patients with “complicated” dissections who have evidence of malperfusion or rupture had a 31.4% mortality, while those who had “uncomplicated” dissections treated medically had 10.7% in-hospital mortality.5
An acute aortic dissection has long been classified as one that occurs within 2 weeks of symptom onset, while chronic dissections are those that persist greater than 2 months following symptoms. These long accepted definitions, however, have been called into question as our understanding of this disease is evolving. Recently, Kasirajan and colleagues suggested that the term acute aortic dissections include the first 3 months of disease from the onset of symptoms, sub-acute aortic dissections have symptoms persist from 3 through 12 months, and chronic aortic dissections have symptoms persist for >1 year.6 This suggested classification is based on the behavior of the aortic septum as the disease progresses, from dynamic (acute), to compliant (sub-acute), and eventually non-compliant (chronic).
Aortic dissections affect 5 to 30 cases per million people each year.7, 8 Men typically are affected 5 times more frequently than women.5 The incidence of Type III aortic dissections peaks in the 6th and 7th decade of life.7 Aortic dissections follow a circadian and a seasonal chronobiologic pattern with the highest occurrences noted in the fall, winter, and spring with the most frequent events reported during the early morning hours.9 Based on the established risk factors for aortic dissection, Marfan syndrome, hypertension, pre-eclampsia and cocaine abuse are special groups that have been exclusively identified for their unique predisposition to aortic dissection. A comprehensive list of risk factors for aortic dissection is listed in Table 1.
Aortic dissection occurs due to a defect in the intimal and medial layers of the aorta allowing blood to propagate within the arterial wall creating a false lumen. The elasticity and mobility of the aorta is challenged by the force (ΔP/ΔTmax) at a local area leading to an intimal tear.10, 11 The stress on the aortic wall during a cardiac cycle is greatest in the ascending and the proximal descending thoracic aorta. As such, the most common patterns of a Type III aortic dissection begin at a tear within 2 cm of the left subclavian artery, extending through the thoracoabdominal aorta with the right renal artery and visceral trunks originating from the true lumen and the left renal artery rising from the false lumen.12–15
Patients with aortic aneurysms are predisposed to dissections with evidence of medial degeneration. Much of our understanding of the pathophysiology of aneurysms stems from several murine models.16 Most relevant, mice with a defect in fibrillin (so-called Marfan mouse) develop aortic root dilation and dissection.17 Rapid expansion in this model and others is believed to progress as a result of proteolytic enzymes and inflammatory mediators that are produced in the diseased aorta.
Patients with acute Type III aortic dissection should be classified by their anatomic sequelae from the dissection and are stratified as complicated or uncomplicated. A “complicated” Type III aortic dissection is defined as one that causes malperfusion of the viscera, kidneys or lower extremities, or as evidenced by rupture and occurs in 25–40% of cases.18, 19 The majority of patients do not have these complications at presentation, and are designated “uncomplicated.” The main factors that establish the degree of malperfusion are: (I) the proportion of the aortic circumference that is dissected, (II) the extent of distal reentry from the false lumen into the true lumen, and (III) the topography of the ostia of the visceral branches off the true lumen.18, 19 Back pain, acute mesenteric ischemia, cerebrovascular incidents, renal failure and spinal cord ischemia are commonly associated with Type III aortic dissections. For those patients who do not undergo immediate surgical intervention, dissected aortas tend to expand at variable rates ranging from 2.1–15.4mm/year.
There is evidence that a subgroup of patients with acute uncomplicated aortic dissection are at higher risk of developing late aortic complications including aortic dilation.20, 21 Specifically, one study demonstrated that >50% patients with a patent false lumen and an aortic diameter ≥40mm had a dissection related death, rupture, malperfusion or late aneurysmal dilation within 5 years (p=0.04).22 This risk factor was confirmed in another retrospective review of 141 patients, 42% of which had a late aortic event.23 In this study, false lumen ≥40mm was highly predictive or late complications (OR=3.18).23 In another study of 100 patients, aortic diameter was not associated with late complications but rather false lumen diameter ≥22mm at the proximal descending aorta.24 Patients with false lumen diameter ≥22mm had higher mortality (17% vs. 5%, p=0.09) and late aneurysmal formation (42% vs. 5%, p<0.001) compared to patients with a <22mm initial false lumen.24 Based on these retrospective studies, there may be a subgroup of patients at highest risk of complications for dissection.
Computed tomographic aortography (CTA) is the gold standard for diagnosing aortic dissections and has supplanted other imaging tests.5, 25 CTA is rapid, widely available, and is highly sensitive (98–100%) and specific (95–98%).26 A dedicated CTA protocol with 2.5–5.0mm section collimation and 1.5–2.5mm spacing during 120–200ml iodinated contrast administration produces optimum aortic definition, and allows for high resolution anatomic detail and reconstructions.27 Transesophageal echocardiography (TEE) administered by experienced operators can be a rapid and highly sensitive test to diagnose an aortic dissection, but is limited by poor visualization of the descending aorta distal to the diaphragm. In spite of these limitations, TEE is the second most utilized diagnostic modality, especially given the ability to diagnose an acute dissection rapidly in an unstable patient.25 Additionally, magnetic resonance angiography (MRA) with a sensitivity of 95–100% and specificity of 95–98% is an alternative diagnostic imaging modality, but requires availability and patient stability.25 Currently, most diagnostic algorithms de-emphasize the role of catheter based aortography due to the lower sensitivity and catheter related risks. Intravascular ultrasound (IVUS), an evolving technology, has demonstrated 100% sensitivity and specificity during endovascular diagnosis and treatment of complicated acute Type III aortic dissection.28
Complicated acute aortic dissection requires urgent or emergent intervention by an open surgical or endovascular approach (Figure 2). Patients with uncomplicated acute aortic dissection have typically been treated with medical therapy. The advent of endovascular therapy is challenging the management of complicated and uncomplicated acute aortic dissection, and offers an opportunity for improved outcomes and definitive treatment of this disease.
In the absence of radiographic deterioration of the aorta or evidence of malperfusion, intense antihypertensive therapy has long been accepted as the initial management of patients with Type III uncomplicated aortic dissection (Table 2).29 Preexisting hypertension is seen in 70% of patients with Type III aortic dissection compared to only 25–35% in patients with Type I or II aortic dissections. The overall goal in the acute management of dissection is blood pressure reduction and minimizing ΔP/ΔTmax on the aortic wall. Close observation should be performed in an intensive care unit with systolic blood pressure regulation <140mmHg, minimizing the risk of a progressive dissection while preventing low flow states to the central nervous system and viscera.30 Ideally, the systolic blood pressure should be titrated to 90–110mmHg as tolerated by the patient. Beta-blockade, with esmolol or labetolol, should be the initial therapy to directly decrease ΔP/ΔTmax. Once maximal beta-blockade has been achieved, secondary agents like nitroprusside, nitroglycerin, fenoldopam, and/or calcium channel antagonists may be administered for enhanced control of hypertension. Management of patients with uncomplicated Type III aortic dissection remains contentious. Despite aggressive medical management patients remain at a risk for late aneurysmal dilatation, with 30-day mortality ranging from 1.6% in some series to 30% in others (Table 2). Late rates of aortic dilation requiring intervention exceed 40% in some series.24 Collectively, these studies highlight the high long term mortality and morbidity rates for medical therapy, despite its accepted practice, for uncomplicated Type III dissections.
Indications for an operation or intervention in acute Type III aortic dissection include: (I) a rapidly enlarging aortic diameter (aneurysmal degeneration), (II) progression of the dissection, (III) aortic rupture, (IV) malperfusion syndrome, and (V) persistent or recurrent intractable pain.21, 31, 32 Many centers argue that pain is a soft indication for an operation in Type III aortic dissection, with only 4% of patients with recurrent pain developing complications.33 The method of open surgical repair typically involves identifying the entry tear and replacing the involved dissected aorta with a graft sewn to the true and false lumens. Pledgets and/or biologic glues are often used to reinforce the suture lines to help obliterate the false lumen. Open surgery in the acute setting is not without significant risk even at experienced centers. Baylor University (Houston, TX) reported a 22% operative mortality in their series of patients who underwent early surgical intervention.34 Similarly, the University of Texas (Houston, TX) reported a 17% operative mortality for aortic dissection with organ or limb ischemia.35 At the Massachusetts General Hospital (Boston, MA), the mortality in patients with acute complicated Type III aortic dissection improved from 37% from 1965–1986 to 18% from 1990–1999.32 This reduction was attributed to earlier diagnosis, aggressive medical care, and adherence to complication-specific-therapy. Moreover, while open surgery addresses the life-threatening event, 25–50% of patients still have persistent flow in the false lumen.36 With residual dissected aorta, patients remain at risk for late aneurysmal degeneration, requiring serial imaging surveillance, close blood pressure monitoring, and medical therapy.36 Clearly, open surgery has significant risks and carries a high morbidity.
Endovascular repair is rapidly becoming a preferred approach by many centers for treatment of distal dissections. The anatomic goal with endovascular repair is to cover the entry tear preventing blood from entering the proximal dissection resulting in false lumen thrombosis and aortic remodeling (Figure 3). In the setting of malperfusion (complicated dissection), reestablishing flow in the true lumen allows blood flow to enter the visceral vessels and often treats the malperfusion. There is further evidence that this approach may reduce the potential for early and late complications of aortic dissections.37 The long term aim of endovascular repair of dissections is to prevent: (I) early rupture of the dissection, (II) progression of the dissection, and (III) late aneurysmal formation. The precise timing of endovascular therapy remains unclear and often is dependent upon the patient’s premorbid condition and dissection anatomy.
Two recent large meta-analyses support endovascular therapy in the management of acute Type III aortic dissection. In a report by Parker and Golledge, 942 patients from 29 studies had an in-hospital mortality of 9% with reintervention rate of 10.4%.38 Emergency surgical conversion was rare (0.6%), and major complications were uncommon, although periprocedural stroke occurred in 3.1% of patients.38 Retrograde Type I aortic dissection developed in 2% of patients, while 1.9% developed paraplegia.38 Survival was an impressive 88% at mean follow-up of 20 months. Endovascular re-intervention was performed in 7.6% of patients, while surgical re-intervention was necessary in 2.8% of patients.38 The authors suggested that endovascular treatment for acute Type III aortic dissection as favorable given their initial outcomes.38
A second meta-analysis by Xiong et al. was just published in 2009, studied 1304 patients (n=39 studies) from China who underwent endovascular repair of acute complicated Type III aortic dissections. A technical success rate of ≥99%, with 30-day mortality of 2.6% was reported.39 At late follow-up, false lumen thrombosis occurred in 92.9% of patients, and surgical conversion was required in 0.8% of cases, with endovascular re-interventions performed in 1.6% of patients.39 In this report, retrograde extension into the ascending aorta (0.4%) and neurologic complications (0.6%) were both rare.39
The significance of these two studies is that they were performed in two different patient populations (Western centers versus Chinese centers), had excellent technical success, low complication rates, and high success rates of false lumen thrombosis. Collectively, these findings support the use of endovascular management in acute Type III aortic dissection.
Typically, dynamic obstruction responds to aortic endografting while static obstruction may require additional branch artery stent grafting or aortic fenestration.11 Thus, additional endovascular procedures may be required in patients with malperfusion on presentation. Patients with a dominant left vertebral artery or patent left internal mammary artery bypass should undergo carotid to subclavian bypass, and carotid-carotid bypass if necessary, if coverage of the origin of the left subclavian or left common carotid arteries is necessary.16, 40 Additionally, carotid to subclavian bypass should be considered when extensive coverage of the descending thoracic aorta is required in patients with concomitant abdominal aortic aneurysm due to the loss of the intercostal artery supply to the spinal cord. This results in preservation of the vertebral artery and the thyrocervical trunk collaterals to the anterior and posterior spinal arteries that may be critical to decreasing the risk of paraplegia.11 Similarly, a planned laparotomy either immediately or 48 hours after acute complicated Type III aortic dissection stabilization, to ensure adequate visceral perfusion has been suggested. Finally, femoral to femoral cross-over bypass may be required if the dissection flap occludes flow unilaterally.
Currently, there are no devices approved by the United States (US) Food and Drug Administration (FDA) for use in aortic dissections. There are several trials scheduled to begin enrolling patients in early 2010 to evaluate the use of newer devices on the horizon in the management of aortic dissection. Today, devices approved for thoracic aneurysm repair are used off-label for aortic dissections. Wall stents are preferred since the goal is to cover the entry tear. Uncovered stents that are available at present are less ideal due to higher radial forces on the aortic wall. Aortic dissection anatomy (tortuosity and involvement of branch vessels), adequate peripheral access, and assessment for extent of coverage necessary are important for technical success. Preoperative carotid subclavian bypass should be considered in the setting of dominant left vertebral artery or patent left internal mammary artery bypass, if coverage of the left subclavian orifice is likely. In addition, extensive coverage of the thoracic aorta should warrant preoperative lumbar drain insertion. Femoral artery access is the most common approach for stent delivery. Through-and-through access via the brachial and femoral arteries is often useful as well. The intimal tear location must be clearly identified both on preoperative imaging and prior to device deployment.41 The utility and benefit of IVUS to localize the entry tear is well established. Confirming wire placement in the true lumen by IVUS is paramount, as endograft deployment in the false lumen is catastrophic. Adequate proximal and distal landing zones are required, often necessitating coverage of the left subclavian artery. Pharmacologic hypotension is used to minimize migration during device deployment. A properly positioned stent graft occludes the proximal tear, results in decompression and thrombosis of the false lumen (typically within 3 months), and re-expands the true lumen (Figure 4). Residual malperfusion following coverage of the proximal entry tear may require further fenestration of the dissection flap and direct stenting of branch ostia.
Although there are no completed randomized studies in the treatment of complicated aortic dissection, a number of small series have lent to the collective experience (Table 3). Slonim et al. were one of the first groups to report endovascular treatment of ischemic complications caused by true lumen obliteration in aortic dissection, and found stent placement to be effective with a 9.1% 30-day mortality.42 Dake et al. and Nienaber et al. in two landmark papers presented the concept of endovascular stenting for acute Type III aortic dissection. Dake and colleagues reported the Stanford experience with their first generation homemade device deployed in 19 patients with 79% rate of false lumen thrombosis at 13 months follow-up.43 Nienaber and colleagues reported superior outcomes and more rapid functional recovery in 12 consecutive endovascular patients with acute Type III aortic dissection compared to matched open surgical cases.44 These two early series marked the shift in endovascular treatment paradigms of acute Type III aortic dissections.
In 2001, an Austrian group was one of the first to report successful false lumen obliteration in four patients with complicated Type III aortic dissection with a commercially available device.45 In a subsequent study, a group from Tulane (New Orleans, LA) reported false lumen thrombosis in nine out of ten patients with Type III aortic dissections; however, 30% of these patients still developed late abdominal aortic aneurysm.46 In 2005, Dialetto et al. compared 28 complicated acute Type III aortic dissection patients to a similarly matched uncomplicated medically managed cohort.47 The stented group had a 10.7% in-hospital mortality, compared to none in-house deaths in the medical group (p=0.24).47 However, mid-term mortality was equivalent: 10.7% in the endovascular group versus 14.3% in the medically managed group (p=0.71).47 Importantly, follow-up imaging demonstrated complete thrombosis of the false lumen in 75% of cases in the endovascular group compared to only 10.7% in the medical group (p=0.0001).47 These initial trials supported stent grafting in complicated acute Type III aortic dissection. Then in 2006, Resch et al. presented their results from a multi-center Swedish trial and included 115 patients who underwent stent placement for acute complicated Type III aortic dissection.48 Thrombosis of the false lumen at the level of the stent graft occurred in 80% of patients while 50% of patients still had persistent filling of the false lumen in the abdominal aorta.48
Analysis in many of these studies is limited due to the inclusion of both acute and chronic aortic dissection patients. In one of the largest series in the US, Szeto et al. reported the experience from the University of Pennsylvania (Philadelphia, PA). In this series of 35 patients all of whom underwent thoracic stenting for acute complicated Type III aortic dissection, a 97.1% technical success rate and 93.4% one-year survival rate was achieved.49 Similarly, the group from Stanford (Stanford, CA) reported their experience in 16 patients with acute complicated dissection who underwent endovascular treatment with custom made grafts within 48 hours of presentation.50 The study reported an actuarial survival of 73% and 67% at 1 and 5 years, respectively.50 A recent study from Malmö University Hospital, Sweden reported perioperative morbidity and mortality of 18% and 11%, and a re-intervention rate of 22% in 52 patients who underwent stent repair for complicated acute Type III aortic dissection.51
Endovascular stent graft repair appears to be a feasible approach in patients with acute complicated aortic dissection. In the setting of malperfusion, some centers advocate percutaneous fenestration of the dissection flap by a number of approaches including the scissor technique.42, 52 With this approach, stiff guide wires are placed in each lumen from a single femoral access, then by advancing a single long sheath over the two wires, the dissection flap is fenestrated.53–55 In 2009, Patel and Williams reported their experience from the University of Michigan (Ann Arbor, MI) in 69 patients with acute malperfusion.56 Fenestration of the dissection flap followed by expectant management resulted in freedom from aortic rupture or open repair at 1, 5, and 8 years of 80.2%, 67.7%, and 54.2%, respectively.56 It should be noted that these outstanding results reported in this series are center-specific, attributable to their endovascular/interventional radiology technical expertise in aortic fenestration, and have not been replicated by other centers.
Cumulatively, these studies support the concept of aggressive endovascular management of acute complicated Type III aortic dissection to address the acute ischemia and/or rupture and minimize the risk of long term complications. Currently trials are underway to evaluate this therapy prospectively. For example, the Zenith ® Dissection Endovascular System (William Cook Inc., Europe) trial is enrolling patients who have ”complicated” dissections with branch vessel compromise, peri-aortic effusion, transaortic growth ≥5mm within 3 months, or transaortic diameter ≥40 mm and is expected to be completed by 2013. The outcomes of this trial will include mortality and complications at 30 days and 1 year. Similarly, the VIRTUE trial and registry recently completed enrollment to evaluate the Valiant™ Endoluminal Stent-Graft and Xcelerant ® Delivery System (Medtronic Inc., Santa Rosa, CA) in complicated Type III aortic dissection patients. One hundred patients at 14 European centers will participate in this registry to collect health economic, and clinical performance data on Valiant™ for the treatment of acute dissections, complicated or expanding sub-acute dissections, and expanding chronic dissections. Other clinical trials to evaluate the role of endovascular therapy in complicated acute Type III aortic dissection are awaiting FDA approval.
While the majority of data investigating the efficacy of stent grafts for uncomplicated aortic dissections is retrospective, one prospective randomized trial has been completed. The INvestigation of STEnt Grafts in Patients With Type B Aortic Dissection (INSTEAD) is a well known European trial comparing endograft and adjunctive medical therapy, versus optimal medical therapy alone in patients with uncomplicated Type III aortic dissections. A total of 136 patients with uncomplicated Type III aortic dissections and a patent false lumen between 2 and 52 weeks after diagnosis were enrolled, and followed at 3, 12, and 24 month intervals. This trial exclusively used the Talent™ graft (Medtronic Inc., Santa Rosa, CA). Though final results from the INSTEAD trial have not been published, preliminary results demonstrated 7% one-year mortality in the stent arm, and 3% one-year mortality in the medical arm (p=NS). Importantly, medically treated patients had a 10% cross-over to the surgical arm in this intent-to-treat analysis.57 The INSTEAD trial has been criticized due to: (I) the inclusion of all patients with Type III aortic dissection rather than a subgroup of “high risk” patients, (II) the inclusion of patients who received stent graft therapy from 2 weeks to 1 year following the acute presentation, potentially resulting in a selection bias of survivors of acute events in the medical arm, and (III) a high mortality in the stent graft arm compared to other reports.
An additional trial entitled the Acute Dissection Stent-grafting Or Best Medical Treatment (ADSORB) trial is currently enrolling patients in Europe. This trial will randomize 270 patients with acute (<2 weeks) uncomplicated Type III aortic dissection to stent graft (Gore TAG®, WL Gore, Flagstaff, AZ) versus best medical therapy (strict blood pressure goal <125/80mmHg). No cross-over is planned in this multicenter prospective trial. This crucial study is expected to be completed in late 2012 and will evaluate false lumen thrombosis, aortic rupture, and aortic dilatation as the primary end points.
As the collective experience of endovascular repair of aortic dissections grows, there are many reports that include patients with both uncomplicated and complicated presentations. A review of the studies reporting their experience in both complicated and uncomplicated patients is presented in Table 4 with notable studies discussed below.
Between 1996 and 2001, Palma et al. found that 58 of 70 patients treated for an aortic dissection underwent elective stent graft placement for Type III dissection.58 Custom stents were used and no paraplegia occurred despite the use of multiple stents in many patients.58 Subsequently, in 2002 Kato et al. reported outcomes from 38 patients stented for uncomplicated and complicated Type III aortic dissection.59 Compared to the uncomplicated group, patients with complicated dissection had higher 30-day mortality and postoperative complication rates (p<0.05).
In 2002, Bortone et al. described outcomes of acute and chronic Type III (electively treated) aortic dissections, laying the foundation for the aforementioned INSTEAD trial.60 The group reported that immediate endovascular management avoided the potential of high-risk surgical procedures, had lower postoperative complications with a shorter hospitalization, when performed within 2 weeks of the initial diagnosis of aortic dissection. These findings contrast the 2005 study from the IRAD database by Tsai et al., suggesting patients with uncomplicated Type III aortic dissections are best treated by medical therapy given the 10% 30-day mortality.61
Leurs et al. in 2004 presented the experience with endovascular treatment of thoracic aortic aneurysms (n=249) and dissections (n=131) collated by the European Collaborators on Stent Graft Techniques for Thoracic Aortic Aneurysm and Dissection Repair (EUROSTAR) and the United Kingdom Thoracic Endograft registries.62 They reported 89% technical success in dissection patients, with 6.5% mortality compared to 9.3% all-cause mortality for the study.62 Studies that have compared endovascular treatment of acute versus chronic aortic dissections indicate more risk in the acute setting. Eggebrecht et al. performed a meta-analysis compiled from 609 patients, and reported that overall complications were significantly higher in patients undergoing stent-graft placement for acute aortic dissection than in patients with chronic aortic dissection (p=0.005).63 The overall 30-day mortality was three-fold higher in patients with acute aortic dissection when compared with the chronic group (p=0.015).63 These reports have yielded caution in early manipulation and stenting of dissected aortas with the present devices.
Acute Type III aortic dissection is associated with high morbidity and mortality. Treatment options include medical management, open surgery and endovascular techniques. There is an accumulating body of evidence suggesting poor long term outcomes with medical therapy for uncomplicated dissection. The growing experience with endovascular management of both complicated and uncomplicated Type III aortic dissection suggests this approach has promising results. As such, there is a shifting paradigm in the management and treatment of acute Type III aortic dissection at many centers towards an endovascular approach. This review provides the cardiovascular surgeon an updated perspective on the management of acute complicated and uncomplicated Type III aortic dissection. While currently there are no FDA approved devices for aortic dissection, newer devices designed to treat this disease are on the horizon. Medical therapy in uncomplicated patients has unacceptable long term mortality and late complications. Randomized multicenter trials comparing endovascular therapy to medical management and open surgery for the treatment of acute complicated and uncomplicated Type III aortic dissection are necessary.
Supported in part by National Institutes of Health Cardiovascular Surgery Research Training Grant T32/HL007849 (CMB)