Static Branch Involvement

One manifestation that may arise when the advancing dissection septum intersects an aortic branch is static branch vessel involvement (Fig. 36-4). In static involvement, the aortic dissection flap extends directly into the branch for a variable distance. In contrast to the geometry described earlier, orientation of the septal trajectory is such that the branch ostium is incompletely engaged by the edge of the dissection plane. Rather than being circumferentially shorn by the septum, there is only partial circumferential involvement of the branch by the dissection. The aortic flap extends into the branch, creating a false lumen within the artery. As a result, the individual branch has both a true and false lumen like the aorta.

Figure 36-4 Magnetic resonance imaging (MRI) of static branch vessel involvement of iliac arteries.

A, Coronal image of pelvis identifies a flap extending into right common iliac artery (CIA) and down to distal external iliac artery (EIA). Flow is evident in both true and false lumens within the right iliac artery. This is an example of static involvement with direct flap extension from the aorta into a branch. At the end of the dissection within the EIA, there is a distal reentry tear. This terminal tear establishes double-barrel flow within the iliac artery, which is rarely associated with ischemic symptoms. B-C, Similar coronal views show flap extension into left iliac system, but a segmental flow void (black segment) within false lumen of left CIA. This is associated with a no-reentry situation at distal extent of false lumen. No reentry within a branch is typically associated with obstruction of true lumen by a dilated false lumen cul-de-sac. The flow void noted may represent thrombosis in a blind channel or simply no blood flow. The usual consequence of this phenomenon of no reentry is branch vessel ischemia due to a lack of false lumen perfusion and compromised true lumen branch flow.

Similar to the aorta, a branch affected by static involvement may have multiple fates. At the end of the dissection where the flap terminates in the branch, a reentry tear in the false lumen may or may not occur. If a reentry tear occurs at the end of the false lumen, branch perfusion results from blood flow in both the true and false lumens. In many such cases, dual lumen perfusion is not associated with ischemic branch vessel symptoms. If reentry does not occur in cases of static branch vessel involvement, however, the false lumen within the branch has no outflow. The absence of a distal tear to allow communication with the vascular bed beyond the dissection may impair blood flow significantly. This no reentry state within the branch’s false lumen renders perfusion limited to that contributed by the true lumen. Unfortunately, the true lumen may be compromised by the engorged false lumen. The blind pouch of the false lumen, without outflow, swells to a maximum dimension at its distal end. The pressure exerted by the false lumen severely distorts and compresses the true lumen to markedly reduce branch vessel flow. Commonly, the degree of ischemia experienced by the involved vascular bed may be significant and can lead to irreversible tissue necrosis if not relieved quickly.

In no-reentry situations, a local solution directed at improving flow within the affected artery is required because the problem is localized within the specific branch. Two options for endovascular treatment are possible. Resistance to outflow within the false lumen may be decreased by creating a distal tear or fenestration within the blind channel. This can be accomplished with the end of a guidewire or other endovascular probe placed within the false lumen through the aortic false lumen. This approach is associated with practical challenges, including the avoidance of distal extension of the dissection process, safe penetration of the false lumen wall to create an effective outflow tear, and determination of the presence of thrombus within the blind sac of stagnant false lumen blood to avoid its distal embolization.

In most cases, the preferred strategy involves increasing branch flow by decreasing the resistance to true lumen blood flow. This is performed by placing a stent in the true lumen of the branch through catheterization from the aortic true lumen. The stent is typically placed from beyond the end of the false lumen in the branch back to the aortic true lumen. A self-expanding nitinol stent is commonly employed because this distance is frequently greater than 2 cm and because there is a risk of squeezing any existing clot out of the false lumen with a balloon-expandable stent. These stents are sized to the total transarterial diameter of the branch and allowed to progressively expand on their own (post deployment) without supplemental balloon dilation. There are many successful reports of this approach in mesenteric, renal, and iliac arteries affected by no-reentry or static involvement.^8,9,18,19

Occasionally, static branch vessel involvement with reentry anatomy and double-barrel flow may require endovascular intervention. The most common indication for stent placement in this setting occurs with involvement of a renal artery (Fig. 36-5). The kidney supplied by a dissected renal artery may be affected by the physical presence of a flap within the branch. The variable flow reduction caused by the flap, and resultant disrupted pattern of true and false lumen perfusion, may contribute to an exacerbation of hypertension. In cases where high blood pressure is sustained and recalcitrant to numerous intravenous (IV) medications, endovascular intervention may be warranted to restore a single lumen without flap. The approach to treatment involves placement of a balloon-expandable renal stent within the true lumen of the renal artery through the aortic true lumen. In most cases, this type of reentry involvement does not extend into the branch as far as the no-reentry extension. Thus, stents less than 2-cm long are typically implanted. This technique is well established at most centers that manage cases of aortic dissection frequently.

Figure 36-5 Computed tomography (CT) images of true and false lumen relationships to renal arteries.

Axial (A) and coronal (B) CT images at level of left renal artery show that left renal artery is supplied by the false lumen (left aortic lumen). True lumen is located along right wall of aorta, and flap shows a characteristic natural fenestration or defect corresponding to left renal ostium. Flap around fenestration has small tail-like extensions pointing to the left that represent the initial few millimeters of left renal intimal lining that were torn away with the retracted aortic septum.

Dynamic Branch Involvement

In addition to primary branch pathology that occurs as a complication of aortic dissection, another mechanism, dynamic branch vessel involvement, may be responsible for organ ischemia. Dynamic branch involvement is a phenomenon associated with obstruction to branch vessel flow by an aortic septum that has prolapsed over the branch ostia like a curtain. In contrast to static involvement, where the aortic flap extends directly into a branch, dynamic obstruction occurs as an aortic process exclusively without an associated branch lesion. Propagation of the aortic flap may create a circumferential cleavage of the aortic wall surrounding the branch ostium (Fig. 36-6). Factors associated with this event include the flap trajectory, the resultant orientation of the septal plane proximal to the branch, and the inclusion of the ostium by the cleaved flap as it extends past. In this situation, the dissection septum surrounds the branch ostium as it tears distally. The cleavage plane extends 1 to 2 mm into the branch, and then circumferentially reenters, creating a cylindrical tear, coring out a short segment of the intimal/medial lining of the most proximal aspect of the branch. The septum retracts into the aortic lumen with a fenestration corresponding to the branch orifice. This gives the flap a stencil-like appearance when viewed en face, with the number of holes related to the number of branch vessels involved by this phenomenon. When imaged in an axial plane, the affected artery appears to originate exclusively from the aortic false lumen. Closer inspection usually allows identification of a tear in the flap at the level or adjacent to the level of the branch. The flap often displays small projections angled from the edge of the tear, giving its outline on axial imaging an appearance similar to the contour of a metal rivet, the short-legged extensions corresponding to the amputated proximal lining of the branch.

Figure 36-6 Magnetic resonance imaging (MRI) demonstrates dynamic branch vessel involvement, with aortic true lumen collapse and accompanying static no-reentry obstruction of left common iliac artery (CIA).

A-C, Axial MRI shows wafer-thin crescent-shaped true lumen collapsed against anterior aortic wall at level of visceral arteries. Aortic septum prolapses like a curtain across the origins of branches originating from true lumen, with resultant malperfusion and multiorgan ischemia. D, At the level just below aortic bifurcation, there is marked asymmetry in appearance of CIAs. Lumen of left CIA has a flow void (black circle) due to static involvement without reentry that coexists with the dynamic process observed more proximally.

In dynamic branch obstruction, hemodynamic flow patterns result in a large aortic false lumen with a diminutive or collapsed true lumen. There is variability, however, in the degree of true lumen obliteration related to the dynamic compromise.

In the majority of aortic dissection cases with true and false lumen aortic flow (often called double-barrel flow), the process described does not cause critical branch perfusion abnormalities. Flow to the branch originates primarily from the false lumen, with a small contribution from the true lumen through the corresponding fenestration in the aortic septum. Most of the false lumen flow usually occurs in diastole. During systole, the small contribution from the true lumen arrives through the septal window into the false lumen and branch. If the proximal primary tear is very large or the entry tear is in close proximity to the branch, the dominant flow pattern supplying branch perfusion may be in systole. In general, a branch that originates exclusively from the aortic false lumen is rarely affected by an ischemic complication.

Consistently, the aortic septum prolapses with a convex contour toward a compromised crescent-shaped true lumen. Consequently, all branches originating from the true lumen are at risk of obstruction. In this regard, the aortic septum in a dynamic obstructive process often assumes a coronal position, oriented across the aorta from left to right, in the distal descending thoracic proximal abdominal aortic segments. Consequently, the anteriorly oriented mesenteric vessels are in peril of ischemia because they frequently originate exclusively from a miniscule aortic true lumen. The likelihood of developing clinically relevant dynamic branch vessel compromise appears related in part to the area of the proximal entry tear. Although the process of dynamic involvement is dependent on multiple factors, as a general rule, the more severe the true lumen collapse, the larger or more circumferential the size of the proximal primary entry tear. Management of more than one ischemic vascular bed related to dynamic branch involvement and an obliterated aortic true lumen that supplies the compromised branches is most expeditiously and effectively approached by an endovascular aortic procedure rather than a strategy directed at the individual branches.

More than one mechanism of branch involvement can coexist in any given patient. The clinical manifestations and the analysis of imaging for any patient requires an individualized approach that must synthesize information and aortic and branch vessel involvement to customize an optimal treatment strategy that will safely, successfully, and durably address the most compelling effects of the dissection.

Stent Grafts for Uncomplicated Type B Dissection

The Investigation of Stent Grafts in Patients with Type B Aortic Dissection (INSTEAD) trial observed that elective stent graft placement in survivors of uncomplicated chronic type B dissection does not improve 1-year survival and adverse event rates compared with medical therapy. Among the 140 patients randomized in this prospective trial, 1-year survival was 91% compared with 97% in patients randomized to medical therapy.^34,35 Moreover, aorta-related mortality was not different, and the risk for the combined endpoint of aorta-related death (rupture) and progression (including conversion or additional endovascular or open surgical intervention) was similar.

In the setting of complicated aortic dissection, medical management is associated with a high mortality rate, such that most patients will undergo surgery to address life-threatening complications.^2,4 Depending on the patient’s underlying medical conditions and the nature of the complication(s), surgical mortality rates range from between 30% and 60% or higher.^24,25 It is in these high-risk scenarios that an opportunity exists to establish a role for interventional management. Thus the question becomes, What constitutes complicated type B aortic dissection? There is no strict definition for this category of disease, but traditionally it is relegated to two unambiguous disease manifestations: aortic rupture (Fig. 36-7) and symptomatic branch vessel involvement. These conditions are clear and their diagnosis unequivocal. Other adverse effects of the dissection process, such as uncontrollable hypertension, unrelenting pain, and increasing pleural fluid, defy easy classification and do not have uniform criteria for comparative assessment. These so-called softer indications for intervention are commonly included as a surgical indication in most published series of acute complicated dissection.^8,26

Figure 36-7 Acute type B aortic dissection with rupture in a 68-year-old woman.

A, Frontal chest radiograph upon presentation to emergency room with severe back pain and hypertension that occurred while gardening. B, Axial computed tomography (CT) scan after contrast media administration shows typical appearance of aortic dissection in mid-descending aorta. C, Repeat chest radiograph performed after transfer to referral facility 4 hours after initial study, with marked interval change including opacification of left hemithorax from leaking blood.

Endograft Treatment of Complicated Type B Dissection

The procedural goal for endovascular stent grafting in patients with complicated acute type B aortic dissection is endograft elimination of blood flow entry into the proximal entry tear. Obliterating the primary communication between the true lumen and the false redirects pulsatile flow into the true lumen, promotes false lumen thrombosis, and ultimately improves remodeling of the aorta by increasing the dimensions of the true lumen while shrinking the false lumen (Fig. 36-8).

Figure 36-8 Treatment and follow-up imaging of type B aortic dissection with rupture.

A, Aortograms pre- and post placement of a thoracic endograft across mid-descending aorta entry tear of a type B dissection in the 68-year-old woman described in Figure 36-7. B, Series of axial computed tomography (CT) images obtained 1 week postendograft management of a type B dissection with rupture. Stent graft is in good position, and false lumen is thrombosed. Residual extravascular blood and hematoma are evident.

Specific procedural techniques vary depending on the precise complication. Faced with dynamic branch vessel involvement and clinically relevant obstruction compromising flow to one or multiple branches, the procedural strategy focuses on unloading the aortic false lumen by increasing resistance to false lumen inflow or decreasing resistance to its outflow. The former is attempted by deploying an endograft over the proximal primary entry tear and rechanneling all flow into the true lumen. Logistically, this typically involves placement of a 15-cm-long (range 12-20 cm) stent graft from the nondissected segment of aorta proximal to the primary intimal tear, commonly between the origins of the left carotid and left subclavian arteries. This may require intentional partial or complete coverage of the left subclavian origin. The distal extent of the device usually remains above the diaphragm. The diameter of the implant selected is based on the transaortic dimension of the nondissected aorta just proximal to the dissection, rather than the size of the true lumen or transaortic diameter of the dissected segment.

Endovascular Treatment of Branch Vessel Involvement

The outcomes of stent graft therapy for reversal of dynamic branch vessel involvement are excellent, with procedural success in up to 95% of cases and complete false lumen thrombosis in 85% of patients.^7–9 These procedures are associated with 67.7% 5-year freedom from aortic rupture and open repair.⁹ Additionally, static branch involvement remote from the covered proximal aortic entry tear may require separate targeted intervention to manage residual ischemic compromise. This is especially important in cases with no-reentry anatomy complicating static branch involvement. In these situations, endovascular branch intervention should be provided emergently.

An alternative to endograft placement in dynamic branch compromise is distal flap fenestration.^9,27 Percutaneous balloon fenestration of the aortic septum has replaced the operative procedure. Balloon fenestration of the septum is designed to unload the aortic false lumen by decreasing the resistance to outflow. Technically, initial transgression of the aortic flap with a small cardiac transseptal TIPS needle and cannula usually is performed from the small true lumen into the larger target of the false channel. The site of the needle puncture commonly lies within the infrarenal aorta at the level of the aortic bifurcation. Once successful transgression of the septum is confirmed, a wire is advanced across the flap and well into the targeted lumen. Sequentially larger balloon dilation of the flap is performed until a final size of between 20 and 25 mm is obtained.

Balloon fenestration causes a linear transverse tear in the flap that allows greater mixture of blood between the two aortic channels and decompresses the true lumen. These effects must be confirmed by aortography or intravascular ultrasound (IVUS) to ensure relief of the dynamic pattern of branch obstruction. After these two endovascular (endograft or fenestration) procedures, imaging comparisons of the anatomical effects (with computed tomography [CT], magnetic resonance imaging [MRI] or IVUS), including changes in the size of the aortic lumens, typically demonstrate a more dramatic result following endograft management. Specifically, the magnitude of true lumen expansion with stent grafting is greater than that observed after distal flap fenestration. Because false lumen fenestration promotes flow in the false lumen, whereas endograft placement promotes false lumen thrombosis, the latter is thought to be a superior method to minimize late aneurysm formation. Consequently, the opportunities for percutaneous balloon fenestration are decreasing now that thoracic endograft availability has improved. Fenestration is typically limited to situations when stent grafts are unavailable or when the specific aortic anatomy is unsuitable for endograft placement.

Aortic Rupture

Rupture that complicates aortic dissection is an interventional imperative.^28,29 The procedural considerations for aortic rupture focus on preventing exsanguination. Both open surgical and endovascular therapies are associated with high mortality and morbidity rates in the presence of aortic rupture. Recent reports suggest that endovascular approaches permit treatment of more patients, including older and less fit individuals whose operative risk in this setting is prohibitive.^21,28,29

Localizing the precise site of rupture noninvasively is not always possible. The point of rupture through the false lumen wall may be evident by the presence of contrast enhancement beyond the anticipated aortic border, though this occurs typically in the setting of severe hemodynamic instability or shock (Fig. 36-9). More commonly, a periaortic, mediastinal, and/or pleural collection is evident on CT imaging, which has an appearance and attenuation value consistent with hematoma or complex fluid. This abnormality may be most prominent around a focal aortic segment or extend diffusely over a wider zone.

Figure 36-9 Endograft management of aortic dissection with rupture.

A, Axial and sagittal computed tomography (CT) images of 59-year-old man with acute type B dissection with primary tear distal to left subclavian artery and retrograde extension into the proximal arch (DeBakey class IIID) complicated by rupture. Axial projection shows a large quantity of extravascular fluid, and sagittal image shows a faint wisp of contrast extravasation above aorta, just distal to subclavian artery. B, Three views from the stent graft procedure, with the left and middle panels before device placement, and the right panel after deployment. A good result is evident, with contrast opacification of the true lumen only.

The goal of endograft management for aortic rupture is coverage of the proximal entry tear, with isolation of the false lumen, to ensure false lumen obliteration and expeditious thrombosis. It is thrombosis of the false lumen that prevents aortic leakage of blood. To facilitate rapid false lumen thrombosis, the overall endograft coverage of the aorta is often longer than that used for other thoracic pathologies. By extending the length of coverage (20-30 cm) to at least the level of the diaphragm or celiac trunk, the aortic septum is braced by the stent in the true lumen, and the thoracic false lumen is converted to a long, inverted cul-de-sac, or blind pouch. Then with flap pulsation limited by the buttressing stent, blood in the false lumen becomes stagnant and prone to thrombosis.^30–32

False lumen thrombosis is critical because the precise rupture point in any individual patient is frequently unknown, and the breech may exist well below the entry tear. Simple coverage of the proximal entry may then eliminate direct flow into the false lumen, but if distal retrograde flow from abdominal sources persists, the risk of a continued leak exists and morbidity remains. Although this strategy is associated with considerable mortality and procedural complications, it represents an addition to the existing treatment armamentarium.

Other Indications for Aortic Endografts

The question of unidentified patient subgroup(s) who present with uncomplicated acute type B aortic dissection who may benefit from endograft placement remains. Some investigators have identified certain high-risk features in patients with acute uncomplicated type B dissection that may portend an increased risk of early aneurysm formation and increased mortality. These features include measurements of various aortic dimensions at the time of initial diagnosis. Initial attempts to propose high-risk criteria from CT imaging considered descriptive features associated with a poor prognosis and disease progression, such as a patent false lumen, a gaping and circumferential entry tear with resultant small true lumen, and a dominant false lumen with early fusiform expansion of the proximal descending aorta within 3 months of initial symptoms.

Marui et al. proposed that patients with uncomplicated aortic dissection and transaortic diameter greater than 40 mm were at high risk of rapid aortic expansion.³⁶ When applied to larger groups of patients with dissection, this benchmark provided modest prognostic value. The poor results encouraged others to focus on the issues and pursue more in-depth imaging analysis. Thereafter, Marui et al. offered an improved prognostic factor that was based on the extent of proximal descending aorta dilation at the time of initial diagnosis³⁷: the fusiform index. This index is defined as the maximum transaortic diameter of the distal aortic arch divided by the sum of the minimum diameter of the proximal aortic arch plus the aortic diameter at the level of the pulmonary artery. A value greater than 0.64 anticipates late aortic events in patients with uncomplicated type B aortic dissection. The investigators recommended that patients with these predictors should undergo early intervention with open surgery or stent graft implantation.

Immer et al. analyzed imaging studies (CT or MRI) over the initial 18 months after diagnosis in 84 patients with acute type A aortic dissection.³⁸ They concluded that a large false lumen at the time of the initial diagnostic scan is the strongest predictor of subsequent downstream aortic enlargement. This was especially true if the true lumen was less than 30% of the overall transaortic area 6 months after aortic surgery for repair of type A dissection.

This concept of the initial false lumen diameter as a determinant of late clinical deterioration was evaluated for type B disease in 2007 by Song et al.³⁹ These authors studied 100 consecutive patients with acute aortic dissection, including 51 with type A dissection and 49 with type B dissection. Over half of the patients underwent CT imaging follow-up through 24 months. Of these, an aneurysm (diameter > 60 mm) was diagnosed in 28%, with the maximal aortic diameter located in the proximal descending segment. A greater than 22-mm initial false lumen diameter of the upper thoracic segment of the descending aorta predicted late aneurysm formation with a sensitivity of 100% and a sensitivity of 76%. The 42 patients with an initial false lumen diameter greater than 22 mm had a higher event rate than the 58 with smaller false lumen aortic diameters (aneurysm, 42% vs. 5%; or death, 12% vs. 5%).

More recently, another predictive feature for early complication and clinical deterioration was described by Tsai et al. after reviewing data from the International Registry of Aortic Dissection (IRAD).⁴⁰ They reviewed 201 cases of type B acute aortic dissection. During the index hospitalization, 114 patients (56.7%) had a patent false lumen, 68 patients (33.8%) had partial thrombosis of the false lumen, and 19 (9.5%) had complete thrombosis of the false lumen. The mean 3-year mortality rate for patients with a patent false lumen was 13.7%, for those with partial thrombosis was 31.6%, and for those with complete thrombosis was 22.6%. Although postdischarge mortality was high among patients with acute type B aortic dissection, partial thrombosis, as compared with complete patency, is a significant independent predictor of postdischarge mortality (relative risk, 2.69; 95% confidence interval [CI], 1.45-4.98; P = 0.002).

In the future, it is likely that more sophisticated analysis will identify additional factors beyond simple dimensional aortic measurements to better predict patients with acute type B aortic dissection who are at increased risk of disease progression, rapid deterioration, or acute rupture. As prognostic evaluation of aortic dissection improves, the use of endovascular approaches will better target and improve outcomes of this disease.