If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
The study objective was to evaluate the management of malperfusion in acute type B aortic dissection with endovascular fenestration/stenting.
Methods
From 1996 to 2018, 182 patients with an acute type B aortic dissection underwent fenestration/stenting for suspected malperfusion based on imaging, clinical manifestations, and laboratory findings. Data were obtained from medical record review and the National Death Index database.
Results
The median age of patients was 55 years. Signs of malperfusion included abdominal pain (61%), lower-extremity weakness (27%), nonpalpable lower-extremity pulses (24%), and abnormal lactate, creatinine, liver enzymes, and creatine kinase levels. Confirmed hemodynamically significant malperfusion affected the spinal cord (2.7%), celiac (24%), superior mesenteric (40%), renal (51%), and iliofemoral (43%) arterial distributions. Of the 182 patients, 99 (54%) underwent aortic fenestration/stenting, 108 (59%) had 1 or multi-branch vessel fenestration/stenting, 5 (2.7%) had concomitant thoracic endovascular aortic repair, 17 (9.3%) had additional thrombolysis or thromboembolectomy, and 48 (26%) received no intervention. After fenestration/stenting, 24 patients (13%) required additional procedures for necrotic bowel or limb and 9 patients (4.9%) had subsequent aortic repair (thoracic endovascular aortic repair, open repair) before discharge. The new-onset paraplegia was 0%. The in-hospital mortality was 7.7% over 20+ years and 0% in the last 8 years. The 5- and 10-year survivals were 72% and 49%, respectively. The significant risk factors for late mortality were age and acute paralysis (hazard ratio, 3.5; both P < .0001). Given death as a competing factor, the 5- and 10-year cumulative incidence of reintervention was 21% and 31% for distal aortic pathology, respectively.
Conclusions
Patients with acute type B aortic dissection with malperfusion can be managed with endovascular fenestration/stenting with excellent short- and long-term outcomes. This approach is particularly helpful to patients with static malperfusion of aortic branch vessels.
Endovascular fenestration/stenting can effectively resolve dynamic and static malperfusion in ATBAD with favorable short- and long-term outcomes (survival and reoperation).
Endovascular fenestration/stenting effectively and timely resolves dynamic and static malperfusion in ATBAD with minimal risk of paraplegia and retrograde type A dissection, and excellent in-hospital mortality, cumulative incidence of reintervention, and long-term survival in this patient population as combined with TEVAR or open repair when indicated.
See Commentaries on pages 1162 and 1164.
Malperfusion, a feared complication of aortic dissection, is present in approximately 20% of type B aortic dissections
Clinical profiles and outcomes of acute type B aortic dissection in the current era: lessons from the international registry of aortic dissection (IRAD).
Clinical profiles and outcomes of acute type B aortic dissection in the current era: lessons from the international registry of aortic dissection (IRAD).
Clinical profiles and outcomes of acute type B aortic dissection in the current era: lessons from the international registry of aortic dissection (IRAD).
An International Registry of Acute Aortic Dissection study showed 28% mortality in acute type B aortic dissection (ATBAD) with malperfusion compared with 9.6% in those without malperfusion.
Clinical profiles and outcomes of acute type B aortic dissection in the current era: lessons from the international registry of aortic dissection (IRAD).
Malperfusion is defined as inadequate flow to a vascular territory, whereas malperfusion syndrome (MPS) is decreased flow to a vascular territory resulting in tissue/organ necrosis and end-organ dysfunction, both because of dissection-related obstruction of the aorta and its branch vessels. Patients with malperfusion more often undergo thoracic endovascular aortic repair (TEVAR) than open repair
Thoracic endovascular aortic repair for acute complicated type B aortic dissection: superiority relative to conventional open surgical and medical therapy.
to alleviate the end-organ ischemia. However, TEVAR in ATBAD cannot reliably resolve static malperfusion of aortic branch vessels, which results from thrombosis of the false lumen and compression of the true lumen of the aortic branch vessels, has the potential risk of acute paraplegia due to false lumen thrombosis of the descending thoracic aorta and its intercostal arteries when the entire descending thoracic aorta is covered, and carries approximately a 2% to 5%
Since 1996, we have adopted the approach of endovascular reperfusion via fenestration/stenting by interventional radiology (IR) of the critically malperfused organ systems for patients with ATBAD and malperfusion, which can resolve both dynamic and static malperfusion of aortic branch vessels. Previously we reported our 10-year experience of treating malperfusion in ATBAD.
In this study, we report the short- and long-term outcomes of endovascular fenestration/stenting to treat malperfusion in ATBAD over the past 20 years.
Materials and Methods
This study was approved by the Institutional Review Board at the University of Michigan, Michigan Medicine (Ann Arbor, Mich).
Study Population
From February 1996 to May 2018, 182 patients presented with spontaneous ATBAD and suspected malperfusion and proceeded to angiography for diagnosis and fenestration/stenting. Patients with ATBAD and malperfusion due to trauma (n = 4) or malperfusion treated with isolated TEVAR (n = 8) or open surgical repair were excluded. ATBAD was defined as onset within 14 days of admission with dissection confined to the aorta distal to the left subclavian artery. Investigators used medical record review to obtain preprocedural, intraprocedural, and postprocedural characteristics. Reinterventions included open or endovascular aortic repair of the aorta distal to the left subclavian artery and were collected from a thorough medical record review. The National Death Index database through December 31, 2015,
and medical record review were used to determine survival. Loss of follow-up was treated as a censor during the time-to-event analysis.
Diagnosis of Dynamic and Static Malperfusion
Malperfusion, inadequate blood flow to the end organs, could be diagnosed with radiographic findings consistent with reduced or absent flow to an end-organ or complete true lumen collapse on computed tomography, including disappearance of the aortic double lumen indicating elimination of the true lumen with the dissection flap being pushed against the aortic wall causing obstruction of flow to branch vessels, continuation of dual lumen patency with absent flow in a branch vessel (dynamic malperfusion), and dissection into a branch vessel or a thrombosed false lumen in a branch vessel (static malperfusion), with or without clinical evidence of end-organ dysfunction (Figure 1). MPS involves tissue/organ necrosis and end-organ dysfunction as a result of inadequate blood flow (malperfusion) and requires clinical features (abdominal pain, bloody diarrhea, tenderness to palpation, decreased urine output, absence of peripheral pulses, motor or sensory deficit of the lower extremity) and laboratory findings (elevated lactate, liver enzymes, metabolic acidosis, elevated creatinine) in addition to radiographic findings. The etiology of the malperfusion can be static, dynamic, or both static and dynamic obstruction of a branch vessel.
Dynamic malperfusion results from the dissection flap of a collapsed true lumen prolapsing across the origin of the branch vessel and obstructing flow and can vary in severity depending on variations of pressure in the false lumen. Dynamic obstruction can usually be resolved with restoration of the true lumen with a TEVAR endograft covering the intimal tear or aortic fenestration/stenting. Static obstruction results from extension of the dissection flap into a branch vessel, frequently accompanied by false lumen thrombosis due to no or very small reentry tear and occlusion of the true lumen, and is present throughout the cardiac cycle. Total occlusion of a vessel, like the superior mesenteric artery (SMA), by a thrombosed false lumen can lead to thrombosis of the true lumen distal to the dissection. Furthermore, collateral flow to an obstructed SMA is often compromised by dissection-related compromise of the celiac trunk or inferior mesenteric artery. Static obstruction usually requires stenting or other intervention (fenestration/thromboembolectomy/thrombolysis) of the affected branch vessel to restore flow. Both static and dynamic obstruction can be resolved with endovascular reperfusion via fenestration/stenting. In contrast to our management in type A aortic dissection, where MPS is the indication for IR procedures, in type B aortic dissection suspected malperfusion (not MPS) is an indication for IR.
Figure 1Computed tomography (CT) angiogram of a 55-year-old man with an ATBAD and static malperfusion of the celiac artery and SMA before and 8 years after endovascular fenestration/stenting. Axial CT at the level of the SMA (A) shows thrombosed false lumen (arrowhead) just beyond the SMA origin. Sagittal CT of the upper abdomen through the false lumen of the aorta (B) again shows thrombus in the terminal portion of the false lumen of the celiac trunk and the SMA, resulting in arterial occlusion. Three-dimensional reconstruction of the abdominal aorta (C) shows the dissection flap cleaving the celiac and SMA origins with thrombosed false lumen (red), with the inferior mesenteric artery (IMA) supplied by the false lumen. Superior mesenteric arteriogram (D), approximately 28 hours after symptom onset, shows proximal occlusion of the SMA trunk (arrowheads), absent filling of jejunal and ileal branches, and retrograde filling of the celiac distribution through pancreatic collaterals. After stenting of the SMA and celiac trunk, distal SMA pressure was 57/43 mm Hg, 32 mm Hg lower than aortic true lumen pressure. Sheath injection at the celiac origin after stenting (E) fills hepatic and splenic arteries and refluxes into the abdominal aorta, filling stented SMA (arrowhead) with jejunal and ileal branches (asterisks). CT with 3-dimensional reconstruction 8 years later (F) shows the celiac artery (C) stent extending into the common hepatic artery (H) with jailed but patent splenic artery (S). The SMA stent (arrowhead) is patent down to the ileocolic artery. Several small jejunal and ileal branches continue to fill through stent interstices. A prominent IMA (not shown) supports flow through the middle colic (MC) artery. F, False lumen; T, true lumen; SMA, superior mesenteric artery; IMA, inferior mesenteric artery; H, hepatic artery; S, splenic artery.
In angiography, treatable malperfusion was indicated by ongoing arterial obstruction and was confirmed by a systolic blood pressure gradient greater than 15 mm Hg between the ascending aorta and the branch vessel. If a branch artery is dissected, branch artery manometry is performed distal to the dissection (confirmed by intravascular ultrasound [IVUS]). The gradient of 15 mm Hg was chosen as the criterion for malperfusion based on the customary acceptance of a blood pressure differential of greater than 20 mm Hg as indicating hemodynamic significance in patients with aortic coarctation.
Fenestration and stenting were performed by creating a tear in the dissection flap 2 to 4 cm above the celiac artery using a 16-mm diameter balloon, thereby permitting flow between the true and false lumens, followed by deployment of a 16- to 18-mm diameter closed-cell self-expanding stent (Wallstent, Boston Scientific Corporation, Marlborough, Mass, off-label application) exclusively in the aortic true lumen (Figure 2), as previously described,
if the true lumen remains collapsed or a gradient between the aortic root and the abdominal aorta persists after aortic fenestration. Blood pressure gradients between the aorta and the branch vessels were determined both before and after fenestration/stenting (Wallstents). If after aortic fenestration/stenting a significant gradient persisted between the aorta and a branch vessel, then branch vessel fenestration/stenting, thrombolysis, or thromboembolectomy was performed as appropriate, based on angiographic and IVUS findings (Figure 2). Complete resolution of malperfusion was defined as blood pressure gradient decreased to less than 15 mm Hg. In dissected vessels with thrombosed false lumens, gradients after stenting might exceed 15 mm Hg, defined as partial resolution of malperfusion, but as long as absolute perfusion pressure was viable (ie, >60 mm Hg), postdilation of stents was not performed.
Concomitant (n = 5) or post-IR TEVAR (n = 4) or open aortic repair (n = 5) was indicated for pending rupture, refractory back pain, uncontrollable severe hypertension, and aortic aneurysm. Concomitant TEVAR includes patients who initially had TEVAR for back pain/impending rupture with persistent postoperative static malperfusion subsequently treated by fenestration/stenting. All open procedures were performed before 2004. Postprocedure management consisted of aspirin therapy, blood pressure control, standard management of end-organ dysfunction, and adequate analgesia and sedation. When bowel ischemia or extremity ischemia was present, general or vascular surgery was consulted, respectively, to determine if exploratory laparotomy or fasciotomies were indicated.
Statistical Analysis
Continuous variables were summarized by median (25%, 75%), and categoric variables were reported as n (%) in frequency tables. Crude survival curves since admission were estimated using the nonparametric Kaplan–Meier method. Multivariable logistic regression was performed to calculate the odds ratio (OR) of risk factors for in-hospital mortality adjusting for age, gender, coronary artery disease, acute myocardial infarction, acute renal failure on dialysis, acute paralysis, celiac malperfusion, mesenteric malperfusion, renal malperfusion, and extremity malperfusion. Cox proportional hazard regression was performed to calculate the hazard ratio (HR) for late mortality by stepwise selection of variables including age, gender, coronary artery disease, chronic renal failure on dialysis, acute myocardial infarction, acute paralysis, acute renal failure requiring dialysis, MPS found, bowel resection, amputation, and fasciotomy. Because patients may experience death before reintervention was indicated, cumulative incidence curves adjusting for death as the competing risk were generated to assess cumulative incidence of reintervention over time. Cox regression was used to calculate the risk factors of reintervention adjusting for age, gender, connective tissue disease, aortic flap fenestration without TEVAR or open aortic repair, and hypertension. All statistical calculations used SAS 9.4 (SAS Institute, Inc, Cary, NC).
Results
Demographics and Preprocedural Data
The median age was 55 years, and most patients (88%) had hypertension. The majority (93%) of patients were transferred from another hospital. Patients frequently presented with abdominal pain (61%), lower-extremity weakness (27%) with nonpalpable pulses (24%), and elevated creatinine and liver enzymes. Some already had certain vascular procedures for limb ischemia before presentation (Table 1).
Table 1Demographics and characteristics before procedures by interventional radiology
ATBAD noted intraoperatively during aortic root repair.
1 (0.5)
Open TAA
1 (0.5)
Data presented as median (25%, 75%) for continuous data and n (%) for categoric data. BMI, Body mass index; COPD, chronic obstructive pulmonary disease; CAD, coronary artery disease; PVD, peripheral vascular disease; MI, myocardial infarction; IR, interventional radiology; AST, aspartate aminotransferase; ALT, alanine aminotransferase; CK, creatine kinase; SMA, superior mesenteric artery; TAVR, transcatheter aortic valve replacement; TAA, thoracic aortic aneurysm.
∗ ATBAD noted intraoperatively during aortic root repair.
Renal, extremity, and mesenteric malperfusions were the most frequently suspected and confirmed malperfusions in ATBAD. Multiple vascular beds were frequently affected. Overall, 74% of patients had interventions in the angiography suite, including aortic fenestration/stenting (54%), branch vessel fenestration/stenting (59%), and thrombolysis/thrombectomy/embolectomy (9.3%) for static malperfusion (Table 2, Figure 1). Branch artery stenting was performed in the iliac, renal, superior mesenteric, and celiac arteries 56, 47, 37, and 8 times, respectively. Overall, 93% of malperfusion was completely resolved and 5% was partially resolved (Table 3). After endovascular fenestration/stenting, 13 patients required laparotomy for suspected bowel ischemia, including 9 patients who had bowel resection, and 12 patients required additional vascular procedures during the current hospital stay. Nine patients required additional TEVAR (n = 4) or open aortic repair (n = 5) because of aortic aneurysm, pending rupture, and persistent symptoms (Table 2).
Table 2Interventional radiology indications and procedures and subsequent procedures during hospital stay
Thrombolysis, thrombectomy, embolectomy of aortic branch vessels.
17 (9.3)
Subsequent procedures during hospital stay
Exploratory laparotomy for suspected ischemia
13 (7.1)
Bowel resection
9 (4.9)
Vascular surgery for suspected ischemia
12 (6.6)
Thrombectomy/embolectomy
6 (3.3)
Fem-fem bypass
2 (1.1)
Fasciotomy
5 (2.7)
Amputation
3 (1.6)
Aortic surgery
9 (4.9)
TEVAR
4 (2.2)
Open TAA/A
4 (2.2)
Open AAA
1 (0.5)
Time from IR (d)
12 (9, 14)
Data presented as n (%). TEVAR was performed by both IR faculty and cardiac surgeons together. IR, Interventional radiology; TEVAR, thoracic endovascular aortic replacement; TAA/A, thoracic aortic aneurysm or thoracoabdominal aortic aneurysm; AAA, abdominal aortic aneurysm.
∗ Thrombolysis, thrombectomy, embolectomy of aortic branch vessels.
Complete resolution of malperfusion was defined as the systolic blood pressure gradient between the branch vessel and the ascending aorta less than 15 mm Hg after endovascular fenestration/stenting. Partial resolution of malperfusion was defined as systolic blood pressure gradient greater than 15 mm Hg.
Complete resolution of malperfusion was defined as the systolic blood pressure gradient between the branch vessel and the ascending aorta less than 15 mm Hg after endovascular fenestration/stenting. Partial resolution of malperfusion was defined as systolic blood pressure gradient greater than 15 mm Hg.
Among 47 patients receiving renal artery stenting, 38 (81%) received unilateral stenting (right renal, 38%; left renal, 43%) and 9 (19%) had bilateral renal artery stenting. Among 56 patients receiving iliac stenting, 42 (75%) had aorto-iliac stenting, 32 (57%) had common iliac artery stenting, 33 (59%) had external iliac artery stenting, and 7 (12.5%) had common femoral artery stenting. Four renal artery malperfusions and 1 iliac malperfusion were unable to be treated because of an inability to cannulate the specific branch vessel (n = 1, renal artery), anatomy unsuitable for stenting (n = 3, renal artery), or a significant gradient without symptoms of malperfusion (n = 1, iliac).
Among 47 patients receiving renal artery stenting, 38 (81%) received unilateral stenting (right renal, 38%; left renal, 43%) and 9 (19%) had bilateral renal artery stenting. Among 56 patients receiving iliac stenting, 42 (75%) had aorto-iliac stenting, 32 (57%) had common iliac artery stenting, 33 (59%) had external iliac artery stenting, and 7 (12.5%) had common femoral artery stenting. Four renal artery malperfusions and 1 iliac malperfusion were unable to be treated because of an inability to cannulate the specific branch vessel (n = 1, renal artery), anatomy unsuitable for stenting (n = 3, renal artery), or a significant gradient without symptoms of malperfusion (n = 1, iliac).
-
-
2
56
79
74
4
In columns 2 to 5, n is the number of patients. In column 6, n is the total malperfusion found in different vascular beds, including celiac artery, SMA, renal arteries, and common iliac arteries and their branches (external iliac arteries, femoral arteries). Aortic stenting: If there was a compression of the true lumen by the thrombosed false lumen in the descending aorta, we placed a 16-mm self-expanding bare stent in the descending aorta (descending thoracic aortic stenting). After fenestration of the aortic dissection flap, we place the same self-expanding bare stent in the distal descending thoracic aorta if needed to keep the true lumen open. The distal descending thoracic aortic stent was placed frequently above the SMA (supramesenteric aortic stenting). If we had to place stents in the SMA, then we placed the aortic stent above the celiac artery (supraceliac aortic stenting). The goal of aortic stenting is to achieve adequate expansion of the true lumen of dissected aorta and eliminate blood pressure gradient between the distal aorta and the ascending aorta.
∗ Complete resolution of malperfusion was defined as the systolic blood pressure gradient between the branch vessel and the ascending aorta less than 15 mm Hg after endovascular fenestration/stenting. Partial resolution of malperfusion was defined as systolic blood pressure gradient greater than 15 mm Hg.
† Among 47 patients receiving renal artery stenting, 38 (81%) received unilateral stenting (right renal, 38%; left renal, 43%) and 9 (19%) had bilateral renal artery stenting. Among 56 patients receiving iliac stenting, 42 (75%) had aorto-iliac stenting, 32 (57%) had common iliac artery stenting, 33 (59%) had external iliac artery stenting, and 7 (12.5%) had common femoral artery stenting. Four renal artery malperfusions and 1 iliac malperfusion were unable to be treated because of an inability to cannulate the specific branch vessel (n = 1, renal artery), anatomy unsuitable for stenting (n = 3, renal artery), or a significant gradient without symptoms of malperfusion (n = 1, iliac).
Postprocedural new-onset paraplegia (0%), retrograde type A dissection (0%), and new-onset acute renal failure requiring dialysis (1.6%) were low. In 5 patients with spinal cord malperfusion, 2 had complete resolution and 1 had partial resolution of paraplegia. Overall in-hospital mortality was 7.7% over 20 years, 11.3% in the first decade (1996-2007), 3.5% in the second decade (2008-2018), and 0% in the last 8 years (Table 4). The significant risk factors for in-hospital mortality were age (OR, 1.15), acute myocardial infarction (OR, 8.6), acute paralysis (OR, 11.5), and extremity malperfusion (OR, 8.8) (Table 5). Detailed causes of death included aortic rupture, extensive necrotic intestine, arrhythmia, and stroke (Table E1).
Table 4Postprocedural outcomes
Variables
Total (n = 182)
Stroke
10 (5.5)
Continued acute renal failure requiring new dialysis
Five patients had preoperative paraplegia due to spinal cord malperfusion, 2 patients' paraplegia resolved completely, and 1 patient's paraplegia resolved partially.
3 (60)
GI bleed
1 (0.5)
Groin hematoma
8 (4.4)
Length of stay (d)
11 (8, 18)
In-hospital mortality
14 (7.7)
Data presented as median (25%, 75%) for continuous data and n (%) for categoric data. GI, Gastrointestinal.
∗ Five patients had preoperative paraplegia due to spinal cord malperfusion, 2 patients' paraplegia resolved completely, and 1 patient's paraplegia resolved partially.
The 5- and 10-year survivals were 72% (95% confidence interval [CI], 64-78) and 49% (95% CI, 39-58), respectively (Figure 3, A). The significant risk factors for late mortality were age (HR, 1.04; 95% CI, 1.02-1.06) and acute paralysis (HR, 3.5; 95% CI, 1.8-6.8; P < .001).
Figure 3A, Kaplan–Meier survival analysis of patients with an ATBAD and MPS undergoing endovascular fenestration/stenting. The 5- and 10-year survivals were 72% (95% CI, 64-78) and 49% (95% CI, 39-58), respectively. B, The cumulative incidence of reintervention of the whole cohort (n = 182) for pathology of the descending thoracic or thoracoabdominal aorta after hospital discharge, adjusting for death as the competing event. The 5- and 10-year cumulative incidence of reintervention was 21% (95% CI, 15-28) and 31% (95% CI, 23-40), respectively. C, The cumulative incidence of reintervention for patients who had only fenestration/stenting (n = 125) after hospital discharge, adjusting for death as the competing event. The 5- and 10-year cumulative incidence of reintervention was 20% (95% CI, 13-28) and 31% (95% CI, 21.5-41), respectively.
Of the 182 patients, 14 died in the hospital before discharge and 12 (6.6%) were lost to follow-up for reintervention. The mean follow-up time was 6.2 years. The 5- and 10-year cumulative incidence of reintervention for pathology of the descending or thoracoabdominal aorta adjusted for death as a competing factor was 21% (95% CI, 15-28) and 31% (95% CI, 23-40), respectively (Figure 3, B). The significant risk factors for reintervention were connective tissue disease (HR, 3.4; 95% CI, 1.4-8.3; P = .007) and male gender (HR, 3.2; 95% CI, 1.3-8.0; P = .014). Fenestration without TEVAR or open repair was not a significant risk factor (HR, 0.8; 95% CI, 0.4-1.5; P = .49). The 5- and 10-year cumulative incidence of reintervention for patients undergoing fenestration/stenting only, without TEVAR or open repair during the hospital stay, was 20% (95% CI, 13-28) and 31% (95% CI, 21.5-41), respectively (Figure 3, C). The primary indication for reintervention was aortic aneurysm (91%) and primarily done through open repair (76%). The median interval time to reintervention was 2 years.
Discussion
In this study, we managed ATBAD complicated by malperfusion with endovascular fenestration/stenting, which enables resolution of both dynamic and static malperfusion. The in-hospital mortality was 7.7% over 20+ years and 0% in the last 8 years; the cumulative incidence of reintervention was 21% at 5 years and 31% at 10 years; and the 5- and 10-year survivals were 72% and 49%, respectively.
Aortic dissection, both type A and B, can be complicated by malperfusion due to dissection-related obstruction of aortic branch vessels. In acute type A aortic dissection (ATAAD), because of the risks of aortic rupture, acute heart failure and aortic insufficiency, acute myocardial infarction, pericardial effusion and tamponade, and neurologic complications,
we pursue upfront angiography with endovascular reperfusion only for patients with MPS (malperfusion with tissue/organ necrosis and end-organ dysfunction). In patients with ATBAD, we extend this strategy to patients with malperfusion unresponsive to blood pressure and heart rate control and patients with a documented history of poor compliance with antihypertensive medication in addition to patients with MPS. Patients with ATBAD have a lower risk of aortic rupture with adequate blood pressure management
Malperfusion due to dynamic obstruction, which is generally corrected by proximal aortic repair in patients with ATAAD, may persist in patients with ATBAD unless dealt with directly by fenestration/stenting or TEVAR. Malperfusion, in addition to MPS, is an indication for angiography for diagnosis and potential treatment in patients with ATBAD. Because clinical manifestation of mesenteric and renal malperfusion may lag their computed tomography demonstration and unsuspected vascular beds with malperfusion are frequently identified when we investigate malperfusion of suspected vascular beds, we consider the endovascular evaluation of patients with ATBAD an angiographic emergency and have a low threshold for performing it. The angiographic evaluation of these patients typically involves manometry, IVUS examination, and hand injection of 7 mL of contrast material diluted 1:1 with normal saline into the SMA, bilateral renal arteries, and external iliac arteries. In the case of SMA or renal artery dissection associated with a significant pressure deficit, IVUS examination is performed to determine radiographic landmarks of the dissection terminus to aid in stent placement.
Management options for ATBAD complicated with malperfusion have generally included open surgical repair and more recently TEVAR. Although TEVAR has reduced early mortality to approximately 10% in all ATBAD,
we still use fenestration/stenting as our mainstream treatment for malperfusion in ATBAD for the following reasons: (1) TEVAR has risks of retrograde type A dissection
); (2) TEVAR alone cannot reliably resolve static malperfusion; (3) TEVAR sometimes has to cover the left subclavian artery to cover the primary intimal tear, which requires additional procedures (eg, left carotid artery-subclavian artery bypass) to preserve blood flow to the left subclavian artery; and (4) when patients have necrotic bowel or limb and sepsis, TEVAR has a higher risk of graft infection. With endovascular fenestration/stenting, we can avoid all the risks from TEVAR and adequately treat static malperfusion with branch vessel stenting, fenestration, thromboembolectomy, or thrombolysis. In this study, 59% of patients had fenestration/stenting of an aortic branch vessel and 9.3% had thrombolysis or thromboembolectomy for static malperfusion that could not be resolved by TEVAR alone. Because we did not cover any intercostal arteries, protecting the spinal cord from ischemic injury, postprocedural new-onset paraplegia due to ischemic spinal cord injury was 0%, which is lower than in those treated with TEVAR alone (2%-10%
). The in-hospital mortality rate was 7.7% with this approach in this subpopulation with ATBAD and malperfusion with 0% mortality in the last 8 years as we became more experienced with fenestration/stenting and with use of TEVAR for aortic pending rupture or rupture, which is lower than that seen with open repair and with TEVAR alone,
possibly because TEVAR alone does not reliably correct static obstruction. Other reasons for improved mortality include better imaging, prompt diagnosis, treating suspected malperfusion in an acute dissection as an angiographic emergency, and better intensive care unit management (blood pressure control). Endovascular fenestration/stenting of ATBAD with malperfusion combined with TEVAR and open repair achieved favorable survival (5- and 10-year survivals: 72% and 49%, respectively), which was better than or similar to those treated with TEVAR or open repair alone.
Thoracic endovascular aortic repair for acute complicated type B aortic dissection: superiority relative to conventional open surgical and medical therapy.
The significant risk factors for late mortality were age and acute paralysis (HR, 3.5). By decreasing the risk of new-onset paraplegia, endovascular fenestration/stenting could decrease the late mortality.
Our approach is based on risk stratification to determine the best management. In the setting of ATBAD with malperfusion without signs of rupture (persistent or increasing back pain), we think the malperfusion is the most immediate concern and elect to treat the malperfusion with percutaneous fenestration/stenting. This approach accomplishes the goal of resolving the malperfusion and essentially “converts” a complicated ATBAD to an uncomplicated ATBAD and allows patients to recover with medical management afterward. If patients had rupture/pending rupture, refractory back pain, uncontrollable hypertension, or large aortic aneurysm, a concomitant or delayed TEVAR or open aortic repair was performed as is seen in a small portion of this cohort (n = 14, 7.7%) (Table 2). Managing ATBAD with malperfusion via fenestration/stenting does not prevent aortic rupture, as would open surgery or TEVAR. In this study, 7 patients (3.8%) possibly died of aortic rupture a median of 3 days (interquartile range, 2-4.5 days) after angiography with fenestration/stenting, which all happened before 2011 when TEVAR was not commonly used at our institution. Aortic rupture may have been prevented in these patients if they had undergone TEVAR, although cases of rupture have been reported during
After 2010, we applied endovascular fenestration/stenting, combined with TEVAR and open repair as appropriate, to all patients with ATBAD and malperfusion, with a 0% aortic rupture rate and 0% in-hospital mortality. Endovascular fenestration/stenting is an effective tool to treat malperfusion (dynamic and static) in ATBAD and is a valuable adjunct to both medical and surgical therapy (TEVAR and open repair). The fenestration/stenting approach does not exclude TEVAR or open repair of the aorta. If needed, all 3 approaches can be used to treat ATBAD with different complications.
For patients with MPS (late-stage malperfusion with tissue/organ necrosis and dysfunction), endovascular fenestration/stenting resolves the malperfusion with minimal operative trauma and provides the opportunity for patients to recover from MPS. Additional intervention may be needed to recover from severe MPS. In our study, 24 patients (13%) required general or vascular surgery intervention for bowel and extremity necrosis after angiography, including bowel resection, fasciotomy, and amputations (Table 2). Of these 24 patients, 20 had branch vessel stenting during angiography for static malperfusion, which could not have been treated with open repair or TEVAR alone. This highlights the gravity of the MPS; despite initial reperfusion of affected vascular territories, patients may still have complications of the preexisting malperfusion and subsequent reperfusion. With prolonged static malperfusion, as would be with initial open repair or TEVAR, it is suspected that more patients would experience irreversible, unsalvageable end-organ death.
One concern about leaving a patent false lumen after fenestration/stenting is that it could increase the risk of reintervention. TEVAR could facilitate aortic remodeling by thrombosing and stabilizing the size of the thoracic false lumen due to closure of the primary intimal tear with a covered stent graft.
We think the aortic flap fenestration/stenting approach achieves the same goal by creating a distal fenestration as outflow for the false lumen to decompress and prevent the dilation of the false lumen even though the proximal primary intimal tear is open. Burris and colleagues
found that in chronic type B aortic dissection the false lumen dilates quickly with a large proximal primary intimal tear and a small distal reentry tear due to high pressure in the false lumen during diastole evident as regurgitant blood flow from the false into the true lumen through both proximal and distal intimal tears. If the distal reentry tear is large, there is no regurgitant flow from false to true lumen and there is minimal growth of the false lumen and the dissected aorta.
Our approach, endovascular fenestration of the distal aortic flap, serves exactly the same purpose. As a result, the 5- and 10-year cumulative rate of reintervention with fenestration/stenting alone was 20% and 31%, respectively, adjusting for death as a competing factor (Figure 3, B and C), which is similar if not better than that reported with TEVAR
alone, and we had longer follow-up than most studies because TEVAR is a more recent technology. Most of the studies using TEVAR in the literature use freedom from reintervention and do not adjust for late death as a competing factor, which could underestimate the rate of reintervention. We do not think endovascular fenestration/stenting alone treating malperfusion in ATBAD increases the risk of reintervention compared with TEVAR.
Study Limitations
Our study is limited by a single-center and retrospective experience. The management strategy of angiography evaluation and endovascular fenestration/stenting has a learning curve. Because the follow-up of reintervention was 93.4% complete, we could underestimate the rate of reintervention. This study is also limited by lack of a direct comparison group, such as TEVAR alone.
Conclusions
In patients with ATBAD complicated by malperfusion, endovascular fenestration/stenting can effectively resolve the malperfusion and achieve favorable short- and long-term results with additional indicated TEVAR or open aortic repair. We recommend endovascular fenestration/stenting when treating ATBAD with malperfusion, especially in patients with static malperfusion.
Dr Williams is on the Medical Advisory Board of Boston Scientific. Drs Williams and Patel are consultants with WL Gore and Associates on an unrelated device. All other authors have nothing to disclose with regard to commercial support.
The authors thank Eric Wizauer, Vanessa Allen, and Sarah Abate for their help in composing and annotating the figures.
Appendix
Table E1Detailed cause of death in patients with in-hospital mortality
Case
Age, y
Year of treatment
Locations of malperfusion
Cause of death
1
56
1999
Celiac, mesenteric, renal
Unstable after IR, returned to cardiac care unit for continued dialysis, correction of coagulopathy, and correction of acidosis. Given grave prognosis, family decided to designate patient DNR with comfort measures.
2
53
2000
Celiac, mesenteric, renal, extremity
After bowel resection, family decided to place on comfort care.
3
51
2001
Celiac, mesenteric, renal, extremity
Second exploratory laparotomy revealed extensive necrosis - nothing to be done; no resection. Family withdrew treatment.
4
61
2002
Celiac, mesenteric, renal
Mechanical ventilation, continued dialysis for ARF, many old and new lacunar strokes, withdrawal of life support.
Evaluated for hypotension and left neck discomfort, stabilized with limited volume infusion, ∼ 1 h later she suffered bradycardia and unresponsiveness and a code was carried out. Patient died.
8
67
2006
Spinal cord, celiac, mesenteric, renal, extremity
After finishing amputation but before leaving the OR, patient went into ventricular tachycardia then ventricular fibrillation with return of pulsatile rhythm and taken to surgical intensive care unit where the patient again entered a pulseless rhythm. Regained a BP for a brief period of time, and emergency hemodialysis was attempted. He again entered into a pulseless rhythm, and angiocaths were inserted. Pulse could not be regained. Patient died.
Loss of consciousness with acute decrease in BP, with subsequent PEA arrest. CPR and ACLS protocols were initiated, a stat TEE showed a large, false aortic lumen and a large, echo-free region posterior to the descending aorta, which was likely free blood due to aortic rupture. CPR was stopped; patient died.
11
71
2007
Mesenteric, renal, extremity
Ischemia of bilateral lower extremities, became hypotensive, made DNR
Clinical profiles and outcomes of acute type B aortic dissection in the current era: lessons from the international registry of aortic dissection (IRAD).
Thoracic endovascular aortic repair for acute complicated type B aortic dissection: superiority relative to conventional open surgical and medical therapy.
Sources of Funding: Dr Yang is supported by the National Heart, Lung, and Blood Institute of the National Institutes of Health K08HL130614 and R01HL141891, Phil Jenkins, and Darlene & Stephen J. Szatmari Funds. Dr Patel is supported by the Joe D. Morris Collegiate Professorship, the David Hamilton Fund, and the Phil Jenkins Breakthrough Fund in Cardiac Surgery. Dr Deeb is supported by the Herbert Sloan Collegiate Professorship, Jamie Buhr Fund, and Richard Nerod Fund.
Date and Number of Institutional Review Board Approval: October 29, 2018, and HUM00152881.
I read with great interest the article by Norton and colleagues,1 which reported the efficacy of endovascular fenestration/stenting in patients with acute type B aortic dissection (ATBAD) with malperfusion. In their study cohort, only a limited number (4.9%) of ATBAD patients with malperfusion underwent thoracic endovascular aortic repair (TEVAR).1 However, I believe that primary entry closure with TEVAR is the first-line treatment for malperfusion, especially due to dynamic occlusion, for the following 2 reasons.
The interest for the endovascular treatment of acute type B aortic dissection (ATBAD) has been increasing during the last decades as a result of improved imaging technologies, prompt diagnosis, improved preoperative and postoperative management, and improved experience of the aortic team. This evolution has determined reduced early morbidity and mortality and improved long-term survival.1 Advanced endovascular techniques, such as thoracic endovascular aortic repair (TEVAR) with stent-graft deployment and dissection flap fenestration with branch stenting repair, are considered the treatment of choice in ATBAD with malperfusion syndrome.
Early mortality for acute type B aortic dissection remains significant,1 and malperfusion is often implicated in poor early outcomes. Malperfusion after the initial acute postdissection period is uncommon and can indicate a new dissection process, especially when accompanied by severe back pain.
32070-7&id=fx1.jpg){kind=link}
32070-7&id=gr1.jpg){kind=link}
32070-7&id=gr2.jpg){kind=link}
32070-7&id=gr3.jpg){kind=link}
32070-7&id=fx2.jpg){kind=link}