The Safety and Efficacy of Extended TEVAR in Acute Type B Aortic Dissection

Abstract

Background.

Thoracic endovascular aortic repair (TEVAR) with endograft coverage from the left subclavian artery to the celiac artery has been hypothesized to increase spinal cord ischemia. This study analyzes the impact of extended coverage on adverse outcomes and aortic remodeling in patients with complicated acute type B aortic dissection (aTBAD).

Methods.

From January 2012 to October 2018, 91 patients underwent TEVAR for aTBAD. Median follow-up was 3.1 (interquartile range, 1.2-4.9) years and was complete in 94% of patients. The extent of aortic endograft coverage was categorized as standard (n = 39) or extended (n = 52). Contrast-enhanced imaging scans were analyzed to determine length of coverage, maximum aortic diameters, and false lumen (FL) status.

Results.

The mean age was 52.6 ± 13.9 years, and 66% were men. The most common indications for intervention were malperfusion (42%) and refractory pain (34%). Thirteen (14%) patients required a lumbar drain (preoperative: n = 3; postoperative: n = 10). Mean duration between scans was 2.0 ± 1.9 years. Length of aortic coverage was significantly longer in the extended group (241.7 ± 29.2 mm vs 180.8 ± 22.3 mm in the standard group; P < .001). In-hospital and overall mortality were 6% and 11%, respectively. There were no cases of paraplegia, and the incidence of spinal cord ischemia was 3%. After TEVAR, there was a higher incidence of FL obliteration or thrombosis at the distal descending thoracic aorta in the extended group (53% vs 16% in the standard group; P = .004).

Conclusions.

Extended TEVAR carries a low risk of spinal cord ischemia and improves FL remodeling of the descending thoracic aorta in patients with aTBAD. This strategy may decrease the need for reinterventions on the thoracic aorta in the chronic phase of TBAD.

Thoracic endovascular aortic repair (TEVAR) is well recognized as the optimal therapy for patients presenting with complicated acute type B aortic dissection (aTBAD). TEVAR has been shown to decrease mortality, compared with open surgery or medical management, and effectively remodel the aorta by eliminating antegrade false lumen (FL) flow, promoting FL thrombosis, and reexpanding the true lumen.^1–3 In experienced, high-volume centers, the in-hospital mortality rate of patients presenting with complicated aTBAD ranges from 0% to 8%.^1,4,5

Although elimination of antegrade FL flow can be achieved with limited endograft coverage of the thoracic aorta, this strategy often leaves large secondary thoracic aortic tears untreated. As a result, FL flow within the thoracic aorta can persist, which may cause FL aneurysmal degeneration requiring open or endovascular reintervention. Ideally, all fenestrations between the true lumen and FL in the thoracic aorta would be covered at the time of the initial endovascular procedure by extending the stent graft coverage to the level of the celiac artery. While this strategy does not address infradiaphragmatic fenestrations and the subsequent risk of aneurysmal degeneration of the abdominal aorta, it does offer the potential of eliminating further interventions in the thoracic aorta. Previous reports in the literature have identified extended aortic coverage as a risk factor for postoperative spinal cord ischemia (SCI), a highly morbid complication that can result in paralysis and increased mortality.^6–9 Although drainage of cerebrospinal fluid (CSF) can mitigate the risk of SCI, lumbar drains carry their own set of complications including epidural hematomas, cerebrospinal fluid leaks, meningitis, nerve injuries, and intracranial hemorrhage.

After establishing the safety and efficacy of TEVAR for the treatment of complicated aTBAD in our initial 30 patients with endograft coverage extending to the mid-descending thoracic aorta (DTA), we observed suboptimal aortic remodeling in our postoperative surveillance imaging. Therefore, we began a more aggressive strategy consisting of extended endograft coverage to the celiac artery with selective CSF drainage. The purpose of this study was to evaluate the safety and efficacy of this novel aggressive strategy by comparing adverse outcomes and aortic remodeling in patients presenting with complicated aTBAD who received either limited or extended aortic coverage.

Patients and Methods

This study was conducted with the approval of the Institutional Review Board at Emory University in compliance with Health Insurance Portability and Accountability Act regulations and the Declaration of Helsinki. The institutional review board waived the need for individual patient consent. After a detailed retrospective review of the Emory Aortic Database, 91 consecutive patients from 2012 to 2018 with complicated aTBAD who underwent TEVAR at index hospitalization were identified. The indications for TEVAR were rupture, malperfusion, intractable back pain despite adequate blood pressure control, or the rapid progression of disease within 7 days of the diagnosis of aTBAD. Patients with uncomplicated TBAD, chronic dissection, residual distal dissections after proximal aortic repair for type A aortic dissection, and all other acute aortic syndromes involving the DTA were excluded from analysis. Adverse clinical outcomes analyzed included the incidence of spinal cord ischemia (paraparesis and paraplegia), stroke, renal failure requiring dialysis, retrograde type A dissection, reinterventions, and mortality. Overall mortality was defined as all-cause mortality over the follow-up period. Aortic remodeling, defined as a thrombosed or obliterated FL, and a reduction in aortic size on follow-up imaging analysis were also analyzed.

Endografts used in this series were oversized by 5% to 10% and consisted of the Medtronic Valiant with Captivia (n = 74; Medtronic Endovascular, Santa Rosa, CA), Zenith TX 2 (n = 11; Cook Medical, Bloomington, IN), and Gore C-TAG (n = 6; W.L. Gore and Associates, Flagstaff, AZ). The extent of aortic endograft coverage was categorized as standard (endograft coverage from the left subclavian artery [LSA] to the mid-DTA; n = 39) or extended (endograft coverage of the entire DTA extending below the diaphragm to the level of the celiac artery; n = 52). In cases requiring LSA coverage in order to exclude the primary intimal tear, revascularization of the LSA before TEVAR was performed based on the following anatomic criteria: (1) patent left internal mammary artery to left anterior descending coronary artery bypass graft, (2) dominant left vertebral artery with incomplete circle of Willis or atretic right vertebral artery, or (3) patent left axillary-femoral artery bypass graft. In the current series, 5 patients underwent left carotid–LSA bypass and 2 underwent left vertebral artery transposition for aberrant arch anatomy. No patient required postoperative LSA revascularization.

In DeBakey type IIIa patients (n = 10), the operative strategy was to cover all proximal and distal intimal tears with TEVAR. In DeBakey type IIIb patients (n = 81), endograft coverage was initially performed with a single 150-mm-long or 200-mm-long endograft. After our initial 30 patients, we began extending endograft coverage to the origin of the celiac artery in the majority of patients.

Per institutional protocol, preoperative lumbar drains were inserted in patients with a history of open or endovascular abdominal aortic surgery or in patients presenting with signs of SCI (eg, urinary retention, lower extremity weakness). Patients presenting with paraplegia (n = 2) underwent lumbar drain insertion after TEVAR, so that spinal cord reperfusion would not be delayed. Our detailed technique of TEVAR for complicated aTBAD has been previously described.¹⁰ After deployment of the proximal endograft, systolic blood pressure was augmented to greater than 160 mm Hg. Patients were extubated in the operating room and a neurologic exam was performed. Patients demonstrating evidence of lower extremity weakness underwent lumbar drain placement before leaving the operating room. Postoperative intensive care unit management consisted of maintenance of systolic blood pressure in the range of 160 mm Hg to 200 mm Hg, including use of vasopressors as necessary for 24 hours. Patients were discharged from the hospital with permissive hypertension, and blood pressure control was gradually optimized over a 3-month period.

Imaging Analysis

Patients underwent contrast-enhanced computed tomographic or magnetic resonance angiograms at the time of clinical intervention and follow-up. The preoperative imaging scan and the most recent postoperative scan were utilized to analyze aortic remodeling. Mean duration between scans was 2.0 ± 1.9 years.

Centerline analysis was performed using the TeraRecon Aquarius iNutrition 3D-workstation (TeraRecon, San Mateo, CA) to determine the length of aortic coverage for all postoperative scans. The maximum diameter of the aorta orthogonal to the centerline was measured at the following locations: (1) the DTA, (2) the level of the celiac artery, (3) the stent end 1 cm below the distal edge of the stent graft, and (4) the infrarenal aorta. The FL status was assessed at the following locations: (1) proximal DTA, (2) mid-DTA, (3) distal DTA, (4) stent end, (5) celiac artery, and (6) infrarenal aorta (Figure 1). Arterial and delayed venous phase images were qualitatively assessed to determine whether the residual FL within the stented segment of the aorta was (1) patent (evidence of contrast without evidence of thrombus), (2) partially thrombosed (evidence of both contrast and thrombus), (3) completely thrombosed (evidence of thrombus without evidence of contrast), or (4) obliterated (no evidence of residual FL). All imaging analyses were conducted by the primary and senior authors.

Figure 1.

Centerline analysis of maximum aortic diameters and false lumen status. (CELIAC, celiac artery; DTA, descending thoracic aorta; INFRA-RENAL, infrarenal aorta; PROX, proximal.)

Follow-up

Follow-up data were obtained via office visits, telephone calls, queries of the Social Security Death Index, or Internet obituary searches. Median follow-up was 3.1 (interquartile range, 1.2-4.9) years and was complete in 94% of patients within the last 12 months.

Statistical Analysis

Categorical variables were summarized using frequencies and percentages. Continuous variables were analyzed with mean ± SD if normally distributed; otherwise, median with interquartile range was used. Comparisons between groups were performed using chi-square analysis for qualitative variables and Student’s t test for continuous variables. When the frequency of any nominal variable was less than or equal to 5, a Fisher exact test was used. Reintervention rates were estimated using Kaplan-Meier methodology and compared between groups using the log-rank test.

Results

Table 1 lists baseline characteristics of patients with complicated aTBAD based on extent of graft coverage at time of initial presentation. There were no differences in major comorbidities between the groups. Radiographic characteristics at the time of presentation are reported in Table 2. In the initial scans before TEVAR, the maximum aortic diameter was 4.1 ± 0.7 cm at the DTA. There was no difference in the maximum aortic diameters at all analyzed levels between groups. The thoracic FL was partially thrombosed in 36% and patent in 64% of patients, and the abdominal FL was partially thrombosed in 24% and patent in 76% of patients. There was no difference in the FL status in the thoracic or abdominal aorta between the standard and extended groups. No patient presented with a completely thrombosed FL.

Table 1.

Baseline Characteristics

Variable	All (N = 91)	Standard Group (n = 39)	Extended Group (n = 52)	P Value
Age, y	52.6 ± 13.9	52.1 ± 15.1	52.9 ± 13.0	.80
Male	66 (60)	62 (24)	69 (36)	.40
Hypertension	89 (81)	87 (34)	90 (47)	.70
COPD	6 (5)	5 (2)	6 (3)	.90
Diabetes	4 (4)	8 (3)	2 (1)	.30
Dyslipidemia	26 (24)	26 (10)	27 (14)	.90
Current smoker	34 (31)	39 (15)	31 (16)	.40
History of stroke	8 (7)	5 (2)	10 (5)	.70
Prior abdominal aortic surgery	2 (2)	3 (1)	2 (1)	.90
Congestive heart failure	7 (6)	5 (2)	8 (4)	.70
Renal failure on dialysis	3 (3)	3 (1)	4 (2)	.90

Values are mean ± SD or % (n).

COPD, chronic obstructive pulmonary disease.

Table 2.

Radiographic Characteristics at Presentation

Characteristic	All (N = 91)	Standard Group (n = 39)	Extended Group (n = 52)	P Value
DeBakey type IIIb aTBAD	89 (81/91)	87 (34/39)	90 (47/52)	.70
Maximum aortic diameter, cm
DTA	4.1 ± 0.7	4.2 ± 0.8	4.0 ± 0.7	.20
Celiac artery	3.0 ± 0.4	3.0 ± 0.4	3.0 ± 0.4	1.0
Infrarenal aorta	2.5 ± 0.6	2.6 ± 0.8	2.4 ± 0.3	.40
Thoracic aorta FL status				.40
Partial	36 (21/59)	41 (11/27)	31 (10/32)
Patent	64 (38/59)	59 (16/27)	69 (22/32)
Abdominal aorta FL status				.20
Partial	24 (13/54)	32 (8/25)	17 (5/29)
Patent	76 (41/54)	68 (17/25)	83 (24/29)

Values are mean ± SD or % (n/N).

aTBAD, acute type B aortic dissection; FL, false lumen.

The most common indications for intervention were malperfusion (42%) and refractory pain (34%) (Table 3). Thirteen (14%) patients required a lumbar drain (preoperative: n = 3, postoperative: n = 10). In the 2 patients presenting with paraplegia, motor function did not improve despite preoperative lumbar drain insertion, and both were discharged to acute rehab facilities after index hospitalization. Rates of postoperative lumbar drain insertion were higher in the extended group (17% vs 3% in the standard group; P = .039), and no patients experienced complications as a result of utilization. The length of aortic coverage was significantly longer in the extended group (241.7 ± 29.2 mm vs 180.8 ± 22.3 mm in the standard group; P < .001). The mean length of aorta left uncovered relative to the celiac axis was 101.6 ± 65.1 mm in the standard group and 30.0 ± 23.0 mm in the extended group (P < .001).

Table 3.

Perioperative Variables

Variable	All (N = 91)	Standard Group (n = 39)	Extended Group (n = 52)	P Value
Indications for intervention				.80
Malperfusion	42 (38)	44 (17)	40 (21)
Lower extremity	17	11	6
Renal	6	0	6
Mesenteric	2	0	2
>1 vascular system	13	6	7
Refractory pain	34 (31)	36 (14)	33 (17)
Impending or contained rupture	12 (11)	8 (3)	15 (8)
>1 indication	12 (11)	13 (5)	12 (6)
Lumbar drain use	14 (13)	8 (3)	19 (10)	.10
Preoperative	3 (3)	5 (2)	2 (1)	.60
Postoperative	11 (10)	3 (1)	17 (9)	.039a
Length of stent graft coverage, mm	215.6 ± 40.1	180.8 ± 22.3	241.7 ± 29.2	<.001a

^aStatistically significant (P<0.05).

Values are % (n), n, or mean ± SD.

Table 4 presents postoperative outcomes based upon extent of stent graft coverage. Among the entire cohort, there were 5 (6%) in-hospital deaths (6% in the extended group vs 5% in the standard group). There were 3 inhospital mortalities in the extended group. One patient died from cardiac arrest on the sixth postoperative day after undergoing TEVAR for a ruptured DeBakey type IIIb aortic dissection. The other 2 patients had delayed presentations with mesenteric ischemia, underwent TEVAR with angiographic confirmation of reperfusion of the celiac and superior mesenteric artery vascular territories and died from mesenteric ischemia. In the standard group, 1 patient who presented with mesenteric ischemia died after TEVAR from respiratory failure secondary to acute respiratory distress syndrome. The other mortality was a patient early in our series with a thoracoabdominal aneurysm with contained rupture of the proximal DTA.

Table 4.

Postoperative Outcomes

Outcome	All Group (N = 91)	Standard Group (n = 39)	Extended Group (n = 52)	P Value
In-hospital mortality	6 (5)	5 (2)	6 (3)	1.0
Overall mortality	11 (10)	13 (5)	10 (5)	.70
Stroke	6 (5)	5 (2)	6 (3)	1.0
Paralysis	0 (0)	0 (0)	0 (0)	1.0
Paraparesis	3 (3)	0 (0)	6 (3)	.30
Renal failure requiring dialysis	0 (0)	0 (0)	0 (0)	1.0
Retrograde type A dissection	3 (3)	3 (1)	4 (2)	1.0

Values are % (n).

On postoperative day 3, he ruptured his abdominal aorta and expired. The overall mortality for the series was 11% and was similar between groups throughout the duration of follow-up (Figure 2). There were no differences between groups in the rates of stroke, new-onset renal failure requiring dialysis, and retrograde type A dissection. The rate of new-onset SCI was 6% in the extended group and 0% in the standard group. No patients experienced paraplegia. In the extended group, 3 patients had postoperative paraparesis that improved after emergent lumbar drain placement and blood pressure augmentation.

Figure 2.

Overall survival among groups.

Reinterventions

Table 5 lists all aortic reinterventions. One patient in each group underwent proximal aortic replacement for retrograde type A dissection or intramural hematoma after initial TEVAR deployment. Both of these cases involved ascending aortic pathology secondary to wire-induced injury early in the series when institutional experience was limited. Nine patients, all in the standard group, required open proximal aortic replacement for either de novo acute type A dissection or intramural hematoma (n = 4) or preexisting proximal aortic aneurysms (n = 5). Other reinterventions procedures included left carotid–LSA bypass (n = 2), LSA embolization (n = 4), and renal or iliac artery stenting for malperfusion (n = 3).

Table 5.

Late Aortic Reinterventions

Variable	All	Standard Group	Extended Group	P Value
Distal aortic reintervention rate, %	9.9	17.9	3.8	.038a
Distal aortic reinterventions
Open DTA/TAAA replacement	2	2	0
Distal endovascular extension	7	5	2
Proximal aortic reintervention rate, %	12.1	25.6	1.9	.002a
Proximal aortic reinterventions
Retrograde type A/IMH	2	1	1
De novo acute type A/IMH	4	4	0
Proximal aortic aneurysm	5	5	0
Left carotid–subclavian artery bypass	2	2	0
Left subclavian artery embolization	4	2	2
Branch artery stenting for malperfusion	3	1	2
Total aortic/branch vessel reinterventions	29	22	7

^aStatistically significant (p<0.05).

Values are n, unless otherwise indicated.

DTA/TAAA, descending thoracic aortic or thoracoabdominal aortic aneurysm; IMH, intramural hematoma.

Among the entire cohort, there were 9 distal aortic reinterventions (Table 5). All distal aortic reinterventions were due to progression of disease, and the reintervention rate was higher in the standard group (18% vs 4% in the extended group; P = .038). Of the 7 distal reinterventions in the standard group, 2 patients underwent open aortic replacement and 5 underwent endovascular stent graft extensions to the celiac artery. In the extended group, 1 patient underwent complete DTA relining for a type IV endoleak; the other underwent an FL embolization procedure. Kaplan-Meier estimates demonstrated greater freedom from distal aortic reinterventions in the extended group at 5-years (92% vs 68% in the standard group; P = .044) (Figure 3). There were no mortalities after any of the reintervention procedures.

Figure 3.

Freedom from distal aortic reinterventions among groups.

Aortic Remodeling Outcomes

Imaging follow-up was 90% complete. There was no statistical difference in aortic remodeling between groups. On the most recent post-TEVAR imaging study, maximum aortic diameters at all analyzed levels of the thoracic and abdominal aorta were equivalent between patients undergoing standard vs extended coverage (Table 6). Within the stented segment of the proximal and mid DTA, TEVAR was successful in eliminating FL flow in 90% and 73% of all patients, respectively. These results were statistically equivalent between groups. However, at the distal DTA, the proportion of FL obliteration or thrombosis was significantly higher in the extended coverage group (extended 53% vs standard 16%; P = .004).

Table 6.

Aortic Remodeling Outcomes

Outcome	All	Standard Group	Extended Group	P Value
Maximum aortic diameter, cm
DTA	4.4 ± 0.9	4.4 ± 0.8	4.4 ± 0.9	.80
Stent end	3.5 ± 0.6	3.5 ± 0.6	3.4 ± 0.6	.30
Celiac artery	3.2 ± 0.5	3.2 ± 0.5	3.2 ± 0.5	.60
Infrarenal aorta	2.6 ± 0.4	2.6 ± 0.5	2.5 ± 0.4	.40
Proximal DTA FL status				.70
Obliterated	54 (44/82)	49 (17/35)	57 (27/47)
Thrombosed	37 (30/82)	40 (14/35)	34 (16/47)
Partial	10 (8/82)	11 (4/35)	9 (4/47)
Mid-DTA FL status				.30
Obliterated	42 (34/82)	31 (11/35)	49 (23/47)
Thrombosed	32 (26/82)	34 (12/35)	30 (14/47)
Partial	26 (21/82)	31 (11/35)	21 (10/47)
Patent	1 (1/82)	3 (1/35)	0 (0/47)
Distal DTA FL status				.004a
Obliterated	20 (15/76)	3 (1/31)	31 (14/45)
Thrombosed	18 (14/76)	13 (4/31)	22 (10/45)
Partial	45 (34/76)	58 (18/31)	36 (16/45)
Patent	17 (13/76)	26 (8/31)	11 (5/45)
Stent end FL status				.90
Obliterated	7 (5/71)	7 (2/29)	7 (3/42)
Thrombosed	24 (17/71)	28 (8/29)	21 (9/42)
Partial	37 (26/71)	38 (11/29)	36 (15/42)
Patent	32 (23/71)	28 (8/29)	36 (15/42)
Celiac artery FL status				.80
Obliterated	1 (1/69)	0 (0/28)	2 (1/41)
Thrombosed	7 (5/69)	4 (1/28)	10 (4/41)
Partial	33 (23/69)	32 (9/28)	34 (14/41)
Patent	58 (40/69)	64 (18/28)	54 (22/41)
Infrarenal aorta FL status				.80
Thrombosed	6 (4/67)	4 (1/28)	8 (3/39)	.80
Partial	24 (16/67)	21 (6/28)	26 (10/39)
Patent	70 (47/67)	75 (21/28)	67 (26/39)

^aStatistically significant (P<0.05).

Values are mean ± SD or % (n/n).

DTA, descending thoracic aorta; FL, false lumen.

Comment

By directly treating the aortic pathology responsible for inducing malperfusion or rupture, TEVAR has significantly reduced mortality in patients with complicated aTBAD, thus justifying its role as the gold standard treatment for this presentation of aortic dissection.¹¹ The in-hospital mortality of 6% observed in the current series is consistent with the mortality rates of 0% to 8% reported in other contemporary reports of complicated aTBAD treated with TEVAR.^5,12 Yet, despite the widespread consensus of using endovascular therapy as first-line therapy for complicated aTBAD, there remains a lack of agreement regarding several important details of the procedure. This study addresses one of these unresolved issues, namely the safety and efficacy of extended aortic coverage.

In the current study, the average length of aortic coverage was 180.8 ± 22.3 mm in the standard group and 241.7 ± 29.2 mm (P < .001) in the extended group. There were no cases of paraplegia after TEVAR, and the overall incidence of stroke and paraparesis were 6% and 3%, respectively. Three patients, all in the extended group, developed postoperative paraparesis. Only 1 of these patients emerged from anesthesia with clinical signs of SCI, and the paraparesis resolved after 1 hour of CSF drainage and blood pressure augmentation. The other 2 patients awoke from the procedure with normal neurologic function; however, they developed delayed paraparesis in the intensive care unit secondary to inappropriate blood pressure management. Both of these patients underwent blood pressure augmentation and CSF drainage and were ambulating at hospital discharge. The 3% overall incidence of SCI for the series is consistent with the 2% to 7% rate reported in other series for patients undergoing TEVAR for either dissection or aneurysmal disease.^13–15

The goal of extensive aortic endograft coverage is to maximize the aortic remodeling benefits of TEVAR, as this has been shown to reduce the progression of aortic disease and aortic-specific mortality.¹⁶ Extending the covered endografts to the celiac axis provides coverage of all secondary tears in the DTA, thus further reducing both antegrade and retrograde perfusion of the FL. Our hypothesis is that increased thrombosis and obliteration of the FL will translate into a reduction in aortic growth and distal aortic reintervention and ultimately improve long-term survival. The data from the current study demonstrate that while both limited and extended endograft coverage of the DTA achieve complete FL thrombosis and sobliteration throughout the proximal and mid-DTA in greater than 70% of patients, there is a distinct advantage with the extended strategy with regard to the complete elimination of FL flow throughout the entire DTA (distal DTA FL thrombosis and obliteration: 53% in the extended group vs 16% in the standard group; P = .004). The status of the FL is an important radiographic feature in TBAD patients, and FL patency has been identified as an independent risk factor for aneurysm growth in patients with TBAD.^17,18

Despite this difference in the distal DTA FL blood flow, the length of stent graft coverage did not impact aortic size throughout the entire thoracoabdominal aorta, as both groups demonstrated minimal aortic growth in all segments measured. However, there did appear to be a statistically significant benefit in the need for distal aortic reinterventions with the extended endografting strategy. In the standard group, 7 patients demonstrated significant FL growth requiring either open (n = 2) aortic replacement or distal stent graft extension to the celiac artery (n = 5). All of the patients had robust DTA FL blood flow originating from uncovered secondary tears in the DTA. In the extended group, 2 patients required distal aortic reinterventions. One patient underwent relining of previous endografts in the DTA for persistent FL perfusion due to suspected endograft porosity, as no obvious antegrade or retrograde FL perfusion could explain her aortic growth. The other patient received an FL embolization procedure in an attempt to arrest retrograde FL perfusion originating from patent visceral segment and infrarenal aortic intimal tears that were causing rapid aortic expansion with total aortic diameter greater than 6.0 cm. Retrograde FL perfusion is the Achilles heel of TEVAR for TBAD and currently remains the biggest challenge to overcome with endovascular therapy.

Regardless of which strategy is chosen, elimination of FL flow distal to the stent graft remains a barrier. Our results demonstrated progressively declining rates of FL obliteration/thrombosis of 31% at a level 1 cm distal to the edge of the stent graft to 6% at the level of the infrarenal aorta (Table 6). Thus, the abdominal aorta remains at risk and warrants continued surveillance.

Lessons learned from the Investigation of Stent Grafts in Aortic Dissection with extended length of follow-up (INSTEAD-XL) trial are germane to the interpretation of the results of the current study. In the initial INSTEAD trial, there was no difference between TEVAR and optimal medical therapy for the treatment of TBAD in allcause mortality, aortic-related mortality, or progression of aortic disease at 2 years, despite positive aortic remodeling in greater than 90% of the TEVAR patients. However, when the endpoints were analyzed after 5 years, there was a significant reduction in the progression of aortic disease and aortic-related mortality with an improvement (nonsignificant) in all-cause mortality in the TEVAR group.¹⁶ Therefore, while the radiographic benefits of FL thrombosis are clearly seen within the first year after TEVAR, the clinical benefits of aortic remodeling may not become evident until 3 to 5 years after the index procedure.^1,16 The lack of significant differences observed in the current study between the standard and extended groups in aortic size may be explained from the lack of long-term follow-up. These patients will continue to undergo lifelong aortic surveillance per our institutional protocol, and the long-term outcomes will be analyzed in a follow-up study.

Limitations of the current study include a lack of complete radiographic follow-up, a lack of long-term follow-up, a treatment bias depending on the time period of treatment within the series, and a limited sample size. Although all imaging analyses were conducted by the primary and senior authors and crosschecked with written attending radiologist reports, we did not perform interobserver or intraobserver analyses. Additionally, although we analyzed aortic size and FL status in 90% of patients, the missing data from 10% of patients could have influenced our results. After the first 30 patients in this series (standard coverage), we began our extended endograft protocol in the vast majority of cases. Therefore, the clinical outcomes could have been impacted by the time of treatment within the series. Finally, we acknowledge the limitations in the generalizability of our study findings given the small sample size. Because the rates of adverse events in our series were low, our analysis was not sufficiently powered to detect a difference in the incidence of spinal cord ischemia between groups.

In conclusion, the data from this series demonstrate the safety and efficacy of TEVAR coverage of the entire DTA in patients with complicated aTBAD. Complete elimination of thoracic FL blood flow can be accomplished in the majority of patients with a low risk of spinal cord ischemia using a selective CSF drainage protocol combined with permissive hypertension. TEVAR remains ineffective in treating the abdominal aortic FL, as it remains patent in the vast majority of patients. However, we have not observed significant abdominal aortic growth requiring reintervention regardless of strategy. The long-term benefit of extended TEVAR coverage may be a decreased need for future interventions on the DTA. Longer-term follow-up will provide conclusive evidence.