- Email: [email protected]
Remodeling of abdominal aortic branch perfusion after thoracic endovascular aortic repair for aortic dissections
From the Society for Vascular Surgery
Remodeling of abdominal aortic branch perfusion after thoracic endovascular aortic repair for aortic dissections Sukgu M. Han, MD, Eric C. Kuo, MD, Karen Woo, MD, Ramsey Elsayed, MD, B. Sean Nguyen, BS, Sung W. Ham, MD, Vincent L. Rowe, MD, and Fred A. Weaver, MD, Los Angeles, Calif Objective: The fate of the abdominal aorta and its branches after thoracic endovascular aortic repair for aortic dissection (TEVAR-AD) has not been studied. The objective of this study was to describe the midterm changes in abdominal aortic branch perfusion after TEVAR-AD. Methods: A retrospective analysis of TEVAR-AD at a single institution from December 1, 2008, to March 31, 2015, was performed. Computed tomography angiography (CTA) images were reviewed to characterize the perfusion pattern changes of the celiac, superior mesenteric, inferior mesenteric, bilateral renal, and common iliac arteries. Risk factors associated with branch interventions were identified. Results: During the study period, 68 patients underwent TEVAR-AD, 46 of whom had pre-TEVAR and post-TEVAR CTA images available for review. For post-TEVAR CTA, the most recent scans were selected for analysis. The mean period between CTA studies was 371 days. Indications for TEVAR-AD were persistent pain (41%), malperfusion (15%), rupture (6%), and aneurysmal degeneration (33%). Twenty-five patients (54%) were treated during the acute phase (<14 days). All patients had dissections extending to the paravisceral aorta. Of the 304 abdominal aortic branches analyzed, 8 required intervention (2.6%). Branch events requiring intervention included malperfusion (two) and aneurysms involving the branches (three). No intervention was performed for one asymptomatic inferior mesenteric artery occlusion. Of the remaining 295 branches, changes in perfusion patterns were observed in 16 (5.4%). Twelve branches (75%) demonstrated an increased true lumen contribution to perfusion. Four branches (25%) had increased false lumen contribution, without clinical evidence of malperfusion. Patients requiring branch interventions were more likely to have severe chronic kidney disease (P [ .012) and more extensive aortic zone coverage during TEVAR (P [ .003). On multivariable Cox proportional hazards analysis, coverage of four or more zones during TEVAR-AD was associated with branch intervention (odds ratio, 6.44; 95% confidence interval, 1.01-40.8). The estimated intervention-free patency of the abdominal aortic branches was 89% at 5 years. Conclusions: Perfusion patterns of abdominal aortic branches remain largely stable after TEVAR-AD. The need for branch intervention is rare and associated with extensive aortic coverage. (J Vasc Surg 2016;-:1-10.)
Since its first reported use in aortic dissections in 1999, thoracic endovascular aortic repair (TEVAR) has become the first-line treatment modality for complicated type B aortic dissections.1,2 In uncomplicated cases, recent randomized trials have suggested that TEVAR may improve long-term aorta-specific survival and delay disease progression compared with medical management alone.3,4 These long-term benefits have been ascribed to the favorable aortic remodeling that results from TEVAR. “Favorable”
From the Division of Vascular Surgery and Endovascular Therapy, Keck Medical Center of University of Southern California. Author conflict of interest: none. Presented during a plenary session at the 2015 Vascular Annual Meeting of the Society for Vascular Surgery, Chicago, Ill, June 17-20, 2015. Correspondence: Sukgu M. Han, MD, Division of Vascular Surgery and Endovascular Therapy, University of Southern California, 1520 San Pablo St, Ste 4300, Los Angeles 90033, CA (e-mail: sukgu.han@med. usc.edu). The editors and reviewers of this article have no relevant financial relationships to disclose per the JVS policy that requires reviewers to decline review of any manuscript for which they may have a conflict of interest. 0741-5214 Copyright Ó 2016 by the Society for Vascular Surgery. Published by Elsevier Inc. http://dx.doi.org/10.1016/j.jvs.2016.03.441
or “positive” aortic remodeling includes true lumen expansion, false lumen regression, complete false lumen thrombosis, and stabilization of overall aortic diameter.3-8 Previous studies on aortic remodeling after TEVAR for aortic dissection (TEVAR-AD) have focused on the changes along the thoracic aorta.3-5,7-10 The data on the effects of TEVAR-AD on the abdominal aorta have been limited to diameter measurements at few selected abdominal aortic levels and the total volume measurements. However, little information, with the exception of occasional reports of branch stenting, is available on the fate of the abdominal aortic branches after TEVAR-AD when it is used in both acute and chronic phases.7,8,11-13 Therefore, the aim of our study was to assess changes in the perfusion patterns of the abdominal aortic branches after TEVARAD. Furthermore, we sought to identify risk factors associated with the need for branch interventions. METHODS Patients. A retrospective review was conducted to identify all patients undergoing TEVAR-AD at the Keck Medical Center of University of Southern California from December 1, 2008, through March 31, 2015. Clinical data were collected from hospital electronic medical records and 1
JOURNAL OF VASCULAR SURGERY --- 2016
2 Han et al
entered into a designated database using the Research Electronic Data Capture (REDCap) system. The were 118 variables, including patient demographics, medical comorbidities, symptoms on index admission, dissection chronicity, operative details, and postoperative course, collected. The study was approved by the Institutional Review Board of the University of Southern California, and patient consents were waived. The diagnosis of aortic dissection was made by computed tomography angiography (CTA). In addition to the de novo type B aortic dissections, patients with prior ascending aortic repair for type A dissections with residual dissection distal to the left subclavian were included. Degenerative aneurysms, penetrating aortic ulcers, and traumatic transections were excluded. Patients in whom the aortic dissection did not extend to the paravisceral aorta were excluded. Other exclusion criteria included patients who underwent total visceral debranching or open arch reconstruction in frozen elephant trunk configuration with antegrade deployment of TEVAR. Finally, patients who did not have post-TEVAR CTA images of the abdominal segments available for analysis were excluded. TEVAR procedure. Lumbar drains were placed preoperatively in nonemergent cases. Stent graft placement was performed through either open or percutaneous bilateral femoral access. For percutaneous procedures, two ProGlide percutaneous closure devices (Abbott Vascular, Abbott Park, Ill) were deployed perpendicular to each other, as described in preclose technique.14 Both intravascular ultrasound (IVUS) and transesophageal echocardiography were used in the majority of cases. IVUS was used for confirmation of true lumen cannulation, identification of proximal entry tear, and measurement of the proximal seal zone diameters. IVUS-measured proximal seal zone diameters were used to confirm stent graft sizing. Transesophageal echocardiography was used to monitor for development of retrograde dissection after stent graft deployment in addition to intraoperative cardiac function and fluid status. The operating surgeon’s preference determined the device choice. In all cases, the primary goal of TEVAR was to cover the proximal entry tear in the hope of achieving true lumen expansion. The decision to place additional distal stent grafts was individualized according to the existence of more distal septal tears within the thoracic aorta and was at the discretion of the operating surgeon. Definitions. Aortic dissections managed by TEVAR within 14 days from symptom onset were defined as acute. Subacute was defined as 15 to 90 days and chronic as >90 days from the initial symptom. The baseline extent of the dissection was assessed by the number of aortic zones involved, as defined by the standards established by the Society for Vascular Surgery Ad Hoc Committee on Reporting Standards for TEVAR.15 The total number of aortic zones involved was calculated for each patient and categorized into quartiles (Table I, A). The top quartile was defined as extensive aortic dissection (eight or more zones involved). Similarly, the number of aortic zones covered determined the extent of aortic coverage by
Table I. A, Extent of aortic dissection at baseline Percentile 0-24 25-74 75-99
No. of zones 0-5 6-7 8-11
Table I. B, Aortic coverage by thoracic endovascular aortic repair (TEVAR) Percentile 0-49 50-74 75-99
No. of zones 0-2 3 4-6
TEVAR. The top quartile of the aortic coverage by TEVAR was defined as extensive aortic coverage (four or more aortic zones covered; Table I, B). Branch vessel analysis. Axial cuts from CTA immediately before TEVAR were compared with the most recent post-TEVAR CTA. Celiac, superior mesenteric artery (SMA), inferior mesenteric artery (IMA), right renal artery, left renal artery, right common iliac artery, and left common iliac artery were analyzed. The perfusion pattern for each branch was characterized as being supplied by true lumen, false lumen, or both lumens (Fig 1). Remodeling in branch perfusion was determined by comparing the baseline and post-TEVAR CTA images. Remodeling was defined as positive when there was increased true lumen and loss of false lumen contribution to branch perfusion. For example, a branch whose perfusion pattern changed from both lumens to only true lumen after TEVAR was characterized as positive (Fig 2, A and B). Likewise, branches changed from only false lumen perfusion to both true and false lumens were also classified as positive (Fig 2, C and D). Conversely, remodeling was defined as negative when false lumen perfusion increased with loss of true lumen perfusion or branches with true or false lumen contributions to perfusion occluded after TEVAR. Statistical analysis. Clinical data were imported from REDCap to Stata 12 (Stata Corp, College Station, Tex). To analyze branch remodeling, a branch-specific data set was created with an individual branch as a unit of analysis. Univariable analyses were performed to identify clinical and operative risk factors associated with increased risk of branch intervention. Continuous variables were compared using Student t-test; categorical variables were compared using c2 or Fisher exact test. The intervention-free patency of abdominal aortic branches was estimated using Kaplan-Meier survival analysis. A Cox proportional hazards model was created for branch intervention by including variables with P values < .2 from univariable analyses. Both univariable
JOURNAL OF VASCULAR SURGERY Volume -, Number -
Han et al 3
Fig 1. Branch perfusion classification. Branches perfused by (A) true lumen, (B) false lumen, and (C) both lumens at baseline.
and multivariable analyses were performed using individual branches as a unit of analysis. RESULTS Sixty-four patients who underwent TEVAR-AD satisfied the described inclusion criteria. Two patients died postoperatively before CTA imaging. One died of a ruptured type B aortic dissection despite TEVAR. The other died of visceral malperfusion, which was manifested by abdominal pain and the associated CTA finding of the celiac artery and SMA being supplied entirely by an extremely compressed true lumen. TEVAR followed by immediate SMA stenting was performed. On postoperative day 2, the patient underwent exploration and was found to have necrosis of the entire small bowel. He died shortly thereafter. Of the remaining 62 patients, 16 patients did not have satisfactory post-TEVAR CTA of the chest, abdomen, and pelvis; 2 patients had a non-contrast-enhanced CT scan only because of chronic renal insufficiency, and 14 patients had a CTA scan of the chest only (Fig 3). This left a cohort of 46 patients with adequate post-TEVAR imaging for analysis. The 46 patients provided 304 branches for review and analysis.
Baseline characteristics of these 46 patients are shown in Table II. The majority of the patients were male with a history of hypertension. Other cardiovascular comorbidities, such as diabetes, coronary artery disease, congestive heart failure, and prior stroke, were rare. Approximately half were former or active smokers, whereas two carried the diagnosis of chronic obstructive pulmonary disease. Twenty-five patients underwent TEVAR for an acute aortic dissection. Pain (12/25 [48%]) and malperfusion (7/25 [28%]) were the major indications. Malperfused organs included bowel (four), lower extremity (two), and kidney (one). Two additional patients with acute dissections were asymptomatic but had high-risk features on CTA, namely, a proximal entry tear >1 cm and transaortic diameters >4 cm (Table III). TEVAR was performed in 21 patients, with a chronic aortic dissection with aneurysmal degeneration (13/21 [62%]) as the most common indication. Operative details are shown in Table IV. A variety of commercially available stent grafts were deployed, and in one third of cases, more than one stent graft was used. The aortic zones covered by TEVAR are shown in Fig 4. Three patients had zone 1 coverage after arch debranching, whereas 16 had left subclavian coverage
4 Han et al
JOURNAL OF VASCULAR SURGERY --- 2016
Fig 2. Examples of positive branch remodeling. Celiac artery (A) perfused by both lumens at baseline and (B) perfused by true lumen after thoracic endovascular aortic repair for aortic dissection (TEVAR-AD). Left renal artery (C) perfused by false lumen at baseline (arrow on the true lumen) and (D) perfused by both lumens after TEVAR-AD.
with revascularization in 11. In all cases, TEVAR was limited to the thoracic aorta. Zone 5 was covered in 32 cases (69%). There were no cases of distal bare-metal stent placement or adjunctive false lumen treatment. After TEVAR, there was a single case (2%) of spinal cord ischemia in a patient who presented with a 7-cm aneurysmal degeneration due to a chronic dissection, involving segments 4 to 7. One patient developed a loculated hemothorax after presenting with a rupture, requiring thoracoscopic evacuation. Two patients had access site complications, one brachial sheath hematoma and one femoral artery thrombosis after percutaneous TEVAR. One patient developed hand ischemia after left subclavian
TEVAR coverage that resolved after a left carotid to subclavian bypass. Two late deaths occurred, one following a grand mal seizure 6 weeks after TEVAR-AD and the other of an unknown cause 2 months after TEVAR-AD. Aortic morphologic findings and branch remodeling. The aortic zones involved by the dissection at baseline are illustrated in Fig 4. Two patients had residual dissection involving zone 2 after an arch repair of a type A aortic dissection. Zones 4 to 6 were involved in all patients. The majority of the patients had zone 3 to 9 involvement. Common iliac arteries were involved in half of the patients, whereas external iliacs were involved in a quarter of the patients.
JOURNAL OF VASCULAR SURGERY Volume -, Number -
Han et al 5
Fig 3. Patient selection. TEVAR, Thoracic endovascular aortic repair.
Table II. Baseline characteristics of patients Patients Age, years, mean 6 SD Male gender HTN DM CVA CAD CHF COPD Smoking ESRD Prior ascending aortic repair
Table III. Preoperative presentation 46 58.63 6 12.1 35 (76.1) 42 (91.3) 1 (2.2) 1 (2.2) 4 (8.7) 2 (4.3) 2 (4.3) 24 (52.2) 4 (8.7) 9 (19.6)
CAD, Coronary artery disease; CHF, congestive heart failure; COPD, chronic obstructive pulmonary disease; CVA, cerebrovascular accident; DM, diabetes mellitus; ESRD, end-stage renal disease; HTN, hypertension; SD, standard deviation. Data are presented as number (%) unless otherwise indicated.
The baseline branch perfusion pattern is shown in Fig 5. The majority of branches were perfused entirely by a true lumen (69% [211/304]). Among the abdominal aortic branches, the left renal artery perfusion was most frequently involved with dissection (43% by a false lumen or both lumens), whereas the SMA was most often perfused entirely by a true lumen (83%). Branch remodeling after TEVAR was characterized as positive in 12 branches: 3 celiac, 1 right renal, 4 left renal, 1 IMA, 2 right common iliac, and 1 left common iliac. Ten (83%) branches were in patients with acute aortic dissections. Negative branch remodeling was seen in five branches: SMA, right renal artery, IMA, and two left common iliac arteries (Fig 6). There were no clinical sequelae related to negative branch remodeling. On the KaplanMeier survival analysis, the estimated intervention-fee patency rates of abdominal aortic branches were 98% at 1 year and 89% at 5 years (Fig 7).
No. Pain Malperfusion Renal Visceral Lower extremities Aneurysmal degeneration Rupture Other
Acute (n ¼ 25), No. (%)
Chronic (n ¼ 21), No. (%)
25 12 (48) 7 (28) 1 4 2 2 (8) 2 (8) 2 (8)
21 7 (33)
13 (62) 1 (5) 0
Branch interventions. In the seven patients presenting with malperfusion, only two patients required branch interventions. In one patient with abdominal pain, the SMA was perfused entirely by the false lumen on preoperative CTA. Poor visualization of the SMA on post-TEVAR angiography resolved after SMA stent placement. The patient’s subsequent postoperative course was uncomplicated. In a second patient with lower extremity pain, weakness, and absent pulses, the left common iliac artery on preoperative CTA was perfused entirely by the false lumen. After a femoralfemoral bypass at the time of TEVAR, the symptoms resolved. In the remaining five patients with malperfusion (three SMA, one renal, one lower extremity), TEVAR resulted in an immediate true lumen expansion and improvement in branch perfusion with postoperative resolution of malperfusion. In four additional patients, six branch interventions were performed. In two, celiac stents were “prophylactically” placed because of poor visualization of the celiac artery on post-TEVAR angiography. Preoperatively,
JOURNAL OF VASCULAR SURGERY --- 2016
6 Han et al
Table IV. Thoracic endovascular aortic repair (TEVAR) details OR time, minutes Fluoroscopy time, minutes EBL, mL Transfusion required, % (No.) Preoperative lumbar drain, % (No.)
129.7 6 68.1 18.17 6 12.13 203 6 313.76 15.2 (7) 68.9 (31)
EBL, Estimated blood loss; OR, operating room.
Fig 5. Baseline branch perfusion pattern. IMA, Inferior mesenteric artery; L CIA, left common iliac artery; LRA, left renal artery; R CIA, right common iliac artery; RRA, right renal artery; SMA, superior mesenteric artery.
DISCUSSION
Fig 4. Baseline extent of aortic dissection and coverage by thoracic endovascular aortic repair for aortic dissection (TEVAR-AD).
neither patient had signs of visceral malperfusion. In a third patient, stenting of the right renal artery was performed for resistant hypertension 2 months after TEVAR. Poor renal artery perfusion due to significant false lumen flow was hypothesized as the cause of the hypertension. However, the hypertension persisted despite successful improvement of true lumen perfusion after stent placement. Finally, a fourth patient required endovascular aneurysm repair for aneurysmal degeneration of the dissected infrarenal aorta (one IMA coverage, two iliac stents) 3 years after TEVAR-AD. Thus, a total of six patients ultimately required branch interventions, with only two of these being associated with pre-TEVAR evidence of malperfusion. On univariable analyses, severe chronic kidney disease (P ¼ .012) and extensive aortic zone coverage (P ¼ .003) were associated with increased risk of needing branch intervention. There was a trend toward pain as the indication for repair (P ¼ .07) being protective. Chronicity or the extent of dissection at baseline was not associated with branch interventions (Table V). A Cox proportional hazards regression model was constructed including age, severe chronic kidney disease, indication for repair, extensive aortic coverage, and extensive aortic dissection. Among these variables, extensive aortic coverage by TEVAR-AD was the only independent factor associated with increased risk of branch interventions (hazard ratio, 6.44; 95% confidence interval, 1.01-40.8; Table VI).
The indications for TEVAR in patients with type B aortic dissections continue to expand. Poor long-term outcomes of patients with medically managed type B dissections combined with recent studies demonstrating favorable aortic remodeling after TEVAR-AD have challenged the traditional paradigm of medical management as the first-line therapy for acute uncomplicated type B aortic dissections.3-5,7-10,16,17 To date, studies on aortic remodeling after TEVAR-AD have focused on true and false lumen changes as measured by cross-sectional diameters at select aortic levels and total volumetric analysis.3-10,18,19 By sealing the proximal tear, TEVAR rechannels blood flow into the true lumen while depressurizing the false lumen. This hemodynamic alteration, in turn, raises a concern for the potential adverse impact of proximal TEVAR on the downstream aortic branches, particularly the ones supplied by the false lumen. Whereas technical aspects of TEVAR-AD, such as the optimal length of coverage and the role of pre-emptive left subclavian revascularization, continue to be a subject of debate,13,16,20-22 no study has specifically investigated the impact of TEVAR-AD on the branches of the abdominal aorta. Therefore, this study analyzed branch perfusion patterns at baseline and the changes after TEVAR-AD. Furthermore, factors associated with the need for branch intervention were evaluated, thereby potentially guiding clinicians in the decision-making process. Most importantly, we have demonstrated that the natural course of the abdominal branches after TEVAR-AD is largely benign. After TEVAR-AD, 92% (279/304) of the branches remained unchanged in their perfusion pattern. Of the remaining 25 branches, 8 (32%) had branch interventions. Furthermore, two celiac artery branch interventions occurred on the basis of the radiographic appearance of branch perfusion in the absence of clinical malperfusion, and one for renal perfusion was nontherapeutic. This low rate of branch interventions (3% [8/304]) is consistent with anecdotal reports of
JOURNAL OF VASCULAR SURGERY Volume -, Number -
Han et al 7
Fig 6. Thoracic endovascular aortic repair (TEVAR)-induced branch perfusion changes in (A) celiac artery, (B) superior mesenteric artery (SMA), (C) right renal artery (RRA), (D) left renal artery (LRA), (E) right common iliac artery (R CIA), (F) left common iliac artery (L CIA), and (G) inferior mesenteric artery (IMA).
Fig 7. Kaplan-Meier analysis of intervention-free patency of abdominal aortic branches after thoracic endovascular aortic repair for aortic dissection (TEVAR-AD).
secondary branch interventions included in the Study of Thoracic Aortic Type B Dissection Using Endoluminal Repair (STABLE) trial (two renal artery stents; N ¼ 40) and a series from Scali et al (four visceral, renal, and iliac stents; N ¼ 80).7,8,13 Therefore, loss of patency in the
downstream aortic branches as a consequence of proximal TEVAR in aortic dissections appears exceedingly rare. Preservation of abdominal branch patency after TEVAR-AD was also observed in patients with clinical malperfusion. Of the seven patients with clinical malperfusion who underwent TEVAR, only two (28%) required branch intervention. This is consistent with a recent series by Ryan et al, who reviewed 61 patients with malperfusion, 41% of whom required a branch intervention after TEVAR.11 Similarly, Alsac et al observed that 40% of the patients with visceral and renal malperfusion required branch stenting after TEVAR.12 TEVAR-AD appears to obviate adjunctive branch intervention by expeditiously increasing true lumen perfusion to the malperfused branches in a majority of patients. In the absence of clinical malperfusion, our data suggest that pre-emptive branch stenting may not be necessary. Furthermore, most of the branches, including those perfused partially or entirely by the false lumen, will remain patent and in some cases remodel positively. Of the branches that changed perfusion pattern after TEVAR-AD, negative remodeling did not result in adverse clinical events during the study period, with only one asymptomatic branch (IMA) thrombosis. In addition, positively remodeled branches were mostly seen in acute
JOURNAL OF VASCULAR SURGERY --- 2016
8 Han et al
Table V. Univariate analysis of risk factors associated with branch interventions
Age, years, mean 6 SD Male gender HTN DM CVA CAD CHF COPD Smoking Severe CKD (stage 4 or greater) Acute dissection Indication for repair Pain Malperfusion Aneurysmal degeneration Rupture Unknown No. of zones involved in dissection No. of zones covered
Nonevent (n ¼ 296)
Intervened (n ¼ 8)
58.3 6 11.8 226 (76) 282 (95) 4 (1.4) 7 (2.4) 28 (9.5) 14 (4.7) 14 (4.7) 151 (51) 36 (12)
58.4 6 12.1 8 (100) 8 (100) 0 0 0 0 0 2 (25) 4 (50)
.98 .12 .53
164 (55)
3 (38)
.32 .07 .07 .64 .07 .34
131 (44) 46 (15) 91 (31) 14 (4.7) 14 (4.7) 7.56 6 2.61
1 1 5 1
(13) (13) (63) (13) 0 8.38 6 2.26
2.86 6 0.94
3.88 6 1.13
P
.46 .012
.39 .003
CAD, Coronary artery disease; CHF, congestive heart failure; CKD, chronic kidney disease; COPD, chronic obstructive pulmonary disease; CVA, cerebrovascular accident; DM, diabetes mellitus; ESRD, end-stage renal disease; HTN, hypertension; SD, standard deviation. Data are presented as number (%) unless otherwise indicated.
Table VI. Cox proportional hazards analysis of risk factors associated with branch interventions
Age Severe CKD Pain as indication for repair Extensive aortic coverage (4 or more) Extensive aortic dissection (8 or more)
HR
95% CI
0.95 2.58 0.25 6.44 0.89
0.89-1.02 0.42-15.8 0.02-2.24 1.01-40.8 0.66-1.2
CI, Confidence interval; CKD, chronic kidney disease; HR, hazard ratio.
dissections (10/12 [83%]) that did not require extensive aortic coverage (11/12 [92%]). These were likely seen in patients who had robust true lumen expansion down to the abdominal aorta as well as into the involved branches after TEVAR. Future studies are needed to confirm this apparent relationship between the aortic and the branch perfusion remodeling after TEVAR-AD. The significant association between extensive aortic coverage and branch interventions may be explained by two different but not mutually exclusive possibilities. First, extensive aortic coverage may be a surrogate marker for a complex dissection process involving multiple distal fenestrations. It is possible that initial completion angiography showed persistent robust filling of the false lumen through the distal fenestrations, prompting the operating surgeon to place additional distal devices. In turn, the same underlying dissection process may have led to the failure of true
lumen expansion in the involved branches, thereby prompting branch interventions. Second, there may be the confounding component of operator decisionmaking. Only two branch interventions occurred in the clinical setting of malperfusion. Others were based on subjective interpretation of the angiographic appearance of branch perfusion. It is possible that more aggressive surgeons tended to cover more aortic zones and stented more branches. This scenario is further supported by our observation that those who underwent TEVAR-AD for pain as the indication were less likely to undergo branch interventions. Unlike in those who presented with clinical malperfusion, the operating surgeons are unlikely to have been compelled to perform pre-emptive branch intervention in this subset of patients with persistent pain. This highlights the lack of consensus recommendations regarding the optimal length of coverage and the threshold for secondary branch interventions.23-25 Despite the likely heterogeneous threshold for aortic stent graft extension, our group’s approach remained consistent regarding the primary goal of TEVAR-AD during the study period. The initial priority was coverage of the proximal entry tear with a minimum of 150-mm length stent graft. The decision to perform distal extension was typically based on the degree of persistent false lumen perfusion in the context of presenting symptoms. As such, extensive coverage was more often seen in aneurysmal degenerations and ruptures. However, after adjusting for the indication and extent of dissection, we found that the only independent factor for branch interventions was extensive coverage. Limitations of our study include those inherent to a single-institution, retrospective review with a relatively small sample size. It is possible that there was a selection bias toward favorable clinical outcomes by exclusion of 16 patients without adequate imaging. Particularly, exclusion of one patient who died of bowel necrosis after TEVAR-AD and SMA stenting limits generalizability of our study to only those who survive the initial ischemic insult from complicated aortic dissection. Challenges with following up this population of patients are not unique to our study. The International Registry of Acute Aortic Dissection follow-up rate of 70% at 2 years also reflects this challenge.4 Our study had a 24% (16/68) exclusion rate, based on stricter imaging criteria. The consecutive nature of our series may mitigate some of the selection bias. No correlation between remodeling of the aorta and abdominal branches was performed because of the limited availability of imaging studies that allow detailed anatomic analyses of the entire aorta as well as the branches. Longitudinal follow-up results with multiple surveillance imaging studies are needed to assess the long-term clinical significance of branch remodeling observed in our study. CONCLUSIONS After TEVAR-AD, abdominal branch perfusion remains largely stable with high midterm interventionfree patency. The need for branch interventions was rare
JOURNAL OF VASCULAR SURGERY Volume -, Number -
and associated with extensive aortic coverage. In the absence of clinical malperfusion, pre-emptive branch stenting may not be necessary. AUTHOR CONTRIBUTIONS Conception and design: SMH, SWH, FW Analysis and interpretation: SMH, EK, KW, FW Data collection: SMH, EK, RE, BN Writing the article: SMH, EK, RE Critical revision of the article: SMH, EK, KW, RE, BN, SWH, VL, FW Final approval of the article: SMH, EK, KW, RE, BN, SWH, VL, FW Statistical analysis: SMH, EK, KW Obtained funding: Not applicable Overall responsibility: SMH REFERENCES 1. Dake MD, Kato N, Mitchell RS, Semba CP, Razavi MK, Shimono T, et al. Endovascular stent-graft placement for the treatment of acute aortic dissection. N Engl J Med 1999;340:1546-52. 2. Nienaber CA, Fattori R, Lund G, Dieckmann C, Wolf W, von Kodolitsch Y, et al. Nonsurgical reconstruction of thoracic aortic dissection by stent-graft placement. N Engl J Med 1999;340:1539-45. 3. Nienaber CA, Kische S, Rousseau H, Eggebrecht H, Rehders TC, Kundt G, et al; INSTEAD-XL trial. Endovascular repair of type B aortic dissection: long-term results of the randomized investigation of stent grafts in aortic dissection trial. Circ Cardiovasc Interv 2013;6: 407-16. 4. Heijmen R, Fattori R, Thompson M, Dai-Do D, Eggebrecht H, Degrieck I, et al. Mid-term outcomes and aortic remodelling after thoracic endovascular repair for acute, subacute, and chronic aortic dissection: the VIRTUE Registry. Eur J Vasc Endovasc Surg 2014;48: 363-71. 5. Sigman MM, Palmer OP, Ham SW, Cunningham M, Weaver FA. Aortic morphologic findings after thoracic endovascular aortic repair for type B aortic dissection. JAMA Surg 2014;149:977-83. 6. Watanabe Y, Shimamura K, Yoshida T, Daimon T, Shirakawa Y, Torikai K, et al. Aortic remodeling as a prognostic factor for late aortic events after thoracic endovascular aortic repair in type B aortic dissection with patent false lumen. J Endovasc Ther 2014;21:517-25. 7. Lombardi JV, Cambria RP, Nienaber CA, Chiesa R, Teebken O, Lee A, et al. Prospective multicenter clinical trial (STABLE) on the endovascular treatment of complicated type B aortic dissection using a composite device design. J Vasc Surg 2012;55:629-40.e2. 8. Lombardi JV, Cambria RP, Nienaber CA, Chiesa R, Mossop P, Haulon S, et al. Aortic remodeling after endovascular treatment of complicated type B aortic dissection with the use of a composite device design. J Vasc Surg 2014;59:1544-54. 9. Kim KM, Donayre CE, Reynolds TS, Kopchok GE, Walot I, Chauvapun JP, et al. Aortic remodeling, volumetric analysis, and clinical outcomes of endoluminal exclusion of acute complicated type B thoracic aortic dissections. J Vasc Surg 2011;54:316-25. 10. Andacheh ID, Donayre C, Othman F, Walot I, Kopchok G, White R. Patient outcomes and thoracic aortic volume and morphologic changes
Han et al 9
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
following thoracic endovascular aortic repair in patients with complicated chronic type B aortic dissection. J Vasc Surg 2012;56:644-50. Ryan M, Vargas L, Mastracci T, Srivastava S, Eagleton M, Kelso R, et al. Progress in management of malperfusion syndrome from type B dissections. J Vasc Surg 2013;57:1283-90. Alsac JM, Girault A, El Batti S, Abou Rjilli M, Alomran F, Achouh P, et al. Experience with the Zenith Dissection Endovascular System in the emergency setting of malperfusion in acute type B dissections. J Vasc Surg 2014;59:645-50. Scali ST, Feezor RJ, Chang CK, Stone DH, Hess PJ, Martin TD, et al. Efficacy of thoracic endovascular stent repair for chronic type B aortic dissection with aneurysmal degeneration. J Vasc Surg 2013;58:10-7. Krajcer Z, Nelson P, Bianchi C, Rao V, Morasch M, Bacharach J. Percutaneous endovascular abdominal aortic aneurysm repair: methods and initial outcomes from the first prospective, multicenter trial. J Cardiovasc Surg (Torino) 2011;52:651-9. Fillinger MF, Greenberg RK, McKinsey JF, Chaikof EL. Reporting standards of thoracic aortic endovascular repair (TEVAR). J Vasc Surg 2010;52:1022-33. Lee CJ, Rodriguez HE, Kibbe MR, Malaisrie SC, Eskandari MK. Secondary interventions after elective thoracic endovascular aortic repair for degenerative aneurysms. J Vasc Surg 2013;57:1269-74. Kusagawa H, Shimono T, Ishida M, Suzuki T, Yasuda F, Yuasa U, et al. Changes in false lumen after transluminal stent-graft placement in aortic dissections: six years’ experience. Circulation 2005;111:2951-7. Jia X, Guo W, Li TX, Guan S, Yang RM, Liu XP, et al. The results of stent graft versus medication therapy for chronic type B dissection. J Vasc Surg 2013;57:406-14. Melissano G, Bertoglio L, Rinaldi E, Civilini E, Tshomba Y, Kahlberg A, et al. Volume changes in aortic true and false lumen after the “PETTICOAT” procedure for type B aortic dissection. J Vasc Surg 2012;55:641-51. Manning BJ, Dias N, Ohrlander T, Malina M, Sonesson B, Resch T, et al. Endovascular treatment for chronic type B dissection: limitations of short stent-grafts revealed at midterm follow-up. J Endovasc Ther 2009;16:590-7. Szeto WY, McGarvey M, Pochettino A, Moser GW, Hoboken A, Cornelius K, et al. Results of a new surgical paradigm: endovascular repair for acute complicated type B aortic dissection. Ann Thorac Surg 2008;86:87-93; discussion: 93-4. Flecher E, Cluzel P, Bonnet N, Aubert S, Gaubert A, Pavie A, et al. Endovascular treatment of descending aortic dissection (type B): shortand medium-term results. Arch Cardiovasc Dis 2008;101:94-9. Grabenwöger M, Alfonso F, Bachet J, Bonser R, Czerny M, Eggebrecht H, et al. Thoracic Endovascular Aortic Repair (TEVAR) for the treatment of aortic diseases: a position statement from the European Association for Cardio-Thoracic Surgery (EACTS) and the European Society of Cardiology (ESC), in collaboration with the European Association of Percutaneous Cardiovascular Interventions (EAPCI). Eur J Cardiothorac Surg 2012;42:17-24. Svensson LG, Kouchoukos NT, Miller DC, Bavaria JE, Coselli JS, Curi MA, et al. Expert consensus document on the treatment of descending thoracic aortic disease using endovascular stent-grafts. Ann Thorac Surg 2008;85:S1-41. Thrumurthy SG, Karthikesalingam A, Patternson BO, Hold PJ, Hinchliffe RJ, Loftus IM, et al. A systematic review of mid-term outcomes of thoracic endovascular repair (TEVAR) of chronic type B aortic dissection. Eur J Vasc Endovasc Surg 2011;42:632-47.
Submitted Dec 23, 2015; accepted Mar 15, 2016.
DISCUSSION Dr Mark Farber (Chapel Hill, NC). I appreciate the talk that you gave. A few more questions: do you have any information in terms of differences as to whether you went to zone 3 or zone 4 in your thoracic coverage? Because your timeframe is about 5 years long (2009 to 2014) and somewhere in the middle of that most of us switched to bringing our endograft coverage for the dissections
down to within 5 cm of the celiac artery, which would put you more in zone 5, it would be nice to know whether that was a difference in any of your outcomes. And, the second question I have: you said there were no adverse events associated with doing these procedures, but you had two deaths that you threw out. Would those two death
10 Han et al
patients have anything to do with branch vessel occlusions and multisystem organ failure? Because if that’s the case, I don’t think that that would be a valid conclusion. Dr Sukgu M. Han. Dr Farber, thanks for your questions. The reason why we categorized the aortic coverage and the extent of dissection according to the aortic zones as opposed to the centimeters was in an attempt to answer that question: is there an association between the coverage of specific zone and adverse branch events? We found no such association. The only significant independent risk factor for adverse branch events was coverage of four or more aortic zones. We have not made changes in our approach regarding the length of coverage over time. Rather, the length of coverage was determined by each patient’s anatomy and indication. For
JOURNAL OF VASCULAR SURGERY --- 2016
example, patients with rupture or large aneurysmal degeneration were more commonly covered down to the celiac. However, after controlling for indications, extensive aortic coverage was the only risk factor for adverse branch events. Regarding the two deaths that were excluded, one patient died of rupture and the second patient died of bowel necrosis despite superior mesenteric artery stenting that was performed concurrently to thoracic endovascular aortic repair. However, I do not believe that exclusion of these patients invalidates our conclusion. Including the single patient would increase our adverse branch event total to 9 out of 304. The adverse branch events were still rare, and this particular event was due to the presenting pathology, not a result of thoracic endovascular aortic repair induced branch remodeling.
Remodeling of abdominal aortic branch perfusion after thoracic endovascular aortic repair for aortic dissections Sukgu M. Han, MD, Eric C. Kuo, MD, Karen Woo, MD, Ramsey Elsayed, MD, B. Sean Nguyen, BS, Sung W. Ham, MD, Vincent L. Rowe, MD, and Fred A. Weaver, MD, Los Angeles, Calif Objective: The fate of the abdominal aorta and its branches after thoracic endovascular aortic repair for aortic dissection (TEVAR-AD) has not been studied. The objective of this study was to describe the midterm changes in abdominal aortic branch perfusion after TEVAR-AD. Methods: A retrospective analysis of TEVAR-AD at a single institution from December 1, 2008, to March 31, 2015, was performed. Computed tomography angiography (CTA) images were reviewed to characterize the perfusion pattern changes of the celiac, superior mesenteric, inferior mesenteric, bilateral renal, and common iliac arteries. Risk factors associated with branch interventions were identified. Results: During the study period, 68 patients underwent TEVAR-AD, 46 of whom had pre-TEVAR and post-TEVAR CTA images available for review. For post-TEVAR CTA, the most recent scans were selected for analysis. The mean period between CTA studies was 371 days. Indications for TEVAR-AD were persistent pain (41%), malperfusion (15%), rupture (6%), and aneurysmal degeneration (33%). Twenty-five patients (54%) were treated during the acute phase (<14 days). All patients had dissections extending to the paravisceral aorta. Of the 304 abdominal aortic branches analyzed, 8 required intervention (2.6%). Branch events requiring intervention included malperfusion (two) and aneurysms involving the branches (three). No intervention was performed for one asymptomatic inferior mesenteric artery occlusion. Of the remaining 295 branches, changes in perfusion patterns were observed in 16 (5.4%). Twelve branches (75%) demonstrated an increased true lumen contribution to perfusion. Four branches (25%) had increased false lumen contribution, without clinical evidence of malperfusion. Patients requiring branch interventions were more likely to have severe chronic kidney disease (P [ .012) and more extensive aortic zone coverage during TEVAR (P [ .003). On multivariable Cox proportional hazards analysis, coverage of four or more zones during TEVAR-AD was associated with branch intervention (odds ratio, 6.44; 95% confidence interval, 1.01-40.8). The estimated intervention-free patency of the abdominal aortic branches was 89% at 5 years. Conclusions: Perfusion patterns of abdominal aortic branches remain largely stable after TEVAR-AD. The need for branch intervention is rare and associated with extensive aortic coverage. (J Vasc Surg 2016;-:1-10.)
Since its first reported use in aortic dissections in 1999, thoracic endovascular aortic repair (TEVAR) has become the first-line treatment modality for complicated type B aortic dissections.1,2 In uncomplicated cases, recent randomized trials have suggested that TEVAR may improve long-term aorta-specific survival and delay disease progression compared with medical management alone.3,4 These long-term benefits have been ascribed to the favorable aortic remodeling that results from TEVAR. “Favorable”
From the Division of Vascular Surgery and Endovascular Therapy, Keck Medical Center of University of Southern California. Author conflict of interest: none. Presented during a plenary session at the 2015 Vascular Annual Meeting of the Society for Vascular Surgery, Chicago, Ill, June 17-20, 2015. Correspondence: Sukgu M. Han, MD, Division of Vascular Surgery and Endovascular Therapy, University of Southern California, 1520 San Pablo St, Ste 4300, Los Angeles 90033, CA (e-mail: sukgu.han@med. usc.edu). The editors and reviewers of this article have no relevant financial relationships to disclose per the JVS policy that requires reviewers to decline review of any manuscript for which they may have a conflict of interest. 0741-5214 Copyright Ó 2016 by the Society for Vascular Surgery. Published by Elsevier Inc. http://dx.doi.org/10.1016/j.jvs.2016.03.441
or “positive” aortic remodeling includes true lumen expansion, false lumen regression, complete false lumen thrombosis, and stabilization of overall aortic diameter.3-8 Previous studies on aortic remodeling after TEVAR for aortic dissection (TEVAR-AD) have focused on the changes along the thoracic aorta.3-5,7-10 The data on the effects of TEVAR-AD on the abdominal aorta have been limited to diameter measurements at few selected abdominal aortic levels and the total volume measurements. However, little information, with the exception of occasional reports of branch stenting, is available on the fate of the abdominal aortic branches after TEVAR-AD when it is used in both acute and chronic phases.7,8,11-13 Therefore, the aim of our study was to assess changes in the perfusion patterns of the abdominal aortic branches after TEVARAD. Furthermore, we sought to identify risk factors associated with the need for branch interventions. METHODS Patients. A retrospective review was conducted to identify all patients undergoing TEVAR-AD at the Keck Medical Center of University of Southern California from December 1, 2008, through March 31, 2015. Clinical data were collected from hospital electronic medical records and 1
JOURNAL OF VASCULAR SURGERY --- 2016
2 Han et al
entered into a designated database using the Research Electronic Data Capture (REDCap) system. The were 118 variables, including patient demographics, medical comorbidities, symptoms on index admission, dissection chronicity, operative details, and postoperative course, collected. The study was approved by the Institutional Review Board of the University of Southern California, and patient consents were waived. The diagnosis of aortic dissection was made by computed tomography angiography (CTA). In addition to the de novo type B aortic dissections, patients with prior ascending aortic repair for type A dissections with residual dissection distal to the left subclavian were included. Degenerative aneurysms, penetrating aortic ulcers, and traumatic transections were excluded. Patients in whom the aortic dissection did not extend to the paravisceral aorta were excluded. Other exclusion criteria included patients who underwent total visceral debranching or open arch reconstruction in frozen elephant trunk configuration with antegrade deployment of TEVAR. Finally, patients who did not have post-TEVAR CTA images of the abdominal segments available for analysis were excluded. TEVAR procedure. Lumbar drains were placed preoperatively in nonemergent cases. Stent graft placement was performed through either open or percutaneous bilateral femoral access. For percutaneous procedures, two ProGlide percutaneous closure devices (Abbott Vascular, Abbott Park, Ill) were deployed perpendicular to each other, as described in preclose technique.14 Both intravascular ultrasound (IVUS) and transesophageal echocardiography were used in the majority of cases. IVUS was used for confirmation of true lumen cannulation, identification of proximal entry tear, and measurement of the proximal seal zone diameters. IVUS-measured proximal seal zone diameters were used to confirm stent graft sizing. Transesophageal echocardiography was used to monitor for development of retrograde dissection after stent graft deployment in addition to intraoperative cardiac function and fluid status. The operating surgeon’s preference determined the device choice. In all cases, the primary goal of TEVAR was to cover the proximal entry tear in the hope of achieving true lumen expansion. The decision to place additional distal stent grafts was individualized according to the existence of more distal septal tears within the thoracic aorta and was at the discretion of the operating surgeon. Definitions. Aortic dissections managed by TEVAR within 14 days from symptom onset were defined as acute. Subacute was defined as 15 to 90 days and chronic as >90 days from the initial symptom. The baseline extent of the dissection was assessed by the number of aortic zones involved, as defined by the standards established by the Society for Vascular Surgery Ad Hoc Committee on Reporting Standards for TEVAR.15 The total number of aortic zones involved was calculated for each patient and categorized into quartiles (Table I, A). The top quartile was defined as extensive aortic dissection (eight or more zones involved). Similarly, the number of aortic zones covered determined the extent of aortic coverage by
Table I. A, Extent of aortic dissection at baseline Percentile 0-24 25-74 75-99
No. of zones 0-5 6-7 8-11
Table I. B, Aortic coverage by thoracic endovascular aortic repair (TEVAR) Percentile 0-49 50-74 75-99
No. of zones 0-2 3 4-6
TEVAR. The top quartile of the aortic coverage by TEVAR was defined as extensive aortic coverage (four or more aortic zones covered; Table I, B). Branch vessel analysis. Axial cuts from CTA immediately before TEVAR were compared with the most recent post-TEVAR CTA. Celiac, superior mesenteric artery (SMA), inferior mesenteric artery (IMA), right renal artery, left renal artery, right common iliac artery, and left common iliac artery were analyzed. The perfusion pattern for each branch was characterized as being supplied by true lumen, false lumen, or both lumens (Fig 1). Remodeling in branch perfusion was determined by comparing the baseline and post-TEVAR CTA images. Remodeling was defined as positive when there was increased true lumen and loss of false lumen contribution to branch perfusion. For example, a branch whose perfusion pattern changed from both lumens to only true lumen after TEVAR was characterized as positive (Fig 2, A and B). Likewise, branches changed from only false lumen perfusion to both true and false lumens were also classified as positive (Fig 2, C and D). Conversely, remodeling was defined as negative when false lumen perfusion increased with loss of true lumen perfusion or branches with true or false lumen contributions to perfusion occluded after TEVAR. Statistical analysis. Clinical data were imported from REDCap to Stata 12 (Stata Corp, College Station, Tex). To analyze branch remodeling, a branch-specific data set was created with an individual branch as a unit of analysis. Univariable analyses were performed to identify clinical and operative risk factors associated with increased risk of branch intervention. Continuous variables were compared using Student t-test; categorical variables were compared using c2 or Fisher exact test. The intervention-free patency of abdominal aortic branches was estimated using Kaplan-Meier survival analysis. A Cox proportional hazards model was created for branch intervention by including variables with P values < .2 from univariable analyses. Both univariable
JOURNAL OF VASCULAR SURGERY Volume -, Number -
Han et al 3
Fig 1. Branch perfusion classification. Branches perfused by (A) true lumen, (B) false lumen, and (C) both lumens at baseline.
and multivariable analyses were performed using individual branches as a unit of analysis. RESULTS Sixty-four patients who underwent TEVAR-AD satisfied the described inclusion criteria. Two patients died postoperatively before CTA imaging. One died of a ruptured type B aortic dissection despite TEVAR. The other died of visceral malperfusion, which was manifested by abdominal pain and the associated CTA finding of the celiac artery and SMA being supplied entirely by an extremely compressed true lumen. TEVAR followed by immediate SMA stenting was performed. On postoperative day 2, the patient underwent exploration and was found to have necrosis of the entire small bowel. He died shortly thereafter. Of the remaining 62 patients, 16 patients did not have satisfactory post-TEVAR CTA of the chest, abdomen, and pelvis; 2 patients had a non-contrast-enhanced CT scan only because of chronic renal insufficiency, and 14 patients had a CTA scan of the chest only (Fig 3). This left a cohort of 46 patients with adequate post-TEVAR imaging for analysis. The 46 patients provided 304 branches for review and analysis.
Baseline characteristics of these 46 patients are shown in Table II. The majority of the patients were male with a history of hypertension. Other cardiovascular comorbidities, such as diabetes, coronary artery disease, congestive heart failure, and prior stroke, were rare. Approximately half were former or active smokers, whereas two carried the diagnosis of chronic obstructive pulmonary disease. Twenty-five patients underwent TEVAR for an acute aortic dissection. Pain (12/25 [48%]) and malperfusion (7/25 [28%]) were the major indications. Malperfused organs included bowel (four), lower extremity (two), and kidney (one). Two additional patients with acute dissections were asymptomatic but had high-risk features on CTA, namely, a proximal entry tear >1 cm and transaortic diameters >4 cm (Table III). TEVAR was performed in 21 patients, with a chronic aortic dissection with aneurysmal degeneration (13/21 [62%]) as the most common indication. Operative details are shown in Table IV. A variety of commercially available stent grafts were deployed, and in one third of cases, more than one stent graft was used. The aortic zones covered by TEVAR are shown in Fig 4. Three patients had zone 1 coverage after arch debranching, whereas 16 had left subclavian coverage
4 Han et al
JOURNAL OF VASCULAR SURGERY --- 2016
Fig 2. Examples of positive branch remodeling. Celiac artery (A) perfused by both lumens at baseline and (B) perfused by true lumen after thoracic endovascular aortic repair for aortic dissection (TEVAR-AD). Left renal artery (C) perfused by false lumen at baseline (arrow on the true lumen) and (D) perfused by both lumens after TEVAR-AD.
with revascularization in 11. In all cases, TEVAR was limited to the thoracic aorta. Zone 5 was covered in 32 cases (69%). There were no cases of distal bare-metal stent placement or adjunctive false lumen treatment. After TEVAR, there was a single case (2%) of spinal cord ischemia in a patient who presented with a 7-cm aneurysmal degeneration due to a chronic dissection, involving segments 4 to 7. One patient developed a loculated hemothorax after presenting with a rupture, requiring thoracoscopic evacuation. Two patients had access site complications, one brachial sheath hematoma and one femoral artery thrombosis after percutaneous TEVAR. One patient developed hand ischemia after left subclavian
TEVAR coverage that resolved after a left carotid to subclavian bypass. Two late deaths occurred, one following a grand mal seizure 6 weeks after TEVAR-AD and the other of an unknown cause 2 months after TEVAR-AD. Aortic morphologic findings and branch remodeling. The aortic zones involved by the dissection at baseline are illustrated in Fig 4. Two patients had residual dissection involving zone 2 after an arch repair of a type A aortic dissection. Zones 4 to 6 were involved in all patients. The majority of the patients had zone 3 to 9 involvement. Common iliac arteries were involved in half of the patients, whereas external iliacs were involved in a quarter of the patients.
JOURNAL OF VASCULAR SURGERY Volume -, Number -
Han et al 5
Fig 3. Patient selection. TEVAR, Thoracic endovascular aortic repair.
Table II. Baseline characteristics of patients Patients Age, years, mean 6 SD Male gender HTN DM CVA CAD CHF COPD Smoking ESRD Prior ascending aortic repair
Table III. Preoperative presentation 46 58.63 6 12.1 35 (76.1) 42 (91.3) 1 (2.2) 1 (2.2) 4 (8.7) 2 (4.3) 2 (4.3) 24 (52.2) 4 (8.7) 9 (19.6)
CAD, Coronary artery disease; CHF, congestive heart failure; COPD, chronic obstructive pulmonary disease; CVA, cerebrovascular accident; DM, diabetes mellitus; ESRD, end-stage renal disease; HTN, hypertension; SD, standard deviation. Data are presented as number (%) unless otherwise indicated.
The baseline branch perfusion pattern is shown in Fig 5. The majority of branches were perfused entirely by a true lumen (69% [211/304]). Among the abdominal aortic branches, the left renal artery perfusion was most frequently involved with dissection (43% by a false lumen or both lumens), whereas the SMA was most often perfused entirely by a true lumen (83%). Branch remodeling after TEVAR was characterized as positive in 12 branches: 3 celiac, 1 right renal, 4 left renal, 1 IMA, 2 right common iliac, and 1 left common iliac. Ten (83%) branches were in patients with acute aortic dissections. Negative branch remodeling was seen in five branches: SMA, right renal artery, IMA, and two left common iliac arteries (Fig 6). There were no clinical sequelae related to negative branch remodeling. On the KaplanMeier survival analysis, the estimated intervention-fee patency rates of abdominal aortic branches were 98% at 1 year and 89% at 5 years (Fig 7).
No. Pain Malperfusion Renal Visceral Lower extremities Aneurysmal degeneration Rupture Other
Acute (n ¼ 25), No. (%)
Chronic (n ¼ 21), No. (%)
25 12 (48) 7 (28) 1 4 2 2 (8) 2 (8) 2 (8)
21 7 (33)
13 (62) 1 (5) 0
Branch interventions. In the seven patients presenting with malperfusion, only two patients required branch interventions. In one patient with abdominal pain, the SMA was perfused entirely by the false lumen on preoperative CTA. Poor visualization of the SMA on post-TEVAR angiography resolved after SMA stent placement. The patient’s subsequent postoperative course was uncomplicated. In a second patient with lower extremity pain, weakness, and absent pulses, the left common iliac artery on preoperative CTA was perfused entirely by the false lumen. After a femoralfemoral bypass at the time of TEVAR, the symptoms resolved. In the remaining five patients with malperfusion (three SMA, one renal, one lower extremity), TEVAR resulted in an immediate true lumen expansion and improvement in branch perfusion with postoperative resolution of malperfusion. In four additional patients, six branch interventions were performed. In two, celiac stents were “prophylactically” placed because of poor visualization of the celiac artery on post-TEVAR angiography. Preoperatively,
JOURNAL OF VASCULAR SURGERY --- 2016
6 Han et al
Table IV. Thoracic endovascular aortic repair (TEVAR) details OR time, minutes Fluoroscopy time, minutes EBL, mL Transfusion required, % (No.) Preoperative lumbar drain, % (No.)
129.7 6 68.1 18.17 6 12.13 203 6 313.76 15.2 (7) 68.9 (31)
EBL, Estimated blood loss; OR, operating room.
Fig 5. Baseline branch perfusion pattern. IMA, Inferior mesenteric artery; L CIA, left common iliac artery; LRA, left renal artery; R CIA, right common iliac artery; RRA, right renal artery; SMA, superior mesenteric artery.
DISCUSSION
Fig 4. Baseline extent of aortic dissection and coverage by thoracic endovascular aortic repair for aortic dissection (TEVAR-AD).
neither patient had signs of visceral malperfusion. In a third patient, stenting of the right renal artery was performed for resistant hypertension 2 months after TEVAR. Poor renal artery perfusion due to significant false lumen flow was hypothesized as the cause of the hypertension. However, the hypertension persisted despite successful improvement of true lumen perfusion after stent placement. Finally, a fourth patient required endovascular aneurysm repair for aneurysmal degeneration of the dissected infrarenal aorta (one IMA coverage, two iliac stents) 3 years after TEVAR-AD. Thus, a total of six patients ultimately required branch interventions, with only two of these being associated with pre-TEVAR evidence of malperfusion. On univariable analyses, severe chronic kidney disease (P ¼ .012) and extensive aortic zone coverage (P ¼ .003) were associated with increased risk of needing branch intervention. There was a trend toward pain as the indication for repair (P ¼ .07) being protective. Chronicity or the extent of dissection at baseline was not associated with branch interventions (Table V). A Cox proportional hazards regression model was constructed including age, severe chronic kidney disease, indication for repair, extensive aortic coverage, and extensive aortic dissection. Among these variables, extensive aortic coverage by TEVAR-AD was the only independent factor associated with increased risk of branch interventions (hazard ratio, 6.44; 95% confidence interval, 1.01-40.8; Table VI).
The indications for TEVAR in patients with type B aortic dissections continue to expand. Poor long-term outcomes of patients with medically managed type B dissections combined with recent studies demonstrating favorable aortic remodeling after TEVAR-AD have challenged the traditional paradigm of medical management as the first-line therapy for acute uncomplicated type B aortic dissections.3-5,7-10,16,17 To date, studies on aortic remodeling after TEVAR-AD have focused on true and false lumen changes as measured by cross-sectional diameters at select aortic levels and total volumetric analysis.3-10,18,19 By sealing the proximal tear, TEVAR rechannels blood flow into the true lumen while depressurizing the false lumen. This hemodynamic alteration, in turn, raises a concern for the potential adverse impact of proximal TEVAR on the downstream aortic branches, particularly the ones supplied by the false lumen. Whereas technical aspects of TEVAR-AD, such as the optimal length of coverage and the role of pre-emptive left subclavian revascularization, continue to be a subject of debate,13,16,20-22 no study has specifically investigated the impact of TEVAR-AD on the branches of the abdominal aorta. Therefore, this study analyzed branch perfusion patterns at baseline and the changes after TEVAR-AD. Furthermore, factors associated with the need for branch intervention were evaluated, thereby potentially guiding clinicians in the decision-making process. Most importantly, we have demonstrated that the natural course of the abdominal branches after TEVAR-AD is largely benign. After TEVAR-AD, 92% (279/304) of the branches remained unchanged in their perfusion pattern. Of the remaining 25 branches, 8 (32%) had branch interventions. Furthermore, two celiac artery branch interventions occurred on the basis of the radiographic appearance of branch perfusion in the absence of clinical malperfusion, and one for renal perfusion was nontherapeutic. This low rate of branch interventions (3% [8/304]) is consistent with anecdotal reports of
JOURNAL OF VASCULAR SURGERY Volume -, Number -
Han et al 7
Fig 6. Thoracic endovascular aortic repair (TEVAR)-induced branch perfusion changes in (A) celiac artery, (B) superior mesenteric artery (SMA), (C) right renal artery (RRA), (D) left renal artery (LRA), (E) right common iliac artery (R CIA), (F) left common iliac artery (L CIA), and (G) inferior mesenteric artery (IMA).
Fig 7. Kaplan-Meier analysis of intervention-free patency of abdominal aortic branches after thoracic endovascular aortic repair for aortic dissection (TEVAR-AD).
secondary branch interventions included in the Study of Thoracic Aortic Type B Dissection Using Endoluminal Repair (STABLE) trial (two renal artery stents; N ¼ 40) and a series from Scali et al (four visceral, renal, and iliac stents; N ¼ 80).7,8,13 Therefore, loss of patency in the
downstream aortic branches as a consequence of proximal TEVAR in aortic dissections appears exceedingly rare. Preservation of abdominal branch patency after TEVAR-AD was also observed in patients with clinical malperfusion. Of the seven patients with clinical malperfusion who underwent TEVAR, only two (28%) required branch intervention. This is consistent with a recent series by Ryan et al, who reviewed 61 patients with malperfusion, 41% of whom required a branch intervention after TEVAR.11 Similarly, Alsac et al observed that 40% of the patients with visceral and renal malperfusion required branch stenting after TEVAR.12 TEVAR-AD appears to obviate adjunctive branch intervention by expeditiously increasing true lumen perfusion to the malperfused branches in a majority of patients. In the absence of clinical malperfusion, our data suggest that pre-emptive branch stenting may not be necessary. Furthermore, most of the branches, including those perfused partially or entirely by the false lumen, will remain patent and in some cases remodel positively. Of the branches that changed perfusion pattern after TEVAR-AD, negative remodeling did not result in adverse clinical events during the study period, with only one asymptomatic branch (IMA) thrombosis. In addition, positively remodeled branches were mostly seen in acute
JOURNAL OF VASCULAR SURGERY --- 2016
8 Han et al
Table V. Univariate analysis of risk factors associated with branch interventions
Age, years, mean 6 SD Male gender HTN DM CVA CAD CHF COPD Smoking Severe CKD (stage 4 or greater) Acute dissection Indication for repair Pain Malperfusion Aneurysmal degeneration Rupture Unknown No. of zones involved in dissection No. of zones covered
Nonevent (n ¼ 296)
Intervened (n ¼ 8)
58.3 6 11.8 226 (76) 282 (95) 4 (1.4) 7 (2.4) 28 (9.5) 14 (4.7) 14 (4.7) 151 (51) 36 (12)
58.4 6 12.1 8 (100) 8 (100) 0 0 0 0 0 2 (25) 4 (50)
.98 .12 .53
164 (55)
3 (38)
.32 .07 .07 .64 .07 .34
131 (44) 46 (15) 91 (31) 14 (4.7) 14 (4.7) 7.56 6 2.61
1 1 5 1
(13) (13) (63) (13) 0 8.38 6 2.26
2.86 6 0.94
3.88 6 1.13
P
.46 .012
.39 .003
CAD, Coronary artery disease; CHF, congestive heart failure; CKD, chronic kidney disease; COPD, chronic obstructive pulmonary disease; CVA, cerebrovascular accident; DM, diabetes mellitus; ESRD, end-stage renal disease; HTN, hypertension; SD, standard deviation. Data are presented as number (%) unless otherwise indicated.
Table VI. Cox proportional hazards analysis of risk factors associated with branch interventions
Age Severe CKD Pain as indication for repair Extensive aortic coverage (4 or more) Extensive aortic dissection (8 or more)
HR
95% CI
0.95 2.58 0.25 6.44 0.89
0.89-1.02 0.42-15.8 0.02-2.24 1.01-40.8 0.66-1.2
CI, Confidence interval; CKD, chronic kidney disease; HR, hazard ratio.
dissections (10/12 [83%]) that did not require extensive aortic coverage (11/12 [92%]). These were likely seen in patients who had robust true lumen expansion down to the abdominal aorta as well as into the involved branches after TEVAR. Future studies are needed to confirm this apparent relationship between the aortic and the branch perfusion remodeling after TEVAR-AD. The significant association between extensive aortic coverage and branch interventions may be explained by two different but not mutually exclusive possibilities. First, extensive aortic coverage may be a surrogate marker for a complex dissection process involving multiple distal fenestrations. It is possible that initial completion angiography showed persistent robust filling of the false lumen through the distal fenestrations, prompting the operating surgeon to place additional distal devices. In turn, the same underlying dissection process may have led to the failure of true
lumen expansion in the involved branches, thereby prompting branch interventions. Second, there may be the confounding component of operator decisionmaking. Only two branch interventions occurred in the clinical setting of malperfusion. Others were based on subjective interpretation of the angiographic appearance of branch perfusion. It is possible that more aggressive surgeons tended to cover more aortic zones and stented more branches. This scenario is further supported by our observation that those who underwent TEVAR-AD for pain as the indication were less likely to undergo branch interventions. Unlike in those who presented with clinical malperfusion, the operating surgeons are unlikely to have been compelled to perform pre-emptive branch intervention in this subset of patients with persistent pain. This highlights the lack of consensus recommendations regarding the optimal length of coverage and the threshold for secondary branch interventions.23-25 Despite the likely heterogeneous threshold for aortic stent graft extension, our group’s approach remained consistent regarding the primary goal of TEVAR-AD during the study period. The initial priority was coverage of the proximal entry tear with a minimum of 150-mm length stent graft. The decision to perform distal extension was typically based on the degree of persistent false lumen perfusion in the context of presenting symptoms. As such, extensive coverage was more often seen in aneurysmal degenerations and ruptures. However, after adjusting for the indication and extent of dissection, we found that the only independent factor for branch interventions was extensive coverage. Limitations of our study include those inherent to a single-institution, retrospective review with a relatively small sample size. It is possible that there was a selection bias toward favorable clinical outcomes by exclusion of 16 patients without adequate imaging. Particularly, exclusion of one patient who died of bowel necrosis after TEVAR-AD and SMA stenting limits generalizability of our study to only those who survive the initial ischemic insult from complicated aortic dissection. Challenges with following up this population of patients are not unique to our study. The International Registry of Acute Aortic Dissection follow-up rate of 70% at 2 years also reflects this challenge.4 Our study had a 24% (16/68) exclusion rate, based on stricter imaging criteria. The consecutive nature of our series may mitigate some of the selection bias. No correlation between remodeling of the aorta and abdominal branches was performed because of the limited availability of imaging studies that allow detailed anatomic analyses of the entire aorta as well as the branches. Longitudinal follow-up results with multiple surveillance imaging studies are needed to assess the long-term clinical significance of branch remodeling observed in our study. CONCLUSIONS After TEVAR-AD, abdominal branch perfusion remains largely stable with high midterm interventionfree patency. The need for branch interventions was rare
JOURNAL OF VASCULAR SURGERY Volume -, Number -
and associated with extensive aortic coverage. In the absence of clinical malperfusion, pre-emptive branch stenting may not be necessary. AUTHOR CONTRIBUTIONS Conception and design: SMH, SWH, FW Analysis and interpretation: SMH, EK, KW, FW Data collection: SMH, EK, RE, BN Writing the article: SMH, EK, RE Critical revision of the article: SMH, EK, KW, RE, BN, SWH, VL, FW Final approval of the article: SMH, EK, KW, RE, BN, SWH, VL, FW Statistical analysis: SMH, EK, KW Obtained funding: Not applicable Overall responsibility: SMH REFERENCES 1. Dake MD, Kato N, Mitchell RS, Semba CP, Razavi MK, Shimono T, et al. Endovascular stent-graft placement for the treatment of acute aortic dissection. N Engl J Med 1999;340:1546-52. 2. Nienaber CA, Fattori R, Lund G, Dieckmann C, Wolf W, von Kodolitsch Y, et al. Nonsurgical reconstruction of thoracic aortic dissection by stent-graft placement. N Engl J Med 1999;340:1539-45. 3. Nienaber CA, Kische S, Rousseau H, Eggebrecht H, Rehders TC, Kundt G, et al; INSTEAD-XL trial. Endovascular repair of type B aortic dissection: long-term results of the randomized investigation of stent grafts in aortic dissection trial. Circ Cardiovasc Interv 2013;6: 407-16. 4. Heijmen R, Fattori R, Thompson M, Dai-Do D, Eggebrecht H, Degrieck I, et al. Mid-term outcomes and aortic remodelling after thoracic endovascular repair for acute, subacute, and chronic aortic dissection: the VIRTUE Registry. Eur J Vasc Endovasc Surg 2014;48: 363-71. 5. Sigman MM, Palmer OP, Ham SW, Cunningham M, Weaver FA. Aortic morphologic findings after thoracic endovascular aortic repair for type B aortic dissection. JAMA Surg 2014;149:977-83. 6. Watanabe Y, Shimamura K, Yoshida T, Daimon T, Shirakawa Y, Torikai K, et al. Aortic remodeling as a prognostic factor for late aortic events after thoracic endovascular aortic repair in type B aortic dissection with patent false lumen. J Endovasc Ther 2014;21:517-25. 7. Lombardi JV, Cambria RP, Nienaber CA, Chiesa R, Teebken O, Lee A, et al. Prospective multicenter clinical trial (STABLE) on the endovascular treatment of complicated type B aortic dissection using a composite device design. J Vasc Surg 2012;55:629-40.e2. 8. Lombardi JV, Cambria RP, Nienaber CA, Chiesa R, Mossop P, Haulon S, et al. Aortic remodeling after endovascular treatment of complicated type B aortic dissection with the use of a composite device design. J Vasc Surg 2014;59:1544-54. 9. Kim KM, Donayre CE, Reynolds TS, Kopchok GE, Walot I, Chauvapun JP, et al. Aortic remodeling, volumetric analysis, and clinical outcomes of endoluminal exclusion of acute complicated type B thoracic aortic dissections. J Vasc Surg 2011;54:316-25. 10. Andacheh ID, Donayre C, Othman F, Walot I, Kopchok G, White R. Patient outcomes and thoracic aortic volume and morphologic changes
Han et al 9
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
following thoracic endovascular aortic repair in patients with complicated chronic type B aortic dissection. J Vasc Surg 2012;56:644-50. Ryan M, Vargas L, Mastracci T, Srivastava S, Eagleton M, Kelso R, et al. Progress in management of malperfusion syndrome from type B dissections. J Vasc Surg 2013;57:1283-90. Alsac JM, Girault A, El Batti S, Abou Rjilli M, Alomran F, Achouh P, et al. Experience with the Zenith Dissection Endovascular System in the emergency setting of malperfusion in acute type B dissections. J Vasc Surg 2014;59:645-50. Scali ST, Feezor RJ, Chang CK, Stone DH, Hess PJ, Martin TD, et al. Efficacy of thoracic endovascular stent repair for chronic type B aortic dissection with aneurysmal degeneration. J Vasc Surg 2013;58:10-7. Krajcer Z, Nelson P, Bianchi C, Rao V, Morasch M, Bacharach J. Percutaneous endovascular abdominal aortic aneurysm repair: methods and initial outcomes from the first prospective, multicenter trial. J Cardiovasc Surg (Torino) 2011;52:651-9. Fillinger MF, Greenberg RK, McKinsey JF, Chaikof EL. Reporting standards of thoracic aortic endovascular repair (TEVAR). J Vasc Surg 2010;52:1022-33. Lee CJ, Rodriguez HE, Kibbe MR, Malaisrie SC, Eskandari MK. Secondary interventions after elective thoracic endovascular aortic repair for degenerative aneurysms. J Vasc Surg 2013;57:1269-74. Kusagawa H, Shimono T, Ishida M, Suzuki T, Yasuda F, Yuasa U, et al. Changes in false lumen after transluminal stent-graft placement in aortic dissections: six years’ experience. Circulation 2005;111:2951-7. Jia X, Guo W, Li TX, Guan S, Yang RM, Liu XP, et al. The results of stent graft versus medication therapy for chronic type B dissection. J Vasc Surg 2013;57:406-14. Melissano G, Bertoglio L, Rinaldi E, Civilini E, Tshomba Y, Kahlberg A, et al. Volume changes in aortic true and false lumen after the “PETTICOAT” procedure for type B aortic dissection. J Vasc Surg 2012;55:641-51. Manning BJ, Dias N, Ohrlander T, Malina M, Sonesson B, Resch T, et al. Endovascular treatment for chronic type B dissection: limitations of short stent-grafts revealed at midterm follow-up. J Endovasc Ther 2009;16:590-7. Szeto WY, McGarvey M, Pochettino A, Moser GW, Hoboken A, Cornelius K, et al. Results of a new surgical paradigm: endovascular repair for acute complicated type B aortic dissection. Ann Thorac Surg 2008;86:87-93; discussion: 93-4. Flecher E, Cluzel P, Bonnet N, Aubert S, Gaubert A, Pavie A, et al. Endovascular treatment of descending aortic dissection (type B): shortand medium-term results. Arch Cardiovasc Dis 2008;101:94-9. Grabenwöger M, Alfonso F, Bachet J, Bonser R, Czerny M, Eggebrecht H, et al. Thoracic Endovascular Aortic Repair (TEVAR) for the treatment of aortic diseases: a position statement from the European Association for Cardio-Thoracic Surgery (EACTS) and the European Society of Cardiology (ESC), in collaboration with the European Association of Percutaneous Cardiovascular Interventions (EAPCI). Eur J Cardiothorac Surg 2012;42:17-24. Svensson LG, Kouchoukos NT, Miller DC, Bavaria JE, Coselli JS, Curi MA, et al. Expert consensus document on the treatment of descending thoracic aortic disease using endovascular stent-grafts. Ann Thorac Surg 2008;85:S1-41. Thrumurthy SG, Karthikesalingam A, Patternson BO, Hold PJ, Hinchliffe RJ, Loftus IM, et al. A systematic review of mid-term outcomes of thoracic endovascular repair (TEVAR) of chronic type B aortic dissection. Eur J Vasc Endovasc Surg 2011;42:632-47.
Submitted Dec 23, 2015; accepted Mar 15, 2016.
DISCUSSION Dr Mark Farber (Chapel Hill, NC). I appreciate the talk that you gave. A few more questions: do you have any information in terms of differences as to whether you went to zone 3 or zone 4 in your thoracic coverage? Because your timeframe is about 5 years long (2009 to 2014) and somewhere in the middle of that most of us switched to bringing our endograft coverage for the dissections
down to within 5 cm of the celiac artery, which would put you more in zone 5, it would be nice to know whether that was a difference in any of your outcomes. And, the second question I have: you said there were no adverse events associated with doing these procedures, but you had two deaths that you threw out. Would those two death
10 Han et al
patients have anything to do with branch vessel occlusions and multisystem organ failure? Because if that’s the case, I don’t think that that would be a valid conclusion. Dr Sukgu M. Han. Dr Farber, thanks for your questions. The reason why we categorized the aortic coverage and the extent of dissection according to the aortic zones as opposed to the centimeters was in an attempt to answer that question: is there an association between the coverage of specific zone and adverse branch events? We found no such association. The only significant independent risk factor for adverse branch events was coverage of four or more aortic zones. We have not made changes in our approach regarding the length of coverage over time. Rather, the length of coverage was determined by each patient’s anatomy and indication. For
JOURNAL OF VASCULAR SURGERY --- 2016
example, patients with rupture or large aneurysmal degeneration were more commonly covered down to the celiac. However, after controlling for indications, extensive aortic coverage was the only risk factor for adverse branch events. Regarding the two deaths that were excluded, one patient died of rupture and the second patient died of bowel necrosis despite superior mesenteric artery stenting that was performed concurrently to thoracic endovascular aortic repair. However, I do not believe that exclusion of these patients invalidates our conclusion. Including the single patient would increase our adverse branch event total to 9 out of 304. The adverse branch events were still rare, and this particular event was due to the presenting pathology, not a result of thoracic endovascular aortic repair induced branch remodeling.