If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Department of Vascular Surgery, University Medical Center Utrecht, Utrecht, the NetherlandsThoracic Aortic Research Center, Policlinico San Donato IRCCS, San Donato Milanese, ItalyDepartment of Cardiac Surgery, Frankel Cardiovascular Center, University of Michigan, Ann Arbor, Michigan
Department of Vascular Surgery, University Medical Center Utrecht, Utrecht, the NetherlandsDepartment of Cardiac Surgery, Frankel Cardiovascular Center, University of Michigan, Ann Arbor, Michigan
Address reprint requests to: Himanshu J. Patel, MD, Adult Cardiac Surgery, Joe D. Morris Collegiate Professor, Department of Cardiac Surgery, University of Michigan Hospitals, 1500 E. Medical Center Drive, 5144 Frankel Cardiovascular Center/SPC 5864, Ann Arbor, MI 48109-5864.
We performed a contemporary assessment of clinical and radiographic factors of stroke after thoracic endovascular aortic repair (TEVAR). Patients undergoing TEVAR from 2006 to 2017 were identified. We assessed clinical and radiographic data, including preoperative head and neck computed tomography, Doppler ultrasonography, and intraoperative angiography. Our primary outcome was stroke after TEVAR. Four hundred seventy-nine patients underwent TEVAR, mean age 68.1 ± 19.5 years, 52.6% male. Indications for TEVAR included aneurysms (n = 238, 49.7%) or dissections (n = 152, 31.7%). Ishimaru landing zones were Zone 2 (n = 225, 47.0%), Zone 3 (n = 151, 31.5%), or Zone 4 (n = 103, 21.5%). Stroke occurred in 3.8% (n = 18) of patients, with 1.9% (8) major events (modified Rankin Scale >3). Pathophysiology was predominantly embolic (n = 14), and occurred in posterior (n = 6), anterior (n = 6), or combined circulation (n = 4), and in the left hemisphere (n = 10) or bilateral (n = 6). Univariate analysis suggested use of lumbar drain (33.3% versus 57.2%, P = 0.04), inability to revascularize the left subclavian artery (16.7% vs 5.2%, P = 0.04) and number of implanted components (2.5 ± 1.2 vs 2.0 ± 0.97, P = 0.03) were associated with stroke. Multivariable analysis identified number of implanted components (OR 1.7, 95%CI 1.17–2.67 P = 0.00) and inability to revascularize the left subclavian artery as independent predictors of stroke. Stroke was associated with a higher perioperative mortality (27.8% vs 3.9%, P < 0.01). Stroke after TEVAR is primarily embolic in nature and related to both anatomic and procedural factors. This may have important implications for device development in the era of endovascular arch repair.
Stroke after TEVAR is primarily embolic in nature, associated with reduced survival and related to implantation of additional components and inability to revascularize the left subclavian artery.
Perspective Statement
This clinical and radiographic study contributes to better understanding of stroke after TEVAR. Realizing that stroke is primarily embolic in nature, and its risk factors relate to the location of underlying aortic pathology and the need for additional endografts, both careful patient selection and continuing improvement of endograft technology are necessary to reduce postoperative stroke after TEVAR.
INTRODUCTION
Thoracic endovascular aortic repair (TEVAR) has been increasingly utilized for the treatment of complex thoracic aortic pathology. One of the most serious complications after TEVAR is stroke, with reported rates in contemporary series between 3% and 10%.
Stroke pathophysiology is diverse and is thought to occur most often from embolic phenomena and vascular insufficiency. Prior studies have suggested risk factors to include treatment in the aortic arch with extensive atheroma,
With the increasing frequency of aortic intervention in the arch aorta and the advent of branched endograft technology, an evaluation of potential clinical, anatomic and radiographic factors important in occurrence of postoperative stroke is timely, warranted and the focus of this retrospective cohort study.
METHODS
Patient Population
We retrospectively reviewed hospital data for all patients from the University of Michigan Adult Cardiac Surgery database undergoing TEVAR in the distal arch or descending thoracic aorta for any pathology between 2006 and 2017, with follow-up extended until March 2020. This timeframe was chosen to ensure adequate and similar imaging quality and patient-information for all patients. Those patients who required Zone 0 and Zone 1 coverage and those who underwent total arch debranching or frozen elephant trunk procedures to facilitate an adequate landing zone for stent grafting of descending aortic disease were excluded. No exclusions were made based on pathology or urgency of the procedure.
Indications for TEVAR included aneurysms (fusiform or saccular), acute or chronic dissection, traumatic aortic injury and other indications. See Figure 1 for a flowchart of the patient selection. Procedures were characterized as elective, urgent or emergent
Figure 1Flow diagram of patient selection. (Color version of figure is available online.)
Preoperative imaging was performed with contrast enhanced computed tomography (CT) scans. Operative planning was done with 3-dimensional reconstructions obtained from these CT scans with adjunctive information via intravascular ultrasound where necessary for appropriate sizing and to assess feasibility of a TEVAR procedure. Proximal and distal landing zones were evaluated, as well as access vessels. Cerebrospinal fluid drainage was utilized when extensive coverage (>20 cm) was expected or in cases of prior infrarenal aortic repair, as described previously. Intraoperative endoleaks were identified during the intervention and treated accordingly. Implanted stent grafts included TAG or CTAG (W.L. Gore, Flagstaff, AZ), TX2 (Cook Medical Inc., Bloomington, IN) and Talent or Valiant (Medtronic Inc., Minneapolis, MN).
Radiologic Characteristics
We used both clinical data and a comprehensive evaluation of preoperative CT-scans, Doppler ultrasonography and intraoperative angiography. Information obtained included assessment of presence and degree of arch atheroma (≤ or > 5 mm) and extracranial and intracranial vasculature. We assessed atheroma by evaluating the CT-scan and reviewing the evaluation of a radiologist. Assessment of extracranial vasculature comprised of data from preoperative carotid Duplex studies, including peak systolic velocities and flow direction in bilateral vertebral and internal carotid arteries, and the presence of any significant carotid disease, defined as a stenosis of >49% as identified with Doppler ultrasonography. Assessment of intracranial vasculature (basilar, anterior cerebral, middle cerebral, posterior cerebral and cerebellar arteries) by preoperative CT angiography and preoperative CT angiography of extracranial vessels (supra-aortic branches, common carotid, internal carotid, and vertebral arteries). Intracranial vessel assessment included an evaluation of the circle of Willis by a dedicated neuroradiologist. Incompleteness of circle of Willis was defined by the absence of communicating anterior or posterior arteries and/or fetal origin of the posterior cerebral arteries. Furthermore, intracranial arteries, were evaluated for irregularities, including aneurysmal enlargement, stenosis and/or occlusion.
Outcomes
The primary outcome of this report was the occurrence of perioperative stroke, defined as a new focal or global neurologic deficit lasting more than 24 hours and occurring within 30 days after the operative procedure. Stroke was defined as embolic, hemorrhagic or secondary to hypoperfusion. The potential diagnosis was confirmed with either magnetic resonance imaging (MRI) or CT-imaging of the brain and assessment by a neurologist. Strokes were stratified on the basis of the modified Rankin score (mRS) with major (mRS > 3) and minor (mRS ≤ 3) classes. The mRS scale is assessed as follows: 0-No symptoms at all; (1) No significant disability despite symptoms, able to carry out all usual duties and activities; (2) Slight disability, unable to carry out all previous activities but able to look after own affairs without assistance; (3) Moderate disability, requiring some help but able to walk without assistance; (4) Moderately severe disability, unable to walk without assistance and unable to attend to own bodily needs without assistance; (5) Severe disability, bedridden, incontinent and requiring constant nursing care and attention; (6) Dead.
Secondary outcomes included perioperative mortality and other morbidity (transient neurologic deficit, renal failure, spinal cord ischemia, limb and visceral ischemia, and myocardial infarction), as well as incidence of late neurologic events (defined as stroke or transient neurologic deficit) and vital status. Early outcomes were defined as <30 days, late as occurring after 30 days. Follow-up was performed by determining the last outpatient visit to the University of Michigan Hospital. If a patient was deceased, the date of death was noted as last follow-up. If a patient was still alive, but did not have an outpatient or hospital visit, the last known date of visit was noted and these patients were censored in the survival analysis.
Statistical Analysis
Univariate analysis included use of Chi-square tests (and Fisher exact tests where applicable) to analyze categorical variables. Continuous variables were analyzed using Student's T-Test or Mann-Whitney U Test, when appropriate. To identify independent associations for early outcomes, a binary logistic regression was used. Variables with a P-value of less than 0.20 in univariate analysis were included in the model (see supplemental list). We used forward selection to make sure that the model was not overpowered. After 4 steps the regression was finished and included the following variables: Number of grafts implanted, LSA plugged or coiled, inability to revascularize LSA and use of lumbar drain. There was no multicollinearity between the variables used in the regression model. Survival analysis was performed by Kaplan-Meier curves, and log-rank tests when appropriate. Missing values were coded differently and were censored in time-to-event analysis. A P-value < 0.05 was considered significant. Data analysis was performed with the use of IBM SPSS Statistics for Windows, Version 25.0. (IBM Corp., Armonk, NY)
RESULTS
Patient Characteristics
The mean age of the entire cohort was 68.1 ± 19.5 years (52.6% male). Comorbidities were prevalent and included hypertension in 79.0% (n = 377), coronary artery disease in 29.9% (n = 143), as well as a prior history of stroke in 9.8% (n = 47). Demographics stratified by occurrence of stroke are listed in Table 1. Indications for TEVAR were fusiform aneurysm (n = 200, 41.6%) or saccular aneurysm (n = 38, 7.9%), acute (n = 98, 20.4%) or chronic (n = 54, 11.2%) dissection, traumatic aortic injury (n = 42, 8.7%) and other indications (n = 46, 9.6%). The majority of procedures were elective (61.7%, n = 295); urgent and emergent cases were performed in 38.3% (n = 184). Pathology involving the arch aorta was observed in 128 patients (26,8%). Inability to perform LSA bypass for Zone 2 TEVAR was not associated with increased stroke rates (25% vs 9.8%, P = 0.09). A lumbar drain was used in 56.3% (n = 268). Access for TEVAR was predominantly via standard femoral access (87.2%, n = 418). Adjunctive surgical procedures were performed in 15.7% (n = 75) of the patients, adjunctive endovascular procedures were performed in 19.6% (n = 93) and any LSA revascularization strategy was performed in 41.6% (n = 198) of patients. In addition, more devices were implanted in the patients with stroke (2.5 ± 1.2 vs 2.0 ± 0.97, P = 0.03). Intraoperative transfusion was more prevalent in patients who had a postoperative stroke (41.2% (n = 7) vs 19.6% (n = 89), P = 0.03) All procedural details are listed in Table 2.
Table 1Demographics and Patient History Stratified by Occurrence of Postoperative Stroke
Variable
Total
Stroke
No Stroke
P-Value
Total number of patients
479 (100)
18 (3.8)
461 (96.2)
Demographics
Age (mean)
68.1 ± 19.5
71.5 ± 13.9
68.0 ± 19.7
0.46
Male
252 (52.6)
12 (66.7)
240 (52.1)
0.22
Caucasian
389 (82.4)
15 (83.3)
374 (82.4)
0.91
BMI (mean)
28.0 ± 6.1
28.1 ± 7.2
28.0 ± 6.0
0.95
ASA I/II/III
199 (41.6)
5 (27.8)
194 (42.2)
0.22
Smoker current/former
353 (78.1)
11 (73.3)
342 (78.3)
0.65
Medical History
Chronic renal insufficiency
85 (17.7)
6 (33.3)
79 (17.1)
0.08
Connective tissue disease
3 (0.6)
0 (0.0)
3 (0.7)
>0.99
Diabetes
71 (14.9)
1 (5.6)
70 (15.3)
0.49
Hypertension
377 (79.0)
14 (77.8)
363 (79.1)
0.89
Hyperlipidemia
229 (48.1)
7 (38.9)
222 (48.5)
0.42
COPD
124 (26.1)
4 (22.2)
120 (26.2)
>0.99
PVD
80 (16.8)
2 (11.1)
78 (17.0)
0.75
CHF
32 (6.7)
0 (0.0)
32 (7.0)
0.62
Prior Stroke
47 (9.8)
3 (16.7)
44 (9.6)
0.40
CAD
143 (29.9)
4 (22.2)
139 (30.2)
0.60
Prior MI
48 (10.1)
1 (5.6)
47 (10.3)
>0.99
Prior aortic dissection
66 (13.8)
2 (11.1)
64 (13.9)
>0.99
Prior aortic aneurysm
165 (34.5)
6 (33.3)
159 (34.6)
0.91
Family history aortic disease
40 (8.7)
2 (11.8)
38 (8.6)
0.65
Surgical History
PCI
52 (10.9)
1 (5.6)
51 (11.1)
0.70
CABG
47 (9.8)
1 (5.6)
46 (10.0)
>0.99
Aortic valve surgery
55 (11.5)
4 (22.2)
51 (11.1)
0.14
Ascending aortic surgery
78 (16.3)
4 (22.2)
74 (16.1)
0.51
Arch surgery
55 (11.5)
4 (16.7)
52 (11.3)
0.45
Descending aortic surgery
34 (7.1)
0 (0.0)
34 (7.4)
0.63
Abdominal aortic surgery
77 (16.1)
3 (16.7)
74 (16.1)
>0.99
ASA, American Society of Anesthesiologists; BMI, body mass index; CABG, coronary artery bypass graft; CAD, coronary artery disease; CHF, congestive heart failure; COPD, chronic obstructive pulmonary disease; MI, myocardial infarction; LSA, left subclavian artery; PCI, percutaneous coronary intervention; PVD, peripheral vascular disease.
We identified a postoperative stroke rate of 3.8% (n = 18), with 10 (2.0%) characterized as a major stroke by modified Rankin criteria. Fourteen (2.9%) strokes occurred after TEVAR and 4 (2.0%) after the LSA revascularization performed in anticipation of TEVAR. The nature of stroke was predominantly embolic (n = 16), while one hemorrhagic and one hypoperfusion stroke were also observed. Affected vascular territories included the posterior (n = 7) anterior (n = 6), or combined circulation (n = 5) and all were either in the left hemisphere (n = 11) or bilateral (n = 7). As expected, postoperative stroke was associated with an increase in 30-day mortality (27.8% vs 3.9%, P < 0.01, Table 4).
Presentation of stroke was heterogeneous and included symptoms such as aphasia, extremity weakness and visual field deficits. All but one patient were fully independent with activities of daily living (ADL) prior to surgery. Of the 8 patients sustaining major strokes, 3 died during the index hospitalization, while the remaining 7 patients were discharged to rehabilitation facilities. In 2 patients, the functional status at discharge was difficult to assess due to multiple traumatic injuries. Three other patients were unable to walk independently and required help with ADL's. One patient had residual bladder dysfunction but showed improvement in upper and lower extremity function, while the eighth patient required full assistance in ADL's secondary to residual cognitive deficit. Full details of each stroke patient are listed in Table 3.
Table 3Functional Status and Outcomes of Patients With Postoperative Stroke
Functional Status Before
Type of Stroke
Symptoms During Hospital Stay
Discharge Location
Discharge Location and Vital Status
Latest Clinical Assessment Neurologic Status
Functional Status Follow-up
M 72y
Fully independent
Anterior, left MCA
Right facial drop, right sided hemitaxia
Acute rehab.
No left and right difference. Reduced balance, strength.
83 mo
Full recovery
F 95y
Fully independent
Posterior, left PCA
Lethargic, right facial droop.
Home
Ambulating independently, ongoing mental status issues
1.5 mo
Deceased after infection left groin
M 84y
Fully independent
Mixed, multiple distributions
Left facial droop, left UE and bilateral LE weakness, left UE numbness, and left Babinski.
Inpatient rehab.
Unable to sense the need to void. Improving strength in UE and LE.
1.4 mo
Due to paralysis and difficulty with transportation, opted follow-up in local hospital.
Two additional patients were identified that had suffered a transient ischemic attack. One patient, who underwent TEVAR for a distal arch and/or proximal descending aneurysm, demonstrated brief right upper extremity weakness that resolved within a day. The second patient with a prior history of type A dissection repair and treatment of a distal descending thoracic pseudoaneurysm presented with a short period of postoperative expressive aphasia. All other postoperative complications are listed in Table 4.
Table 4Postoperative and Follow-up Outcomes Stratified by Occurrence of Postoperative Stroke.
Carotid duplex studies were available in 369 cases (74.4%). Preoperative CT-angiography of the intra-cranial vasculature was available in 193 cases (38.9%) and preoperative CT-angiography of the extracranial vessels was available in 424 cases. Interestingly, no significant difference in preoperative CT or Duplex evaluation was identified between groups. In particular, neither the presence of any arch atheroma (50.0% versus 61.0%, P = 0.35), arch atheroma >5 mm (11.1% (n = 2) vs 8.1% (n = 35), P = 0.65), nor an incomplete circle of Willis was associated with stroke (12.5% (n = 1) vs 17.4% (n = 32), P > 0.99). Mobility of the arch atheroma was not assessed in this study. Similarly, Duplex peak systolic velocities in the extracranial vessels were similar between patients with and without postoperative stroke. However, postoperative flow in the left vertebral artery (LVA) was more frequently antegrade in patients without stroke (50.0% (n = 4) vs 82.4% (n = 108), P = 0.04). All anatomical and imaging outcomes are listed in Table 5.
Table 5CT-scan and Carotid Duplex Outcomes Stratified by Occurrence of Postoperative Stroke.
Stenosis of >49% as found on Duplex study. CT, computed tomography; ICA, internal carotid artery; LVA, left vertebral artery; PSV, peak systolic velocity; RVA, right vertebral artery.
61 (36.1)
2 (28.6)
59 (36.4)
0.55
PSV in LVA
50.8 ± 21.7
55.4 ± 20.4
50.7 ± 21.8
0.55
PSV in RVA
49.6 ± 20.1
49.5 ± 25.9
49.6 ± 20.0
0.54
PSV in right ICA
95.9 ± 41.5
78.1 ± 10.1
96.4 ± 42.0
0.25
PSV in left ICA
97.1 ± 36.3
96.7 ± 17.3
97.2 ± 36.7
0.97
Postoperative Duplex outcomes
Antegrade flow LVA
112 (80.6)
4 (50.0)
108 (82.4)
0.04
Antegrade flow RVA
126 (96.6)
4 (100)
122 (96.8)
>0.99
PSV in LVA
37.9 ± 55.9
22.8 ± 64.5
38.7 ± 55.6
0.49
PSV in RVA
62.4 ± 27.5
73.8 ± 31.2
61.9 ± 27.4
0.35
PSV in right ICA
96.7 ± 47.6
100.5 ± 49.2
96.6 ± 48.1
0.79
PSV in left ICA
97.8 ± 40.1
101.4 ± 22.7
97.6 ± 41.0
0.84
Stenosis of >49% as found on Duplex study.CT, computed tomography; ICA, internal carotid artery; LVA, left vertebral artery; PSV, peak systolic velocity; RVA, right vertebral artery.
A multivariable model (Hosmer and Lemeshow test P = 0.84, area under the curve: 0.77) identified number of implanted grafts (OR 1.73, 95% CI 1.16–2.59, P < 0.01), LSA plugged or coiled (OR 3.71, 95% CI 1.25–11.07 P = 0.02), use of lumbar drain (OR 3.12, 95% CI 1.11–8.76) and inability to revascularize LSA (OR 6.07, 95% CI 1.46–25.27, P = 0.01) as predictors of postoperative stroke (Table 6).
Table 6Analysis of Risk Factors Influencing Stroke.
The mean follow-up for the entire cohort was 52.6 ± 37.2 months. The overall 5-year survival was 75.6% (Fig. 2.). Five patients sustaining a postoperative stroke expired during follow-up. Additional causes of late mortality included sepsis in 2, pneumonia in 1 and unknown causes in 2. Full details regarding follow-up outcomes for each patient sustaining a stroke can be found in Table 3.
Figure 2Kaplan-Meier analysis of the entire cohort estimates a 5-year survival of 75.6 ± 2.2%. Mean survival time for the entire cohort was 60.4 ± 35.2 months. (Color version of figure is available online.)
The cumulative incidence of neurologic events at 5 years was 12.4%. These late neurologic events were embolic (n = 3), hemorrhagic (n = 1), or unknown (n = 2). Other late complications are listed in Table 4.
DISCUSSION
The introduction of thoracic endovascular aortic repair created a less invasive treatment paradigm for complex distal arch and descending aortic pathology with expected decreases in mortality and early morbidity rates as well as a reduction in duration of hospital stay.
While its long-term outcomes when compared to open surgery continue to be debated, its application has been broadly accepted and it is now considered as a first line therapy.
However, it is interesting to note that while this paradigm shift has consistently been associated with a reduced incidence of spinal cord ischemia, improvements have not been consistently seen in the rates of cerebrovascular events.
Indeed, recent United States Food and Drug Administration sponsored clinical trials comparing TEVAR with open surgery did not identify a reduction in stroke rates despite a higher frequency of distal thoracic aortic pathology in the TEVAR subset in one study.
With the evolution of TEVAR to treat more proximal aortic segments including the ascending and arch aorta, we chose to focus an evaluation of potential clinical, anatomic and radiographic characteristics that may predict the occurrence of stroke after TEVAR.
Our findings suggest that the occurrence of stroke is predicted by operative factors, namely the number of implanted components, inability to revascularize the LSA, use of a lumbar drain and performance of occlusive interventions on the LSA. Its origin was predominantly embolic in nature, equally distributed for anterior and posterior circulations and typically involving the left hemisphere of the brain. Importantly, specific anatomic features such as an incomplete Circle of Willis and the presence of radiographically evident atheromatous material were not identified as predictors of stroke in this study. Our findings also suggest that stroke occurrence portends a poorer prognosis, with a significant early (<18 month) impact on mortality. Liu et al. found in a meta-analysis there was no difference in the occurrence of stroke between open aortic repair and TEVAR.
Thoracic endovascular aortic repair versus open chest surgical repair for patients with type B aortic dissection: A systematic review and meta-analysis.
Finally, the observation that future neurological events occurred infrequently and not necessarily in patients who sustained a perioperative stroke was interesting and warrants further study.
Other studies have also shown that a more proximal landing zone is associated with increased stroke rates.
This finding along with our observation of the embolic predominance of its etiology and its positive relationship with implanted components leads us to the conclusion that extensive wire manipulation and device positioning may be the primary reason for stroke occurrence. It was therefore surprising that the presence of radiographically evident arch atheroma was not associated with occurrence of stroke in our study. When we subsequently re-evaluated this to stratify arch atheroma based on size (>5 mm), we again noted its lack of significant association with stroke. These findings may reflect the known frequency and presence of atherosclerotic disease in distal aortic pathology regardless of its identification on preoperative imaging, and therefore its potential to embolize during device deployment.
Furthermore, hemodynamic analysis suggests that wave reflections that occur at the aortic bifurcation can potentially carry atheromatous emboli retrograde in the arch aorta resulting in stroke.
Aortic stiffness determines diastolic blood flow reversal in the descending thoracic aorta: Potential implication for retrograde embolic stroke in hypertension.
It might also be expected that certain anatomical parameters, such as an incomplete Circle of Willis correlate with the occurrence of stroke. Mariscalco et al. proposed this as a predictor but the confounding factor in that study was the finding that patients with an incomplete circle of Willis also had the highest grade of arch atheroma. Furthermore, unintentional coverage of the left common carotid artery was performed in one out of 6 patients in that study, and one patient developed posterior stroke after LSA revascularization.
A potential explanation as to why we did not identify anatomical predictors could reflect our propensity to routinely revascularize the LSA when clinically able in instances of Zone 2 coverage. An important finding however, was that 22% (4 of 18 total) of the observed strokes occurred after LSA revascularization procedures and before endograft implantation. The rate of stroke was higher in the group where the LSA could not be revascularized due to clinical circumstances. Literature remains unclear on the reason for an increased risk of having a stroke in case of an inability to revascularize the LSA, one reason for that is the heterogeneity of study populations. It has been suggested that a reason for the increased risk of stroke occurred because of hemodynamic changes at the vertebral circulation. We have suggested this in our work where cerebrovascular hemodynamics were altered in the setting of arch vessel debranching.
Noting however in our study that the majority of strokes are embolic in etiology, it is possible that cerebrovascular atheroembolism is less frequent or less extensive when the arch vessel origin is more proximally located (ie, LSA bypass), thus rendering deployment in Ishimaru zone 2 different in stroke risk when no direct arch branch originates in this proximal landing zone (i.e. embolizing into a LSA residual stump). It is unknown whether a larger sample size would allow validation of this finding, as suggested in a recent meta-analysis and further research should be conducted to answer this question.
Editor's choice – incidence of stroke following thoracic endovascular aortic repair for descending aortic aneurysm: A systematic review of the literature with meta-analysis.
Chimney, snorkel and in-situ-fenestration techniques have been proposed to treat complex arch aortic pathologies, but have reported rates of mortality and stroke of up to 18%.
The recent advent of branched arch endograft technology appears promising, and its attendant neurologic risk is not yet fully understood. In a recent evaluation of a dual branched device for treatment into zone 0 in 38 patients, Haulon et al., described a neurologic event rate of 15.8% (n = 6). These were further stratified as transient ischemic attack in 4, ischemic stroke in 1, and hemorrhagic in 1.
All patients in this series required LSA revascularization. In another multicenter feasibility study of a single branch device used for aortic pathology in zone 2, Patel et al. reported no observed stroke in a cohort of 22 patients.
Branched endovascular therapy of the distal aortic arch: Preliminary results of the feasibility multicenter trial of the gore thoracic branch endoprosthesis.
This report has potential implications for design and application of this emerging technology, as well as operative technical adjustments to reduce the risk for stroke after TEVAR. Given the predominant embolic nature of perioperative stroke, consideration may need to be given for the potential of embolic protection devices to reduce its occurrence. Device modification to reduce interaction with the potentially atherosclerotic aortic wall itself may also be helpful. Finally, given the inherent neurologic risk of a revascularization procedure, branched technology itself may be important to reduce risk for stroke in the future, after carefully accessing the risks of branch vessel intervention.
Our study has several limitations. First, this is a retrospective study from a single institution, limited by sample bias and size. Furthermore, the number of CT scans and Duplex studies that were available were incomplete. Most importantly however, routine preoperative and postoperative neurologic examinations were not performed by a neurologist, which may have resulted in a lower overall reported stroke rate. The low overall stroke rate makes it difficult to use statistics to visualize the mortality after stroke in a plot. Given the small postoperative stroke rate it is difficult to make a strong prediction model for risk factors. As suggested by Messe et al., prospective evaluation by a neurologist resulted in higher identified stroke rates after cardiovascular procedures than if reported otherwise.
We believe that our reported frequency of major strokes is likely to be accurate but identification of less severe events could be under reported in the absence of routine pre- and postoperative neurologic assessment by a neurologist. Despite these limitations, we believe this report provides new information regarding risk stratification for perioperative stroke after TEVAR, and highlights the need for future investigation regarding its occurrence, both during the operative procedure and in late follow-up.
CONCLUSION
Stroke occurs infrequently after TEVAR but is associated with reduced early and late survival. In this clinical and radiographic evaluation, the finding that it is primarily embolic in nature, and its risk factors relate to the need for additional component implantation as well as inability to revascularize the LSA may have important implications for patient selection and device development for the evolving technology of branched endovascular arch repair.
Thoracic endovascular aortic repair versus open chest surgical repair for patients with type B aortic dissection: A systematic review and meta-analysis.
Aortic stiffness determines diastolic blood flow reversal in the descending thoracic aorta: Potential implication for retrograde embolic stroke in hypertension.
Editor's choice – incidence of stroke following thoracic endovascular aortic repair for descending aortic aneurysm: A systematic review of the literature with meta-analysis.
Branched endovascular therapy of the distal aortic arch: Preliminary results of the feasibility multicenter trial of the gore thoracic branch endoprosthesis.
Read at the American Association for Thoracic Surgery 97th Annual Meeting, Boston, Massachusetts, April 29 – May 3, 2017.
Funding: The Bob and Ann Aikens Aortic Program at the University of Michigan provided support for this research. Dr. Patel was generously supported by the David Hamilton Fund, the Phil Jenkins Breakthrough Fund in Cardiac Surgery and the Joe D. Morris Collegiate Professorship at the University of Michigan Frankel Cardiovascular Center.
This single-center retrospective study has been approved by the Institutional Review Board of the University of Michigan Medical School (HUM00111566).
This paper was presented at the AATS Centennial April 29- May 3, 2017, Boston, MA, USA
Conflicts of Interest: Dr. Patel has consulting relationship with W.L. Gore, Medtronic and Terumo. Dr. Williams discloses financial relationships with W.L. Gore and Boston Scientific. Dr. Trimarchi is a consultant for Medtronic and W.L. Gore. Dr. Eagle has a scientific Grant from W.L. Gore and Medtronic. Dr. Moll is a consultant for Medtronic and Philips Medical Systems. Dr. van Herwaarden is a proctor for Bolton Medical and Cook Medical and is a consultant for Philips Medical Systems. The other authors have no relevant disclosures to report.
Since its development in 1995, del Nido cardioplegia (DC) has become widely used in the congenital cardiac sphere and has subsequently been increasingly used in adult patients. The theory behind its composition is based on functional preservation with metabolic arrest by preventing cardiomyocyte swelling and contraction whilst inhibiting intracellular calcium accumulation, all while promoting anaerobic glycolysis and cardiomyocyte depolarization.1 These properties have motivated investigations regarding DC's efficacy in adult cardiac surgery including aortic valve replacements2,3 and coronary artery bypass grafts.
Stroke has long been a serious complication of catheterization procedures.1 The main cause is thought to be embolism, and even thoracic endovascular aortic repair (TEVAR), which has been developing remarkably recently, cannot be completely escaped. For the minimally invasive TEVAR, stroke can be called the Achilles tendon.2
Stenosis of >49% as found on Duplex study.CT, computed tomography; ICA, internal carotid artery; LVA, left vertebral artery; PSV, peak systolic velocity; RVA, right vertebral artery.
00002-8&id=gr3.jpg){kind=link}
00002-8&id=gr1.jpg){kind=link}
00002-8&id=gr2.jpg){kind=link}