Skip main navigation

Stroke and Cardiovascular Outcomes in Patients With Carotid Disease Undergoing Transcatheter Aortic Valve Replacement

Originally published Cardiovascular Interventions. 2018;11:e006322



    Stroke is a serious complication of both transcatheter aortic valve replacement (TAVR) and carotid artery disease (CD). The implications of CD in patients undergoing TAVR are unclear.

    Methods and Results—

    The Society of Thoracic Surgeons and American College of Cardiology Transcatheter Valve Therapies Registry, consisting of data from consecutive US TAVR cases during the years 2013 to 2015, was linked to Medicare claims data to ascertain 30-day and 1-year cumulative incidence rates of stroke and all-cause mortality. We compared 30-day and 1-year stroke and mortality outcomes between patients with no-CD and patients with moderate, severe, and occlusive CD and adjusted for baseline covariates using proportional hazards models. Among 29 143 patients undergoing TAVR across 390 US sites, 22% had CD. Patients with CD had higher rates of prior hypertension, diabetes mellitus, stroke, and myocardial infarction. Observed in-hospital stroke rates were 2.0% among no-CD, 2.5% among moderate CD, 3.0% among severe CD, and 2.6% among occlusive CD. There was no association between the presence of CD and 30-day stroke (adjusted hazard ratio, 1.16; 95% confidence interval, 0.94–1.43) or mortality (adjusted hazard ratio, 1.10; 95% confidence interval, 0.95–1.28). There was no association between CD and 1-year stroke (adjusted hazard ratio, 1.03; 95% confidence interval, 0.86–1.24) or mortality (adjusted hazard ratio, 1.02; 95% confidence interval, 0.93–1.12). Furthermore, there was no significant risk-adjusted association between severity of CD and 30-day or 1-year stroke or mortality.


    CD is common among TAVR patients, present in 1 of 5. CD was not associated with an increased risk of stroke or mortality at 30 day or 1 year. Post-TAVR stroke seems to be because of mechanisms other than CD.



    • Stroke is a serious complication of transcatheter aortic valve replacement; it is associated with reduced quality of life and higher mortality.

    • Despite the development of embolic protection devices, there is a residual risk of stroke which may be related to other pathogeneses of stroke, such as carotid artery disease.


    • Carotid artery disease is common among patients undergoing transcatheter aortic valve replacement present in 1 in 5 patients.

    • There was no association between presence of carotid artery disease and either 30-day or 1-year rates of stroke or mortality.

    • The lack of association between carotid artery disease and risk of stroke was consistent even after accounting for the severity of carotid stenosis.

    Transcatheter aortic valve replacement (TAVR) has transformed the management of severe aortic stenosis; hence, TAVR volumes have increased dramatically during the past decade.15 Stroke is a serious complication of TAVR procedures and is often associated with reduced quality of life measures and survival.6,7 The biological mechanisms underpinning periprocedural stroke in TAVR patients are incompletely understood. A proposed mechanism includes periprocedural debris embolization.8 However, despite the use of a dual-filter device in the SENTINEL trial, the 30-day stroke rates were still 5.6% in the device arm versus 9.1% in the control group.9 These findings suggest either current technologies do not sufficiently reduce the risk of periprocedural debris embolization or there are other mechanisms for TAVR-associated stroke.

    See Editorial by Lansky et al

    A potential alternative cause for TAVR-associated stroke is carotid artery disease (CD). The presence of CD has been associated with an increased risk of stroke in patients undergoing surgical aortic valve replacement.10,11 Yet, there are few analyses exploring the relationship between CD and stroke among TAVR patients. Therefore, our aim was to use data from a national registry to (1) evaluate the national prevalence of CD stratified by severity among patients undergoing TAVR; (2) assess the rates of cardiovascular outcomes stratified by severity of CD; and (3) examine whether there is an association of CD with 30-day and 1-year stroke and mortality outcomes.


    Society of Thoracic Surgeons (STS) and American College of Cardiology (ACC) Transcatheter Valve Therapies (TVT) Registry

    The data, analytic methods, and study materials will be made available to other researchers for purposes of reproducing the results or replicating the procedure. The STS/ACC TVT Registry is a collaborative registry program that collects clinical data from consecutive TAVR cases across all US TAVR sites. Centers participating in the registry gather clinical data using standardized definitions.12,13 The National Cardiovascular Data Registry warehouse and Duke Clinical Research Institute have performed data quality checks to optimize data accuracy.7 The ACC has designated Chesapeake Research Review Incorporated as its internal review board; TVT has submitted a protocol to this institutional review board, and it has been granted a waiver of informed consent.

    Definition of CD Severity

    The TVT data collection form provided CD status and stenosis severity by the following classification scheme: no CD (≤50 stenosis), moderate (50%–79%), severe (80%–99%), occlusive (100%). The stenosis designation is based on the best estimate of the highest value of stenosis recorded between birth and the procedure. The registry allows determination of CD by any diagnostic test. We allocated patients with differing stenoses in their right and left carotid arteries to the higher stenosis group. Carotid revascularization was defined as patients with a history of previous carotid artery surgery or stenting using current procedural terminology codes: 35301, 37215, and 37216.

    Study Cohort

    The study period was from October 2013 to September 2015. The initial cohort included 36 340 patients across 391 US centers. Patients with missing data on CD (n=5864) and patients with missing data on CD severity (n=1333) were excluded, yielding an analysis population of 29 143 TAVR patients from 390 sites. To ascertain 30-day and 1-year outcomes, the Centers for Medicare and Medicaid Services (CMS) linked the registry to CMS claims data through direct patient identifiers (name and social security number). Eligible patients for this linked sample included individuals aged ≥65 years with Medicare Part A or B fee-for-service insurance. Patients were excluded from the final linked sample because of ineligibility (n=3729) and an unsuccessful linkage process (n=5451). The final sample of patients with successful linkage to CMS data for 30-day and 1-year outcomes was 19 963 patients from 387 sites (Figure 1). There were no clinically significant differences in baseline characteristics between patients in the STS/ACC TVT Registry who were successfully linked and those who were unable to be linked (Data Supplement I).

    Figure 1.

    Figure 1. Patient flow diagram. CD indicates carotid artery disease; CMS, Centers for Medicare and Medicaid Services; TAVR, transcatheter aortic valve replacement; and TVT, transcatheter valve therapy.

    End Point Definitions

    All in-hospital events were derived from TVT Registry data; the data definitions were harmonized with Valve Academic Research Consortium 2 criteria.14 Postdischarge events were ascertained using International Classifications for Diseases, Ninth Revision codes (Data Supplement II). The primary end point for this analysis was the 1-year incidence of stroke. There were several secondary end points, including 30-day incidence of stroke, 30-day and 1-year incidence of all-cause mortality, 30-day and 1-year incidence of the composite of stroke and all-cause mortality, and 30-day and 1-year incidence of myocardial infarction (MI) and bleeding. Censoring for nonfatal end points was performed at the earliest of the following events: at 30 days for the 30-day end point and at 1 year for the 1-year end point after TAVR, the end of CMS follow-up (October 1, 2015), or the end of Medicare Part A and B fee-for-service eligibility.

    Statistical Analysis

    We stratified baseline features and procedural characteristics by CD versus no-CD and by CD severity: none, moderate, severe, and occlusive. Continuous variables were reported as medians with 25th and 75th percentiles, and categorical variables were presented as frequencies with percentages. Carotid severity was considered ordinal, and thus the association between carotid severity and patient characteristics was tested through the Spearman correlation coefficient for continuous variables and the Wilcoxon rank-sum test or Kruskal–Wallis test for categorical variables. We compared the cumulative incidence of 30-day and 1-year outcomes post-TAVR stratified by varying degrees of carotid severity using Gray test.15 For nonfatal outcomes, death was treated as a competing risk. We adjusted for baseline covariates with Cox proportional hazards models for mortality and composite outcomes. We used Fine and Gray proportional subdistribution hazards models for stroke.16 The covariates (listed in Data Supplement III) were identified from the validated risk prediction model for in-hospital mortality after TAVR.17 We generated 2 additional sets of sensitivity models: (1) to assess possible collinearity between CD and peripheral artery disease (PAD) and (2) to account for the possible influence of discharge medications on 30-day or 1-year stroke and mortality outcomes. Both CD and PAD are caused by atherosclerosis that affects arteries in various parts of the body; thus, it might be redundant to have them both in the model if they carry the same information in our models. To examine such a collinearity between CD and PAD, we removed PAD from the original multivariable models to determine whether the association between CD and the outcomes would be affected by the exclusion of PAD. Furthermore, to assess the impact of discharge medications, we included discharge medications in the model and changed the starting date for event ascertainment accordingly from the TAVR procedure date to the discharge date instead.

    All covariates had a missing rate of <2%, so missing values were imputed to the median for continuous variables and the mode for categorical variables. Clustering of patients within sites was accounted for by the robust variance estimation method. All analyses were performed using SAS, version 9.4 (SAS Institute, Inc, Cary, NC).


    Study Cohort

    From October 2013 to September 2015, 29 143 patients underwent TAVR, of which 6410 patients (22%) had CD. When stratified by CD severity, 5001 (17.2%) were moderate CD, 940 (3.2%) were severe CD, and 469 (1.6%) were occlusive CD. Baseline features and characteristics are presented in Table 1. Patients with CD were slightly younger and more likely to be male and white. Furthermore, CD patients had a higher prevalence of prior transient ischemic attack, stroke, diabetes mellitus, MI, and PAD. CD patients had higher STS risk scores, more frequent hostile chest, and prior open heart surgery compared with no-CD patients. Regarding procedural features, the transfemoral approach was used less frequently in patients with CD as opposed to patients without CD. The transcarotid approach was used infrequently (<1%) in both groups.

    Table 1. Baseline Characteristics Stratified by Severity of CD

    No CD (N=22 733)CD (N=6410)P Value*Moderate CD (N=5001)Severe CD (N=940)Occlusive CD (N=469)P Value
     Age, y83 (77–88)82 (77–87)<0.00183 (77–87)82 (76–86)81 (75–86)<0.001
     Female49.0% (11 126)43.8% (2806)<0.00146.4% (2317)38.2% (359)27.8% (130)<0.001
     White94.4% (21 316)96.4% (6142)<0.00196.1% (4775)97.4% (913)97.2% (454)<0.001
     TIA8.0% (1804)12.6% (804)<0.00111.6% (580)14.4% (135)19.0% (89)<0.001
     Stroke11.1% (2515)17.1% (1093)<0.00115.6% (780)18.2% (171)30.3% (142)<0.001
     Diabetes mellitus36.7% (8337)41.4% (2650)<0.00141.2% (2057)40.3% (378)45.8% (215)<0.001
     Hypertension88.9% (20 201)92.8% (5943)<0.00192.9% (4642)93.1% (875)90.8% (426)<0.001
     MI23.3% (5274)32.0% (2047)<0.00130.8% (1535)37.2% (350)34.6% (162)<0.001
     PAD25.1% (5698)48.7% (3112)<0.00146.5% (2322)55.2% (519)59.9% (281)<0.001
     Atrial fibrillation/flutter41.5% (9430)39.3% (2517)0.00138.9% (1941)40.5% (381)41.6% (195)0.002
    Surgical features
     STS risk score6.3 (4.1–9.6)7.1 (4.7–11.0)<0.0017.1 (4.7–11.1)7.2 (4.6–10.9)6.7 (4.3–10.3)<0.001
     Hostile chest7.3% (1654)9.4% (600)<0.0019.1% (452)9.6% (90)12.4% (58)<0.001
     Porcelain aorta5.2% (1185)9.0% (578)<0.0019.3% (463)7.14 (67)10.2% (48)<0.001
     Prior open heart surgery28.7% (6501)42.0% (2689)<0.00140.5% (2020)48.6% (457)45.3% (212)<0.001
     Access site<0.001<0.001
     Femoral access78.6% (17 765)64.8% (4129)66.8% (3324)57.3% (536)58.0% (269)
     Transapical access13.1% (2063)21.6% (1379)20.2% (1007)27.2% (254)25.4% (118)
     Transcarotid access0.2% (39)0.3% (17)0.2% (12)0.4% (4)0.2% (1)
    Discharge medications
     Aspirin86.4% (18 779)88.4% (5347)<0.00188.3% (4175)88.9% (785)87.4% (387)<0.001
     P2Y12 inhibitor65.1% (14 150)69.7% (4217)<0.00169.8% (3298)71.4% (630)65.2% (289)<0.001
     Warfarin24.5% (5328)22.8% (1381)0.00622.3% (1052)23.4% (207)27.5% (122)<0.013
     Dabigatran1.1% (236)0.8% (48)0.0450.8% (37)0.7% (6)1.1% (5)0.049
     Factor Xa inhibitor5.1% (1099)4.1% (250)0.0033.9% (184)5.1% (45)4.7% (21)0.005

    CD indicates carotid artery disease; MI, myocardial infarction; PAD, peripheral artery disease; STS, Society of Thoracic Surgeons; and TIA, transient ischemic attack.

    *P value for no CD vs CD.

    P value for no CD vs moderate CD vs severe CD vs occlusive CD.

    Presented as median (25th percentile, 75th percentile).

    Observed In-Hospital Outcomes

    The in-hospital stroke rates were 2.0% in no-CD patients, 2.5% in moderate CD patients, 3.0% in severe CD patients, and 2.6% in occlusive CD patients. In-hospital mortality rates were 3.4% in no-CD, 4.2% in moderate CD, 4.8% in severe CD, and 4.9% in occlusive CD patients. Rates of major or life-threatening bleeding increased based on the severity of CD (Table 2).

    Table 2. Observed In-Hospital Outcomes Stratified by Severity of CD

    OutcomesNo CD (N=22 733)CD (N=6410)P Value*Moderate CD (N=5001)Severe CD (N=940)Occlusive CD (N=469)P Value
    Stroke2.0% (452)2.6% (167)0.0032.5% (127)3.0% (28)2.6% (12)0.002
    TIA0.2 (38)0.3% (19)0.0390.2% (11)0.8% (7)0.2% (1)0.022
    Death3.4% (784)4.4% (279)0.0014.2% (211)4.8% (45)4.9% (23)<0.001
    MI0.4% (82)0.5% (34)0.0570.5% (26)0.3 (3)1.1% (5)0.052
    VARC major or life-threatening bleeding7.2% (1625)8.7% (552)<0.0018.4% (417)9.1% (85)10.7% (50)<0.001
    Major access site complication1.3% (287)1.2% (75)0.5521.1% (55)1.2% (11)1.9% (9)0.632

    CD indicates carotid artery disease; MI, myocardial infarction; TIA, transient ischemic attack; and VARC, Valve Academic Research Consortium.

    *P value for no CD vs CD.

    P value for no CD vs moderate CD vs severe CD vs occlusive CD.

    Unadjusted 30-Day and 1-Year Outcomes by Presence of CD

    The 30-day and 1-year unadjusted outcomes are depicted in Table 3. The 30-day cumulative incidence of stroke was lower for patients with no CD 2.4% when compared with patients with CD 3.1% (P=0.01). Alternatively, the 1-year rate of stroke was not different between patients with no CD 4.1% when compared with patients with CD 4.5% (P=0.16). The 30-day mortality rate was lower for no-CD patients 4.9% when compared with CD patients 6.1% (P<0.01). At 1 year, mortality remained lower for the no-CD cohort 19.9% when compared with the CD group 21.5% (P<0.01). We also examined clinically relevant secondary outcomes, such as MI and bleeding. The 1-year unadjusted rates were lower among no-CD patients when compared with CD patients for both MI (1.9% versus 2.8%) and any bleeding (22.4% versus 24.6%; Table 3). The unadjusted event rates of any bleeding were high in the periprocedural period and continued to rise in the subsequent follow-up year (Figure 2B).

    Table 3. Unadjusted Cumulative Incidence of 30-Day and 1-Year Outcomes Stratified by Severity of CD

    OutcomesNo CDCDP ValueModerate CDSevere CDOcclusive CDP Value*
    N % (95% CI)N% (95% CI)N% (95% CI)N% (95% CI)N% (95% CI)
    30 d
     Stroke3662.4 (2.2–2.7)1363.1 (2.6–3.7)0.0111023.0 (2.5–3.6)243.8 (2.6–5.7)103.3 (1.8–6.0)0.048
     Mortality7404.9 (4.6–5.2)2686.1 (5.4–6.9)0.0012146.2 (5.5–7.1)355.5 (4.0–7.6)196.2 (4.0–9.6)0.011
     Composite of mortality or stroke10277.0 (6.6–7.5)3708.8 (8.0–9.7)<0.0012908.8 (7.9–9.8)548.8 (6.8–11.4)268.8 (6.1–12.8)0.002
     Myocardial infarction860.6 (0.5–0.7)400.9 (0.7–1.3)0.011320.9 (0.7–1.3)50.8 (0.3–2.0)31.1 (0.3–3.3)0.085
     Any bleeding167411.2 (10.7–11.8)57413.4 (12.4–14.4)<0.00144313.1 (12.0–14.3)8313.4 (10.9–16.3)4816.2 (12.4–21.0)<0.001
    1 y
     Stroke5314.1 (3.8–4.5)1714.5 (3.9–5.3)0.1551284.3 (3.6–5.1)305.5 (3.8–8.0)134.8 (2.8–8.4)0.324
     Mortality235919.9 (19.1–20.6)74921.5 (20.1–23.0)0.00257421.3 (19.7–23.0)11721.8 (18.4–25.7)5823.9 (18.8–30.5)0.015
     Composite of mortality or stroke254222.6 (21.8–23.5)79924.5 (23.0–26.2)0.00161224.2 (22.5–26.1)12725.4 (21.7–29.8)6026.1 (20.6–33.0)0.009
     Myocardial infarction2041.9 (1.6–2.2)892.8 (2.3–3.5)<0.001672.6 (2.1–3.4)153.4 (2.0–5.8)73.2 (1.5–6.7)0.005
     Any bleeding279422.4 (21.7–23.2)89324.6 (23.2–26.2)<0.00168824.3 (22.7–26.1)13224.5 (21.0–28.7)7328.4 (23.2–34.9)0.002

    CD indicates carotid artery disease; and CI, confidence interval.

    *P value for no CD vs moderate CD vs severe CD vs occlusive CD.

    Figure 2.

    Figure 2. The cumulative incidence of stroke, mortality, and the composite of stroke and mortality during a 1-year period stratified by severity of carotid artery disease (CD). Mod indicates moderate; Occ, occlusive; Sev, severe; and TAVR, transcatheter aortic valve replacement.

    Unadjusted 30-Day and 1-Year Outcomes by Severity of CD

    There was no ascending trend in stroke rates based on the increasing degree of stenosis (Table 3). Similarly, there was no increase in 30-day unadjusted rates of the composite of mortality or stroke associated with increasing CD stenosis. Alternatively, the rates of 1-year mortality or stroke increased with greater CD severity. The 1-year unadjusted rate of mortality or stroke was 22.6% among no CD patients, 24.2% among moderate CD, 25.4% among severe CD, and 26.1% with occlusive CD (P=0.009; Table 3).

    The stroke curves across the CD severity spectrum demonstrate a high proportion of stroke events in the early period post-TAVR with a relative plateauing of events up to 1 year (Figure 3A). In contrast, mortality rates across the CD severity spectrum steadily rise throughout the first year (Figure 3B). Among all the patients on the CD severity spectrum, patients with bilateral severe CD could be at the highest risk for stroke and mortality. Yet, there was no difference in the 1-year composite end point of stroke/mortality between patients with bilateral severe stenosis 20.7% (95% confidence interval [CI], 12.9%–33.1%) when compared with no-CD patients 22.6% (95% CI, 21.8%–23.5%; P=0.91). There were higher bleeding rates associated with an increasing degree of CD stenosis when compared with no-CD (Figure 2B).

    Figure 3.

    Figure 3. The cumulative incidence of myocardial infarction and any bleeding during a 1-year period stratified by severity of carotid artery disease (CD). Mod indicates moderate; Occ, occlusive; Sev, severe; and TAVR, transcatheter aortic valve replacement.

    Adjusted 30-Day and 1-Year Primary End Points

    Patients with CD had higher observed cumulative incidence rates for stroke and mortality at 30 days and 1 year. However, after adjustment for patient characteristics, these observed differences were no longer significant. Among CD patients and no-CD patients, there was no difference in the risk of 30-day stroke (adjusted hazard ratio, 1.16; 95% CI, 0.94–1.43) or mortality (adjusted hazard ratio, 1.10; 95% CI, 0.95–1.28; Figure 4). There was no association between presence of CD and 1-year risk of stroke (adjusted hazard ratio, 1.03; 95% CI, 0.86–1.24) or mortality (adjusted hazard ratio, 1.02; 95% CI, 0.93–1.12). Even after incorporating the degree of stenosis, there was no significant adjusted association between CD severity and 30-day or 1-year risks of stroke or mortality (Table 4).

    Table 4. Adjusted Associations of CD Severity With 30-Day and 1-Year Cardiovascular Outcomes

    OutcomePredictorAdjusted* HR (95% CI)P Value
    30 d
     Composite of stroke or mortalityModerate vs no carotid disease1.14 (0.99–1.30)0.069
    Severe vs no carotid disease1.11 (0.83–1.48)0.487
    Occlusive vs no carotid disease1.10 (0.73–1.65)0.644
     StrokeModerate vs no carotid disease1.12 (0.89–1.41)0.351
    Severe vs no carotid disease1.39 (0.88–2.19)0.159
    Occlusive vs no carotid disease1.14 (0.60–2.15)0.695
     MortalityModerate vs no carotid disease1.13 (0.97–1.32)0.123
    Severe vs no carotid disease0.94 (0.67–1.32)0.731
    Occlusive vs no carotid disease1.11 (0.68–1.81)0.682
    1 y
     Composite of stroke or mortalityModerate vs no carotid disease1.01 (0.92–1.11)0.838
    Severe vs no carotid disease1.10 (0.91–1.34)0.313
    Occlusive vs no carotid disease1.07 (0.82–1.39)0.635
     StrokeModerate vs no carotid disease1.00 (0.82–1.22)0.982
    Severe vs no carotid disease1.22 (0.82–1.82)0.332
    Occlusive vs no carotid disease0.99 (0.58–1.71)0.979
     MortalityModerate vs no carotid disease1.01 (0.91–1.11)0.868
    Severe vs no carotid disease1.05 (0.86–1.28)0.630
    Occlusive vs no carotid disease1.14 (0.87–1.48)0.346

    CD indicates carotid artery disease; CI, confidence interval; and HR, hazard ratio.

    *Adjusted for the variables enumerated in Data Supplement III.

    Figure 4.

    Figure 4. Forest plot depicting the adjusted association of carotid artery disease (CD) with 30-day and 1-year risk of stroke and mortality. CI indicates confidence interval; and HR, hazard ratio.

    Sensitivity Analyses

    Among patients with any CD, 48.7% also had evidence of PAD. There was no marked change in the relationship between CD and outcomes with or without adjustment for PAD, except for the 30-day composite end point of mortality and stroke where the P value went from 0.069 to 0.041 (Data Supplement IV). We created a new model to include discharge antiplatelet and anticoagulation medications. After adjusting for these medications, there was no significant relationship between CD and 30-day or 1-year stroke and mortality outcomes (Data Supplement V).

    Carotid Revascularization

    The overall rate of carotid revascularization within 90 days before or after the index-TAVR procedure was low at 0.6% (n=114). However, the rate of prior revascularization was higher among patients with severe CD at 10.8% (n=63). Table 5 describes the outcomes of patients with severe CD with and without prior revascularization. The 30-day rates of stroke were similar for patients with prior revascularization 3.2% (95% CI, 0.8–12.8) when compared with patients without prior revascularization 3.9% (95% CI, 2.5–6.0).

    Table 5. Cumulative Incidence of Adverse Outcomes in Patients With or Without Prior Carotid Revascularization

    Severe CD With Prior Carotid Revascularization (N=63)Severe CD Without Prior Carotid Revascularization (N=518)
    N% (95% CI)N% (95% CI)
    30-d stroke23.2 (0.8–12.8)203.9 (2.5–6.0)
    30-d mortality00305.9 (4.2–8.4)
    30-d composite of mortality or stroke23.2 (0.8–12.8)469.1 (6.9–12.0)
    1-y stroke37.7 (2.2–27.1)255.3 (3.6–7.9)
    1-y mortality814.6 (7.6–28.2)10023.8 (19.9–28.5)
    1-y composite of mortality or stroke1020.0 (10.7–37.4)11126.0 (21.9–30.8)

    CD indicates carotid artery disease; and CI, confidence interval.


    This analysis from the STS/ACC TVT Registry of patients undergoing TAVR examined the relationship between CD and the risk of stroke and mortality. Our study revealed 3 major findings: (1) 22% of patients undergoing TAVR had CD and the majority of CD was of moderate severity (50%–79% by any diagnostic test); (2) patients with CD had higher observed rates of in-hospital and 30-day stroke and mortality; and (3) after adjusting for covariates, there was no significant relationship between the presence of CD or CD severity and the 30-day or 1-year risk of stroke or mortality.

    Prior studies examining the relationship between CD and TAVR have been limited. A single-center analysis of 294 consecutive cases of TAVR identified 19% of patients had CD or vertebral artery disease and a 6.8% 30-day post-TAVR rate of stroke. There was no association between the outcomes of stroke and mortality with CD.18 Another retrospective analysis of both TAVR and surgical aortic valve patients demonstrated that there was no association between carotid stenosis severity and stroke.19 Our study builds on these previous studies by significantly expanding the sample size, presenting data from a national cohort of consecutive TAVR patients, and extending the follow-up to 1 year. We discovered a similar rate of CD among TAVR patients (22%); we also did not appreciate an association between adjusted 30-day or 1-year stroke and mortality outcomes and CD. Thus, CD is possibly a marker for patients with a higher comorbidity burden. However, CD itself does not result in a higher risk of stroke or mortality. Further supporting this notion that CD is a marker for a large comorbidity burden are the higher unadjusted 1-year cumulative incidence rates for both MI and bleeding rates among CD patients compared with no-CD patients.

    It is imperative to understand the underlying pathophysiology of TAVR-related cerebral events to reduce the risk of stroke. Yet, the biological mechanisms driving increased strokes post-TAVR are not fully understood. The cumulative incidence curves from our analysis show a sharp increase in stroke events in the periprocedural time period, with a relatively plateauing of stroke events for the subsequent year. This suggests that the mechanism of stroke could be related to the TAVR procedure. For example, periprocedural debris can arise from either aortic atheromas during the transfemoral access or the calcified valve itself during valve manipulation. The SENTINEL trial examined the efficacy of the Sentinel transcatheter cerebral embolic protection device (Claret Medical, Santa Rosa, CA). The Sentinel device is comprised 2 filters that are deployed percutaneously and positioned in the brachiocephalic and left common carotid arteries before a TAVR procedure. Debris was found in 99% of filters, and the debris components included acute thrombus, calcification, valve tissue, arterial wall, and foreign material.9 Despite the captured debris, strokes at 30 days still occurred in 5.6% of patients randomized to the device group. In addition to the Sentinel device, several other embolic protection devices have been developed and tested. For example, TriGuard HDH (Keystone Heart, Caesarea, Israel) is a device that resides in the aortic arch and deflects embolic particles toward the descending aorta, protecting the entire cerebral circulation while maintaining adequate blood flow to the brain. The DEFLECT-III trial (A Prospective, Randomized Evaluation of the TriGuard™ HDH Embolic Deflection Device During Transcatheter Aortic Valve Implantation) randomized 85 patients undergoing TAVR to TriGuard HDH versus standard of care. The trial results demonstrated that complete cerebral vessel coverage occurred in 89% of patients; further confirmation of the neuroprotective effects of this device is being examined in the pivotal REFLECT trial (A Randomized Evaluation of the TriGuard Embolic Deflection Device to Reduce the Impact of Cerebral Embolic Lesions After Transcatheter Aortic Valve Implantation; NCT02536196).20 Other examples of embolic protection devices include the Embrella embolic deflector system (Embrella Cardiovascular, Inc, Wayne, PA) and the Shimon embolic filter (SMT Research and Development, Herzliya, Israel).21 Importantly, despite the growing field of embolic protection devices, even after the use of these technologies, there are residual strokes. Thus, we explored the hypothesis that CD may play a pathophysiological role in some TAVR-related strokes. We postulated that patients with severe or occlusive CD undergoing TAVR may develop stroke symptoms from hypoperfusion because of rapid pacing or hypotension leading to watershed infarcts.22,23 In addition, a proinflammatory state during the periprocedural period may lead to increased likelihood of carotid plaque rupture. However, our results suggest that there is no significant association between CD and 30-day or 1-year stroke rates. Our findings are in contrast with previously published results that demonstrated a relationship between CD and increased risk of stroke in surgical aortic valve replacement patients.10,11,24 We also examined a particularly high-risk group for watershed infarcts—patients with bilateral severe carotid artery stenosis. Prior data reveal patients with bilateral severe CD have impaired dynamic cerebral autoregulation.25 This disrupted blood flow among patients with bilateral severe CD may clinically manifest as a higher risk for cognitive impairment and stroke.26 Even among this high-risk group, there was no association between CD and 30-day or 1-year stroke or mortality.

    Our analysis has important clinical and scientific implications. These neutral findings are helpful in focusing attention toward other stroke reduction strategies, such as improving distal embolization devices, facilitating the diagnosis and management of postprocedure atrial fibrillation, and optimizing antiplatelet therapies.21,23 As we did not find a difference in stroke risks among patients with varying degrees of carotid stenosis, we think CD in TAVR does not require a specific set of management instructions. Screening carotid ultrasounds are commonly ordered before cardiovascular surgeries; this practice is not supported by our findings for TAVR patients.

    We postulated a potential approach to reducing the risk of stroke in TAVR patients might involve pre-TAVR carotid revascularization. Therefore, we performed an analysis examining the rate of carotid revascularization within 90 days of TAVR. The rate of revascularization within 90 days of TAVR for patients with severe CD was low at 10.8%. Although we describe 30-day and 1-year outcomes of severe CD patients with and without prior revascularization, these numbers were too small to perform formal hypothesis testing to determine the role of pre-TAVR revascularization. Moreover, there are several confounders, such as a patient’s symptomatic status, to determine whether there is a role for carotid revascularization in reducing TAVR-related stroke rates. However, our overall findings demonstrating the lack of association between CD and 1-year risk of stroke and mortality suggest it is unlikely that pre-TAVR carotid revascularization will influence the risk of stroke in TAVR patients.

    This study has several limitations. First, there was no protocoled approach to imaging the carotid arteries of TAVR patients which may have resulted in an underestimation of CD, especially asymptomatic CD. Moreover, there was no evaluation of carotid plaque morphology, which modulates stroke risk.27 Second, linking the STS/ACC TVT Registry to Medicare claims data may limit the generalizability of our results to non-Medicare patients although there were no clinically significant differences in the baseline characteristics of patients with and without available CMS linkage (Data Supplement I). Third, we did not obtain data on stroke severity, cognitive function, or other quality of life metrics. Fourth, TVT is an observational data set, thus there is the possibility of unmeasured confounding during comparisons. Fifth, because the TVT Registry is a real-world data set, it lacks protocoled postoperative evaluation by a neurologist after TAVR. The use of administrative claims data for 30-day and 1-year stroke determination may not correspond with rigorous physician adjudication of strokes. Sixth, our study could be underpowered to detect the impact of CD severity as only 22% of CD patients had severe or occlusive disease although our analysis is the first large, multicenter, nationally representative study of CD in TAVR patients. Furthermore, it is possible that CD acts as an effect-modifier for other end points, such as cognitive impairment, which we could not evaluate.28,29 Finally, we were interested in the relationship between carotid revascularization and post-TAVR outcomes; however, we were unable to pursue an adjusted analysis because of the small sample size (n=63) of patients with severe CD who underwent carotid revascularization.


    One fifth of TAVR patients have CD; however, there was no association between CD and the 30-day and 1-year risk of stroke or mortality. Despite technological improvements, post-TAVR stroke is still a major complication; our results suggest CD does not influence this risk of stroke.


    The Data Supplement is available at

    Correspondence to Ajar Kochar, MD, Duke University Medical Center, Box 3126, 2301 Erwin Rd, Durham, NC 27710. E-mail


    • 1. Adams DH, Popma JJ, Reardon MJ, Yakubov SJ, Coselli JS, Deeb GM, Gleason TG, Buchbinder M, Hermiller J, Kleiman NS, Chetcuti S, Heiser J, Merhi W, Zorn G, Tadros P, Robinson N, Petrossian G, Hughes GC, Harrison JK, Conte J, Maini B, Mumtaz M, Chenoweth S, Oh JK; U.S. CoreValve Clinical Investigators. Transcatheter aortic-valve replacement with a self-expanding prosthesis.N Engl J Med. 2014; 370:1790–1798. doi: 10.1056/NEJMoa1400590.CrossrefMedlineGoogle Scholar
    • 2. Leon MB, Smith CR, Mack M, Miller DC, Moses JW, Svensson LG, Tuzcu EM, Webb JG, Fontana GP, Makkar RR, Brown DL, Block PC, Guyton RA, Pichard AD, Bavaria JE, Herrmann HC, Douglas PS, Petersen JL, Akin JJ, Anderson WN, Wang D, Pocock S; PARTNER Trial Investigators. Transcatheter aortic-valve implantation for aortic stenosis in patients who cannot undergo surgery.N Engl J Med. 2010; 363:1597–1607. doi: 10.1056/NEJMoa1008232.CrossrefMedlineGoogle Scholar
    • 3. Reardon MJ, Van Mieghem NM, Popma JJ, Kleiman NS, Søndergaard L, Mumtaz M, Adams DH, Deeb GM, Maini B, Gada H, Chetcuti S, Gleason T, Heiser J, Lange R, Merhi W, Oh JK, Olsen PS, Piazza N, Williams M, Windecker S, Yakubov SJ, Grube E, Makkar R, Lee JS, Conte J, Vang E, Nguyen H, Chang Y, Mugglin AS, Serruys PW, Kappetein AP; SURTAVI Investigators. Surgical or transcatheter aortic-valve replacement in intermediate-risk patients.N Engl J Med. 2017; 376:1321–1331. doi: 10.1056/NEJMoa1700456.CrossrefMedlineGoogle Scholar
    • 4. Smith CR, Leon MB, Mack MJ, Miller DC, Moses JW, Svensson LG, Tuzcu EM, Webb JG, Fontana GP, Makkar RR, Williams M, Dewey T, Kapadia S, Babaliaros V, Thourani VH, Corso P, Pichard AD, Bavaria JE, Herrmann HC, Akin JJ, Anderson WN, Wang D, Pocock SJ; PARTNER Trial Investigators. Transcatheter versus surgical aortic-valve replacement in high-risk patients.N Engl J Med. 2011; 364:2187–2198. doi: 10.1056/NEJMoa1103510.CrossrefMedlineGoogle Scholar
    • 5. Grover FL, Vemulapalli S, Carroll JD, Edwards FH, Mack MJ, Thourani VH, Brindis RG, Shahian DM, Ruiz CE, Jacobs JP, Hanzel G, Bavaria JE, Tuzcu EM, Peterson ED, Fitzgerald S, Kourtis M, Michaels J, Christensen B, Seward WF, Hewitt K, Holmes DR; Registry SAT. 2016 annual report of the Society of Thoracic Surgeons/American College of Cardiology Transcatheter Valve Therapy Registry.J Am Coll Cardiol. 2017; 69:1215–1230. doi: 10.1016/j.jacc.2016.11.033.CrossrefMedlineGoogle Scholar
    • 6. Daneault B, Kirtane AJ, Kodali SK, Williams MR, Genereux P, Reiss GR, Smith CR, Moses JW, Leon MB. Stroke associated with surgical and transcatheter treatment of aortic stenosis: a comprehensive review.J Am Coll Cardiol. 2011; 58:2143–2150. doi: 10.1016/j.jacc.2011.08.024.CrossrefMedlineGoogle Scholar
    • 7. Holmes DR, Brennan JM, Rumsfeld JS, Dai D, O’Brien SM, Vemulapalli S, Edwards FH, Carroll J, Shahian D, Grover F, Tuzcu EM, Peterson ED, Brindis RG, Mack MJ; Registry SAT. Clinical outcomes at 1 year following transcatheter aortic valve replacement.JAMA. 2015; 313:1019–1028. doi: 10.1001/jama.2015.1474.CrossrefMedlineGoogle Scholar
    • 8. Van Mieghem NM, Schipper ME, Ladich E, Faqiri E, van der Boon R, Randjgari A, Schultz C, Moelker A, van Geuns RJ, Otsuka F, Serruys PW, Virmani R, de Jaegere PP. Histopathology of embolic debris captured during transcatheter aortic valve replacement.Circulation. 2013; 127:2194–2201. doi: 10.1161/CIRCULATIONAHA.112.001091.LinkGoogle Scholar
    • 9. Kapadia SR, Kodali S, Makkar R, Mehran R, Lazar RM, Zivadinov R, Dwyer MG, Jilaihawi H, Virmani R, Anwaruddin S, Thourani VH, Nazif T, Mangner N, Woitek F, Krishnaswamy A, Mick S, Chakravarty T, Nakamura M, McCabe JM, Satler L, Zajarias A, Szeto WY, Svensson L, Alu MC, White RM, Kraemer C, Parhizgar A, Leon MB, Linke A; SENTINEL Trial Investigators. Protection against cerebral embolism during transcatheter aortic valve replacement.J Am Coll Cardiol. 2017; 69:367–377. doi: 10.1016/j.jacc.2016.10.023.CrossrefMedlineGoogle Scholar
    • 10. Hogue CW, Murphy SF, Schechtman KB, Dávila-Román VG. Risk factors for early or delayed stroke after cardiac surgery.Circulation. 1999; 100:642–647.LinkGoogle Scholar
    • 11. Likosky DS, Marrin CA, Caplan LR, Baribeau YR, Morton JR, Weintraub RM, Hartman GS, Hernandez F, Braff SP, Charlesworth DC, Malenka DJ, Ross CS, O’Connor GT; Northern New England Cardiovascular Disease Study Group. Determination of etiologic mechanisms of strokes secondary to coronary artery bypass graft surgery.Stroke. 2003; 34:2830–2834. doi: 10.1161/01.STR.0000098650.12386.B3.LinkGoogle Scholar
    • 12. Carroll JD, Edwards FH, Marinac-Dabic D, Brindis RG, Grover FL, Peterson ED, Tuzcu EM, Shahian DM, Rumsfeld JS, Shewan CM, Hewitt K, Holmes DR, Mack MJ. The STS-ACC transcatheter valve therapy national registry: a new partnership and infrastructure for the introduction and surveillance of medical devices and therapies.J Am Coll Cardiol. 2013; 62:1026–1034. doi: 10.1016/j.jacc.2013.03.060.CrossrefMedlineGoogle Scholar
    • 13. Leon MB, Piazza N, Nikolsky E, Blackstone EH, Cutlip DE, Kappetein AP, Krucoff MW, Mack M, Mehran R, Miller C, Morel MA, Petersen J, Popma JJ, Takkenberg JJ, Vahanian A, van Es GA, Vranckx P, Webb JG, Windecker S, Serruys PW. Standardized endpoint definitions for Transcatheter Aortic Valve Implantation clinical trials: a consensus report from the Valve Academic Research Consortium.J Am Coll Cardiol. 2011; 57:253–269. doi: 10.1016/j.jacc.2010.12.005.CrossrefMedlineGoogle Scholar
    • 14. Kappetein AP, Head SJ, Généreux P, Piazza N, van Mieghem NM, Blackstone EH, Brott TG, Cohen DJ, Cutlip DE, van Es GA, Hahn RT, Kirtane AJ, Krucoff MW, Kodali S, Mack MJ, Mehran R, Rodés-Cabau J, Vranckx P, Webb JG, Windecker S, Serruys PW, Leon MB. Updated standardized endpoint definitions for transcatheter aortic valve implantation: the Valve Academic Research Consortium-2 consensus document.J Am Coll Cardiol. 2012; 60:1438–1454. doi: 10.1016/j.jacc.2012.09.001.CrossrefMedlineGoogle Scholar
    • 15. Gray RJ. A class of k-sample tests for comparing the cumulative incidence of a competing risk.Ann Stat. 1988; 16:1141–1154.CrossrefGoogle Scholar
    • 16. Fine JP,, Gray RJ. A proportional hazards model for the subdistribution of a competing risk.J Am Stat Assoc. 1999; 94:496–509.CrossrefGoogle Scholar
    • 17. Edwards FH, Cohen DJ, O’Brien SM, Peterson ED, Mack MJ, Shahian DM, Grover FL, Tuzcu EM, Thourani VH, Carroll J, Brennan JM, Brindis RG, Rumsfeld J, Holmes DR; Steering Committee of the Society of Thoracic Surgeons/American College of Cardiology Transcatheter Valve Therapy Registry. Development and validation of a risk prediction model for in-hospital mortality after transcatheter aortic valve replacement.JAMA Cardiol. 2016; 1:46–52. doi: 10.1001/jamacardio.2015.0326.CrossrefMedlineGoogle Scholar
    • 18. Huded CP, Youmans QR, Puthumana JJ, Sweis RN, Ricciardi MJ, Malaisrie SC, Davidson CJ, Flaherty JD. Lack of association between extracranial carotid and vertebral artery disease and stroke after transcatheter aortic valve replacement.Can J Cardiol. 2016; 32:1419–1424. doi: 10.1016/j.cjca.2016.03.010.CrossrefMedlineGoogle Scholar
    • 19. Condado JF, Jensen HA, Maini A, Ko YA, Rajaei MH, Tsai LL, Devireddy C, Leshnower B, Mavromatis K, Sarin EL, Stewart J, Guyton RA, Babaliaros V, Chen EP, Halkos M, Simone A, Keegan P, Block PC, Thourani VH; Emory Structural Heart and Valve Center. Should we perform carotid Doppler screening before surgical or transcatheter aortic valve replacement?Ann Thorac Surg. 2017; 103:787–794. doi: 10.1016/j.athoracsur.2016.06.076.CrossrefMedlineGoogle Scholar
    • 20. Lansky AJ, Schofer J, Tchetche D, Stella P, Pietras CG, Parise H, Abrams K, Forrest JK, Cleman M, Reinöhl J, Cuisset T, Blackman D, Bolotin G, Spitzer S, Kappert U, Gilard M, Modine T, Hildick-Smith D, Haude M, Margolis P, Brickman AM, Voros S, Baumbach A. A prospective randomized evaluation of the TriGuard™ HDH embolic DEFLECTion device during transcatheter aortic valve implantation: results from the DEFLECT III trial.Eur Heart J. 2015; 36:2070–2078. doi: 10.1093/eurheartj/ehv191.CrossrefMedlineGoogle Scholar
    • 21. Fassa AA, Himbert D, Vahanian A. Mechanisms and management of TAVR-related complications.Nat Rev Cardiol. 2013; 10:685–695. doi: 10.1038/nrcardio.2013.156.CrossrefMedlineGoogle Scholar
    • 22. Momjian-Mayor I, Baron JC. The pathophysiology of watershed infarction in internal carotid artery disease: review of cerebral perfusion studies.Stroke. 2005; 36:567–577. doi: 10.1161/01.STR.0000155727.82242.e1.LinkGoogle Scholar
    • 23. Hynes BG, Rodés-Cabau J. Transcatheter aortic valve implantation and cerebrovascular events: the current state of the art.Ann N Y Acad Sci. 2012; 1254:151–163. doi: 10.1111/j.1749-6632.2012.06477.x.CrossrefMedlineGoogle Scholar
    • 24. Shishehbor MH, Venkatachalam S, Sun Z, Rajeswaran J, Kapadia SR, Bajzer C, Gornik HL, Gray BH, Bartholomew JR, Clair DG, Sabik JF, Blackstone EH. A direct comparison of early and late outcomes with three approaches to carotid revascularization and open heart surgery.J Am Coll Cardiol. 2013; 62:1948–1956. doi: 10.1016/j.jacc.2013.03.094.CrossrefMedlineGoogle Scholar
    • 25. Reinhard M, Müller T, Roth M, Guschlbauer B, Timmer J, Hetzel A. Bilateral severe carotid artery stenosis or occlusion— cerebral autoregulation dynamics and collateral flow patterns.Acta Neurochir (Wien). 2003; 145:1053–1059; discussion 1059–1060. doi: 10.1007/s00701-003-0137-8.CrossrefMedlineGoogle Scholar
    • 26. Buratti L, Balucani C, Viticchi G, Falsetti L, Altamura C, Avitabile E, Provinciali L, Vernieri F, Silvestrini M. Cognitive deterioration in bilateral asymptomatic severe carotid stenosis.Stroke. 2014; 45:2072–2077. doi: 10.1161/STROKEAHA.114.005645.LinkGoogle Scholar
    • 27. Brott TG, Halperin JL, Abbara S, Bacharach JM, Barr JD, Bush RL, Cates CU, Creager MA, Fowler SB, Friday G, Hertzberg VS, McIff EB, Moore WS, Panagos PD, Riles TS, Rosenwasser RH, Taylor AJ; American College of Cardiology Foundation; American Stroke Association; American Association of Neurological Surgeons; American College of Radiology; American Society of Neuroradiology; Congress of Neurological Surgeons; Society of Atherosclerosis Imaging and Prevention; Society for Cardiovascular Angiography and Interventions; Society of Interventional Radiology; Society of NeuroInterventional Surgery; Society for Vascular Medicine; Society for Vascular Surgery. 2011 ASA/ACCF/AHA/AANN/AANS/ACR/ASNR/CNS/SAIP/SCAI/SIR/SNIS/SVM/SVS guideline on the management of patients with extracranial carotid and vertebral artery disease: executive summary. A report of the American College of Cardiology Foundation/American Heart Association Task Force on Practice Guidelines, and the American Stroke Association, American Association of Neuroscience Nurses, American Association of Neurological Surgeons, American College of Radiology, American Society of Neuroradiology, Congress of Neurological Surgeons, Society of Atherosclerosis Imaging and Prevention, Society for Cardiovascular Angiography and Interventions, Society of Interventional Radiology, Society of NeuroInterventional Surgery, Society for Vascular Medicine, and Society for Vascular Surgery.Circulation. 2011; 124:489–532. doi: 10.1161/CIR.0b013e31820d8d78.LinkGoogle Scholar
    • 28. Johnston SC, O’Meara ES, Manolio TA, Lefkowitz D, O’Leary DH, Goldstein S, Carlson MC, Fried LP, Longstreth WT. Cognitive impairment and decline are associated with carotid artery disease in patients without clinically evident cerebrovascular disease.Ann Intern Med. 2004; 140:237–247.CrossrefMedlineGoogle Scholar
    • 29. Romero JR, Beiser A, Seshadri S, Benjamin EJ, Polak JF, Vasan RS, Au R, DeCarli C, Wolf PA. Carotid artery atherosclerosis, MRI indices of brain ischemia, aging, and cognitive impairment: the Framingham study.Stroke. 2009; 40:1590–1596. doi: 10.1161/STROKEAHA.108.535245.LinkGoogle Scholar