Abstract
BACKGROUND
Contralateral carotid occlusion (CCO) is an established high-risk feature for undergoing carotid endarterectomy (CEA) and is traditionally an indication for carotid artery stenting (CAS). Recent observational data have called into question whether CCO remains a high-risk feature for CEA.
OBJECTIVES
The purpose of this study was to determine the clinical impact of CCO among patients undergoing CEA and CAS in a contemporary nationwide registry.
METHODS
All patients undergoing CEA or CAS from 2007 to 2019 in the NCDR CARE (National Cardiovascular Data Registry Carotid Artery Revascularization and Endarterectomy) and PVI (Peripheral Vascular Intervention) registries were included. The primary exposure was the presence of CCO. The outcome was a composite of in-hospital death, stroke, and myocardial infarction. Multivariable logistic regression and inverse-probability of treatment weighting were used to compare outcomes.
RESULTS
Among 58,423 patients who underwent carotid revascularization, 4,624 (7.9%) had a CCO. Of those, 68.9% (n = 3,185) underwent CAS and 31.1% (n = 1,439) underwent CEA. The average age of patients with CCO was 69.5 ± 9.7 years, 32.6% were women, 92.8% were Caucasian, 51.7% had a prior transient ischemic attack or stroke, and 45.4% presented with symptomatic disease. Over the study period, there was a 41.7% decrease in the proportion of patients with CCO who underwent revascularization (p < 0.001), but CAS remained the primary revascularization strategy. Unadjusted composite outcome rates were lower after CAS (2.1%) than CEA (3.6%). Following adjustment, CCO was associated with a 71% increase in the odds of an adverse outcome after CEA (95% confidence interval: 1.27 to 2.30; p < 0.001) compared with no increase after CAS (adjusted odds ratio: 0.94; 95% confidence interval: 0.72 to 1.22; p = 0.64).
CONCLUSIONS
CCO remains an important predictor of increased risk among patients undergoing CEA, but not CAS.
Keywords: contralateral carotid occlusion, carotid artery stenting, carotid endarterectomy
Ischemic stroke is the fifth leading cause of death and the greatest cause of long-term disability among adults in the United States (1). Extracranial internal carotid stenosis is associated with approximately 11.5% of all ischemic strokes, or 41,000 strokes annually in the United States (2). The 2 predominant interventional strategies for the treatment and prevention of stroke caused by carotid artery stenosis are carotid endarterectomy (CEA) and carotid artery stenting (CAS). Previous research has demonstrated that patients with different risk profiles experience significantly different outcomes following CEA relative to CAS. Thus, it is critical to select optimal interventional strategies for patients based on their comorbidities (3).
One particularly high-risk subset of patients is those with a contralateral carotid occlusion (CCO), which comprises approximately 8% to 10% of patients with carotid artery stenosis (4,5). CCO is historically considered a high-risk feature for undergoing CEA as it confers an elevated risk of adverse events, primarily that of stroke, in the perioperative period (3,6). In contrast, CCO has not been shown to increase the risk of adverse events in patients undergoing CAS (3,5). Therefore, CCO is traditionally a reason for referral for CAS, and is a reimbursable condition for CAS in symptomatic patients by the Centers for Medicare and Medicaid Services (7).
Recent observational data have called into question whether CEA continues to be inferior to CAS among patients with CCO in contemporary practice (8,9); however, these analyses have several important limitations. First, these studies were performed in surgical quality improvement registries in which CAS is often under-represented relative to CEA (10). In addition, these registries primarily include surgeon operators, with less diversity in clinical specialties (11). Last, patients with CCO referred for CEA may be a healthier subset than those treated with CAS; therefore, an analysis comparing adjusted outcomes between these patient groups is likely subject to unmeasured confounding.
We aimed to examine the association between CCO and rates of in-hospital stroke, death, and myocardial infarction (MI) among patients who underwent CAS and, separately, among patients who underwent CEA in the NCDR (National Cardiovascular Data Registry) CARE (Carotid Artery Revascularization) and PVI (Peripheral Vascular Intervention) registries. The NCDR carotid registries represent more than 200 sites, with near equal contribution of CAS and CEA procedures, and include procedures from a diversity of clinical specialties (12).
METHODS
STUDY POPULATION.
The study cohort was derived from the American College of Cardiology’s NCDR CARE and PVI registries. The CARE registry is comprised of data from 186 centers in the United States from 2005 until 2014 (13). After 2014, the CARE registry was redesigned and incorporated into the larger NCDR PVI Registry, which includes peripheral vascular procedures in addition to carotid procedures. The PVI registry includes data from more than 200 institutions and began collecting data in April 2014 (12). Institutional Review Board (IRB) approval was not required for this study, as each NCDR registry has an IRB-approved protocol that governs human subject research conducted by that registry. The American College of Cardiology has designated Chesapeake Research Review Incorporated as its IRB of record.
We analyzed data beginning in January of 2007 and ending in July of 2019. The study included 288 sites that performed CAS and 129 sites that performed CEA. The centers voluntarily participated in collection and validation of demographic, medical history, and procedural data.
We identified all carotid interventions performed for carotid artery disease within the NCDR CARE and PVI registries during the study period, which totaled 63,348 procedures. We excluded patients with an acute evolving stroke (n = 1,884), spontaneous carotid artery dissection (n = 302), or fibromuscular dysplasia (n = 340); patients for whom data about the presence of a CCO were unavailable (n = 116); and patients who required general anesthesia for undergoing CAS (n = 2,283) (5) (Supplemental Figure 1). After exclusions, there were 58,423 patients in the analytical cohort.
PRIMARY EXPOSURE AND OUTCOME.
The primary exposure was the presence of CCO, which was defined by the registry as a 100% occluded internal carotid artery contralateral to the site being intervened upon. The primary endpoint was the composite of in-hospital death, nonfatal MI, and nonfatal ischemic stroke. Secondary endpoints included the components of the composite outcome as well as transient ischemic attack (TIA). MI was defined in 2 ways: 1) a rise and/or fall in cardiac biomarkers with at least 1 value above the 99th percentile upper reference limit and at least 1 of the following: symptoms of ischemia, new significant ST-T changes or new left bundle branch block, development of pathologic Q waves, imaging evidence of new loss of viable myocardium or new regional wall motion abnormality, or identification of coronary thrombus by angiography or autopsy; or 2) cardiac death with symptoms suggestive of myocardial ischemia and presumed new electro-cardiographic changes or new left bundle branch block , but with death occurring before cardiac biomarkers were obtained or before cardiac biomarker values would be increased. Intraprocedural and post-procedural TIA was defined in the registry as a transient episode of focal neurological dysfunction caused by brain, spinal cord, or retinal ischemia without acute infarction. An ischemic stroke was defined as an acute episode of focal or global neurological dysfunction caused by brain, spinal cord, or retinal injury as a result of infarction of central nervous system tissue.
CLINICAL CHARACTERISTICS.
Sociodemographic data were reported for all patients including age, sex, and ethnicity. Patient characteristics included cardiovascular comorbidities as well as the presence of symptoms, target vessel, and history of neurological events. Procedural characteristics included the presence of a chronic total occlusion, restenosis, or thrombus.
STATISTICAL METHODS.
All metrics and normally distributed variables are reported as mean ± SD and were compared using the Student’s t-test. Categorical variables are presented as frequency and percentage and were compared with chi-square tests. To avoid the influence of treatment selection bias, the population was divided into patients who underwent CAS and patients who underwent CEA. Among these groups, patients were stratified by the presence or absence of CCO.
To account for differences in patient populations between those with and without CCO, we used 2 different approaches. First, we included all clinical characteristics in a multivariable logistic regression model to identify factors most associated with the primary endpoints. Second, we used propensity score-based inverse probability of treatment weighting (IPTW) to examine the adjusted association between CCO and outcomes (14). Propensity scores were calculated by generating a logistic regression model that included all clinical characteristics to predict the probability of each patient having CCO. Propensity scores were calculated separately for patients who underwent CAS and those who underwent CEA. Logistic regressions models applied to the weighted population were then used to calculate odds ratios (ORs) for the primary endpoints. As a pre-specified subgroup analysis, we stratified patients by whether they were symptomatic or asymptomatic. Due to the smaller sample size after stratifying, we only used IPTW to adjust for differences in baseline characteristics.
We analyzed temporal trends in CAS and CEA performed among patients with CCO from 2007 to 2019 and evaluated differences in rates using the Cochran-Armitage test. We used the median OR to analyze heterogeneity in regard to revascularization of patients with CCO across different centers in the analysis. A median OR >1.2 was considered to suggest a high degree of heterogeneity in practice (15).
A p value <0.05 was considered significant for all analyses. SAS version 9.4 (SAS Institute, Cary, North Carolina) was used for all analysis.
RESULTS
TOTAL POPULATION.
Of the 58,423 patients who underwent CAS or CEA during the study period, 4,624 (7.9%) had CCO. Among those with CCO, 3,185 patients (68.9%) underwent CAS and 1,439 (31.1%) underwent CEA (Supplemental Table 1). The average age of patients with CCO was 69.5 ± 9.7 years. Patients with CCO were predominantly Caucasian (92.8%) and male (67.4%). Of CCO patients, 51.7% had experienced a prior TIA or stroke, and 45.4% were symptomatic at presentation.
Patients with prior CCO who underwent CAS had more comorbidities than patients who underwent CEA. These included peripheral artery disease, severe lung disease, congestive heart failure, and end-stage renal disease (Supplemental Table 1). Patients with CCO undergoing CAS also had more procedural risk factors, including restenotic lesions, prior neck surgery and radiation, and presence of tracheostomy.
Of the physicians performing revascularization procedures on patients with CCOs in this analysis, 33.0% were cardiologists, 27.7% were vascular surgeons (22.8%), and 9.8% were radiologists (9.8%) (Table 1). Physicians represented both teaching and nonteaching institutions as well as varied community settings. Among the 288 sites that performed CAS during the study period, the median annual procedural volume was 13 procedures (interquartile range: 8 to 25 procedures). Of the 129 sites that performed CEA, the median annual procedural volume was 36 procedures (interquartile range: 17 to 71 procedures).
TABLE 1.
Specialties and Practice Characteristics of Physicians Performing CAS and CEA Among Patients With CCO
| Total (N = 1.588) | CAS (n = 1.135) | CEA (n = 453) | |
|---|---|---|---|
| Specialty | |||
| Cardiology | 524 (33.0) | 514 (45.3) | 10 (2.2) |
| Vascular surgery | 440 (27.7) | 259 (22.8) | 181 (40.0) |
| Other surgery | 231 (14.5) | 60 (5.3) | 171 (37.7) |
| Radiology | 155 (9.8) | 152 (13.4) | 3 (0.7) |
| Other | 238 (15.0) | 150 (13.2) | 88 (19.4) |
| Teaching hospital | 726 (45.7) | 552 (48.6) | 174 (38.4) |
| Urban location | 791 (49.8) | 590 (52.0) | 201 (44.4) |
| Suburban location | 584 (36.8) | 408 (35.9) | 176 (38.9) |
| Rural location | 213 (13.4) | 137 (12.1) | 76 (16.8) |
Values are n (%).
CAS = carotid artery stenting; CCO = contralateral carotid occlusion; CEA = carotid endarterectomy.
CAS SUBGROUP.
Of the 31,685 (54.2%) patients that underwent CAS, 3,185 (10.1%) had CCO. Patients with CCO were younger, but had more comorbidities and risk factors than patients without a CCO, including recent stroke, peripheral artery disease, smoking, and prior neck surgery (Table 2).
Table 2.
Clinical Characteristics of Patients With CCO Compared With Patients Without CCO Who Underwent Either CAS or CEA
| CAS Cohort | CEA Cohort | |||||
|---|---|---|---|---|---|---|
| CCO Status | CCO Status | |||||
| CCO (n = 3,185) | No CCO (n = 28,500) | p Value | CCO (n = 1,439) | No CCO (n = 25,299) | p Value | |
| Age, yrs | 69.1 ± 9.6 | 71.0 ± 9.8 | <0.01 | 70.4 ± 9.9 | 71.0 ± 9.9 | 0.02 |
| Male | 2,160 (67.8) | 17,469 (61.3) | <0.01 | 955 (66.4) | 14,806 (58.5) | <0.01 |
| Caucasian | 2,942 (92.4) | 26,399 (92.6) | 0.60 | 1,351 (93.9) | 23,795 (94.1) | 0.79 |
| Hispanic or Latino ethnicity | 119 (3.7) | 1,258 (4.4) | 0.08 | 51 (3.6) | 961 (3.8) | 0.62 |
| Previous carotid revascularization | 958 (30.1) | 8,483 (29.8) | 0.71 | 207 (14.4) | 3,873 (15.3) | 0.34 |
| ESRD on dialysis | 86 (2.7) | 653 (2.3) | 0.15 | 19 (1.3) | 406 (1.6) | 0.40 |
| Diabetes mellitus | 1,274 (40.0) | 11,206 (39.4) | 0.461 | 508 (35.4) | 9,143 (36.2) | 0.53 |
| Left main coronary artery stenosis ≥50% | 187 (5.9) | 1,608 (5.6) | 0.60 | 89 (6.3) | 1,027 (4.1) | <0.01 |
| Multivessel coronary artery stenoses ≥70% | 906 (28.4) | 8,235 (28.9) | 0.60 | 384 (26.9) | 5,559 (22.2) | <0.01 |
| Prior heart failure | 559 (17.6) | 4,850 (17.0) | 0.44 | 181 (12.6) | 2,634 (10.4) | <0.01 |
| Smoker | 2,530 (79.4) | 21,020 (73.8) | <0.01 | 1,097 (76.2) | 18,113 (71.6) | <0.01 |
| Prior TIA | 1,006(31.6) | 9,146 (32.1) | 0.56 | 368 (25.6) | 6,043 (23.9) | 0.14 |
| Prior ischemic stroke | 725 (22.8) | 4,695 (16.5) | <0.01 | 290 (20.2) | 3,736 (14.8) | <0.01 |
| NYHA functional class III/IV | 174 (5.5) | 1,705 (6.0) | 0.24 | 41 (2.8) | 479 (1.9) | 0.01 |
| Target lesion symptoms | 1,505 (47.3) | 12,469 (43.8) | <0.01 | 593 (41.2) | 9,143 (36.1) | <0.01 |
| CAS restenosis | 152 (4.8) | 709 (2.5) | <0.01 | 6 (0.4) | 39 (0.2) | 0.03 |
| CEA restenosis Urgent | 462 (14.5) | 3,557 (12.5) | <0.01 | 28(1.9) 43 | 303 (1.2) 600 | 0.01 |
| cardiac surgery | 91 (2.9) | 590 (2.1) | <0.01 | (3.0) | (2.4) | 0.14 |
| Peripheral arterial disease | 1,426 (44.8) | 11,425(40.1) | <0.01 | 487 (33.8) | 7,524 (29.8) | <0.01 |
| Laryngeal nerve palsy | 0.54 | |||||
| No | 3,152 (99.1) | 28,197 (99.1) | 1,430 (99.4) | 25,105 (99.4) | 1.0 | |
| Yes, right | 28 (0.9) | 249 (0.9) | 9 (0.6) | 158 (0.6) | ||
| Yes, bilateral | 1 (0.0) | 4 (0.0) | 0 (0.0) | 1 (0.0) | ||
| Neck surgery (other than CEA) | 224 (7.0) | 1,646 (5.8) | <0.01 | 50 (3.5) | 447 (1.8) | <0.01 |
| Tracheostomy present | 42 (1.3) | 264 (0.9) | 0.03 | 2 (0.1) | 41 (0.2) | 1.00 |
| Previous neck radiation | 236 (7.4) | 1,765 (6.2) | <0.01 | 19 (1.3) | 210 (0.8) | 0.05 |
| Ischemic heart disease | 1,623 (51.0) | 15,093 (53.0) | 0.04 | 640 (44.5) | 10,405 (41.1) | 0.01 |
| Ml within 6 weeks | 85 (2.7) | 603 (2.1) | 0.04 | 26 (1.8) 85 | 330 (1.3) | 0.12 |
| LVEF <40% | 329 (10.3) | 2,781 (9.8) | 0.30 | (5.9) | 1,210 (4.8) | 0.05 |
| Thrombus present | 110 (3.5) | 1,037 (3.7) | 0.60 | 84 (5.8) | 1,525 (6.0) | 0.78 |
Values are mean SD or n (%).
ESRD = end-stage renal disease; LVEF = left ventricular ejection fraction; MI = myocardial infarction; NYHA = New York Heart Association; TIA = transient ischemic attack; other abbreviations as in Table 1.
Prior to adjustment, patients with CCO who underwent CAS had no difference in the rate of the combined endpoint of in-hospital death, MI, or stroke (2.1% vs. 2.3%; p = 0.33) or the combined endpoint of death or stroke (1.9% vs. 2.1%; p = 0.40) relative to patients without a CCO (Table 3). After multivariable adjustment, CCO remained unassociated with the combined endpoint (OR: 0.94; 95% confidence interval [CI]: 0.72 to 1.22; p = 0.64) in patients who underwent CAS. There was also no association between CCO and the combined endpoint in the IPTW analysis (Table 4, Central Illustration). Factors that were significantly associated with the combined endpoint of in-hospital death, MI, or stroke in the multivariable analysis included advanced age, diabetes, prior TIA, prior MI, prior stroke, need for urgent cardiac surgery, and presence of thrombus (Figure 1).
TABLE 3.
Unadjusted Outcomes Among Patients With and Without CCO, Stratified by CAS and CEA
| Outcome | CAS COHORT | CEA COHORT | ||||
|---|---|---|---|---|---|---|
| CCO STATUS | CCO STATUS | |||||
| CCO (n = 3.185) | No CCO (n = 28,500) | p Value | CCO (n = 1,439) | No CCO (n = 25,299) | p Value | |
| Death, stroke, and MI | 66 (2.1) | 668 (2.3) | 0.33 | 52 (3.6) | 517 (2.0) | <0.01 |
| Death and stroke | 61 (1.9) | 611 (2.1) | 0.40 | 43 (3.0) | 424 (1.7) 101 | <0.01 |
| Death | 16 (0.5) | 114 (0.4) | 0.40 | 15 (1.0) | (0.4) | <0.01 |
| Stroke | 53(1.7) | 542 (1.9) | 0.35 | 36 (2.5) | 356 (1.4) | <0.01 |
| MI | 7 (0.2) | 72 (0.3) | 0.72 | 12 (0.8) | 111 (0.4) | 0.03 |
| TIA | 35 (1.1) | 309 (1.1) | 0.94 | 5 (0.3) | 83 (0.3) | 0.81 |
TABLE 4.
Results After Inverse Probability of Treatment Weighting of Patients With and Without CCO, Stratified by CAS and CEA
| Outcome | CAS Cohort | CEA Cohort | ||
|---|---|---|---|---|
| OR (95% Cl) (CCO vs. No CCO) | OR (95% Cl) (CCO vs. No CCO) | |||
| Death, stroke, and MI | 0.93 (0.71–1.20) | 0.56 | 1.69 (1.25–2.27) | <0.01 |
| Death and stroke | 0.95 (0.72–1.25) | 0.70 | 1.70 (1.23–2.34) | <0.01 |
| Death | 1.30 (0.75–2.24) | 0.35 | 2.23 (1.26–3.95) | <0.01 |
| Stroke | 0.92 (0.69–1.24) | 0.59 | 1.79 (1.20–2.42) | <0.01 |
| MI | 0.78 (0.34–1.81) | 0.56 | 1.76 (0.96–3.23) | 0.07 |
| TIA | 1.06 (0.73–1.52) | 0.07 | 1.05 (0.42–2.61) | 0.91 |
CENTRAL ILLUSTRATION. Selecting a Revascularization Strategy for Stroke Prevention for Patients With a Contralateral Carotid Occlusion.
Of 58,423 procedures, 7.9% had contralateral carotid occlusion (CCO). Patients with CCO referred for carotid artery stenting (CAS) had significantly more comorbidities than those with CCO referred for carotid endarterectomy (CEA). CCO was associated with an increased risk of in-hospital death, stroke, and myocardial infarction (MI) for patients referred for CEA, but not CAS.
FIGURE 1. Factors Associated With Increased Odds of In-Hospital Events Among Patients Undergoing CAS or CEA.
Displayed are the odds ratios (ORs) of factors predictive of adverse outcomes in patients undergoing CAS or CEA in the setting of a CCO. After adjustment, CCO was not associated with an increase in the odds of the combined endpoint of stroke, MI, and death (OR: 0.94; 95% CI: 0.72 to 1.22; p = 0.64) in patients who underwent CAS. After adjustment, CCO was associated with a 71% increase in the odds of an adverse outcome among patients undergoing CEA (95% CI: 1.27 to 2.30; p < 0.01). *p < 0.05. **p < 0.01. CAS = carotid artery stenting; CCO = contralateral carotid occlusion; CEA = carotid endarterectomy; MI = myocardial infarction; NYHA = New York Heart Association; PAD = peripheral artery disease; TIA = transient ischemic attack.
Baseline characteristics of patients with and without symptomatic disease are displayed in Supplemental Table 2. When stratified by symptom status, there remained no association between CCO and the composite endpoint (symptomatic patients: OR: 1.04; 95% CI: 0.74 to 1.45; p = 0.82; asymptomatic patients: 0.78; 95% CI: 0.51 to 1.20; p = 0.26) (Supplemental Table 3).
CEA SUBGROUP.
Among the 26,738 (45.8%) patients who underwent CEA, 1,439 (5.1%) patients had CCO. These patients were younger and had more comorbidities compared with patients without CCO, including coronary artery disease, prior heart failure, prior stroke, and prior TIA (Table 2). They were also more likely to be symptomatic (41.2% vs. 36.1%; p < 0.01) and to have restenotic lesions following a prior CAS (0.4% for those with CCO vs. 0.2% for those without CCO; p = 0.03) or prior CEA (1.9% for those with CCO vs. 1.2% for those without CCO; p = 0.01).
Unadjusted rates of the combined endpoint of in-hospital death, MI, and stroke were significantly higher in patients with CCO who underwent CEA compared with patients who did not have CCO (3.6% vs. 2.0%; p < 0.01) (Table 3). This was also true for the combined endpoint of death or stroke (3.0% vs. 1.7%; p < 0.001). Stroke was the largest contributor to the difference in outcomes (2.5% with CCO vs. 1.4% without CCO; p < 0.001). After multivariable adjustment, CCO was associated with a 71% increase in the odds of an adverse outcome among patients undergoing CEA (95% CI: 1.27 to 2.3; p < 0.01). Similar findings were observed in the IPTW analysis (Table 4, Central Illustration). After removing hybrid procedures involving concomitant cardiac surgery (n = 634), the findings remained consistent with the primary results (Supplemental Table 4). In addition to CCO, other factors significantly associated with the composite outcome in the multivariable analysis included advanced age, need for dialysis, diabetes, prior stroke, advanced heart failure, presence of symptoms, need for urgent cardiac surgery, and history of peripheral artery disease (Figure 1).
Baseline characteristics of patients with and without symptomatic disease are displayed in Supplemental Table 5. After stratifying the cohort by the presence or absence of symptoms, CCO remained significantly associated with an increase in the odds of the composite outcome of death, stroke, or MI for both groups as well as for the combined outcome of death or stroke (Supplemental Table 6).
TEMPORAL TRENDS IN REVASCULARIZATION OF PATIENTS WITH CCO.
From January 2007 to July 2019, there was a 41.7% decrease in the proportion of patients with CCO who underwent carotid revascularization (from 10.3% to 6.0%; p < 0.001) (Figure 2). At the peak in 2016, over 12% of CAS procedures were performed in patients with CCO. This fell to under 8% by 2019. The proportion of CEAs performed among patients with CCO peaked in 2013 at 9% and declined to under 4% by 2019. Throughout the study period, CAS was the primary revascularization strategy used to treat patients with CCO. The proportion of CAS procedures performed among patients with CCO who were symptomatic increased over time (p < 0.01) (Supplemental Figure 2), whereas there was no temporal change by symptomatic status among patients who underwent CEA (Supplemental Figure 3).
FIGURE 2. Temporal Trends in Carotid Revascularization Procedures in Patients With CCO.
Displayed are changes in the proportion of carotid revascularization procedures performed among patients with CCO from 2007 through 2019. During the study period, there was a significant decline in both CEA and CAS procedures performed on patients with CCO (p < 0.01). CAS = carotid artery stenting. Abbreviations as in Figure 1.
HETEROGENEITY IN INSTITUTIONAL REVASCULARIZATION CHOICE FOR CCO PATIENTS.
Among all institutions performing CAS for patients with CCO in the study, the median OR was 1.71 (95% CI: 1.59 to 1.82) for the likelihood of receiving a CAS at different centers if a patient has a CCO. This is suggestive of a high degree of institutional variation in practice (Figure 3A). Similarly, there was a high degree of heterogeneity among centers treating patients with CCO using CEA, with a median OR of 2.01 (95% CI: 1.76 to 2.25) for the likelihood of receiving a CEA at different centers if a patient has CCO (Figure 3B).
FIGURE 3. Variation Among Hospitals in the Use of CAS or CEA for Carotid Revascularization Among Patients With CCO.
Displayed are histograms representing the distribution of CCO among centers performing CAS (A) and CEA (B). There was a median OR of 1.71 (95% CI: 1.59 to 1.82) for the likelihood of receiving a CAS among different centers if a patient has a CCO and a median OR of 2.01 (95% CI: 1.76 to 2.25) for the likelihood of receiving a CEA among different centers if a patient has a CCO. These values indicate that there is significant variation in practice among centers in the revascularization of patients with CCO. MOR = median odds ratio; other abbreviations as in Figure 1.
DISCUSSION
In this large, contemporary, national registry analysis, CCO remains a prevalent condition among patients who undergo carotid artery revascularization. Among patients with CCO selected to undergo CAS, we did not find differences in in-hospital adverse events, either before or after adjustment. These findings persisted when the analyses were stratified by symptom status. However, for patients with CCO selected to undergo CEA, we found a significant increase in the risk of death, stroke, or MI relative to patients without CCO. This risk remained after adjusting for differences in characteristics between groups and after stratifying by the presence or absence of symptoms. Over time, the proportion of patients with CCO who underwent a revascularization procedure has decreased; however, CAS remains the primary revascularization strategy for patients with CCO.
We found that the risk of an in-hospital event after carotid revascularization among patients with CCO was only increased among those who underwent CEA, but not those who underwent CAS. This is particularly notable as patients with CCO referred for CAS had considerably more comorbidities than patients with CCO referred for CEA (Supplemental Table 1). These include prognostically important characteristics such as peripheral artery disease, severe lung disease, congestive heart failure, and end-stage renal disease. This difference in patient risk profiles is not unexpected given that the CAS referral population is influenced by published guidelines that recommend CAS for patients considered to be at high risk for CEA (16). Despite the greater burden of comorbidities, unadjusted in-hospital event rates were lower among those with CCO who underwent CAS versus CEA. These results are consistent with those of a recent meta-analysis (17) and a single-center retrospective cohort study and meta-analysis (18).
Our results differ from 2 recent studies (9,19). In a retrospective observational analysis by Pothof et al. (19) in the Vascular Study Group of New England registry, the investigators reported that, though there were higher rates of thirty-day stroke and death in patients with CCO undergoing CEA as compared with CAS, the absolute numbers of stroke and death were low. The investigators conclude that CEA is reasonable in these patients because their reported stroke and death rates are within the ranges recommended by surgical guidelines for CEA. In another retrospective observational study by Nejim et al. (9) in the Vascular Quality Initiative registry, the authors found a significant increase in the adjusted risk of stroke or death in both symptomatic and asymptomatic patients with CCO who underwent CAS compared with those who underwent CEA.
We believe our results differ for several reasons. First, as demonstrated in our analysis, the CAS population differs markedly from the CEA population in that they have a higher burden of comorbidities. Furthermore, treatment selection is likely to be biased by unmeasured factors, such as patient frailty or physician intuition about surgical candidacy, which cannot be accounted for in observational data (20). Therefore, analyses directly comparing CAS to CEA are likely to reflect the higher level of illness of the CAS population and not the isolated risk of each procedure. Second, the rates of stroke and death reported by Pothof et al. (19) are unusually low, indicating that the study population may not be representative of real-world populations.
We also found that the proportion of carotid revascularization procedures performed among patients with CCO is steadily decreasing. Goodney et al. (21) showed that rates of all carotid revascularization procedures have fallen by more than 15% in Medicare patients over the last decade. This may be due to the fact that fewer procedures are being performed among asymptomatic patients, particularly when a high-risk feature such as CCO is present. Interestingly, we observed that individual sites varied significantly with regard to how often they treated patients with CCO. In addition, we did not find many centers that treat a high volume of patients with CCO. At most centers, patients with CCO comprise a small proportion of their carotid procedures. An area for future investigation is whether outcomes may improve if high-volume centers were created that focused on patients with CCO.
STUDY LIMITATIONS.
First, this analysis was performed in a large, observational, and voluntary registry. Given that patients were not randomized to treatment groups, unmeasured confounders may be present in our populations. Therefore, we chose not to compare CCO patients who underwent CEA directly to those with CCO who underwent CAS. Second, despite rigorous data collection methods, there could be factors, such as socioeconomic status, that influence volunteerism and thus select for a non-generalizable and potentially biased population of patients and operators. Because the registries are voluntary, we cannot exclude selective entering of data, which may also introduce bias. Third, our population was mostly white and male, and results may not be applicable to patients who do not share this demographic background. Fourth, intermediate- and long-term follow-up for CARE and PVI patients is not available. Patients who undergo CEA incur higher upfront risk, and outcomes might be more similar between groups with more longitudinal follow-up. Last, this is primarily a cardiovascular registry and may not capture data on high-volume vascular surgeons for whom event rates may differ. Furthermore, this is one of several quality registries and represents only a portion of carotid procedures performed nationally.
CONCLUSIONS
In this contemporary registry analysis, CCO remains an important predictor of risk of in-hospital death, MI, or stroke among patients undergoing CEA, but not among those undergoing CAS. These data support the continued use of CCO to guide the selection of carotid revascularization strategies for patients.
Supplementary Material
PERSPECTIVES.
COMPETENCY IN PATIENT CARE AND PROCEDURAL SKILLS:
In patients undergoing carotid endarterectomy, the presence of contralateral arterial occlusion is associated with worse in-hospital outcomes.
TRANSLATIONAL OUTLOOK:
Further studies are needed to define strategies that optimize outcomes for patients undergoing carotid revascularization who have high-risk features such as contralateral arterial occlusion.
AUTHOR DISCLOSURES
This work was funded by the American College of Cardiology’s National Cardiovascular Data Registry. Dr. Rosenfield has served as a consultant to or on the scientific advisory board for Abbott Vascular, Access Closure, BTG, Cordis-Cardinal Health, Eximo Medical, Volcano-Philips, Surmodics, Shockwave, Cruzar, Capture Vascular, Endospan, Janssen, Magneto, MD Insider, Micell, Silk Road, Valcare, Thrombolex, and the University of Maryland; has received grants and contracts from the National Institutes of Health and Inari; has equity in Access Closure, AngioDynamics, Contego, Endospan, Embolitech, Eximo Medical, JanaCare, PQBypass, Primacea, MD Insider, Silk Road, Cruzar Systems, Capture Vascular, Micell, and Valcare; and is a board member for VIVA Physicians and the National PERT Consortium. Dr. Jaff is a part-time employee of Boston Scientific. Dr. Jones has served as principal investigator for a research study for Agency for Health-care Research and Quality, AstraZeneca, American Heart Association, Bristol Myers Squibb, Doris Duke Charitable Foundation, Medtronic, and the Patient-Centered Outcomes Research Institute; and has served as an advisory board member for Bayer, Bristol Myers Squibb, and Janssen Pharmaceuticals. Dr. Secemsky is supported by a National Institutes of Health/National Heart, Lung, and Blood Institutes K23HL150290 award; has served as a consultant and/or on the scientific advisory board of Abbott Vascular, BD, Cook, CSI, Janssen, Medtronic, and Philips; and has received research grants from AstraZeneca, BD, Boston Scientific, Cook, CSI, Medtronic, and Philips. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
ABBREVIATIONS AND ACRONYMS
- CAS
carotid artery stenting
- CCO
contralateral carotid occlusion
- CEA
carotid endarterectomy
- IPTW
inverse-probability of treatment weighting
- MI
myocardial infarction
Footnotes
APPENDIX For supplemental figures and tables, please see the online version of this paper.
REFERENCES
- 1.Virani SS, Alonso A, Benjamin EJ, et al. Heart disease and stroke statistics—2020 update: a report from the American Heart Association. Circulation 2020;141:e139e596. [DOI] [PubMed] [Google Scholar]
- 2.Flaherty ML, Kissela B, Khoury JC, et al. Carotid artery stenosis as a cause of stroke. Neuro-epidemiology 2013;40:36–41. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Touzé E, Trinquart L, Felgueiras R, et al. A clinical rule (sex, contralateral occlusion, age, and restenosis) to select patients for stenting versus carotid endarterectomy: systematic review of observational studies with validation in randomized trials. Stroke 2013;44:3394–400. [DOI] [PubMed] [Google Scholar]
- 4.Baker WH, Howard VJ, Howard G, Toole JF, for the ACAS Investigators. Effect of contralateral occlusion on long-term efficacy of endarterectomy in the asymptomatic carotid atherosclerosis study (ACAS). Stroke 2000;31:2330–4. [DOI] [PubMed] [Google Scholar]
- 5.Mercado N, Cohen DJ, Spertus JA, et al. Carotid artery stenting of a contralateral occlusion and in-hospital outcomes: results from the CARE (Carotid Artery Revascularization and Endarterectomy) registry. J Am Coll Cardiol Intv 2013;6:59–64. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Gasecki AP, Eliasziw M, Ferguson GG, Hachinski V, Barnett HJ, for the North American Symptomatic Carotid Endarterectomy Trial (NAS-CET) Group. Long-term prognosis and effect of endarterectomy in patients with symptomatic severe carotid stenosis and contralateral carotid stenosis or occlusion: results from NASCET. J Neurosurg 1995;83:778–82. [DOI] [PubMed] [Google Scholar]
- 7.https://www.cms.gov, CAG-00085R.
- 8.Ricotta JJ 2nd, Upchurch GR Jr, Landis GS, et al. The influence of contralateral occlusion on results of carotid interventions from the Society for Vascular Surgery Vascular Registry. J Vasc Surg 2014;60:958–64; discussion 964–5. [DOI] [PubMed] [Google Scholar]
- 9.Nejim B, Dakour Aridi H, et al. Carotid artery revascularization in patients with contralateral carotid artery occlusion: stent or endarterectomy? J Vasc Surg 2017;66:1735–17348.e1. [DOI] [PubMed] [Google Scholar]
- 10.Shean KE, O’Donnell TFX, Deery SE, et al. , for the Society for Vascular Surgery Vascular Quality Initiative. Regional variation in patient outcomes in carotid artery disease treatment in the Vascular Quality Initiative. J Vasc Surg 2018;68:749–59. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.SVS Vascular Quality Initiative. 2019 detailed annual report. Available at: https://www.vqi.org/wp-content/uploads/2019DetailedAnnualReport_Final.pdf. AccessedDecember 27, 2020. [Google Scholar]
- 12.Jones WS, Kennedy KF, Hawkins BM, et al. Expanding opportunities to understand quality and outcomes of peripheral vascular interventions: the ACC NCDR PVI Registry. Am Heart J 2019;216: 74–81. [DOI] [PubMed] [Google Scholar]
- 13.White CJ, Anderson HV, Brindis RG, et al. The Carotid Artery Revascularization and Endarterectomy (CARE) registry: objectives, design, and implications. Catheter Cardiovasc Interv 2008;71: 721–5. [DOI] [PubMed] [Google Scholar]
- 14.Austin PC, Stuart EA. Moving towards best practice when using inverse probability of treatment weighting (IPTW) using the propensity score to estimate causal treatment effects in observational studies. Stat Med 2015;34:3661–79. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Chan PS, Maddox TM, Tang F, Spinler S, Spertus JA. Practice-level variation in warfarin use among outpatients with atrial fibrillation (from the NCDR PINNACLE program). Am J Cardiol 2011; 108:1136–40. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Liapis CD, Bell PR, Mikhailidis D, et al. for the ESVS Guidelines Collaborators. ESVS guidelines. Invasive treatment for carotid stenosis: indications, techniques. Eur J Vasc Endovasc Surg 2009;37 Suppl 4:1–19. [DOI] [PubMed] [Google Scholar]
- 17.Kokkinidis DG, Chaitidis N, Giannopoulos S, et al. Presence of contralateral carotid occlusion is associated with increased periprocedural stroke risk following CEA but not CAS: a meta-analysis and meta-regression analysis of 43 studies and 96,658 patients. J Endovasc Ther 2020;27: 334–44. [DOI] [PubMed] [Google Scholar]
- 18.Faggioli G, Pini R, Mauro R, Freyrie A, Gargiulo M, Stella A. Contralateral carotid occlusion in endovascular and surgical carotid revascularization: a single centre experience with literature review and meta-analysis. Eur J Vasc Endovasc Surg 2013;46:10–20. [DOI] [PubMed] [Google Scholar]
- 19.Pothof AB, Soden PA, Fokkema M, et al. The impact of contralateral carotid artery stenosis on outcomes after carotid endarterectomy. J Vasc Surg 2017;66:1727–34.e2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Waldo SW, Secemsky EA, O’Brien C, et al. Surgical ineligibility and mortality among patients with unprotected left main or multivessel coronary artery disease undergoing percutaneous coronary intervention. Circulation 2014;130: 2295–301. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Goodney PP, Travis LL, Malenka D, et al. Regional variation in carotid artery stenting and endarterectomy in the Medicare population. Circ Cardiovasc Qual Outcomes 2010;3: 15–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.