Author + information
- Received July 3, 2013
- Revision received November 1, 2013
- Accepted November 7, 2013
- Published online February 1, 2014.
- Brian G. Hynes, MD∗∗ (, )
- Kevin F. Kennedy, MS†,
- Nicholas J. Ruggiero II, MD‡,
- Thomas J. Kiernan, MD∗,
- Ronan J. Margey, MD∗,
- Kenneth Rosenfield, MD∗ and
- Joseph M. Garasic, MD∗
- ∗Cardiology Division, Department of Medicine, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts
- †Saint Luke's Mid America Heart Institute, Kansas City, Missouri
- ‡Department of Interventional Cardiology, Jefferson University Hospitals, Philadelphia, Pennsylvania
- ↵∗Reprint requests and correspondence:
Dr. Brian G. Hynes, Cardiology Division, Department of Medicine, Massachusetts General Hospital, Harvard Medical School, 55 Fruit Street, Boston, Massachusetts 02114.
Objectives The purpose of this study was to evaluate and compare outcomes of patients undergoing carotid artery stenting (CAS) for ipsilateral restenosis, after either previous CAS or carotid artery endarterectomy (CEA) (CAS-R group), with those of patients who had CAS performed for de novo carotid atherosclerotic stenosis (CAS-DN group).
Background Therapeutic revascularization strategies to reduce stroke include CAS and CEA. Limited data exist concerning the outcomes of CAS in the setting of previous ipsilateral carotid revascularization.
Methods Patients enrolled in the CARE (Carotid Artery Revascularization and Endarterectomy) registry who underwent CAS were identified and separated into 2 groups: those undergoing CAS after previous ipsilateral CEA or CAS (CAS-R group, n = 1,996) and those who had CAS performed for de novo atherosclerotic carotid stenosis (CAS-DN group, n = 10,122). We analyzed the clinical and procedural factors associated with CAS-R and CAS-DN between January 1, 2005, and October 8, 2012. Propensity score matching using 19 clinical and 9 procedural characteristics was used, yielding 1,756 patients in each CAS cohort.
Results The primary endpoint composite of in-hospital death or stroke or myocardial infarction (MI) occurred less often in the CAS-R compared with CAS-DN patients (1.9% vs. 3.2%; p = 0.019). In-hospital adverse cerebrovascular events (stroke or transient ischemic attack) occurred less frequently in the CAS-R cohort (2.2% vs. 3.6%; p < 0.001). However, there was no significant difference in the composite of death, stroke, or MI at 30 days between both groups.
Conclusions Patients who underwent CAS for restenosis after previous ipsilateral revascularization had lower periprocedural adverse event rates and comparable 30-day adverse event rates compared with CAS for de novo carotid artery stenosis.
Carotid artery revascularization strategies include carotid artery endarterectomy (CEA) and carotid artery stenting (CAS) (1). However, the optimal management strategy of obstructive atherosclerotic carotid artery disease after previous ipsilateral revascularization, with either CEA or CAS, remains unclear. Severe carotid artery restenosis or occlusion after previous surgical or endovascular treatment is associated with an increased risk of adverse cerebrovascular events (2). Revascularization may be indicated for CAS in-stent restenosis, failure of CEA, or progression of stenosis adjacent to the previous operative or endovascular interventional site. The purpose of this study, using CARE (Carotid Artery Revascularization and Endarterectomy) registry data, was to evaluate and compare outcomes of patients undergoing CAS for ipsilateral restenosis, either after previous CAS or CEA (CAS-R group), with those of patients who had CAS performed for de novo carotid atherosclerotic stenosis (CAS-DN group). We identified clinical and procedural characteristics of patients requiring CAS after previous ipsilateral carotid revascularization. Within these 2 groups (CAS-R and CAS-DN), we also examined outcomes stratified by the presence of symptomatic carotid artery disease.
The CARE registry is an initiative of the American College of Cardiology Foundation, Society for Cardiovascular Angiography and Interventions, Society of Interventional Radiology, American Academy of Neurology, American Association of Neurological Surgeons/Congress of Neurological Surgeons, Society for Vascular Medicine, and Society of Vascular and Interventional Neurology.
Details pertaining to the CARE registry have been described in depth elsewhere (3). Briefly, this registry was established to document and review clinical outcomes for patients undergoing carotid artery revascularization. Data relating to clinical characteristics, procedural appropriateness, and 30-day outcomes are included. Independent neurological assessment using the National Institutes of Health Stroke Scale performed immediately before and after a procedure, in addition to at a 30-day time point, is a key component. A Web-based data collection tool is used, and the National Cardiovascular Data Registry is responsible for ensuring data consistency and accuracy. Variables and data definitions used for the purpose of the CARE registry were detailed recently (4) and are available at http://www.ncdr.com/WebNCDR/care/home/datacollection.
All study definitions were derived from the CARE registry data dictionary elements and definitions version 1.08. In-hospital adverse events refer to new events occurring during or after CAS but before hospital discharge. Transient ischemic attack (TIA) was defined as a focal neurological abnormality of sudden onset, lasting <24 h, and presumed to be ischemic in origin. Ischemic stroke was defined as the occurrence of a focal neurological abnormality resulting in ischemic symptoms persisting >24 h and leading to impaired functional outcomes. In-hospital mortality was defined as death occurring during the procedure or before hospital discharge. Acute myocardial infarction (MI) was defined as an increase and decrease in cardiac biomarkers with at least 1 of the values above normal for the hospital laboratory (>99th percentile of the upper reference range), together with evidence of myocardial ischemia defined as at least 1 of the following: ischemic symptoms, electrocardiographic changes indicative of new ischemia (new ST–T-wave changes or new left bundle branch block), development of pathological Q waves on the electrocardiogram, and imaging evidence of new loss of viable myocardium or new regional wall motion abnormality.
CARE registry records of patients who underwent CAS from January 1, 2005, through October 8, 2012, were reviewed. The records of all CAS patients enrolled in the registry were included for analysis with the exception of those having an acute evolving stroke, a spontaneous intracranial hemorrhage within the previous 180 days, severe dementia, CAS performed with the patient under general anesthesia, or nonatherosclerotic carotid artery disease. Those undergoing CAS for ipsilateral restenosis after previous carotid artery revascularization (CAS-R) and those who had CAS performed for de novo atherosclerotic carotid artery stenosis (CAS-DN) were identified.
The primary outcome of this study was a composite of in-hospital death/MI/stroke and its individual components. Secondary outcomes included 30-day composite of death/MI/stroke and its individual components.
Initially, we compared patient and procedural characteristics as well as in-hospital outcomes between patients with previous ipsilateral procedures and those without using t tests or chi-square tests. To determine whether differences in outcomes between the previous ipsilateral procedure groups could be accounted for by assessed patient and procedural characteristics, we developed a saturated propensity score model predicting previous ipsilateral procedure conditioned on 33 covariates using logistic regression. Outcomes were compared between the CAS-DN and -R groups by conditional logistic regression to account for the propensity match. The propensity score model had a c-index of 0.715. We used the propensity score to create a balanced cohort by using a 1-to-1 match with a caliper width of 0.2 times the SD of the logit (5). To examine the ability of the propensity score to balance the sociodemographic and clinical characteristics of patients with and without a previous ipsilateral procedure, we compared the overall standardized differences in clinical covariates before and after matching. The standardized difference is 100 times the absolute difference in sample means divided by an estimate of the pooled SD of the variable and represents the difference in means between the 2 groups in units of SD. This represents the preferred metric in assessing the propensity score (6). The results of the propensity score to balance characteristics is shown in Tables 1 and 2; all variables included in the match had an SD <10% indicating quality balance (7). Statistical analyses were performed by the Saint Luke's Mid America Heart Institute Department of Biostatistics using SAS version 9.2 (SAS Institute, Cary, North Carolina).
A total of 12,361 patients who underwent CAS from January 1, 2005, through October 8, 2012, were included in this analysis. Patients undergoing ipsilateral CAS for restenosis of previous carotid artery revascularization (CAS-R, n = 1,996, 16.1%) were identified. Of these, 262 had undergone ipsilateral previous CAS, whereas 1,734 had undergone CAS for post-CEA restenosis. After matching, there was a balanced cohort of 2,512 (1,756 in each group); 30-day follow-up data were available for 2,736 patients (77.90%): CAS-R: n = 1,366 (77.79%) and CAS-DN: n =1,370 (78.02%).
The clinical and procedural characteristics for the unadjusted and propensity score-matched patients are detailed in Tables 1 and 2. As noted in Table 1, there were many significant differences between the unadjusted CAS-R and -DN groups. A greater percentage of females comprised the CAS-R group (45.4% vs. 36.9%; p < 0.001). A history of current or previous tobacco use, hypertension, dyslipidemia, and peripheral arterial disease were more prevalent in those needing repeat revascularization. Rates of heart failure and atrial fibrillation or flutter were less common than in the CAS-DN patients. The proportion of patients with symptomatic target lesions in the 6-month period before CAS was higher in CAS-DN group (41.2% vs. 36.3%; p < 0.001). The pre-procedure neurological status as assessed by the National Institutes of Health Stroke Scale was comparable between both patient groups. Likewise, previous TIA (31.7% vs. 32.7%; p = 0.352) and ischemic stroke rates (14.6% vs. 16.1%; p = 0.81) were similar for the CAS-DN and -R patients, respectively.
Patients in the CAS-DN group were more likely to have their procedure performed by a high-volume operator, defined as >10 CAS procedures per quarter (CAS-DN: 25.2% vs. CAS-R: 15.9%; p < 0.001). Mean fluoroscopy time was longer in the CAS-R group (19.0 ± 15.0 min vs. 17.5 ± 13.1 min; p < 0.001), despite a greater number of patients having a more favorable type 1 aortic arch compared with the CAS-DN group (56.4% vs. 51.3%; p < 0.001). Carotid artery plaque morphology tended to be longer and more complex in the CAS-DN group, with a higher likelihood of ulceration (30.7% vs. 19.1%; p < 0.001) and calcification (67.6% vs. 44.4%; p < 0.001). The high rate of embolic protective device deployment during CAS (CAS-R: 97.4% vs. CAS-DN: 97.1%; p = 0.492) was comparable in both patient subsets. Final minimal luminal diameter was minimally greater in the CAS-R group (5.5 ± 1.7 mm vs. 5.3 ± 1.6 mm; p < 0.001).
The discrimination of the propensity score was adequate, and propensity matching succeeded in matching 1,756 of the 1,996 CAS-R patients with those undergoing CAS for de novo carotid stenosis. As detailed in Tables 1 and 2, the 2 groups were well matched with regard to risk factor prevalence and procedural characteristics, and the standardized difference was below the recommended maximal value of 10% for every factor.
Clinical outcomes in the propensity-matched groups are given in Table 3. The CAS-R group experienced a lower rate of composite of in-hospital death, stroke, or MI compared with the CAS-DN group (1.9% vs. 3.2%; p = 0.019). A lower in-hospital mortality rate was observed for the CAS-R group (0.3% vs. 0.7%; p = 0.012). The adverse neurological event (new stroke or TIA) rate was significantly lower in the repeat revascularization cohort (CAS-R: 2.2% vs. CAS-DN: 3.6%; p < 0.001). No significant differences in MI or TIA rates were recorded.
For asymptomatic patients, there was no significant difference in the primary outcome composite (in-hospital death, stroke, or MI) (CAS-R: 1.7% vs. CAS-DN: 2.2%; p = 0.40). However, among symptomatic patients, the primary outcome composite was lower in those having repeat revascularization (2.3% vs. 4.8%; p = 0.017). Procedure-related stroke or TIA rates were also lower among symptomatic patients in the CAS-R group (3.1% vs. 5.4%; p = 0.041). Within the CAS-R group, no significant difference was found with regard to the composite of in-hospital death, stroke, or MI between patients who underwent repeat CAS (n = 262) and those who had CAS after a previous CEA (n = 1,734) (0.76% vs. 2.19%; p = 0.124) (Online Table 1). However, the numbers in the previous CAS group were low, rendering this an underpowered comparison.
The 30-day composite rate of death, stroke, or MI was not significantly different (CAS-R: 4.61% vs. CAS-DN: 6.28%; p = 0.055) (Table 3). Likewise, the 30-day stroke rate was not significantly lower in those patients undergoing CAS after previous revascularization (3.22% vs. 4.53%; p = 0.077). In addition, there were no significant differences in outcomes between the CAS-R and -DN groups at 30 days stratified by symptom status.
Severe restenosis (≥70%) within the 12-month period after previous carotid artery revascularization has been shown to be associated with significantly higher nonperioperative stroke or TIA rates (2). Secondary analysis of CREST (Carotid Revascularization Endarterectomy Versus Stenting Trial) reported a higher risk of ipsilateral stroke in patients in whom restenosis developed within 2 years after CAS or CEA (hazard ratio: 4.37, 95% confidence interval: 1.91 to 10.03; p = 0.0005, adjusted for age, sex, and symptomatic status) (8). Restenosis after CEA has been reported to occur in 2.8% to 36% of cases, depending on the definition, imaging modality, and length of follow-up used (8–13). However, rates of symptomatic restenosis appear to be much lower (2% to 4%) (14,15). Considerable variation in the documented rates of restenosis after CAS also exists, from <3% to >16% (2,8,16–20). Medium-term follow-up of the CREST trial found that severe carotid artery restenosis (≥70% or occlusion) was infrequent and CAS was as durable as CEA, with restenosis rates of 6.0% and 6.3%, respectively, at 2 years (8).
This study adds to the information concerning CAS after restenosis, and to our knowledge, the cohort described here represents the largest amount of reported data in this clinical setting. After propensity-matched analysis, we found that CAS-R patients, had a lower rate of the composite of in-hospital death, stroke, or MI compared with CAS-DN patients. Although females comprised only 38.54% of the CAS-DN group, this proportion increased to 46.27% of those requiring repeat revascularization (p < 0.001). Previous studies have shown that female sex is associated with a significant risk of restenosis after CEA (21,22). This may be due to smaller arterial dimensions, which also have been implicated in restenosis after endovascular stent placement in other arterial networks (23,24).
Although restenosis after previous CEA or CAS is generally considered a benign entity due to a perceived lower thromboembolic risk, 36% of the lesions in the CAS-R group were symptomatic within 6 months of repeat revascularization. However, this may reflect a selection bias in that symptomatic de novo carotid artery disease patients are referred more frequently for CEA. A greater proportion of the CAS-DN group required CAS for symptomatic carotid diseases (39.5%; p = 0.012), and plaque morphology was more complex, with a greater likelihood of longer and ulcerated stenoses being found. In addition, these patients did have significantly higher rates of composite in-hospital all-cause death, MI, and stroke compared with the CAS-R group after propensity-score matching. In-hospital new adverse neurological event rates were significantly lower among the symptomatic population who had previous ipsilateral revascularization, potentially due to a possible lower relative embolic risk of the restenotic carotid artery versus de novo atheromatous disease at the time of CAS. Within the repeat revascularization group, there was a nonsignificant trend toward a lower rate of composite in-hospital all-cause death, MI, and stroke in patients undergoing repeat CAS versus CAS after CEA, although this comparison was underpowered given the low number in the repeat CAS group.
Despite CAS in the repeat revascularization group being associated with a reduced periprocedural event rate, a 2.2% risk of in-hospital stroke or TIA remained, although the actual recorded in-hospital stroke rate was 1.4%. A very high rate of embolic protection use during CAS was evident in both the CAS-R and -DN groups. Thirty-day follow-up data, which were complete in 77.8% of the CAS-R patients, yielded a stroke rate of 3.2%, which was not significantly lower than that in those undergoing de novo CAS (4.5%; p = 0.77). We found that a large number of documented stroke events were determined to have occurred during the post-hospital discharge to 30-day event period: 56% and 47% for the CAS-R and -DN groups, respectively. It is also worth noting that the risk of stroke, although high in the periprocedural period, persists after successful CAS. The optimal drug therapy post-CAS remains unclear. Many operators routinely continue dual antiplatelet therapy for a number of months. In this study, discharge medications were not significantly different between groups and included aspirin in >90% and clopidogrel in >92% of patients (data not shown). Likewise, rates of warfarin prescribed at the time of hospital discharge were similar (8.2% and 7.7%; p = 0.74, CAS-R vs. CAS-DN). The etiology of these late embolic events is unclear. Previous registry data reported 38% of strokes occurring after 24 h of CAS, with 20% occurring post-discharge from the hospital (25). Minimizing the risk of CAS-related and post-procedure cerebrovascular adverse events is vital to the success of carotid revascularization. Further studies are necessary to investigate the etiology, timing, and features associated with stroke in this setting. In addition, studies may also be needed to ascertain the durability of CAS for in-stent restenosis or restenosis after CEA.
Several limitations, including those related to retrospective design, are inherent to our study findings. The patient and procedural data analyzed are drawn from a large, observational, and self-reported voluntary registry across 141 participating institutions, but may not be applicable to all CAS centers. Although the CARE registry uses standardized data collection in addition to the implementation of quality controls, there are likely to be additional patient characteristics such as socioeconomic and health variables, among others, that were not recorded and may differ significantly between groups. Furthermore, core laboratory evaluation of both noninvasive and invasive carotid imaging was not performed as part of the CARE registry data, thereby potentially leading to additional sources of confounding and inaccuracy. In the previous revascularization group undergoing CAS, it was not possible to differentiate those patients who required repeat revascularization for progressive carotid artery stenosis from the previous CAS or CEA site. Thirty-day follow-up results were available for 78% of the patients.
Analysis of data from the CARE registry indicates that CAS for patients with restenosis after previous ipsilateral carotid revascularization is comparable to CAS for de novo carotid artery disease. CAS-R patients had lower rates of composite in-hospital death, stroke, or MI compared with CAS for de novo disease. This was due primarily to a reduced rate of in-hospital neurovascular complications. Thirty-day outcomes were comparable in both CAS groups.
For a supplemental table, please see the online version of this article.
This research was supported by the National Cardiovascular Data Registry (NCDR) of the American College of Cardiology Foundation. The views expressed in this paper represent those of the authors and do not necessarily represent the official views of the National Cardiovascular Data Registry or its associated professional societies identified at http://www.ncdr.com. Dr. Ruggiero is a consultant for St. Jude Medical. Dr. Margey is a Physician Proctor for Edwards LifeSciences. Dr. Rosenfield has received research grants from Abbott Vascular, Bard Peripheral Vascular, Medtronic/Invatec, Lutonix Bard, and Atrium; has received consulting/advisory board fees from Abbott Vascular, Boston Scientific, Complete Conference Management, Harvard Clinical Research Institute, Contego, Micell, Medicines Company, and Becker Ventures; has equity in Lumen Biomedical, Medical Stimulation Corp., Angioguard (Cordis), and Micell; and has served on the board of directors of VIVA Physicians (501C3). Dr. Garasic has received consulting/advisory board fees from and has an equity interest in Access Closure, Inc. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- carotid artery stenting
- carotid endarterectomy
- carotid artery stenting for ipsilateral restenosis after previous carotid artery revascularization
- carotid artery stenting performed for de novo carotid atherosclerotic stenosis
- myocardial infarction
- transient ischemic attack
- Received July 3, 2013.
- Revision received November 1, 2013.
- Accepted November 7, 2013.
- 2014 American College of Cardiology Foundation
- Brott T.G.,
- Halperin J.L.,
- Abbara S.,
- et al.
- Bonati L.H.,
- Ederle J.,
- McCabe D.J.,
- et al.
- Yeh R.W.,
- Kennedy K.,
- Spertus J.A.,
- et al.
- Arquizan C.,
- Trinquart L.,
- Touboul P.J.,
- et al.
- Frericks H.,
- Kievit J.,
- van Baalen J.M.,
- van Bockel J.H.
- Cunningham E.J.,
- Bond R.,
- Mehta Z.,
- Mayberg M.R.,
- Warlow C.P.,
- Rothwell P.M.
- Yadav J.S.,
- Roubin G.S.,
- Iyer S.,
- et al.
- de Feyter P.J.,
- Kay P.,
- Disco C.,
- Serruys P.W.