Endarterectomy versus stenting in patients with prior ipsilateral carotid artery stenting

From the Society for Vascular Surgery

Endarterectomy versus stenting in patients with prior ipsilateral carotid artery stenting Isibor J. Arhuidese, MD, MPH,a,b Besma Nejim, MD, MPH,a Susruth Chavali, MD,a Satinderjit Locham, MD,a Tammam Obeid, MD,a Caitlin W. Hicks, MD, MS,a and Mahmoud B. Malas, MD, MHS,a Baltimore, Md; and Tampa, Fla

ABSTRACT Objective: In-stent restenosis is a recognized complication of carotid angioplasty and stenting (CAS), and it is associated with an increased risk of stroke. Few case series have reported outcomes separately following carotid endarterectomy (CEA) and CAS for the treatment of in-stent restenosis. In this study, we perform an evaluation of redo-CAS vs CEA in a large contemporary cohort of patients who underwent prior ipsilateral CAS. Methods: We studied all patients in the Vascular Quality Initiative (VQI) database, who underwent CEA or CAS between January 1, 2003, and April 30, 2016, after prior ipsilateral CAS. Univariate methods (c2, t-test), Kaplan-Meier, logistic, and Cox regression analyses adjusting for patient characteristics were employed to evaluate stroke, death, myocardial infarction (MI), stroke/death, and stroke/death/MI within 30 days and up to 1 year following the procedure. Results: There were 645 carotid interventions (CEA, 134 [21%] and redo-CAS, 511 [79%]) performed in this cohort of patients with prior ipsilateral CAS. Postoperative stroke within 30 days comparing CEA vs CAS was 0% vs 0.3% (P ¼ .61) for asymptomatic patients and 4.4% vs 3.5% (P ¼ .79) for symptomatic patients for an overall stroke rate of 1.5% vs 1.4%. MI was 2.3% vs 1.2% (P ¼ .35), 30-day mortality was 3.7% vs 0.9% (P ¼ .02) following CEA vs CAS, whereas the composite of perioperative stroke/death was 4.5% vs 1.9% (P ¼ .09). Freedom from stroke/death at 1 year was 91% for CEA and 92% for redo-CAS (P ¼ .76). After risk adjustment, there was no significant difference in 30-day stroke (odds ratio [OR], 0.82; 95% confidence interval [CI], 0.15-4.48; P ¼ .82), mortality (OR, 2.21; 95% CI, 0.54-9.11; P ¼ .27), or stroke/death (OR, 0.99; 95% CI, 0.26-3.84; P ¼ .99) as well as 1-year stroke (hazard ratio [HR], 0.60; 95% CI, 0.13-2.85; P ¼ .52), mortality (HR, 0.83; 95% CI, 0.42-1.65; P ¼ .60), or stroke/death (HR, 0.80; 95% CI, 0.43-1.49; P ¼ .48) comparing CEA with CAS. The significant predictors of perioperative stroke/death were older age, diabetes, active smoking, and preoperative American Society of Anesthesiologists class IV status (all P < .05). Conclusions: We have reported adverse event rates for CEA and CAS after prior CAS and shown no significant difference in perioperative and 1-year outcomes between both groups. However, CEA is offered to patients who are more severely ill than redo-CAS, resulting in significantly higher absolute mortality. We recommend avoidance of CEA especially in asymptomatic patients with serious systemic disease. Tight management of diabetes and smoking cessation remain potent targets for outcomes improvement in redo-CAS patients. (J Vasc Surg 2017;-:1-11.)

The utilization of carotid artery stenting (CAS) in the United States has increased over the years.1,2 In the wake of results from the Carotid Revascularization Endarterectomy vs Stenting Trial that revealed noninferiority between primary CAS and primary carotid

From the Division of Vascular Surgery, Department of Surgery, Johns Hopkins Medical Institutions, Baltimorea; and the Division of Vascular Surgery, University of South Florida, Tampa.b Author conflict of interest: none. Presented at the plenary session at the 2016 Vascular Annual Meeting of the Society for Vascular Surgery, Washington, D.C., June 10, 2016. Additional material for this article may be found online at www.jvascsurg.org. Correspondence: Mahmoud B. Malas, MD, MHS, Division of Vascular Surgery, Department of Surgery, Johns Hopkins Medical Institution, 4940 Eastern Ave, A 547, Baltimore, MD 21401 (e-mail: [email protected]). The editors and reviewers of this article have no relevant financial relationships to disclose per the JVS policy that requires reviewers to decline review of any manuscript for which they may have a conflict of interest. 0741-5214 Copyright Ó 2016 by the Society for Vascular Surgery. Published by Elsevier Inc. http://dx.doi.org/10.1016/j.jvs.2016.11.041

endarterectomy (CEA) with respect to the composite perioperative outcome,3 the identification of subsets of patients in which one treatment might outperform the other has gained new focus. One such category is patients with in-stent restenosis (ISR) after primary CAS. Restenosis remains an important drawback of intravascular stents, and myointimal hyperplasia has been identified as the major mechanism of this occurrence.4 The incidence of ISR in the coronary arteries ranges from 20% to 50%; however, this complication occurs less frequently in the carotid arteries for reasons yet to be fully understood.5 In the landmark CREST trial,6 the incidence of restenosis (>70%) after CAS and CEA were similar: 6% vs 6.3% at 2 years. Results from other studies range from 1.7% to 21% with wide variation in follow-up and threshold for restenosis on imaging.7-13 Restenosis was also associated with a four-fold increase in the risk of stroke in the CREST trial.6 Thus, hemodynamically significant ISR poses a treatment challenge especially in the presence of symptoms or contralateral disease. 1

Journal of Vascular Surgery

Arhuidese et al

2

---

2017

Table I. Characteristics of patients who underwent redo-carotid angioplasty and stenting (CAS) vs carotid endarterectomy (CEA) after prior ipsilateral CAS Patient characteristics

CAS (n ¼ 511; 79%), No. (%)

CEA (n ¼ 134; 21%), No. (%)

Age, mean 6 SD, years

69.1 6 8.8

68.7 6 9.2

.64

314 (61.5)

79 (59.0)

.60

White

474 (92.8)

126 (94.7)

Black

24 (4.7)

7 (5.3)

Others

13 (2.5)

0 (0.0)

Asymptomatic

338 (66.4)

88 (65.7)

.87

Symptomatic

171 (33.6)

46 (34.3)

.87

338 (66.4)

88 (65.7)

Female sex Race

P value

.17

<.001

Symptom type/interval None TIA

91 (17.9)

15 (11.2)

Minor stroke/<1 month

19 (3.7)

10 (7.5)

Minor stroke/$1 month

45 (8.8)

11 (8.2)

Major stroke/<1 month

2 (0.4)

0 (0.0)

Major stroke/$1 month

14 (2.8)

2 (1.5)

Stroke/unspecified duration or severity

0 (0.0)

8 (6.0)

Hypertension

467 (91.4)

127 (94.8)

.20

Diabetes mellitus

196 (38.4)

64 (47.8)

.05

Active smoking

156 (30.6)

42 (31.3)

.87

History of CAD

168 (32.9)

52 (38.8)

.20

History of CHF History of COPD ESRD

61 (11.9)

19 (14.1)

.49

155 (30.4)

37 (27.6)

.53

8 (1.6)

0 (0.0)

I

4 (0.8)

0 (0.0)

II

92 (19.6)

8 (6.0)

III

308 (65.7)

84 (63.2)

IV V

65 (13.9) -

41 (30.8) -

Degree of ipsilateral stenosis <50%

.09 36 (7.1)

12 (8.9)

>50%

19 (3.7)

8 (5.9)

>60%

18 (3.5)

8 (5.9)

>70%

112 (21.9)

30 (22.4)

>80%

305 (59.7)

64 (47.8)

Occluded

15 (2.9)

8 (5.9)

Unknown

6 (1.2)

4 (3.0)

72 (15.5)

11 (8.9)

Contralateral occlusion

Local/regional

.06 <.001

Anesthesia General

.15 <.001

ASA class

92 (18.1)

121 (90.3)

416 (81.9)

13 (9.7)

Pre-op beta blocker

321 (62.8)

89 (66.4)

.44

Pre-op antiplatelet

495 (96.9)

126 (94.0)

.12

Pre-op statins

436 (85.3)

120 (89.6)

.21

ASA, American Society of Anesthesiologists; CAD, coronary artery disease; CHF, congestive heart failure; COPD, chronic obstructive pulmonary disease; ESRD, end-stage renal disease; pre-op, preoperative; SD, standard deviation; TIA, transient ischemic attack.

Journal of Vascular Surgery Volume

-,

Number

Arhuidese et al

3

-

Table II. Characteristics of patients with complete follow-up data at 30 days, 6 months, and 1 year Thirty days Characteristics Age, mean 6 SD, years Female sex

Six months

One year

CEA CAS CEA CAS CEA CAS (n ¼ 58), % (n ¼ 282), % P value (n ¼ 54), % (n ¼ 237), % P value (n ¼ 40), % (n ¼ 177), % 68.5 6 8.5

.39

66.8 6 9.7

68.2 6 8.4

.25

67.0 6 8.1

68.0 6 8.6

.49

60.3

58.9

.84

62.9

56.5

.39

62.5

57.1

.53

Race

.08

.08

.19

White

93.1

93.6

92.6

93.7

95.0

93.8

Black

6.9

2.5

7.4

2.5

5.0

1.7

Others

P value

67.4 6 9.8

0

3.9

0.0

3.8

0.0

4.5

Symptomatic

41.4

32.9

.21

42.6

31.5

.12

40.0

33.5

.44

Hypertension

93.1

92.6

.88

94.4

92.4

.60

97.5

93.8

.35

Diabetes mellitus

51.7

36.7

.03

51.9

37.7

.06

55.0

36.9

.04 .42

Active smoking

31.0

31.3

.97

29.6

31.4

.81

27.5

34.1

History of CAD

36.2

33.1

.65

35.2

33.9

.86

32.5

32.8

.97

History of CHF

13.8

11.7

.66

14.8

10.2

.33

10.0

10.2

.97

History of COPD

25.9

29.2

.61

25.9

28.8

.67

22.5

27.3

.54

0.0

1.8

.31

0.0

2.1

.28

0.0

1.7

.41

ESRD

CAD, Coronary artery disease; CAS, carotid angioplasty and stenting; CEA, carotid endarterectomy; CHF, congestive heart failure; COPD, chronic obstructive pulmonary disease; ESRD, end-stage renal disease; SD, standard deviation.

The comparative safety and effectiveness of redo-CAS vs CEA for the treatment of ISR is relatively understudied. Prior studies have been based on small samples of patients drawn from single institutions, thus, limiting the generalizability of their findings. The objective of this study is to examine 30-day postoperative and 1-year outcomes after redo-CAS vs CEA for the treatment of ISR in a contemporary and nationally representative cohort of patients. We will also identify predictors of adverse outcomes and targets for improvement.

METHODS We performed a retrospective analysis of all patients in the Vascular Quality Initiative (VQI) who underwent CEA or CAS between January 1, 2003, and April 30, 2016, after a prior ipsilateral CAS. The VQI is a prospectively maintained database approved by the Society for Vascular Surgery. It contains patient- and procedurespecific data from multiple sites across all regions of the United States and incorporates data on mortality from the social security death index. At the end of the study period, there were over 370 hospitals and 2800 participating physicians in the VQI. Details of the data collection and validation process have been published previously.14 The current study was approved by the VQI Research Advisory Committee, and the Johns Hopkins Institutional Review Board waived the need for individual patient consent under the provisions for deidentified human subject and quality improvement research.

Patients who underwent CEA or CAS after ipsilateral CAS were identified directly from variables recorded in the VQI database. The relevant patient- and procedurerelated data assessed are listed in Table I. Symptomatic status was defined as the occurrence of ipsilateral stroke, transient ischemic attack, or amaurosis fugax within 6 months prior to treatment. The interval between symptom and treatment were also examined. Degree of stenosis was the most severe stenosis obtained from duplex, magnetic resonance angiogram, computed tomography angiogram, or arteriogram. The total number of restenotic cases treated by each surgeon/interventionist was computed as a marker for surgeon experience. Operators were subsequently grouped into quartiles of case volume. Postoperative outcomes (30 day) were ipsilateral stroke, myocardial infarction (MI), death, and composite of stroke/death, stroke/death/MI. One-year outcomes were stroke, death, and the composite of stroke/death. Stroke was defined as the clinical occurrence of minor or major cortical or ocular stroke after surgery/intervention. MI was a clinical or electrocardiogram-confirmed diagnosis or an elevation in troponin. Complications examined were cranial nerve injury, wound infection, access site complications including hematoma or arterial occlusion and procedure related arrhythmia. Statistical methods. Descriptive analyses of the study groups were performed using c2 and Student t-tests as appropriate. Univariable and multivariable logistic

4

Journal of Vascular Surgery

Arhuidese et al

---

2017

Table III. Postoperative outcomes following redo-carotid angioplasty and stenting (CAS) vs carotid endarterectomy (CEA) in patients with prior ipsilateral CAS Asymptomatic, % Thirty-day outcome

Symptomatic, %

Total, %

CEA (n ¼ 88)

CAS (n ¼ 338)

P value

CEA (n ¼ 46)

CAS (n ¼ 171)

P value

Stroke

0.0

0.3

.61

4.4

3.5

MI

2.3

0.9

.28

2.2

1.8

Death

3.4

0.3

.01

4.4

2.3

Stroke/death

3.4

0.6

.03

6.5

Stroke/death/MI

3.4

1.5

.24

8.7

CEA (n ¼ 134)

CAS (n ¼ 511)

.79

1.5

1.4

.86

2.3

1.2

.35

.46

3.7

0.9

.02

4.7

.61

4.5

1.9

.09

5.3

.38

5.2

2.7

.15

P value .91

MI, Myocardial infarction.

Table IV. Logistic regression showing 30-day outcomes for redo-carotid endarterectomy (CEA) compared with carotid angioplasty and stenting (CAS) in patients with prior ipsilateral CAS Thirty-day postoperative outcome

Univariable

Multivariable

OR (95% CI)

P value

OR (95% CI)

P value

Stroke

1.10 (0.23-5.35)

.91

0.82 (0.15-4.48)

.82

Death

3.92 (1.12-13.75)

.03

2.21 (0.54-9.11)

.27

MI

1.94 (0.48-7.85)

.35

1.81 (0.42-7.82)

.43

Stroke/death

2.35 (0.84-6.58)

.10

0.99 (0.26-3.84)

.99

Stroke/death/MI 1.96 (0.77-4.95)

.16

1.84 (0.68-4.94)

.23

ASA, American Society of Anesthesiologists; CAD, coronary artery disease; CHF, congestive heart failure; CI, confidence interval; COPD, chronic obstructive pulmonary disease; MI, myocardial infarction; OR, odds ratio. Variables adjusted for age, sex, race, ipsilateral symptoms, degree of ipsilateral stenosis, contralateral occlusion, active smoking, hypertension, diabetes, CAD, CHF, COPD, ASA classification of physical status, and surgeon volume. Bold values are statistically significant.

regression analyses were employed to evaluate 30-day postoperative outcomes and identify their predictors. Kaplan-Meier estimates were computed and related curves were constructed to compare the survival function between the treatment groups. The log-rank and Wilcoxon tests were applied to test the equality of survival functions. Univariable and multivariable Cox regression methods were used to analyze the 1-year outcomes and identify their predictors. Variables included in the multivariate model were based on the univariable analysis, prior literature, guidance of likelihood ratio tests, and Akaike information indices with a goal to achieve model parsimony. Clinically relevant variables and others known to influence outcomes following carotid revascularization were forced into the models obtained from a stepwise selection process. The final variables included in the model were age, sex, race, ipsilateral symptoms, degree of ipsilateral stenosis, contralateral occlusion, active smoking, hypertension, diabetes, coronary artery disease, congestive heart failure, chronic obstructive pulmonary disease, American

Society for Anesthesiologists classification of physical status, and surgeon volume. We performed sensitivity analyses comparing characteristics and outcomes of patients with complete follow-up data at 30 days, 6 months, and 1 year to assess for selection bias because of loss to follow-up. We also performed cluster analyses using unique center identification numbers to assess for the potential impact institutional differences might have on outcomes. All analyses were performed using Stata v 14.1 statistical software (StataCorp, College Station, Texas), and statistical significance was accepted at P value of <.05.

RESULTS Patient characteristics. There were 645 carotid interventions performed between January 1, 2003, and April 30, 2016, in patients with a prior ipsilateral CAS in the VQI database. Of these, 511 (79%) were redo-CAS, whereas 134 (21%) were CEA. The CAS and CEA cohorts were similar about their age (mean, both 69 years; P ¼ .64), female sex (CAS, 62%; CEA, 59%; P ¼ .60) and racial (Caucasian CAS, 93%; CEA, 95%; P ¼ .17) compositions (Table I). Comparing CAS vs CEA, the prevalence of hypertension (91% vs 95%; P ¼ .20), coronary artery disease (33 vs 39%; P ¼ .20), and active tobacco use (both 31%; P ¼ .87) were similar between the groups. However, more American Society of Anesthesiologists (ASA) class IV patients underwent CEA vs CAS (31% vs 14%; P < .001). Secondary revascularization was performed most commonly in asymptomatic patients (both 66%; P ¼ .87) for stenosis greater than 80% (60% vs 48%; P ¼ .09). By augmenting information at patient encounters with data from the social security death index, complete follow-up was achieved for mortality. Assessment for the occurrence of stroke was achieved for 45% at 6 months (CEA, 40%; CAS, 46%; P ¼ .21) and 34% (CEA, 30%; CAS, 35%; P ¼.30) at 1 year. The mean follow-up duration for patients followed up beyond 30 days was 369 (standard deviation, 120; median, 371, interquartile range, 309-447) days for CEA and 354 (standard deviation, 176; median, 367; interquartile range, 288-422) days for CAS (P ¼ .54). The characteristics of patients undergoing redo-CEA vs CAS

Journal of Vascular Surgery Volume

-,

Number

Arhuidese et al

5

-

Table V. Predictors of perioperative stroke/death after carotid endarterectomy (CEA) following prior ipsilateral carotid angioplasty and stenting (CAS) Univariable

Multivariable

P value

HR (95% CI)

P value

.09

1.19 (1.00-1.40)

.047

3.02 (0.53-17.09)

.21

4.60 (0.37-36.5)

.27

1.98 (0.38-10.21)

.42

0.25 (0.02-4.10)

.33

1.09 (0.21-5.65)

.91

1.18 (0.14-10.3)

.88

Active smoking

2.28 (0.44-11.81)

.33

9.17 (0.61-137.0)

.11

History of CAD

0.78 (0.14-4.42)

.78

0.71 (0.75-6.69)

.76

History of CHF

1.22 (0.13-11.08)

.86

0.99 (0.03-28.51)

.99

History of COPD

1.33 (0.23-7.58)

.75

0.88 (0.08-6.86)

.92

11.53 (1.30-102.22)

.03

17.98 (1.22-264.0)

0.53 (0.09-3.06)

.48

0.31 (0.03-3.45)

.99

1.02 (0.09-11.68)

Predictors

HR (95% CI)

Age, mean (SD)

1.09 (0.99-1.21)

Female sex Symptomatic Hypertension Diabetes mellitus

-

-

ASA class III IV Ipsilateral stenosis >70% Contralateral occlusion Beta-blocker

1 (Ref)

1 (Ref)

1.01 (0.18-5.74)

.035 .34

.99

ASA, American Society of Anesthesiologists; CAD, coronary artery disease; CHF, congestive heart failure; CI, confidence interval; COPD, chronic obstructive pulmonary disease; HR, hazard ratio; SD, standard deviation. Outcome occurred exclusively in one of the categories of race, hypertension, and contralateral occlusion, hence, they were not predictive and excluded from the models. Bold values are statistically significant.

patients with complete follow-up data at 30-days, 6 months, and 1 year were similar, thus, minimizing the potential for selection bias because of differential loss to follow-up (Table II). The sensitivity analyses showed no significant difference in results obtained from the subcohorts of patients with complete follow-up at 6 months and 1 year compared with the overall cohort. Herein, we present results from the complete case analyses. Thirty-day postoperative outcomes. Overall, nine (1.4%) patients suffered ipsilateral stroke within 30 days of the procedure. Of these, two (1.5%) occurred following CEA, whereas seven (1.4%) occurred following CAS (P ¼ .91). For asymptomatic patients, no perioperative stroke occurred within 30 days of CEA, whereas one (0.3%) stroke occurred after redo-CAS (P ¼ .61). For symptomatic patients, two (4.4%) strokes for CEA and six (3.5%) for redo-CAS occurred within 30 days of the procedure (P ¼ .79; Table III). Absolute, 30-day mortality was significantly higher after CEA compared with CAS (3.7% vs 0.9%; P ¼ .022). However, MI (2.3% vs 1.2%; P ¼ .35), and the composite of stroke/death (4.5% vs 1.9%; P ¼ .09) and stroke/death/MI (5.2% vs 2.7%; P ¼ .15) rates were statistically similar between the treatment groups (Table III). No wound infection was reported following CEA, whereas cranial nerve injury occurred in 3% of these cases. Technical failure and access site complications occurred in 0.6% and 2.7% of the redo-CAS cases, respectively. The incidence of procedure related arrhythmias was 0.8% for CEA and 1% for CAS (P ¼ .81).

Reperfusion symptoms did not occur after CEA in contrast to 0.6% of CAS cases (P ¼ .38). The multivariable logistic regression analyses showed that there was no significant difference in the odds of perioperative stroke following CEA compared with CAS (odds ratio [OR], 0.82; 95% confidence interval [CI], 0.15-4.48; P ¼ .82) as shown in Table IV. The adjusted odds of stroke remained similar for CEA vs CAS for symptomatic patients (adjusted OR [aOR], 0.85; 95% CI, 0.14-5.19; P ¼ .86) but could not be estimated for asymptomatic patients because there were no stroke events in the asymptomatic CEA group. The risk adjusted odds of 30-day mortality were similar between the CEA and CAS groups (aOR, 2.21; 95% CI, 0.54-9.11; P ¼ .27). There was no difference in the composite of postoperative stroke/death (aOR, 0.99; 95% CI, 0.26-3.84; P ¼ .99) and stroke/death/MI (aOR, 1.84; 95% CI, 0.68-4.94; P ¼ .23) following CEA compared with CAS. The significant predictors of perioperative stroke/death following CEA were older age (aOR, 1.19; 95% CI, 1.00-1.40; P ¼ .047) and ASA class IV (aOR, 17.98; 95% CI, 1.22-264.0; P ¼ .035) compared with patients in ASA class III (Table V). The predictors of perioperative stroke/death following redo-CAS were older age (aOR, 1.23; 95% CI, 1.08-1.40; P ¼ .003), symptomatic status (aOR, 12.7; 95% CI, 1.71-94.2; P ¼ .013), diabetes (aOR, 12.3; 95% CI, 1.88-80.3; P ¼ .01), active smoking (aOR, 9.0; 95% CI, 1.19-67.7; P ¼ .03) and ASA class IV (aOR, 10.0; 95% CI, 1.35-74.2; P ¼ .024) compared with patients in ASA class III (Table VI).

Journal of Vascular Surgery

Arhuidese et al

6

---

2017

Table VI. Predictors of perioperative stroke/death after redo-carotid angioplasty and stenting (CAS) Univariable Predictors Age, mean (SD)

Multivariable

P value

HR (95% CI) 1.14 (1.05-1.24)

HR (95% CI)

P value

.003

1.23 (1.08-1.40)

.003

Female sex

0.39 (0.08-1.87)

.24

0.12 (0.01-1.05)

.055

Symptomatic

8.25 (1.73-39.27)

.008

12.71 (1.71-94.2)

.013

Hypertension

0.84 (0.10-6.83)

.87

Diabetes mellitus

3.84 (0.98-15.03)

.05

12.27 (1.88-80.3)

Active smoking

1.53 (0.42-5.49)

.52

9.0 (1.19-67.7)

.39

0.57 (0.09-3.71)

.56

History of CAD

0.50 (0.11-2.39)

1.19 (0.05-29.77)

.91 .009 .033

History of CHF

0.81 (0.10-6.54)

.85

0.57 (0.03-9.50)

.69

History of COPD

0.98 (0.25-3.85)

.98

0.31 (0.05-1.97)

.22

ASA class III

1 (Ref)

IV Ipsilateral stenosis >70% Contralateral occlusion

1 (Ref)

6.33 (1.65-24.28)

.007

0.38 (0.10-1.52)

.17

10.02 (1.35-74.2)

.024

0.20 (0.03-1.49)

.12

1.37 (0.29-6.59)

.69

8.2 (0.61-110.3)

.11

Beta-blocker

0.89 (0.25-3.18)

.85

2.16 (0.32-14.42)

.43

Aspirin/P2Y2 Antagonist

0.28 (0.03-2.33)

.24

0.26 (0.01-5.52)

.39

Statin

0.39 (0.10-1.55)

.18

0.35 (0.04-3.02)

.34

General anesthesia

4.72 (1.34-16.67)

.016

8.68 (0.99-75.78)

.051

ASA, American Society of Anesthesiologists; CAD, coronary artery disease; CHF, congestive heart failure; CI, confidence interval; COPD, chronic obstructive pulmonary disease; HR, hazard ratio; SD, standard deviation. Outcome occurred exclusively in one category of race, hence it was not predictive and excluded from the models. Bold values are statistically significant.

Table VII. Cox regression showing 1-year term outcomes for carotid endarterectomy (CEA) compared with redocarotid angioplasty and stenting (CAS) in patients with prior ipsilateral CAS Univariable

Multivariable

HR (95% CI)

P value

HR (95% CI)

Stroke

0.82 (0.18-3.75)

.80

0.60 (0.13-2.85)

Death

0.82 (0.43-1.57)

.55

0.83 (0.42-1.65)

.60

Stroke/death

0.91 (0.51-1.63)

.76

0.80 (0.43-1.49)

.48

One-year outcomes

P value .52

ASA, American Society of Anesthesiologists; CAD, coronary artery disease; CHF, congestive heart failure; CI, confidence interval; COPD, chronic obstructive pulmonary disease; HR, hazard ratio. Variables adjusted for age, sex, race, ipsilateral symptoms, degree of ipsilateral stenosis, contralateral occlusion, active smoking, hypertension, diabetes, CAD, CHF, COPD, ASA classification of physical status, and surgeon volume.

One-year outcomes. Overall, there were 12 (1.9%) ipsilateral stroke events over the study period. Of these, two (1.5%) occurred following CEA, and 10 (2%) occurred after CAS (P ¼ .73). Absolute ipsilateral stroke rate was 0.7% (CEA, 0; CAS, 0.9%; P ¼ .38) for asymptomatic patients and 4.2% (CEA, 4.4%; CAS, 4.1%; P ¼ .94) for symptomatic patients. Stroke estimates at 1 year obtained from the Kaplan-Meier analyses were 2.9% (95% CI, 0.7-11.1) for CEA vs 3.9% (95% CI, 2.1-7.5) following CAS (log-rank, P ¼ .80). The multivariable Cox regression analyses showed that there was no significant difference in freedom from stroke

for CEA compared with CAS (adjusted hazard ratio [aHR], 0.60; 95% CI, 0.13-2.85; P ¼ .52; Table VII). Long-term stroke was higher for symptomatic patients compared with asymptomatic patients (aHR, 4.3; 95% CI, 1.14-16.6; P ¼ .03). Absolute all-cause mortality was 10.2% (CEA, 8.2%; CAS, 10.8%; P ¼ .39) over the entire study period. The KaplanMeier estimates of mortality at 1 year were 7% (95% CI, 3.7-13.0) following CEA vs 6.5% (95% CI, 4.6-9.2) following CAS (log-rank, P ¼ .55). Per symptomatic strata, 1-year mortality estimates from Kaplan-Meier for CEA vs CAS were 5.9% (95% CI, 2.5-13.7) vs 6% (3.8-9.4) for asymptomatic patients (log-rank, P ¼ .69) and 9% (95% CI, 3.5-22.2) vs 7.6% (4.4-13.0) for symptomatic patients (log-rank, P ¼ .60). The Cox regression analyses showed similar mortality in the long term for CEA compared with CAS (aHR, 0.83; 95% CI, 0.42-1.65; P ¼ .60). Long-term survival was similar between symptomatic and asymptomatic patients (aHR, 1.27; 95% CI, 0.74-2.17; P ¼ .38). The absolute stroke/death rate over the study period was 12.0% (CEA, 10.5%; CAS, 12.4%; P ¼ .56). Unadjusted Kaplan-Meier estimates of stroke/death at 1 year was 8.6% (95% CI, 4.9-15.0) following CEA vs 8.0% (95% CI, 5.8-10.9) following CAS (log-rank, P ¼ .76). Considering symptomatic status, Kaplan-Meier estimates of stroke/ death at 1 year for CEA vs CAS were 7.3% (95% CI, 3.3-5.5) vs 7% (4.6-10.6) for asymptomatic patients (Fig 1) and 11.1% (95% CI, 4.8-24.7) vs 10.1% (6.3-16.0) for symptomatic patients (Fig 2). There was no difference

Journal of Vascular Surgery Volume

-,

Number

Arhuidese et al

7

-

Fig 1. Freedom from stroke/death in asymptomatic patients for redo-carotid angioplasty and stenting (CAS) vs carotid endarterectomy (CEA) in patients with prior ipsilateral CAS. This figure shows that in asymptomatic patients with prior CAS who develop restenosis, post reintervention stroke-free survival is not statistically different between redo-CAS and CEA. Standard error for both treatments was less than 10%.

Fig 2. Freedom from stroke/death in symptomatic patients for redo-carotid angioplasty and stenting (CAS) vs carotid endarterectomy (CEA) in patients with prior ipsilateral CAS. This figure shows that in symptomatic patients with prior CAS who develop restenosis, post reintervention stroke-free survival is not statistically different between redo-CAS and CEA. Standard error for both treatments was less than 10%.

in this composite outcome in the risk-adjusted analyses comparing CEA to CAS (aHR, 0.80; 95% CI, 0.431.49; P ¼ .48) or symptomatic to asymptomatic patients (aHR, 1.40; 95% CI, 0.86-2.27; P ¼ .17). The clustered analyses revealed no significant change in the magnitude and statistical significance of the estimates reported above. There was no significant difference in outcomes between the clusters (aOR, 1.0; 95% CI, 0.99-1.01; P ¼ .32).

DISCUSSION In this contemporary cohort of patients with restenosis following prior ipsilateral CAS, the majority of patients underwent CAS compared with CEA (79% vs 21%) for the treatment of ISR. There was no significant difference in the adjusted risk of stroke, death, or the composite of stroke/death comparing both treatments at 30 days and at 1 year. A few existing studies have reported results separately for redo-CAS11,15-17 and CEA in patients who

8

Arhuidese et al

Journal of Vascular Surgery ---

underwent prior ipsilateral CAS.18-20 Notably, these studies examined 10 patients or fewer. To our knowledge, this is the first study to perform a direct comparison of perioperative and long-term outcomes following redoCAS and CEA for the treatment of ISR in a relatively large sample of patients using survival analytic methods. In their evaluation of 10 patients treated with redo-CAS, Reimers et al17 reported no strokes or death within 17 months of redo-CAS in contrast to stroke incidence of 25% reported by Willfort-Ehringer et al15 in their assessment of 8 redo-CAS patients. Following operative treatment of ISR, de Borst et al19 reported no strokes at 13 months of follow-up in four patients. Similarly, no events were recorded in case reports of CEA after ipsilateral CAS.18,20 The significantly larger number of patients and multi-institutional nature of the data in the current study provides a more valid and representative estimate of outcomes following these treatments. We acknowledge the efforts of the preceding authors in documenting outcomes for this understudied population of patients. Similar to the proportion of asymptomatic patients in the general and high-risk population of patients who underwent treatment for de novo carotid stenosis, the majority (66%) of patients who underwent treatment for ISR in this study were asymptomatic.21-24 The high proportion of asymptomatic patients in this study was largely driven by severity of ipsilateral stenosis and contralateral occlusion. Over 80% of these patients had moderate (>60%), 57% had severe (>80%) ipsilateral stenosis, and 14% had contralateral occlusions. The threshold for treatment of asymptomatic patients in the current study is higher than that reported from a systematic review of 34 studies that showed only 22% of patients treated for ISR were symptomatic compared to 34% in the current study.25 Follow-up post primary CAS and the cut-offs for the definition of ISR on imaging often attract controversy. At the moment, a consensus is lacking on the parameters for uniform reporting of ISR on duplex ultrasound scan. Understandably, the stent-arterial wall complex differs significantly from native arteries in compliance and other hemodynamic characteristics. Different thresholds of peak systolic velocity yield variations in definition of ISR for therapeutic purposes and several duplex ultrasound grading criteria have been recommended by different authors.6,26,27 The guidelines of the Society for Vascular Surgery indicate that suspicions of ISR on duplex should be confirmed using digital subtraction angiography prior to treatment.28 The management of patients with ISR exemplifies the delicate balance between the need to intervene to prevent further cerebrovascular compromise and cautious conservative management. The precise identification of the point at which intervention is required is key. This is typically based on the constellation of factors including

2017

symptomatic status, the etiology and degree of ipsilateral stenosis, and contralateral disease. In this era of evidence-based practice, an official recommendation of the Society for Vascular Surgery on the treatment of ISR based on these factors is needed; as such a position does not currently exist. We believe the decisions to treat patients with ISR as recorded in this national database were based on the best clinical judgment of vascular surgeon across the United States supported with evidence on imaging. Nonetheless, residual concern exists about the treatment of patients solely on the basis of degree of stenosis; in the absence of symptoms or significant contralateral disease. Although institutional differences in the threshold for revascularization might affect the decision to offer intervention/surgery to individual patients, it is unlikely that these differences will affect the outcomes of intervention/surgery. Not surprisingly, the clustered analyses did not reveal any difference in outcomes. An understanding of the progression of restenotic disease might be useful in guiding follow-up and intervention for ISR. However, there is significant variability in the current literature. In their 20-month duplex ultrasound scan follow-up study, Leger et al29 showed progression from moderate to severe ISR in 75% of cases, thus, warranting treatment. In contrast, WillfortEhringer et al30 showed no progression in neo-intimal hyperplasia of ISR beyond 1 year postprimary CAS, and some regression in patients treated primarily with selfexpanding carotid stents because of positive arterial remodeling.30,31 In this study, we do not have the time interval between the initial CAS and restenosis. Thus, we are unable to ascertain which lesions are due to neointimal hyperplasia. Some authors also recommend early consideration of surgical management in patients with calcified lesions that might interfere with stent expansion.30 These conflicting results and the small number of patients studied leave room for larger studies on ISR progression and the extent to which medical management alters the clinical course as has been done for patients with de novo carotid disease.32 Angioplasty alone was found to be inferior to angioplasty and stenting in the Carotid and Vertebral Artery Transluminal Angioplasty Study of de novo carotid stenosis.33 In practice, patients with ISR are often treated with balloon angioplasty alone, and satisfactory results have been reported in some case series.9,12,13,16 In view of the differences between de novo and restenotic lesions and unique conditions imposed by indwelling stents, objective assessments of large sample of patients are needed to validate and further document the benefit of angioplasty alone for the treatment of ISR. It is possible that some of the patients in the current study failed prior intervention with angioplasty alone. We do not have information on the number and timing of prior angioplasties, hence, we are unable to ascertain the impact

Journal of Vascular Surgery Volume

-,

Number

Arhuidese et al

9

-

they might have on outcomes of subsequent CAS or CEA. The current analysis does not consider factors such as balloon and stent type and size that might contribute to outcomes for CAS but have a limited value in a comparison of CAS vs CEA. Following the decision to revascularize a patient with ISR, nonsignificant difference in outcomes for CEA vs redo-CAS from the current study place emphasis on identification of targets for improvement for either treatment. Symptomatic status did not predict 30-day postoperative stroke/death following CEA possibly because of the relatively small number of patients in this subcohort. In contrast, this characteristic was associated with a 13-fold increase in risk for patients who underwent redo-CAS. The risk of stroke/death was 18-fold greater following CEA (OR, 17.98; 95% CI, 1.22-264.0; P ¼ .035) and 10- fold greater following redo-CAS (OR, 10.02; 95% CI, 1.35-74.2; P ¼ .024) for patients preoperatively classified as having severe systemic disease that is a constant threat to life (ASA class IV) compared with patients with nonincapacitating systemic disease (ASA class III). This finding is similar to that for patients who underwent redo-CEA in a national cohort and questions the value of ISR revascularization in very sick asymptomatic patients.34 There were significantly more patients with ASA class IV in the CEA group (31% vs 14%; P < .001), which could explain the higher absolute mortality in this group (3.7% vs 0.9%; P ¼ .022). The greater proportion (60%) of patients in ASA class IV were treated for asymptomatic disease, and this class of patients was associated with higher odds of stroke/death as previously stated. Thus, we recommend avoidance of CEA in sick asymptomatic patients with ISR. Prior studies6,35 have implicated diabetes mellitus and smoking as risk factors for ISR, and these patient factors negatively influenced outcomes of treatment of ISR in the current study. The current study revealed a 12-fold increase associated with diabetes and a nine-fold increase associated with active smoking in odds of stroke/death within 30 days (Table VI). Notably, there was no significant difference in the prevalence of these predictors between patients with follow-up at 30 days and patients with longer term follow-up. Consequently, aggressive management of diabetes mellitus and smoking cessation remain potent targets for outcomes improvement. The use of beta blockers had no added benefit for these patients in contrast to findings in patients with de novo disease.36 This study has shown that redo-CAS and CEA after prior CAS carries significant risk especially in patients who are high risk at baseline. Vascular surgeons and interventionists are ethically bound to ensure all effective alternatives to treatment are explored before patients are exposed to risks of surgery or intervention. The value of medical management compared with surgery or intervention is currently being evaluated in a randomized trial; however,

compliance and polypharmacy in the elderly remain real threats to true optimal medical therapy in real practice. In addition to the use of statins and traditional antiplatelet medications, phosphodiesterase-3 inhibitors such as cilostazol have been associated with favorable outcomes for the treatment of ISR in the coronaries and lower extremities as well as the prevention of ISR in a limited number of patients with de novo carotid disease.37-39 The mechanism for ISR is primarily myointimal hyperplasia, and the potential benefit of phosphodiesterase-3 inhibition on smooth muscle cell growth deserves full attention and evaluation within the context of carotid restenosis. Effective medical therapy might be most beneficial for patients who progress to occlusion despite multiple attempts at revascularization. We acknowledge that this study is limited in its retrospective design and the duration of follow-up. The similarities between the treatment groups at baseline and over time as shown in Tables I and II, in addition to the modelling techniques for risk adjustment that we have applied, minimize the impact of selection bias on these results. The VQI data does not contain records of the duration between the primary CAS and secondary procedure (redo-CAS or CEA). Hence, we are unable to evaluate the impact of early or late restenosis on these outcomes. In assessing preoperative degree of stenosis, the VQI uses the highest degree of stenosis reported from duplex, magnetic resonance angiogram, computed tomography angiogram, or angiogram, thus, creating a potential for differences in assessment and reporting for degree of stenosis across the multiple centers that contribute to VQI. It is also possible that some of the preceding carotid stents were placed for the treatment of restenosis after CEA, thus, making the index comparison one of a tertiary procedure. We are also unable to ascertain the causes of death and transient or persistent nature of cranial nerve injury. This study contains data up to 1 year. As such our findings might not be generalizable beyond that time point. The relatively small sample of patients in some of the subgroups analyzed in this study attracts wide CIs. These leave room for further study in a larger sample of patients with more complete follow-up. Despite these limitations, this study is novel in its evaluation of contemporary outcomes in an understudied cohort of patients who underwent redo-CAS or CEA after prior CAS.

CONCLUSIONS In this study, we have reported adverse event rates for CEA and CAS after prior CAS and shown no significant difference in perioperative and 1-year outcomes between both groups. However, CEA is offered to patients who are more severely ill than redo-CAS, resulting in significantly higher absolute mortality. We recommend avoidance of CEA especially in asymptomatic patients with serious systemic disease. Tight management of diabetes and

10

Arhuidese et al

Journal of Vascular Surgery ---

smoking cessation remain potent targets for outcomes improvement in redo-CAS patients.

AUTHOR CONTRIBUTIONS Conception and design: IA, MM Analysis and interpretation: IA, BN, SC, SL, TO, CH, MM Data collection: Not applicable Writing the article: IA, BN, MM Critical revision of the article: IA, BN, SC, SL, TO, CH, MM Final approval of the article: IA, BN, SC, SL, TO, CH, MM Statistical analysis: IA Obtained funding: Not applicable Overall responsibility: MM

REFERENCES 1. Goodney PP, Lucas FL, Travis LL, Likosky DS, Malenka DJ, Fisher ES. Changes in the use of carotid revascularization among the Medicare population. Arch Surg 2008;143:170-3. 2. Skerritt MR, Block RC, Pearson TA, Young KC. Carotid endarterectomy and carotid artery stenting utilization trends over time. BMC Neurol 2012;12:17. 3. Brott TG, Hobson RW 2nd, Howard G, Roubin GS, Clark WM, Brooks W, et al. Stenting versus endarterectomy for treatment of carotid-artery stenosis. N Engl J Med 2010;363:11-23. 4. Mintz GS, Popma JJ, Hong MK, Pichard AD, Kent KM, Satler LF, et al. Intravascular ultrasound to discern devicespecific effects and mechanisms of restenosis. Am J Cardiol 1996;78:18-22. 5. Kastrati A, Mehilli J, Dirschinger J, Pache J, Ulm K, Schuhlen H, et al. Restenosis after coronary placement of various stent types. Am J Cardiol 2001;87:34-9. 6. Lal BK, Beach KW, Roubin GS, Lutsep HL, Moore WS, Malas MB, et al. Restenosis after carotid artery stenting and endarterectomy: a secondary analysis of CREST, a randomised controlled trial. Lancet Neurol 2012;11:755-63. 7. Arquizan C, Trinquart L, Touboul PJ, Long A, Feasson S, Terriat B, et al. Restenosis is more frequent after carotid stenting than after endarterectomy: The EVA-3S study. Stroke 2011;42:1015-20. 8. Christiaans MH, Ernst JM, Suttorp MJ, van den Berg JC, Overtoom TT, Kelder JC, et al. Restenosis after carotid angioplasty and stenting: a follow-up study with duplex ultrasonography. Eur J Vasc Endovasc Surg 2003;26:141-4. 9. Setacci C, Pula G, Baldi I, de Donato G, Setacci F, Cappelli A, et al. Determinants of in-stent restenosis after carotid angioplasty: a case-control study. J Endovasc Ther 2003;10: 1031-8. 10. Yadav JS, Roubin GS, Iyer S, Vitek J, King P, Jordan WD, et al. Elective stenting of the extracranial carotid arteries. Circulation 1997;95:376-81. 11. Zhou W, Lin PH, Bush RL, Peden EK, Guerrero MA, Kougias P, et al. Management of in-sent restenosis after carotid artery stenting in high-risk patients. J Vasc Surg 2006;43:305-12. 12. Chakhtoura EY, Hobson RW II, Goldstein J, Simonian GT, Lal BK, Haser PB, et al. In-stent restenosis after carotid angioplasty-stenting: Incidence and management. J Vasc Surg 2001;33:220-5; discussion: 225-6. 13. Lal BK, Hobson RW 2nd, Goldstein J, Geohagan M, Chakhtoura E, Pappas PJ, et al. In-stent recurrent stenosis after carotid artery stenting: Life table analysis and clinical relevance. J Vasc Surg 2003;38:1162-8; discussion: 1169. 14. Cronenwett JL, Kraiss LW, Cambria RP. The Society for Vascular Surgery Vascular Quality Initiative. J Vasc Surg 2012;55:1529-37.

2017

15. Willfort-Ehringer A, Ahmadi R, Gschwandtner ME, Haumer M, Lang W, Minar E. Single-center experience with carotid stent restenosis. J Endovasc Ther 2002;9:299-307. 16. Setacci C, de Donato G, Setacci F, Pieraccini M, Cappelli A, Trovato RA, et al. In-stent restenosis after carotid angioplasty and stenting: a challenge for the vascular surgeon. Eur J Vasc Endovasc Surg 2005;29:601-7. 17. Reimers B, Tubler T, de Donato G, Della Barbera M, Cernetti C, Schluter M, et al. Endovascular treatment of in-stent restenosis after carotid artery stenting: immediate and midterm results. J Endovasc Ther 2006;13:429-35. 18. Reedy FM, Colonna M, Genovese V, Mancuso M, Napoleone M, Notari P, et al. Successful surgical treatment of two patients with restenosis after previous stenting of the carotid artery. Eur J Vasc Endovasc Surg 2000;20:99-101. 19. de Borst GJ, Zanen P, de Vries JP, van de Pavoordt ED, Ackerstaff RG, Moll FL. Durability of surgery for restenosis after carotid endarterectomy. J Vasc Surg 2008;47:363-71. 20. Akin E, Knobloch K, Pichlmaier M, Haverich A. In-stent restenosis after carotid stenting necessitating open carotid surgical repair. Eur J Cardiothorac Surg 2004;26:442-3. 21. Wang FW, Esterbrooks D, Kuo YF, Mooss A, Mohiuddin SM, Uretsky BF. Outcomes after carotid artery stenting and endarterectomy in the Medicare population. Stroke 2011;42:2019-25. 22. Grimm JC, Arhuidese I, Beaulieu RJ, Qazi U, Perler BA, Freischlag JA, et al. Surgeon’s 30-day outcomes supporting the carotid revascularization endarterectomy versus stenting trial. JAMA Surg 2014;149:1314-8. 23. Arhuidese IJ, Obeid T, Hicks CW, Yin K, Canner J, Segev D, et al. Outcomes after carotid artery stenting in hemodialysis patients. J Vasc Surg 2016;63:1511-6. 24. Cooper M, Arhuidese IJ, Obeid T, Hicks CW, Canner J, Malas MB. Perioperative and long-term outcomes after carotid endarterectomy in hemodialysis patients. JAMA Surg 2016;151:947-52. 25. Groschel K, Riecker A, Schulz JB, Ernemann U, Kastrup A. Systematic review of early recurrent stenosis after carotid angioplasty and stenting. Stroke 2005;36:367-73. 26. AbuRahma AF, Abu-Halimah S, Bensenhaver J, Dean LS, Keiffer T, Emmett M, et al. Optimal carotid duplex velocity criteria for defining the severity of carotid in-stent restenosis. J Vasc Surg 2008;48:589-94. 27. Setacci C, Chisci E, Setacci F, Iacoponi F, de Donato G. Grading carotid intrastent restenosis: a 6-year follow-up study. Stroke 2008;39:1189-96. 28. Ricotta JJ, Aburahma A, Ascher E, Eskandari M, Faries P, Lal BK, et al. Updated society for vascular surgery guidelines for management of extracranial carotid disease. J Vasc Surg 2011;54:e1-31. 29. Leger AR, Neale M, Harris JP. Poor durability of carotid angioplasty and stenting for treatment of recurrent artery stenosis after carotid endarterectomy: an institutional experience. J Vasc Surg 2001;33:1008-14. 30. Willfort-Ehringer A, Ahmadi R, Gruber D, Gschwandtner ME, Haumer A, Haumer M, et al. Arterial remodeling and hemodynamics in carotid stents: a prospective duplex ultrasound study over 2 years. J Vasc Surg 2004;39:728-34. 31. Qazi U, Obeid T, Arhuidese I, Malas M. Carotid artery stent continued expansion days after deployment, without post stent deployment angioplasty. Clin Pract 2015;5:767. 32. Hicks CW, Canner JK, Arhuidese I, Glebova NO, Schneider E, Qazi U, et al. Development of a duplex-derived velocity risk prediction model of disease progression in patients with moderate asymptomatic carotid artery stenosis. J Vasc Surg 2014;60:1585-92. 33. McCabe DJ, Pereira AC, Clifton A, Bland JM, Brown MM; CAVATAS Investigators. Restenosis after carotid angioplasty,

Journal of Vascular Surgery Volume

-,

Number

Arhuidese et al

11

-

stenting, or endarterectomy in the carotid and vertebral artery transluminal angioplasty study (CAVATAS). Stroke 2005;36:281-6. 34. Arhuidese I, Obeid T, Nejim B, Locham S, Hicks CW, Malas MB. Stenting versus endarterectomy after prior ipsilateral carotid endarterectomy. J Vasc Surg 2017;65: 1-11. 35. Bonati LH, Ederle J, McCabe DJ, Dobson J, Featherstone RL, Gaines PA, et al. Long-term risk of carotid restenosis in patients randomly assigned to endovascular treatment or endarterectomy in the carotid and vertebral artery transluminal angioplasty study (CAVATAS): long-term follow-up of a randomised trial. Lancet Neurol 2009;8:908-17. 36. Obeid T, Arhuidese I, Gaidry A, Qazi U, Abularrage C, Goodney P, et al. Beta-blocker use is associated with lower stroke and death after carotid artery stenting. J Vasc Surg 2016;63:363-9.

37. Douglas JS Jr, Holmes DR Jr, Kereiakes DJ, Grines CL, Block E, Ghazzal ZM, et al. Coronary stent restenosis in patients treated with cilostazol. Circulation 2005;112:2826-32. 38. Iftikhar O, Oliveros K, Tafur AJ, Casanegra AI. Prevention of femoropopliteal in-stent restenosis with cilostazol: a metaanalysis. Angiology 2016;67:549-55. 39. Miyazaki Y, Mori T, Iwata T, Aoyagi Y, Tanno Y, Kasakura S, et al. Continuous daily use of cilostazol prevents in-stent restenosis following carotid artery stenting: serial angiographic investigation of 229 lesions. J Neurointerv Surg 2016;8:471-5. Submitted Jul 6, 2016; accepted Nov 17, 2016.

Additional material for this article may be found online at www.jvascsurg.org.

DISCUSSION Dr Paul DiMuzio (Philadelphia, Pa). You probably do not have any way of knowing the details of why they received a carotid stent to begin with, but in your experience, what is your algorithm as far as how you decide which method you use to treat a patient with restenosis after stenting? Dr Mahmoud B. Malas. I follow a conservative algorithm. First, I maximize medical management with dual antiplatelet and statin, targeting LDL less than 70 mg/dL. I only intervene on lesions with severe progressive stenosis or on symptomatic patients. I place a new close cell stent if I encounter stent fracture. In situation with intimal hyperplasia, ballooning might be sufficient without adding a second stent. Carotid endarterectomy or bypass is useful in situation where the stent is kinked. Dr Ali AbuRahma (Charleston, WVa). I enjoyed your talk as usual; however, I have a little reservation regarding the data analysis. I have done close to two dozen redo

carotid reinterventions after stenting over the last 10 years and in most of these cases it is necessary to go so high in the carotid artery, that I do not understand how you can only have a 3% cranial nerve injury rate. I appreciate your comment. Dr Malas. Thank you, Dr AbuRahma. There could be an underreporting of cranial nerve injury. However, this does not imply that the entire data is bad. We as the treating vascular surgeons are responsible of reporting our outcomes. It is possible that a few of us might not focus on cranial nerve injury. Maybe because there is no immediate treatment available to make things better and maybe because the majority of these cranial nerve injuries will resolve with time. However, the focus outcomes of this analysis are stroke and death and not cranial nerve injury. While some can under report cranial nerve injuries, it is not possible to under report stroke or death.