You are here

  • Home
  • Archive
  • Volume 11, Issue 9
  • Cumulative incidence of restenosis in the endovascular treatment of extracranial carotid artery stenosis: a meta-analysis

Article Text

Article menu

Download PDFPDF

Head and Neck
Original research
Cumulative incidence of restenosis in the endovascular treatment of extracranial carotid artery stenosis: a meta-analysis
  1. Pierre Clavel1,
  2. Solène Hebert2,
  3. Suzana Saleme1,
  4. Charbel Mounayer1,3,
  5. Aymeric Rouchaud1,3,
  6. Benoit Marin4,5,6
  1. 1 Service de Radiologie, CHU Limoges, Limoges, France
  2. 2 Service de Neurologie, CHU Limoges, Limoges, France
  3. 3 CNRS, XLIM, UMR 7252, Univ. Limoges, Limoges, France
  4. 4 CHU Limoges, Centre d’Epidémiologie de Biostatistique et de Méthodologie de la Recherche, Limoges, France
  5. 5 Tropical Neuroepidemiology, INSERM, UMR1094, Limoges, France
  6. 6 Tropical Neuroepidemiology, Institute of Neuroepidemiology and Tropical Neurology, CNRS FR 3503 GEIST, Univ. Limoges, UMR_S 1094, Limoges, France
  1. Correspondence to Dr Pierre Clavel, Service de radiologie, CHU Limoges, Limoges 87000, France; pierreclavel{at}hotmail.fr

Abstract

Objective To assess the cumulative incidence of restenosis and stroke after stenting for cervical carotid artery stenosis.

Methods We reviewed PubMed, ScienceDirect, and Scopus and included all studies reporting restenosis after stenting. The cumulative incidence of restenosis at 6 and 12 months was calculated. We also estimated the cumulative incidence of ipsilateral stroke within 30 days after stenting. Random effect meta-analysis and meta-regression were performed using relevant study level covariates. Sources of heterogeneity were investigated.

Results Among 7765 records, 40 studies were selected. 15 943 patients and 16 337 carotid arteries were considered. The overall pooled cumulative incidence of restenosis >50% at 12 months was 5.7% (95% CI 3.8% to 8.6%), >70% at 12 months was 5.2% (95% CI 3.3% to 8.2%), >50% at 6 months was 3.9% (95% CI 2.2% to 6.8%), and ipsilateral stroke within 30 days after stenting was 1.6% (95% CI 1.0% to 2.5%) without association with the use of an embolic protection device. We did not identify any relevant source of heterogeneity of the cumulative incidence of restenosis >50% at 12 months. Mean age explained 80.9% (R2=80.9%, p=0.01) of heterogeneities of restenosis >70% at 12 months. The presence of hostile neck explained 53.9% (R2=53.9%, p=0.03) of heterogeneities of restenosis >50% at 6 months.

Conclusion This meta-analysis showed a low cumulative rate of restenosis at 12 months and ipsilateral stroke within 30 days after stenting. Older patients and those with hostile neck present a lower risk of in-stent restenosis. The use of an embolic protection device was not associated with a lower risk of stroke.

  • restenosis
  • extra cranial carotid
  • stent
  • cumulative incidence
  • endovascular

http://dx.doi.org/10.1136/neurintsurg-2018-014534

Statistics from Altmetric.com

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Introduction

    Carotid endarterectomy is the reference treatment in the secondary prevention of stroke for patients with symptomatic carotid stenosis.1 However, the rise of neurointerventional technics and the emergence of new endovascular material have enabled angioplasty with stenting to be a reliable alternative option.2 Endovascular treatment offers a non-inferior treatment for patients with comorbidities that could increase their surgical risk (eg, cardiac failure, respiratory insufficiency, or kidney failure).3

    However, in-stent restenosis after endovascular carotid stenosis treatment is still an issue and a challenging question, and the risk of restenosis at 1 year is not well known.4 Also, embolic complications from the atheromatous plaque are a weakness of the stenting method but are supposed to be prevented when using adjuvant endovascular protection during the procedure.5 Its status is, however, still controversial.6

    Among the different types of stents, closed-cell stents have been found to be superior to open-cell stents in preventing the risk of embolic migration.7

    Even if a prospective clinical trial is the best method to study restenosis and post-stenting complications, no estimated data are accurately known and a meta-analysis is first needed.

    The primary objective of this study was to evaluate the cumulative incidence of restenosis, defined as stenosis >50% at 1 year after angioplasty with stenting, diagnosed using angiography, angio-CT, MR angiography, or Doppler ultrasound.

    Secondary objectives were: (1) identification of the cumulative incidence of severe restenosis (>70%) at 1 year and moderate restenosis (>50%) at 6 months; (2) evaluation of new ipsilateral ischemic stroke within 30 days after stenting; and (3) evaluation of sources of heterogeneity of the abovementioned incidences.

    Methods

    We applied the PRISMA guidelines.8 As this meta-analysis did not include patients but only publications on restenosis, informed consent of patients was not required. Ethics committee approval was not applicable.

    Definitions

    The definitions of stenosis, hostile neck, a representative cohort, and stroke are detailed in the online supplementary Method.9–11

    Search strategy

    A systematic search was performed in PubMed, ScienceDirect, and Scopus up to December 2017. More details are described in the online supplementary Method and Supplementary Panel I.

    All references identified were imported into Zotero Version 5.0.52 and duplicates were deleted.

    Eligibility criteria

    Systematic reviews were not included but their references were examined. Proceedings of conferences were not included. To ensure the highest level of completeness, we included: (1) randomized controlled trials and observational cohorts; and those with (2) a follow-up period of at least 6 months; (3) number of subjects >20; (4) restenosis defined by a reduction in the vessel luminal diameter of at least 50%; and (5) appropriate methodology for restenosis evaluation.

    All studies reporting the incidence of restenosis were included except (1) those that did not report mandatory methodological details (no time to restenosis provided, method for restenosis diagnostic not provided); (2) those with a definition of restenosis that was not adapted; (3) those with the time to restenosis not adapted; (4) studies with ≤20 subjects; and (5) when multiple publications were found (with the same sample), the more recent publication was used.

    Full texts were examined by two independent authors (PC, SH). In case of disagreement, the issue was resolved by discussion.

    Data extraction from included studies

    During the collection of data from articles included in the review, we used a checklist whose items are described in online supplementary table I, focused on all aspects of the study that can influence the quality of the methodology and results.

    Data analysis

    Meta-analysis

    Random effect meta-analyses were performed to produce Forest plots for the cumulative incidence of stenosis at the different time points. Pooled estimates were calculated. Weights were based on the precision of the estimates for each study (ie, SE of the incidence). 95% CI and SE were calculated with exact method; logit transformation was used to stabilize variances. The I2 value was also calculated.12

    Meta-regression

    Potential sources of heterogeneity were evaluated using weighted random effects meta-regression by the DerSimonian and Laird method.13 The Freeman–Tukey arcsin transformation was used.14

    As recommended by Cochrane guidelines, meta-regression was performed only in the presence of significant heterogeneity (ie, Cochran’s Q-test statistically significant) when at least 10 studies were included.15 The proportion of between-study variance, which was explained by the study level covariate, was computed in terms of the R2 index. Two-sided p values <0.05 were considered statistically significant. Analyses were performed using Stata V.11.1 (Stata Corporation, College Station, Texas, USA).

    Results

    Included studies

    The literature search identified 5194 articles. After screening the titles and abstracts and exclusion of duplicates, 131 full-text articles were considered. Following a comprehensive examination of the full texts, 40 articles were finally included (see online supplementary references 26–65) (figure 1, see online supplementary table I). Reasons for non-inclusion of articles are listed in the flowchart (figure 1).

    Figure 1
    Figure 1

    Flow diagram.

    Most studies (92.5%) used Doppler ultrasound criteria for restenosis.

    Patient and treatment characteristics

    A total of 15 943 patients and 16 337 arteries in 40 studies were considered. The mean proportion of men was 68.9±8.7% and the mean age was 70.1±2.7 years (see online supplementary table II). Most of the patients had multiple cardiovascular risk factors with a mean proportion with arterial high blood pressure of 73.1±15.3%, diabetes of 30.6±12.8%, and previous and/or active smoking of 47.6±19.0%. The mean proportion of symptomatic carotid stenosis at baseline was 49.5±19.6%.

    The mean length of follow-up was 23.0±14.8 months, with 11.6±19.4% carotid arteries lost to follow-up.

    These data are also available for the different meta-analyses in online supplementary table II.

    The mean time of double antiplatelet therapy was 2.1±3.3 months, with a large variability from 1 to 12 months. Most of the studies (26/32) chose to maintain double antiplatelet therapy for 1 month. Only three studies maintained it for 12 months.

    In seven studies, only closed-cell stents were used (907 patients). None of the included studies used only open-cell stents.

    The mean proportion of the use of an embolic protection device was 53.3±46.6%.

    Restenosis

    Cumulative incidence of restenosis >50% at 12 months

    A total of 6641 carotid arteries were considered in 22 studies. The overall pooled cumulative incidence of restenosis was 5.7% (95% CI 3.8% to 8.6%; I2=82.8%, p<0.001; figure 2).

    Figure 2
    Figure 2

    Meta-analysis: Forest plots and pooled estimated incidence of restenosis >50% at 12 months.

    A subgroup of studies using only closed-cell stents (Wallstent, Boston Scientific, Natick, Massachusetts, USA; and Palmaz, Cordis Endovascular, a Johnson and Johnson company, Warren, New Jersey, USA) represented a total of 907 carotid arteries in seven studies. In this subgroup, the cumulative incidence of restenosis was 5.2% (95% CI 1.9% to 12.9%; I2=77.1%, p<0.001). There was no statistically significant difference in terms of the cumulative incidence of restenosis >50% at 12 months between this subgroup and the other studies using both open-cell and closed-cell stents.

    No subgroup analysis was possible for the studies using only open-cell stents.

    Cumulative incidence of restenosis at different time points

    The overall pooled cumulative incidence of restenosis >50% at 6 months was 3.9% (95% CI 2.2% to 6.8%; I2=54.4%, p=0.012, n=12 studies, 1835 carotid arteries; figure 3A).

    Figure 3
    Figure 3

    (A) Meta-analysis: Forest plots and pooled estimated incidence for restenosis >50% at 6 months (B) Meta-analysis: Forest plots and pooled estimated incidence for restenosis >70% at 12 months.

    The overall pooled cumulative incidence of restenosis >70% at 12 months was 5.2% (95% CI 3.3% to 8.2%; I2=90.5%, p<0.001, n=10 studies, 7298 carotid arteries; figure 3B).

    Overall, most restenosis was asymptomatic with a mean proportion of 69.7±33.8%.

    Cumulative incidence of ipsilateral stroke within 30 days after stenting

    The overall pooled cumulative incidence of ipsilateral stroke within 30 days after stenting was 1.6% (95% CI 1.0% to 2.5%; I2=87.0%, p<0.001, n=29 studies, 11 740 carotid arteries; figure 4).

    Figure 4
    Figure 4

    Meta-analysis: Forest plots and pooled estimated incidence for ipsilateral stroke within 30 days after stenting.

    In a subgroup with studies using only closed-cell stents, the overall pooled cumulative incidence of ipsilateral stroke within 30 days after stenting was 0.5% (95% CI 0.1% to 1.8%; I2=85%, p<0.001, n=11 studies, 1507 carotid arteries; see online supplementary figure I). No statistically significant difference in the incidence of stroke was observed between the closed-cell stent subgroup and the other subgroup (studies using closed-cell and open-cell stents).

    In a subgroup with studies using systematically an embolic protection device, the overall pooled cumulative incidence of ipsilateral stroke within 30 days after stenting was 0.9% (95% CI 0.4% to 2.2%; I2=91.4%, p<0.001, n=13 studies, 5249 carotid arteries; see online supplementary figure II). In a subgroup with studies in which no embolic protection device was used, the overall pooled cumulative incidence of ipsilateral stroke within 30 days after stenting was 1.2% (95% CI 0.3% to 4.0%; I2=78.8%, p<0.001) in eight studies (948 carotid arteries) (see online supplementary figure II). However, there was no statistically significant difference in the incidence of stroke between these subgroups.

    Evaluation of sources of heterogeneity

    An additional analysis was performed to explore sources of heterogeneity related to (1) the methodology covariates and (2) the mean characteristics of the patients.

    Cumulative incidence of restenosis >50% at 12 months

    We did not identify any significant source of heterogeneity in relation to study methodology or characteristics of the study population. There was, however, a tendency (p=0.08, R2=13.6%) for a lower incidence when the cohort was representative of the usual population treated for carotid stenosis (n=20, pooled incidence 5.22%, 95% CI 3.34% to 8.05%) compared with non-representative cohorts (n=2, pooled incidence 13.36%, 95% CI 7.62% to 22.39%; table 1).

    View this table:
    Table 1

    Meta-regression of cumulative incidence of internal carotid artery restenosis

    Cumulative incidence of restenosis >70% at 12 months

    We identified that heterogeneity in the cumulative incidence of restenosis >70% at 12 months was significantly related to the mean age of the patients, with a lower incidence of restenosis in older patients (R2=80.9%, p=0.01, see online supplementary figure III). For example, the incidence estimated by the meta-regression was 4.8% for 70 years and 1.1% for 80 years. A tendency was identified (p=0.06, R2=34.7%) for a lower incidence for single-center cohorts (n=8, pooled incidence 4.26%, 95% CI 2.53% to 7.10%) compared with multicenter cohorts (n=2, pooled incidence 9.62%, 95% CI 7.35% to 12.51%).

    Cumulative incidence of restenosis >50% at 6 months

    We identified that heterogeneity in the cumulative incidence of restenosis >50% at 6 months was significantly related to the presence of patients with hostile neck in the study populations (R2=53.9%, p=0.03), with a lower incidence of restenosis rates when there was a presence of a hostile neck population in the studies (n=8, pooled incidence 2.77%, 95% CI 1.67% to 4.55%) compared with non-hostile neck population cohorts (n=4, pooled incidence 8.71%, 95% CI 4.38% to 16.51%).

    Cumulative incidence of ipsilateral stroke within 30 days after stenting

    We identified that heterogeneity in the cumulative incidence of ipsilateral stroke within 30 days after stenting was related to the multicenter nature of the cohorts (R2=20.6%, p=0.03), with a lower incidence of ipsilateral stroke rates for single-center cohorts (n=25, pooled incidence 0.99%, 95% CI 0.53% to 1.83%) compared with multicenter cohorts (n=4, pooled incidence 4.76%, 95% CI 2.60% to 8.84%) (table 2).

    Moreover, there was a tendency (p=0.07, R2=12.4%) for a higher incidence when there was the presence of a hostile neck population in the studies (n=24, pooled incidence 2.41%, 95% CI 1.60% to 3.60%) compared with non-hostile neck population cohorts (n=5, pooled incidence 0.04%, 95% CI 0.01% to 0.68%).

    View this table:
    Table 2

    Meta-regression of ipsilateral stroke within 30 days after stenting

    Discussion

    Main results

    This work, based on 15 943 patients and 16 337 arteries, highlights the low cumulative incidence of restenosis of different degrees and at different time points. Indeed, with less than 6% cumulative incidence of restenosis at 1 year, this result makes stenting an efficient treatment compared with surgical endarterectomy, for which restenosis occurs in up to 9% of patients.16 However, active follow-up of all stented arteries seems to be warranted. Also, it is important to keep in mind that most cases of restenosis were clinically asymptomatic (mean proportion of asymptomatic restenosis 69.7±33.8%).

    Stroke during or after carotid artery stenting is the main issue that makes the method still debatable. This meta-analysis shows that the cumulative incidence of ipsilateral stroke within 30 days after stenting is low with 1.6% cumulative incidence at 30 days, and confirms other publications.5

    We did not identify any relevant source of heterogeneity of cumulative incidence >50% at 12 months in relation to study methodology or characteristics of the study population. However, we identified that age explained 80.9% (R2=80.9%, p=0.01) of heterogeneities of restenosis >70% at 12 months and the presence of hostile neck explained 53.9% (R2=53.9%, p=0.03) of heterogeneities of restenosis >50% at 6 months. It appears that endovascular treatment is more efficient in older patients and in the hostile neck population. Hence, an old population could be a preferential target for the endovascular treatment of extracranial carotid artery stenosis. Patients with a hostile neck should also be oriented to an endovascular approach, although a study showed that hostile neck would be a risk factor for in-stent restenosis.17 These points need further investigation.

    The finding that older people had a lower cumulative incidence of restenosis >70% at 12 months might be explained by lower intimal cell proliferation and thus lower intimal hyperplasia.18 This supposition has yet to be proved, but follows the findings of Mousa et al who previously reported that age <65 years was a significant predictor of restenosis.18

    In the current meta-analysis, the use of an embolic protection device was not associated with a lower risk of ipsilateral stroke within 30 days after stenting. However, in the included studies the use of a protection device was not randomly assigned, and we cannot eliminate the potential bias that a protection device might have been preferentially used in challenging cases or associated with other clinical characteristics that may have affected clinical outcomes. This finding needs to be confirmed in a prospective randomized trial with and without the use of a protection device for carotid stenting.

    We identified a lower incidence of ipsilateral stroke within 30 days after stenting in single-center studies than in multicenter studies. This might be related to a tendency to minimize the incidence of complications when self-evaluated by the authors. Since most of the included studies were single-center series, we can assume our results might have been affected by this potential bias.

    Issues in the reporting of previous reports

    This work allowed us to identify recurrent issues in the included reports. Of the 131 full-text articles screened for eligibility, we had to reduce the number of studies to include in the quantitative analysis mainly due to unadapted restenosis definition and time to restenosis (35 studies, 26.7%). Forty-five potential articles were excluded because of unclear methodology. This heterogeneity in definitions and inappropriate methodology reduced our number of treated carotid arteries and hence the precision of our estimates. In those studies, restenosis was not the specific outcome being investigated. Ideally, studies should use a clear and uniform definition of restenosis and identical follow-up schedules. Pizzolato et al suggest, after a systematic literature review, a definition of in-stent restenosis >70% using ultrasound Doppler measures.19

    Among the included studies there was a high rate of patients lost to follow-up (11.6±19.4%), which might affect the reported incidences.

    Limitations of study

    Our study has some limitations. First, while we explored the sources of heterogeneity of our results (design, time period, cohort representativeness, age, the presence of hostile neck in the population studies, cardiovascular risk factors, stent design, and the use of an embolic protection device), we were unable to identify sources of heterogeneity for our main outcome. Second, some subgroup analyses have been performed despite the fact that only a few studies were available (seven studies for the subgroup cumulative incidence of restenosis >50% at 12 months in closed-cell stents, eight studies for the subgroup cumulative incidence of ipsilateral stroke within 30 days after stenting not using an embolic protection device). Third, in the included studies, none used only open-cell stents, so we were not able to perform any statistical analysis. Fourth, potential sources of heterogeneity such as the risk of hemorrhagic complications due to double antiplatelet therapy were not evaluated because of a lack of data in the included studies. Fifth, the number of treated carotid arteries instead of the number of treated patients was used as a denominator, as has been done in the included studies. It would have been interesting to look at these results using results based on a patient approach. Sixth, we focused our study on restenosis at 1 year because the risk of recurrent stenosis after stenting has been reported to be higher in the first year after the procedure and then decreases over time.20 Hence, unlike Texakalidis et al in their recent meta-analysis, we did not report the incidence of restenosis at 2 years.21

    Strengths of study

    Despite these limitations, our study relies on its intrinsic strengths. First, the main strength of this work relies on the literature search, which was broad, without limitations in terms of date of publication or language, and exhaustive. Second, we assessed the impact of study methodology and other characteristics (including diagnostic method, type of stent) on heterogeneity of our estimates and confirmed the robustness of our results. Third, the articles included in this review had none or few missing values. Fourth, we had the possibility to create a database that took the main covariable into account. This allowed us to report accurate and reliable results that are highly clinically relevant. These results are essential when considering a future randomized clinical trial.

    A clue for future research

    The risk of recurrent stenosis after stenting has been reported to be greatest in the first year after the procedure and then decreases over time. It appears that neointimal hyperplasia is the major pathophysiologic process leading to early restenosis after carotid revascularization procedures.20 In support of this notion, neointimal proliferation prevailed up to 12 months after stenting whereas no further relevant changes in the neointima were observed during the second year.22 Neointima hyperplasia is the result of strain injury caused by the stent’s radial force against the artery intima.23 A new generation of stent (with a lower radial force) could potentially reduce the incidence of restenosis after stenting and might be preferred in the youngest population which seems to present a higher risk of restenosis.24 Nevertheless, the cumulative incidence of restenosis at 1 year is low and it would require a very large number of patients to highlight a different rate of restenosis between different stents in a standard population.

    Conclusion

    This meta-analysis showed a low cumulative rate of restenosis at 12 months and ipsilateral stroke within 30 days after stenting. Older patients and those with hostile neck present a lower risk of in-stent restenosis. The use of an embolic protection device was not associated with a lower risk of stroke.

    References

    1. ↵
      2. Ferguson GG ,
      3. Eliasziw M ,
      4. Barr HW , et al
      . The North American Symptomatic Carotid Endarterectomy Trial: surgical results in 1415 patients. Stroke 1999;30:1751–8.
      OpenUrlAbstract/FREE Full Text
    2. ↵
      2. Brott TG ,
      3. Hobson RW ,
      4. Howard G , et al
      . Stenting versus endarterectomy for treatment of carotid-artery stenosis. N Engl J Med 2010;363:11–23.doi:10.1056/NEJMoa0912321
      OpenUrlCrossRefPubMedWeb of Science
    3. ↵
      2. Borenstein M ,
      3. Hedges L ,
      4. Higgins J , et al
      . Introduction to 701 meta-analysis. Chichester: Wiley, 2009.
    4. ↵
      2. Gandini R ,
      3. Del Giudice C ,
      4. Da Ros V , et al
      . Long-term results of drug-eluting balloon angioplasty for treatment of refractory recurrent carotid in-stent restenosis. J Endovasc Ther 2014;21:671–7.doi:10.1583/14-4715MR.1
      OpenUrlCrossRefPubMed
    5. ↵
      2. Kastrup A ,
      3. Gröschel K ,
      4. Krapf H , et al
      . Early outcome of carotid angioplasty and stenting with and without cerebral protection devices: a systematic review of the literature. Stroke 2003;34:813–9.doi:10.1161/01.STR.0000058160.53040.5F
      OpenUrlAbstract/FREE Full Text
    6. ↵
      2. Sztriha LK ,
      3. Vörös E ,
      4. Sas K , et al
      . Favorable early outcome of carotid artery stenting without protection devices. Stroke 2004;35:2862–6.doi:10.1161/01.STR.0000147714.19871.45
      OpenUrlAbstract/FREE Full Text
    7. ↵
      2. Park KY ,
      3. Kim DI ,
      4. Kim BM , et al
      . Incidence of embolism associated with carotid artery stenting: open-cell versus closed-cell stents. J Neurosurg 2013;119:642–7.doi:10.3171/2013.5.JNS1331
      OpenUrlCrossRefPubMed
    8. ↵
      2. Shamseer L ,
      3. Moher D ,
      4. Clarke M , et al
      . Preferred reporting items for systematic review and meta-analysis protocols (PRISMA-P) 2015: elaboration and explanation. BMJ 2015;349:g7647.doi:10.1136/bmj.g7647
      OpenUrlAbstract/FREE Full Text
    9. ↵
      2. Bartlett ES ,
      3. Walters TD ,
      4. Symons SP , et al
      . Quantification of carotid stenosis on CT angiography. AJNR Am J Neuroradiol 2006;27:13–19.
      OpenUrlAbstract/FREE Full Text
    10. ↵
      2. von Reutern GM ,
      3. Goertler MW ,
      4. Bornstein NM , et al
      . Grading carotid stenosis using ultrasonic methods. Stroke 2012;43:916–21.doi:10.1161/STROKEAHA.111.636084
      OpenUrlAbstract/FREE Full Text
    11. ↵
      2. Brott T ,
      3. Adams HP ,
      4. Olinger CP , et al
      . Measurements of acute cerebral infarction: a clinical examination scale. Stroke 1989;20:864–70.doi:10.1161/01.STR.20.7.864
      OpenUrlAbstract/FREE Full Text
    12. ↵
      2. Higgins JP ,
      3. Thompson SG ,
      4. Deeks JJ , et al
      . Measuring inconsistency in meta-analyses. BMJ 2003;327:557–60.doi:10.1136/bmj.327.7414.557
      OpenUrlFREE Full Text
    13. ↵
      2. Thompson SG ,
      3. Higgins JP
      . How should meta-regression analyses be undertaken and interpreted? Stat Med 2002;21:1559–73.doi:10.1002/sim.1187
      OpenUrlCrossRefPubMedWeb of Science
    14. ↵
      2. Freeman MF ,
      3. Tukey JW
      . Transformations related to the angular and the square root. Ann Math Stat 1950;21:607–11.doi:10.1214/aoms/1177729756
      OpenUrlCrossRef
    15. ↵
      2. Higgins J ,
      3. Green S
      . Cochrane handbook for systematic reviews of interventions version 5.1.0. The Cochrane Collaboration. 2011 http://www.cochrane-handbook.org/ (last accessed 9 Jan 2015).
    16. ↵
      2. Oszkinis G ,
      3. Pukacki F ,
      4. Juszkat R , et al
      . Restenosis after carotid endarterectomy: incidence and endovascular management. Interv Neuroradiol 2007;13:345–52.doi:10.1177/159101990701300405
      OpenUrlCrossRefPubMed
    17. ↵
      2. Moon K ,
      3. Albuquerque FC ,
      4. Levitt MR , et al
      . The myth of restenosis after carotid angioplasty and stenting. J Neurointerv Surg 2016;8:1006–10.doi:10.1136/neurintsurg-2015-011938
      OpenUrlAbstract/FREE Full Text
    18. ↵
      2. Mousa AY ,
      3. AbuRahma AF ,
      4. Bozzay J , et al
      . Long-term comparative outcomes of carotid artery stenting following previous carotid endarterectomy vs de novo lesions. J Endovasc Ther 2015;22:449–56.doi:10.1177/1526602815581597
      OpenUrlCrossRefPubMed
    19. ↵
      2. Pizzolato R ,
      3. Hirsch JA ,
      4. Romero JM
      . Imaging challenges of carotid artery in-stent restenosis. J Neurointerv Surg 2014;6:32–41.doi:10.1136/neurintsurg-2012-010618
      OpenUrlAbstract/FREE Full Text
    20. ↵
      2. Toma N ,
      3. Matsushima S ,
      4. Murao K , et al
      . Histopathological findings in a human carotid artery after stent implantation. J Neurosurg 2003;19:199–204.doi:10.3171/jns.2003.98.1.0199
      OpenUrl
    21. ↵
      2. Texakalidis P ,
      3. Giannopoulos S ,
      4. Kokkinidis DG , et al
      . Effect of open- vs closed-cell stent design on periprocedural outcomes and restenosis after carotid artery stenting: a systematic review and comprehensive meta-analysis. J Endovasc Ther 2018;25:523–33.doi:10.1177/1526602818783505
      OpenUrl
    22. ↵
      2. Willfort-Ehringer A ,
      3. Ahmadi R ,
      4. Gschwandtner ME , et al
      . Healing of carotid stents: a prospective duplex ultrasound study. J Endovasc Ther 2003;10:636–42.doi:10.1177/152660280301000333
      OpenUrlCrossRefPubMedWeb of Science
    23. ↵
      2. Siewiorek GM ,
      3. Finol EA ,
      4. Wholey MH
      . Clinical significance and technical assessment of stent cell geometry in carotid artery stenting. J Endovasc Ther 2009;16:178–88.doi:10.1583/08-2583.1
      OpenUrlCrossRefPubMed
    24. ↵
      2. Wissgott C ,
      3. Schmidt W ,
      4. Brandt C , et al
      . Preliminary clinical results and mechanical behavior of a new double-layer carotid stent. J Endovasc Ther 2015;22:634–9.doi:10.1177/1526602815593490
      OpenUrlCrossRefPubMed

    Footnotes

    • AR and BM contributed equally.

    • Contributors Study design: BM, PC, SH, CM, AR, SS. Data collection: PC, SH. Statistical analysis: BM, PC. Critical review: BM, PC, SH, CM, AR, SS.

    • Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.

    • Competing interests None declared.

    • Patient consent Not required.

    • Provenance and peer review Not commissioned; externally peer reviewed.