Article Text
PDF
Abstract
Background Emergent carotid artery stenting (eCAS) is performed during mechanical thrombectomy for acute ischemic stroke due to tandem occlusion. However, the optimal management strategy in this setting is still unclear.
Objective To carry out a systematic review and meta-analysis to investigate the safety and efficacy of eCAS in patients with tandem occlusion.
Methods Systematic review followed the PRISMA guidelines. Medline, EMBASE, and Scopus were searched from January 1, 2004 to March 7, 2022 for studies evaluating eCAS and no-stenting approach in patients with stroke with tandem occlusion. Primary endpoint was the 90-day modified Rankin Scale score 0–2; secondary outcomes were (1) symptomatic intracerebral hemorrhage (sICH), (2) recurrent stroke, (3) successful recanalization (Thrombolysis in Cerebral Infarction score 2b–3), (4) embolization in new territories, and (5) restenosis rate. Meta-analysis was performed using the Mantel-Haenszel method and random-effects modeling.
Results Forty-six studies reached synthesis. eCAS was associated with higher good functional outcome compared with the no-stenting approach (OR=1.52, 95% CI 1.19 to 1.95), despite a significantly increased risk of sICH (OR=1.97, 95% CI 1.23 to 3.15), and higher successful recanalization rate (OR=1.91, 95% CI 1.29 to 2.85). Restenosis rate was lower in the eCAS group than in the no-stenting group (2% vs 9%, p=0.001). Recanalization rate was higher in retrograde than antegrade eCAS (OR=0.51, 95% CI 0.28 to 0.93). Intraprocedural antiplatelets during eCAS were associated with higher rate of good functional outcome (60% vs 46%, p=0.016) and lower rate of sICH (7% vs 11%; p=0.08) compared with glycoprotein IIb/IIIa inhibitors.
Conclusions In observational studies, eCAS seems to be associated with higher good functional outcome than no-stenting in patients with acute ischemic stroke due to tandem occlusion, despite the higher risk of sICH. Dedicated trials are needed to confirm these results.
- Stroke
- Stent
- Stenosis
- Intervention
- Angioplasty
Data availability statement
All data relevant to the study are included in the article or uploaded as supplementary information. Not applicable.
Statistics from Altmetric.com
Introduction
Acute ischemic strokes (AIS) due to tandem occlusions (TOs), defined as cervical internal carotid artery (ICA) high-grade stenosis or occlusion and ipsilateral intracranial occlusion, account for about 10%–15% of all large-vessel occlusions (LVOs). Most of the endovascular randomized trials assessing the efficacy of mechanical thrombectomy (MT) in the treatment of AIS1–5 excluded patients with a TO, mainly due to high heterogeneity in presentation, including diverging sites for intracranial and extracranial occlusion and a broad range of clinical severity, and in treatment approaches.1 6
Despite the HERMES meta-analysis, which showed that the endovascular thrombectomy has an equivalent therapeutic effect in patients with isolated intracranial occlusions and TOs,7 there is a lack of data about the management of the extracranial stenosis or occlusion.8 Overall, although the need for intracranial revascularization is evident, the optimal approach for the treatment of extracranial occlusion is extreme uncertain. Emergent carotid artery stenting (eCAS) is often performed in the context of MT in cases with TO,9 10 even though its safety is unconfirmed. Additionally, the hemorrhagic risk related to periprocedural antithrombotic agents required for eCAS and the risk of early recurrent ischemic events of eCAS and the carotid artery angioplasty (CAA) or no carotid treatment are extremely variable in observational series.11–18
We aimed to provide a systematic review and meta-analysis to assess the efficacy and safety of eCAS in patients with AIS with TOs, in comparison with a no-stenting approach.
Methods
Protocol and registration
This systematic review and meta-analysis was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines.19 The study protocol was registered with PROSPERO: CRD42021279218. The manuscript followed the PRISMA statement.20
Search strategy and selection criteria
We searched Medline, EMBASE and Scopus electronic databases from January 1, 2004 to March 7, 2022, with no language restrictions. Search terms included MeSH and Emtree terms as well as specific keywords for stroke, thrombectomy procedure, and angioplasty/stenting, including: (1) ischemic stroke, stroke, cerebral infarct*, tandem*, tandem occlusion*, (2) thrombectom*, mechanical thrombectomy, (3) carotid, carotid artery, stents, carotid artery stenting, angioplasty, carotid artery angioplasty. Reference lists of relevant studies were also hand searched.
Eligibility criteria and study selection
We included studies assessing clinical and imaging outcomes after eCAS or intracranial MT (alone or in combination with the CAA—overall defined as no-stenting group) in patients with AIS due to a TO. We used the PICOS (Patient population, Intervention, Control, Outcome, Study design) framework for the inclusion criteria: (P) adult patients with AIS-LVO due to a tandem lesion eligible for endovascular thrombectomy, (I) thrombectomy with eCAS, (C) thrombectomy without the eCAS (no-stenting group); (O) good functional outcome, defined as modified Rankin Scale (mRS) score 0–2 after 90 days, mortality, symptomatic intracerebral hemorrhage (sICH), any type of ICH, successful recanalization, clot migration, and early recurrent ischemic stroke. Randomized controlled trials and observational studies were included. We excluded non-English language studies, unpublished studies and conference abstracts, studies including patients with AIS without tandem occlusion, and studies including pediatric patients.
Outcomes
Primary outcome was the proportion of people achieving functional independence, defined as mRS score 0–2 at 90 days.
Secondary outcomes were: (1) rates of sICH, according to study definition, (2) recurrent stroke, (3) successful recanalization, defined as Thrombolysis in Cerebral Infarction score of 2b–3, (4) embolization in new territories (ENT), (5) early restenosis/reocclusion, and (6) stroke recurrence during follow-up.
Study extraction and quality assessment
Two reviewers (FD, GT) independently screened retrieved studies and selected them against prespecified criteria using a check form. Discrepancies were resolved by consensus. Data extraction was performed by two authors (FD, GT) using a predefined electronic data extraction form. Collected data included: study design, sample size, country, patients' baseline characteristics, clinical and neuroradiological data at admission, procedural details, technical and clinical outcomes. A revised Quality Assessment Tool for Diagnostic Accuracy Studies (QUADAS-2) was used to perform the methodological quality assessment of the selected studies (online supplemental table 1) .21 Two authors (FD, GT) performed bias assessment, with controversies resolved by consensus.
Supplemental material
Statistical analysis
For each included study, we calculated prevalence of primary and secondary outcomes in each treatment group (eCAS vs no-stenting). We derived pooled ORs (95% CI) by meta-analysis, using the Mantel-Haenszel method and random-effect modeling. A random effect model was preferred to account for variability in design, sample, variance in study timing and procedural equipment, and timing of ascertainments. Heterogeneity among studies was assessed by the χ² test and I2 statistics, and reported according to guidelines and previous studies.22
Meta-analysis of proportions was implemented whenever fewer than three studies directly comparing eCAS versus no-stenting were available for pooling in meta-analysis of OR. Meta-analysis of proportion was also a priori defined for the primary outcome, to identify potential bias in results due to enrolment of a preferential subpopulation in non-comparative studies. A random-effects model with Freeman-Tukey double arcsine transformation was used for pooling proportions, and the results are presented as prevalence of the chosen outcome in the eCAS group and no-stenting group, compared through a z test.
Meta-regression analysis for the primary outcome was designed for age, Alberta Stroke Program Early CT Score (ASPECTS) score and proportion of people treated with intravenous thrombolysis (IVT). Subgroup analysis was also carried out to assess anterograde (stenting first) and retrograde (MT first) approaches in eCAS cases. R software version 3.3.4 was used for data analysis. Reported probability values are two-sided (significance set at p<0.05).
Results
Characteristics of included studies and bias assessment
Our search detected 1601 citations. Overall, 63 studies were potentially eligible for inclusion, and 46 reached qualitative synthesis11–18 23–60 (figure 1 for PRISMA flow-chart). Among them, 17 observational studies directly compared eCAS and no-stenting, while the remainder focused on a single approach (online supplemental table S1 for included study baseline features).
Supplemental material
Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 flow diagram for new systematic reviews.
Most of the included studies were of moderate to high quality (online supplemental table 1) . The risk of bias was high in some studies, mainly due to ascertainment bias and comparability: 11 studies did not report the intraprocedural or postprocedural medication14 17 25 34 35 44 46 47 49 51 52 (online supplemental figure 1).
Included studies consisted of 6395 patients, 4356 in the eCAS group and 2039 in the no-stenting group. Clinical and technical outcomes are summarized in online supplemental table S2.
Primary outcome
Fifteen studies directly compared the 90-day mRS score 0–2 of both treatment strategies (n=1691). eCAS was associated with a significantly higher good functional outcome than no-stenting (56.3% vs 44.1%, OR=1.52, 95% CI 1.19 to 1.95, figure 2). Among them, restricted analyses of studies reporting the ASPECTS (n=8 studies, OR=1.7, 95% CI 1.26 to 2.29, p<0.001), patients’ age (n=12 studies, OR=1.6, 95% CI 1.17 to 2.19, p<0.001), and proportion of IVT (n=12 studies, OR=1.69, 95% CI 1.34 to 2.13, p<0.001) confirmed this result. Meta-regression analysis revealed the consistency of the point estimates in favor of eCAS, with no impact of age (12 studies, p=0.2), ASPECTS (eight studies, p=0.69), and proportion of people receiving IVT (12 studies, p=0.55).
Meta-analysis of proportions for 90-day modified Rankin Scale score 0–2 of studies directly comparing the emergency carotid artery stenting (eCAS) and no-stenting group.
In an indirect comparison, meta-analysis of proportions for good functional outcome in eCAS and no-stenting groups showed a difference in favor of the former (eCAS, OR=0.54, 95% CI 0.49 to 0.59 vs no-stenting, OR=0.45, 95% CI 0.40 to 0.50, online supplemental figure 2 and 3.
A positive time trend in favor of eCAS emerged for primary outcome and a sensitivity analysis limited to more recent studies (2018 onward) highlighted a consistent benefit of eCAS over no-stenting (OR=1.52, 95% CI 1.19 to 1.95, (online supplemental figure 4).
Clinical secondary outcomes
Thirteen studies reported the occurrence of sICH in the two groups (n=1540). eCAS was associated with a significant increased risk of sICH (OR=1.97, 95% CI 1.23 to 3.15, figure 3).
Meta-analysis of proportions for symptomatic intracerebral hemorrhage of studies directly comparing the emergent carotid artery stenting (eCAS) and no-stenting group.
Only one comparative study reported rates of recurrent stroke.29 Among studies mentioning rates of recurrent stroke after eCAS, only two cases32 of 525 patients from pooled cohorts have been reported. Only two studies reported rates of recurrent stroke in a no-stenting group,15 29 with six events out of 42 patients included. The nature of the data prevented direct comparison(online supplemental figure 5) .
Technical secondary outcomes
Successful recanalization was significantly more frequent in MTs with eCAS than MTs with no-stenting (OR=1.91, 95% CI 1.29 to 2.85, figure 4).
Meta-analysis of proportions for Thrombolysis in Cerebral Infarction 2b–3 of studies directly comparing the emergent carotid artery stenting (eCAS) and no-stenting group.
Rates of ENT were compared in three studies, with no significant differences between the two treatment paradigms, although the sample size consistently limited any conclusion (online supplemental figure 6) .
Rates of cervical ICA restenosis were directly compared in three studies. The eCAS group was associated with a consistently lower risk of carotid restenosis than the no-stenting group (OR=0.39, 95% CI 0.18 to 0.87; online supplemental figure 7). In-stent thrombosis happened in 2% of cases, with data available from eight studies only.
Successful reperfusion was significantly higher in retrograde eCAS than antegrade eCAS (OR=0.51, 95% CI 0.28 to 0.93 in favor of MT first (online supplemental figure 8), although a good functional outcome did not differ between the two groups (online supplemental figure 9).
Antiplatelet therapy
Intraprocedural antiplatelet protocols varied across studies: 15 studies used glycoprotein IIb/IIIa inhibitors (GPI)11 17 18 25 28–31 33 34 37 40 44 48 61 together with the dual antiplatelet therapy after 24 hours, while 13 studies used loading doses of single or dual antiplatelet drugs.4 13 15 23 26 32 36 38 41–43 50 51 Studies using rescue treatment with APTs or GPI after the standard protocol during eCAS were excluded from analysis owing to potential exposure bias.12 16 39 53 59 60 Pooling data from studies reporting sICH rate after intraprocedural use of antithrombotic agents during eCAS (n=20 studies), a marginal non-significant increase in sICH rate emerged in the GPI group compared with the APT group (11% vs 7%, p=0.08,online supplemental figure 10). Meta-regression analysis revealed no effect of previous IVT on the rate of ICH according to antithrombotic use (n=20 studies, p=0.95). Good functional outcome was significantly more frequent in the APT group (OR=60%, 95% CI 53% to 66%) than the GPI group (OR=46%, 95% CI 39% to 54%, ; online supplemental figure 11).
Eleven studies reported the use of intraprocedural heparin in occasional cases, both as e.v. bolus or infusion, ranging from 1000 to 13 000 IU.
Postprocedural APT regimen after eCAS varied among included studies. The association of acetylsalicylic acid and clopidogrel12 13 15 16 23 24 32 41–43 46 50 51 62–64 for 1–6 months was the most frequently used, followed by acetylsalicylic acid in lifelong monotherapy.
Discussion
Despite a consistent amount of evidence on the benefit of MT in AIS due to LVO,1 65 there is still substantial uncertainty about the treatment of TO. Controversies are focused on the treatment of the extracranial ICA stenosis or occlusion,66 since evidence suggests that the eCAS may reduce the risk of early recurrent ischemic events, despite the need for an antiplatelet regimen in the acute settings.
We provided a meta-analysis comparing eCAS with a no-stenting approach in MTs for tandem occlusions. We found 46 observational studies enrolling 6395 with AIS due to TO, reporting clinical and technical outcomes after MTs with eCAS and/or MTs with/without CAA (no-stenting group).
Our meta-analysis suggests a potential benefit of eCAS over a no-stenting approach, with a 12.2% higher chance of reaching good functional outcome. The advantage of eCAS is also confirmed by results of all secondary analyses, including the indirect comparison of the results of meta-analysis of proportions and the meta-regression, which excluded any conditional effect of age, ASPECTS, or previous IVT on such benefit. Such result conflicts with a previous meta-analysis,27 which was limited to data available up to 2018, leaving space for potential improvement of outcomes in the eCAS group over 5 years of refinement of management, device, and technique.
Regarding safety, the risk of sICH remains the main concern of eCAS since the effect of the antiplatelet therapy on patients with AIS, in some cases in conjunction with IVT, is still uncertain. We found an increased risk of sICH with eCAS compared with the no-stenting approach, in line with results of previous studies on eCAS27 51 and of the HERMES meta-analysis.7 However, selection bias for eCAS in patients with a lower risk of sICH is likely, which it is not possible to account for due to the lack of individual patient data. The increase in sICH risk seems counterbalanced by the increase in successful recanalization, which might be a reasonable factor leading to the consistently higher rate of good functional outcome with eCAS found in this study.
Regarding the impact of antiplatelet use on sICH, despite variability in intraprocedural use of antithrombotic agents, our preliminary estimates suggested an increased risk of sICH with GPI compared with APTs in cases of eCAS, with a consistent detrimental impact on long-term good functional outcome. To this extent, the variability of treatment among included studies prevented further analysis, and studies are needed to examine the optimal treatment strategy in this population.
Of note, two-thirds of patients in the eCAS group received bridging IVT. Subgroup analysis of comparative studies reporting proportion of patients receiving IVT and a meta-regression confirmed the clinical benefit of eCAS also in this cohort, without significant impact of previous IVT on the rate of sICH.
Our results add to the previous evidence about the safety and efficacy of eCAS, suggesting that the benefit of this intervention may be driven by the higher recanalization rate seen in the eCAS group. Hence, the increased efficacy of intracranial MT in the eCAS group may be the reason also for the improved clinical outcome, in addition to the reduced risk of early recurrent ischemic events. This meta-analysis seems to confirm the suggestion that this cohort of patients need a dedicated treatment.
Such benefit might be prolonged over time, as demonstrated by the lower risk of cervical ICA restenosis in the eCAS group versus the no-stenting group (2% vs 9%, p=0.001), although the recurrence of restenosis and stroke has been rarely assessed in published studies or limited to 3-months' follow-up.15 32 33 39 43 The need for subsequent extracranial carotid artery treatment in the no-stenting group is not negligible, since 8.3% of patients in that group need non-emergent carotid revascularization during the initial hospital admission, both with CAS or carotid endarterectomy.52
Limitations
Our study has several limitations. First, the observational nature of the studies, with potential selection bias and conflicts due to the devices used in each study. Although we performed meta-regression for some of the variables potentially leading to confounding bias, unmeasured selection bias might still influence in some measure our estimates. Second, although the quality of included studies was acceptable, most of them lacked data for each variable and all outcomes predefined for this meta-analysis. Third, although heterogeneity was acceptable in our analysis, results should be carefully weighted in regard to the evolution of techniques and devices throughout the years, which might produce a progressive increase in quality of eCAS, bringing a potential benefit. Fourth, the endovascular strategies, stenting protocols, and antiplatelet regimens were not standardized. This factor highlights the need for randomized controlled trials, in order to define the optimal treatment strategy.
Conclusions
The results of our study suggest better functional outcome at 3 months in patients with AIS with tandem occlusion receiving eCAS as compared with a no-stenting approach. Emergent CAS may increase the rates of favorable functional outcome and successful recanalization, despite the increased risk of sICH.
Supplemental material
Supplemental material
Supplemental material
Data availability statement
All data relevant to the study are included in the article or uploaded as supplementary information. Not applicable.
Ethics statements
Patient consent for publication
Ethics approval
Not applicable.
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
AZ, DAdS and SP are joint senior authors.
Twitter @fdianamd
Correction notice Since this paper was first published, figure 3 has been updated, because an article was wrongly included in the analysis. The OR, 95 CI and weight categories have been updated. However, point estimate remains unchanged, and only a minor change to 95%CI emerged.
Contributors Conception and design of the work: FD. Data acquisition: FD and GT. Data analysis and interpretation: FD, MR, DAdS. Drafting the work: FD, MR, AR, CM, DGR, FDS, PM. Critical revision: DAdS, AZ, and SP. Final approval: all authors.
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.
Provenance and peer review Not commissioned; internally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.