Endovascular treatment in patients with cervical or intracranial isolated internal carotid artery occlusion: A systematic review and meta-analysis

Abstract

The optimal treatment strategy for isolated internal carotid artery occlusion (IICAO) presenting as acute ischemic stroke (AIS) remains uncertain because these patients were largely excluded from pivotal thrombectomy trials. We compared endovascular treatment (EVT) with best medical treatment (BMT) for IICAO, assessing functional independence, mortality, and safety, and explored outcomes by occlusion site (cervical vs intracranial). Following PRISMA guidelines (PROSPERO CRD420251004624), PubMed, Embase, and Cochrane Library were searched through September 2025. Eligible studies enrolled adults with IICAO treated with EVT or BMT and reported ≥1 predefined outcome: modified Rankin Scale (mRS) 0–2 at 90 days, 90-day mortality, or symptomatic intracranial hemorrhage (sICH). Data were pooled using Mantel–Haenszel random-effects models, reporting odds ratios (ORs) with 95% confidence intervals (CIs). Risk of bias was assessed with ROBINS-I. Five studies including 1531 patients (878 EVT; 653 BMT) met inclusion criteria. EVT patients were younger and had more severe strokes. Pooled analysis showed no significant difference in 90-day functional independence between EVT and BMT (OR 1.78; 95% CI 0.99–3.21; I² = 71%), and adjusted analyses attenuated the effect (OR 1.22; 95% CI 0.82–1.82). No significant differences were found for 90-day mortality (OR 0.84; 95% CI 0.64–1.09; I² = 0%) or sICH (OR 1.48; 95% CI 0.72–3.07; I² = 0%). Subgroup analyses by occlusion site yielded similar neutral results. Current evidence does not demonstrate superiority of EVT over BMT for IICAO, though a possible benefit for intracranial occlusions cannot be excluded. These findings remain hypothesis-generating and emphasize the need for dedicated randomized trials.

Introduction

Isolated internal carotid artery occlusion (IICAO) is a vascular condition where there is a blockage of the internal carotid artery (ICA), representing a complex challenge in the management of acute ischemic stroke (AIS), often leading to substantial neurological impairment. ¹ Unlike individuals presenting with tandem occlusions in the anterior circulation (TL), those with IICAO were not represented in any of the randomized controlled trials (RCTs) investigating the safety and efficacy of endovascular treatment (EVT) for anterior circulation AIS. Specifically, only four patients with IICAO were included in MR CLEAN, one in REVASCAT, and none in the ESCAPE trial. The optimal treatment approach remains uncertain, with EVT and conventional best medical treatment (BMT) being the primary strategies evaluated in recent studies.^1–3 EVT has been established as the standard of care for LVOs; however, its role in IICAO remains less defined due to the anatomical and hemodynamic characteristics of these occlusions. ⁴

Several factors influence the decision-making process, as time from symptom onset, collateral circulation, and individual patient characteristics.^1,5 Observational studies and clinical trials have reported mixed findings regarding the efficacy and safety of EVT compared to BMT in this population. While EVT has demonstrated potential in improving early neurological outcomes, concerns persist regarding its long-term impact on mortality and functional outcomes.^2–5 Given these uncertainties, a comprehensive synthesis of the available evidence is necessary to guide clinical decision-making.

In this context, we conducted a systematic review and meta-analysis to compare EVT and BMT in the management of IICAO. Also, we aimed to analyze the impact of EVT across IICAO subgroups by occlusion site: extracranial cervical ICA (C1 segment) and intracranial ICA (C2–C6 segments).

Methods

This study was conducted and reported in accordance with the Cochrane Collaboration Handbook for Systematic Reviews of Interventions and the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) guidelines. ⁶ The protocol for this review was prospectively registered with the International Prospective Register of Systematic Reviews (PROSPERO) under registration number CRD420251004624.

Inclusion criteria

The inclusion criteria were as follows: (1) adult patients (≥18 years) with IICAO; (2) Intervention: EVT, regardless of device type, thrombectomy technique, or anesthesia modality, reflecting real-world procedural variability; (3) Comparator: BMT; (4) Outcomes: reported at least one predefined outcome of interest, such as favorable functional outcome (mRS 0–2) at 90 days, symptomatic intracranial hemorrhage (ICH) or 90-day mortality; and (5) study design: RCTs or observational studies (retrospective or prospective) published in peer-reviewed journals. BMT, as described in the included studies, comprised intravenous thrombolysis when indicated, antithrombotic therapy, and optimal supportive stroke care based on national and international guidelines.

Exclusion criteria

The exclusion criteria were as follows: (1) case reports, review articles, letters, editorials, and conference abstracts; (2) studies enrolling populations other than adult patients with IICAO; (3) studies that did not include a direct comparison between EVT and BMT; or (4) studies that failed to report any of the predefined outcomes or provide extractable data suitable for quantitative synthesis.

Data source and search strategy

We systematically searched PubMed, Embase, and the Cochrane Central Register of Controlled Trials, with the search updated most recently in September 2025. The complete search strategy included the following terms: (“Internal Carotid Artery” OR “distal ICA” OR “cervical ICA” OR “carotid-T/L” OR “intracranial ICA”) AND (“Endovascular” OR “Thrombectomy”) AND (“Isolated”). All retrieved records were independently reviewed by two authors, who used the Rayyan platform for title and abstract screening. ⁷

Duplicates were removed, and studies that did not directly address the research question were excluded. Full texts of potentially relevant studies were then examined by both authors, and their inclusion or exclusion was discussed based on predefined criteria. Any disagreements were resolved through discussion with a third researcher, to ensure consensus. Additionally, references from eligible papers and systematic reviews were checked for further relevant studies. Searches also included conference abstracts and prospective trials to identify additional studies of interest.

Data extraction

Data extraction was conducted independently by two reviewers, who collected key information from each study, including authors, year of publication, study location, study design, sample size, patient age, and baseline characteristics. When applicable, information on intervention techniques and perioperative complications was also recorded. Discrepancies were resolved with the assistance of a third researcher to ensure consistency and accuracy. Data were managed using an Excel spreadsheet, allowing for systematic organization and comparison across studies.

Outcomes

The outcomes of interest in this study were (1) favorable functional outcomes; (2) mortality at 90 days; and (3) any ICH, including hemorrhagic transformation.

Favorable functional outcomes were defined as an mRS score of 0–2 at 90 days. Symptomatic intracranial hemorrhage (sICH) included any hemorrhagic transformation detected on follow-up imaging with deterioration in NIHSS increase ⩾ 4-points. Mortality at 90 days was considered all-cause mortality occurring within the follow-up period.

Quality assessment

The risk of bias and quality assessment of the included randomized studies were performed using the Risk of Bias in Non-Randomized Studies—of Interventions (ROBINS-I) tool. ⁸ Each study was categorized as “low risk,” “moderate risk,” or “serious risk” based on assessments of specific domains. The evaluations were independently carried out by two authors, with any discrepancies resolved through discussion and consensus.

Statistical analysis

We compared the results using the odds ratio (OR) for binary outcomes with 95% confidence intervals (CI) and p-value <.05 was considered statistically significant. For the pooled analysis of outcomes, we used the Mantel–Haenszel method with a random-effects model. This model was selected a priori, anticipating substantial clinical and methodological heterogeneity among the included studies. Such variability was expected due to differences in study designs, patient populations, eligibility criteria, and treatment practices over time. The random-effects model accounts for between-study variability, providing a more conservative and generalizable estimate of the average effect. We also used the I² statistic and Cochran’s Q test to assess heterogeneity; I² values ≥40% were considered indicative of significant heterogeneity, while values ≥70% were considered high heterogeneity. A leave-one-out sensitivity analysis was performed for outcomes with significant heterogeneity to ensure that the results were not dependent on a single study and to evaluate studies that had high contributions to the heterogeneity of those endpoints when I² ≥ 40%. A meta-regression to formally explore sources of heterogeneity was not performed due to the small number of studies, which would have severely limited the statistical power of such an analysis. All analyses were conducted on version R software, version 4.4.2 (R Foundation for Statistical Computing, Vienna, Austria).

Results

A total of 507 records were initially identified through PubMed (n = 149), Embase (n = 349), and Cochrane (n = 9). After removing 127 duplicate records, 380 studies underwent title and abstract screening, resulting in the exclusion of 174 studies. Full-text assessment was conducted for six studies, of which two were excluded—one due to overlapping data and another for lacking relevant outcomes. Based on Figure 1, five studies met the eligibility criteria and were included in the final review.^9–13

Figure 1.

PRISMA flow diagram.

Baseline characteristics of the included studies and patients

Four observational studies and one post hoc analysis of pooled RCT data were included, published between 2021 and 2025. Study and participant characteristics, including design, interventions, demographics, stroke assessment scores, and comorbidities, are summarized in Table 1. A total of 1531 patients were enrolled across the Netherlands, Germany, USA, Switzerland, Canada, France, the United Kingdom, Australia, and several European, North American, and Asian centers, as well as Israel. Of these, 878 (57.3%) underwent EVT and 653 (42.7%) received BMT. Overall, 550 (62.6%) of patients in the EVT group and 436 (66.8%) in the BMT group were male. The median age of participants ranged from 62 to 72 years across EVT group and from 65 to 83 years in BMT group. Baseline stroke severity, assessed with the NIHSS, varied widely: EVT groups had median scores ranging from 9 (IQR 5–15) to 18 (IQR 16–20), while BMT groups ranged from 5 (IQR 2–13) to 17 (IQR 14–21). ASPECTS was reported in all studies, typically showing medians between 8.5 and 10, with most patients scoring above 7. Baseline disability (pre-stroke mRS) was reported in all studies, though in different formats. In the majority of EVT and BMT patients, the pre-stroke mRS was 0–1. Only Marto et al. ¹³ reported mRS dichotomized as 0–2 versus 3–5, with favorable baseline status in 86.9% of EVT and 85.5% of BMT patients. Regarding comorbidities, atrial fibrillation was present in 14–32% on EVT patients and 23–40% on BMT patients. Diabetes mellitus ranged from 12% to 30.8% in EVT and from 12.5% to 35% in BMT. Dyslipidemia was more heterogeneously reported, ranging from 45.5% to 53% across EVT groups and from 43.5% to 58% in BMT groups. Smoking history varied substantially, between 25.7% and 33.3% in EVT patients and 10% to 36.4% in BMT patients. Arterial hypertension proportion ranged from 23.1% to 78% on the EVT group, and ranged from 40% to 79% on the BMT group. A history of prior stroke or TIA was reported in three studies, ranging from 16.7% to 18.9% in EVT patients and from 12.5% to 26% in BMT patients. Further baseline information is detailed in Table 1.

Table 1.

Baseline characteristics of included studies.

Study, year	Population	Groups	Country	Design	Age, y mean ± SD or median (IQR)	Sex (male, %)	ASPECTS (median, IQR)	Pre-stroke mRS, median (IQR) or %	NIHSS median (IQR)	Atrial fibrillation (%)	Diabetes mellitus (%)	Dyslipidemia (%)	Smoking history (%)	Arterial hypertension (%)	AVC/TIA history (%)
Hoving, 2021	51	EVT = 41	Netherlands	Retrospective cohort	62 (55–76)	30 (73.2%)	10 (9–10)	0 (0–0)	15 (8–21)	6 (14)	5 (12%)	NA	12 (29%)	12 (30%)	NA
		BMT = 10			83 (76–85)	7 (70%)	9 (8–10)	1 (0–3)	17 (14–21)	4 (40)	3 (30%)	NA	1 (10%)	4 (40%)	NA
Kaiser, 2023	38	EVT = 14	Germany, USA, Australia, Netherlands, Canada, France, United Kingdom	Post hoc analysis of pooled RCT data	68.4 (60.3–77.8)	8 (57.1%)	8.5 (6.0–9.8)	mRS = 0 [71.4% (5/7)] mRS = 1 [28.6% (2/7)]	18.0 (16.0–20.0)	2/8 (25)	4/13 (30.8%)	5/11 (45.5%)	NA	3/13 (23.1%)	2/12 (16.7%)
		BMT = 24			72.5 (64–82.5)	8 (33.3%)	9.0 (6.5–10.0)	mRS = 0 [78.6% (11/14)] mRS = 1 [21.4% (3/14)]	16.5 (13.0–19.3)	4 (28.6)	3 (12.5%)	10/23 (43.5%)	NA	17/24 (70.8%)	3 (12.5%)
Kaiser, 2024	90	EVT = 45	Germany, USA, Switzerland	Multicenter retrospective cohort	66 (58–76)	26 (58%)	>7 = 43 (96%)	0 (0–1)	9 (5–15)	12 (27%)	13 (29%)	24 (53%)	13 (29%)	34 (76%)	8 (18%)
		BMT = 45			65 (55–78)	28 (62%)	>7 = 42 (93%)	0 (0–1)	9 (4–15)	14 (31%)	9 (20%)	26 (58%)	13 (29%)	29 (64%)	7 (16%)
Marto, 2025	998	EVT = 487	Multinational (Europe, North America and Israel)	Multicenter retrospective cohort	71.3 ± 13.5	315 (64.7%)	10 (9–10)	mRS 0–2 = 423 (86.9%) mRS 3–5 = 64 (13.1%)	13 (7–18)	147 (30.2%)	129 (26.5%)	224 (46.0%)	125 (25.7%)	360 (73.9%)	92 (18.9%)
		BMT = 511			70.9 ± 12.9	346 (67.7%)	10 (8–10)	mRS 0–2 = 437 (85.5%) mRS 3–5 = 74 (14.5%)	5 (2–13)	118 (23.1%)	136 (26.6%)	286 (56.0%)	186 (36.4%)	382 (74.5%)	133 (26.0%)
Meyer, 2025	354	EVT = 291	Multinational (Europe and Asia)	Multicenter retrospective cohort	72 (60–81)	171 (58.8%)	9.5 (8–10)	0 (0–1)	13 (8–19)	94 (32.3)	26 (18.2%)	NA	75 (33.3%)	226 (78%)	NA
		BMT = 63			72 (59 - 82)	47 (75%)	9 (8–10)	0 (0–1)	11.5 (6–20)	23 (36)	22 (35%)	NA	21 (33)	50 (79%)	NA

Pooled analysis of included studies

Favorable functional outcomes (mRS 0–2) at 90 days

In the meta-analysis comparing EVT and BMT regarding favorable functional outcomes, there was no statistically significant difference between the groups (OR 1.78; 95% CI 0.99–3.21; I² = 71.1%; Figure 2). In the subgroup analysis for statistical adjustment in the included studies, there was no statistically significant difference between the groups (OR 1.22; 95% CI 0.82–1.82; I² = 17.7%; Figure 2) in the adjusted analysis, but there was a statistically significant higher rates of mRS 0–2 in EVT for the unadjusted analysis (OR 3.19; 95% CI 1.40–7.27; I² = 17.7%; Figure 2).

Figure 2.

Meta-analysis comparing EVT versus BMT regarding favorable functional outcome with subanalysis for adjusted vs. unadjusted data.

¹Included extracranial cervical ICA in 100% of patients; ²Included extracranial cervical ICA in 68.64% of patients.

In the subgroup analysis for ICA occlusion site, there was no statistically significant difference between the groups in cervical ICA (OR 1.25; 95% CI 0.36–4.26; I² = 69.2%; Figure 3) or intracranial ICA (OR 2.09; 95% CI 0.97–3.08; I² = 73.9%; Figure 3).

Figure 3.

Meta-analysis comparing EVT versus BMT regarding favorable functional outcome with subanalysis for occlusion sites of ICA.

¹Included extracranial cervical ICA in 100% of patients; ²Included extracranial cervical ICA in 68.64% of patients.

Symptomatic intracranial hemorrhage

In the meta-analysis comparing EVT and BMT regarding symptomatic intracranial hemorrhage, there was no statistically significant difference between the groups (OR 1.48; 95% CI 0.72–3.07; I² = 0.0%; Figure 4). In the subgroup analysis for statistical adjustment in the included studies, there was no statistically significant difference between the groups (OR 1.34; 95% CI 0.56–3.23; I² = 0.0%; Figure 4) in the adjusted analysis or unadjusted analysis (OR 2.01; 95% CI 0.32–12.62; I² = 45.7%; Figure 4).

Figure 4.

Meta-analysis comparing EVT versus BMT regarding sICH with subanalysis for adjusted vs. unadjusted data.

¹Included extracranial cervical ICA in 100% of patients; ²Included extracranial cervical ICA in 68.64% of patients.

In the subgroup analysis for ICA occlusion site, there was no statistically significant difference between the groups in cervical ICA (OR 3.78; 95% CI 0.35–40.70; I² = 57.6%; Figure 5) or intracranial ICA (OR 1.17; 95% CI 0.48–2.82; I² = 0.0%; Figure 5).

Figure 5.

Meta-analysis comparing EVT versus BMT regarding sICH with subanalysis for occlusion sites of ICA.

¹Included extracranial cervical ICA in 100% of patients; ²Included extracranial cervical ICA in 68.64% of patients.

90-day mortality

In the meta-analysis comparing EVT and BMT regarding 90-day mortality, there was no statistically significant difference between the groups (OR 0.84; 95% CI 0.64–1.09; I² = 0.0%; Figure 6). In the subgroup analysis for statistical adjustment in the included studies, there was no statistically significant difference between the groups (OR 0.73; 95% CI 0.31–1.71; I² = 70.2%; Figure 6) in the adjusted analysis or unadjusted analysis (OR 0.74; 95% CI 0.49–1.12; I² = 0.0%; Figure 6).

Figure 6.

Meta-analysis comparing EVT versus BMT regarding 90-day mortality with subanalysis for adjusted vs. unadjusted data.

¹Included extracranial cervical ICA in 100% of patients; ²Included extracranial cervical ICA in 68.64% of patients.

In the subgroup analysis for ICA occlusion site, there was no statistically significant difference between the groups in cervical ICA (OR 0.73; 95% CI 0.31–1.71; I² = 70.2%; Figure 7) or intracranial ICA (OR 0.74; 95% CI 0.49–1.12; I² = 0.0%; Figure 7).

Figure 7.

Meta-analysis comparing EVT versus BMT regarding 90-day mortality with subanalysis for occlusion sites of ICA.

¹Included extracranial cervical ICA in 100% of patients; ²Included extracranial cervical ICA in 68.64% of patients.

Sensitivity analysis through leave-one-out analysis

Notably, omitting the study by Kaiser 2023 ⁹ reduced the heterogeneity to 49.1% (OR, 1.46 [0.79–2.73]; Figure S1), indicating this study was a major contributor to overall heterogeneity. The persistence of wide CIs and lack of statistical significance in all leave-one-out scenarios suggest substantial variability across included studies.

Risk of bias assessment

The individual assessment of each study is shown in Supplemental Material Figures S2 and S3. Hoving et al. ¹¹ had a low risk of bias, while Kaiser et al. (2024) ¹⁰ was rated as serious risk due to confounding and selective reporting. Meyer et al. (2025), ¹² Kaiser et al. (2023), ⁹ and Marto et al. (2025) ¹³ were classified as having a moderate risk of bias, with concerns mainly related to confounding, selection bias, and, in the case of Marto et al., ¹³ also measurement and reporting. Although derived from randomized controlled trials, Kaiser et al. (2023) is a retrospective individual patient data meta-analysis from the HERMES collaboration and was therefore assessed using the ROBINS-I tool, consistent with non-randomized studies. Overall, the risk of bias was mostly moderate, with the exception of Kaiser et al. (2024), ¹⁰ which had a serious risk, and Hoving et al. (2021), ¹¹ which had a low risk. The main concerns across studies were confounding, selection bias, and missing data or non-blinded outcome assessment.

Discussion

This systematic review and meta-analysis evaluated the available evidence comparing EVT with best BMT in patients with IICAO. Overall, pooled analyses did not demonstrate statistically significant differences between EVT and BMT in terms of functional independence at 90 days, mortality, or sICH.^9–13 These findings should be interpreted cautiously, as the available evidence is predominantly observational and characterized by substantial clinical and methodological heterogeneity. This cautious interpretation is further supported by the substantial between-study statistical heterogeneity observed in the primary analysis, indicating considerable variability in treatment effects across cohorts and limiting the generalizability of pooled estimates.

In unadjusted analyses, EVT was associated with numerically higher rates of favorable functional outcomes; however, this association was attenuated and lost statistical significance after adjustment for baseline confounders in studies reporting multivariable analyses.^9,10,12,13 Adjustment methods varied across studies and were limited by the covariates available in each dataset, raising concern for residual confounding. These findings suggest that patient selection plays a major role in observed outcome differences between treatment strategies in IICAO.

Baseline imbalances between EVT and BMT cohorts further complicate interpretation. Patients treated with EVT were generally younger and had better pre-stroke functional status but presented with higher baseline stroke severity, whereas patients managed with BMT more frequently had advanced age and a higher burden of vascular comorbidities.^9–13 Given the strong prognostic influence of these factors, incomplete comparability between groups likely influenced pooled estimates independently of treatment effect.

Subgroup analyses according to the occlusion site suggested numerically higher effect estimates favoring EVT in intracranial compared with cervical ICA occlusions, although these differences did not reach statistical significance.^9–13 Owing to limited sample sizes and wide confidence intervals, these findings should be regarded as hypothesis-generating rather than confirmatory.

Current American Heart Association and American Stroke Association guidelines recommend EVT for selected patients with intracranial internal carotid artery occlusion presenting with acute ischemic stroke. ¹⁴ However, these recommendations are largely derived from randomized trials of terminal ICA or proximal large-vessel occlusions and do not specifically address isolated ICA occlusion without distal middle cerebral or anterior cerebral artery involvement. Consequently, the applicability of existing guideline recommendations to IICAO remains uncertain.

The apparent discrepancy between the neutral findings of the present analysis and the established benefit of EVT in tandem lesions likely reflects differences in pathophysiology and patient selection rather than conflicting evidence. In tandem occlusions, distal large-vessel involvement and larger ischemic territories may increase the potential benefit of reperfusion, whereas IICAO often presents with preserved downstream circulation.^4,15 In this context, clinical outcome in IICAO may depend less on immediate recanalization and more on the capacity to sustain collateral flow through the circle of Willis and external carotid artery pathways, which may partially explain the attenuated treatment effect observed with EVT in this population. ¹⁶

Safety outcomes should also be interpreted with caution. Although no statistically significant difference in sICH was observed between EVT and BMT, definitions of sICH varied across studies, and event rates were low.^9–13 In addition, EVT techniques, device generations, and operator experience evolved substantially over time, introducing heterogeneity and potential confounding by treatment era. Distal embolization during EVT in the setting of preserved downstream circulation has also been reported and may adversely affect outcomes.^17,18

Taken together, the current evidence does not provide definitive support for or against EVT in IICAO. Rather, the findings highlight the limitations of the available observational literature and support the interpretation of this meta-analysis as hypothesis-generating.

Limitations

This study has several limitations. First, the evidence base is predominantly observational, which introduces selection bias and limits causal inference despite statistical adjustment in some studies. Second, substantial clinical heterogeneity was present, including differences in patient characteristics, occlusion site, imaging confirmation, EVT techniques, and outcome definitions.

Incomplete comparability between EVT and BMT cohorts, variability in sICH definitions, and evolution of EVT technology and practice patterns over time may have influenced pooled estimates. Although most studies adhered to definitions based on NIHSS deterioration, nuances in imaging protocols and exact threshold criteria may have varied, thereby limiting the validity of pooling these safety data.

Collectively, these methodological inconsistencies, together with the evolution of endovascular technology and treatment strategies over time, introduce additional sources of bias and constrain external validity. For instance, the included studies spanned periods during which multiple generations of thrombectomy devices and stents were in use, and peri-procedural management practices evolved. Our analysis was not sufficiently powered to stratify results by technological era or device type, representing an unmeasured confounding factor.

In addition, small sample sizes and limited numbers of studies reduced statistical power, particularly in subgroup analyses, and precluded formal assessment of publication bias. These limitations should be considered when interpreting the results.

Conclusion

In patients with isolated internal carotid artery occlusion, this systematic review and meta-analysis did not demonstrate a clear difference between endovascular treatment and best medical treatment in functional outcomes, mortality, or symptomatic intracranial hemorrhage. Given the observational nature of the available data, substantial heterogeneity, and baseline imbalances between treatment groups, these findings should not be interpreted as definitive evidence against EVT. Instead, they should be regarded as hypothesis-generating and underscore the need for well-designed randomized controlled trials to clarify the role of EVT in this distinct clinical entity.

Consent to participate

All participants provided informed consent to participate in this study.

Consent for publication

All authors consent to the publication of this manuscript.

Appendix.

Abbreviations and acronyms

ACA

Anterior cerebral artery

AIS

Acute ischemic stroke

AHA/ASA

American Heart Association/American Stroke Association

ASPECTS

Alberta Stroke Program Early CT Score

BMT

Best medical treatment

CI

Confidence interval

EVT

Endovascular treatment

ICH

Intracranial hemorrhage

ICA

Internal carotid artery

IICAO

Isolated internal carotid artery occlusion

LVO

Large-vessel occlusion

MCA

Middle cerebral artery

MH

Mantel–Haenszel

mRS

Modified Rankin Scale

NIHSS

National Institutes of Health Stroke Scale

OR

Odds ratio

PRISMA

Preferred Reporting Items for Systematic Reviews and Meta-Analyses

PROSPERO

International Prospective Register of Systematic Reviews

RCT

Randomized Controlled Trial

ROBINS-I

Risk of Bias in Non-Randomized Studies—of Interventions

sICH

Symptomatic intracranial hemorrhage

TL

Tandem lesions

Ethics approval

This study is a systematic review and meta-analysis of previously published data and does not involve any new studies with human participants or animals conducted by the authors. Therefore, approval by an Ethics Committee/Institutional Review Board (IRB) was not required.

ORCID iDs

Luciano Falcão https://orcid.org/0000-0003-4521-9041

Lucca Tamara Alves Carretta https://orcid.org/0009-0005-1942-2363

Helvécio Neves Feitosa Filho https://orcid.org/0009-0002-1950-6629

Ahmet Günkan https://orcid.org/0000-0002-6236-5633