Transcarotid Artery Revascularization for Asymptomatic Carotid Stenosis: A Systematic Review and Meta-Analysis Authors: Zhi-Cheng Jiang, Qi-Fei Hu, Jia-Ming Huang, Xian-Wei Wang, Tong Lin, Xi-Lin Gao, Jia-Wen Cui, Jing-Lin Zhao, Dong Chen, Shu-Peng Li Published: 2026 ABSTRACT Abstract Management of asymptomatic carotid stenosis remains controversial despite advances in contemporary best medical therapy. Transcarotid artery revascularization (TCAR) has been introduced as an alternative revascularization option alongside carotid endarterectomy (CEA) and transfemoral carotid artery stenting (TF-CAS). Evidence focused specifically on asymptomatic carotid disease, however, remains limited and is derived mainly from observational analyses of clinical databases, with only a small amount of single-center retrospective evidence. The present study therefore examined the available comparative evidence on TCAR for asymptomatic carotid stenosis. A systematic review and meta-analysis was performed of studies reporting outcomes of TCAR in asymptomatic carotid stenosis. Eligible studies included asymptomatic cohorts or mixed populations with extractable asymptomatic subgroup data and reported outcomes after TCAR with comparison against CEA and/or TF-CAS. The primary pooled outcomes were stroke, death, stroke or death, and myocardial infarction at 30 days, together with the corresponding stroke, death, and stroke or death outcomes at approximately 1 year. For TCAR versus CEA, no significant between-group differences were identified in the pooled 30-day risks of stroke (risk ratio (RR), 1.25; 95% CI, 0.98–1.59; I²=82%), death (RR, 0.94; 95% CI, 0.64–1.38; I²=57%), or stroke or death (RR, 1.10; 95% CI, 0.95–1.26; I²=75%). In contrast, myocardial infarction within 30 days occurred less frequently after TCAR (RR, 0.68; 95% CI, 0.48–0.96; I²=71%). At approximately 1 year, pooled estimates likewise showed no significant difference in stroke (RR, 1.27; 95% CI, 0.70–2.29; I²=0%) or death (RR, 1.14; 95% CI, 0.97–1.35; I²=0%), whereas the pooled estimate for stroke or death favored CEA (RR, 1.25; 95% CI, 1.15–1.36; I²=31%). A different pattern emerged in the TCAR versus TF-CAS comparison. TCAR was associated with significantly lower pooled 30-day risks of stroke (RR, 0.60; 95% CI, 0.48–0.73; I²=0%), death (RR, 0.55; 95% CI, 0.38–0.81; I²=20%), and stroke or death (RR, 0.60; 95% CI, 0.52–0.70; I²=6%), while no significant difference was observed for 30-day myocardial infarction (RR, 0.91; 95% CI, 0.65–1.29; I²=44%). At approximately 1 year, stroke remained comparable between the 2 groups (RR, 0.75; 95% CI, 0.46–1.23; I²=26%), whereas death (RR, 1.09; 95% CI, 1.00–1.20; I²=66%) and stroke or death (RR, 0.79; 95% CI, 0.71–0.87; I²=45%) differed, with the pooled stroke-or-death estimate favoring TCAR. Current observational evidence suggests that TCAR compares favorably with TF-CAS in asymptomatic carotid stenosis. When compared with CEA, TCAR does not appear to provide a clear benefit in pooled perioperative neurologic outcomes, although myocardial infarction may occur less frequently after TCAR.Supplementary InformationThe online version contains supplementary material available at 10.1007/s12975-026-01457-2.Keywords: Asymptomatic carotid stenosis, Transcarotid artery revascularization, Carotid endarterectomy, Transfemoral carotid artery stenting, Stroke Introduction Stroke remains a leading cause of death and long-term disability worldwide, and extracranial carotid atherosclerosis continues to account for a substantial proportion of ischemic cerebrovascular events [1–3]. In selected patients with clinically significant carotid stenosis, carotid revascularization remains an established strategy for stroke prevention [4]. Carotid endarterectomy (CEA) continues to serve as the reference procedure, whereas transfemoral carotid artery stenting (TF-CAS) is used as a less invasive alternative in selected circumstances [5]. Transcarotid artery revascularization (TCAR), which combines direct cervical carotid access with flow-reversal neuroprotection, has more recently emerged as a third revascularization approach [6, 7]. Since receiving US Food and Drug Administration approval in 2015, TCAR has been adopted rapidly in clinical practice [7–9]. Within the Vascular Quality Initiative, it has become a substantial component of carotid revascularization at centers offering all 3 treatment modalities [10]. The role of TCAR is particularly difficult to define in asymptomatic carotid stenosis [11, 12]. Although asymptomatic disease accounts for a large share of carotid interventions, the net benefit of intervention in the setting of contemporary medical therapy remains debated [13]. Current Society for Vascular Surgery documents treat asymptomatic carotid stenosis as a distinct clinical context and include CEA, TCAR, and TF-CAS among revascularization options for selected patients with high-grade disease [14, 15]. From a technical perspective, TCAR is appealing because it may preserve the advantages of a stent-based strategy while limiting the embolic risk associated with transfemoral arch navigation [16]. Contemporary database analyses suggest that TCAR compares favorably with TF-CAS and may approach CEA for selected outcomes [17]. In a propensity score-matched VQI analysis, TCAR was associated with lower perioperative stroke or death than TF-CAS and similar 1-year ipsilateral stroke or death compared with CEA [18]. In a later VISION comparative effectiveness study, asymptomatic patients undergoing TCAR had lower 3-year risks of stroke and stroke or death than those treated with TF-CAS [19]. The available literature, however, remains fragmented. Most TCAR studies are observational, many combine symptomatic and asymptomatic populations, and several arise from overlapping VQI-related data sources [20, 21]. Definitions of asymptomatic status are also not uniform across studies [22]. Turner et al. [21] reported that patients with a remote history of transient ischemic attack or stroke had worse in-hospital outcomes than truly asymptomatic patients, indicating that these groups should not be assumed to carry equivalent risk. The present systematic review and meta-analysis was therefore undertaken to evaluate the available evidence on TCAR in asymptomatic carotid stenosis, with particular emphasis on perioperative stroke, death, myocardial infarction, and intermediate- to longer-term stroke-related outcomes. Methods This systematic review and meta-analysis followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) reporting guideline. The protocol was registered in PROSPERO (CRD420261350973). Data Sources and Search Strategy PubMed and Web of Science Core Collection were searched for studies published in the past 10 years up to the final search date. Search terms combined controlled vocabulary and free-text terms related to carotid stenosis and carotid revascularization, including: transcarotid artery revascularization; TCAR; asymptomatic carotid stenosis and related terms (Supplement 1). All retrieved records were imported into EndNote for deduplication and screening. Study Eligibility Criteria and Study Selection Studies were considered eligible if they included patients with asymptomatic carotid stenosis or mixed symptomatic and asymptomatic cohorts with extractable data for the asymptomatic subgroup, evaluated TCAR, incorporated CEA and/or TF-CAS as comparators for quantitative synthesis when available, and reported at least 1 clinically relevant outcome. These outcomes included perioperative or 30-day stroke, death, stroke or death, myocardial infarction, and stroke-related outcomes assessed at approximately 1 year. Studies were excluded if they were review articles, guidelines, narrative reports, utilization-only analyses not aligned with the comparative review objective, device-focused technical reports that did not evaluate TCAR as a revascularization strategy, or studies that did not provide extractable outcome data for asymptomatic patients. StudY Selection and Data Extraction Titles and abstracts were screened initially, and potentially relevant studies subsequently underwent full-text review. For each eligible study, the extracted data included author, year of publication, country, data source, study design, population characteristics, definition of asymptomatic status, sample size, comparator group, duration of follow-up, reported outcomes, and statistical adjustment strategy. Particular attention was given to whether outcome data for the asymptomatic subgroup were directly extractable rather than described only qualitatively. Quality Assessment As the available evidence was nonrandomized and derived largely from registry-based observational studies, the principal threats to validity included confounding by indication, baseline imbalance, heterogeneity in symptom definitions, variation in follow-up duration, and potential cohort overlap within Vascular Quality Initiative–related data sources. Risk of bias in the included studies was assessed using the Newcastle-Ottawa Scale [23], as all eligible comparative evidence was observational in design. This scale assigns up to 4 points for selection of study groups, 2 points for comparability, and 3 points for outcome assessment, yielding a total score ranging from 0 to 9, with higher scores indicating lower risk of bias. Each study was assessed independently by 2 reviewers, and disagreements were resolved through discussion. Outcomes The primary outcomes selected for quantitative synthesis were 30-day stroke, 30-day death, 30-day composite stroke or death, 1-year stroke, and 1-year death, with myocardial infarction analyzed as an additional safety outcome. Because outcome definitions and follow-up intervals beyond 30 days varied across studies, intermediate- and longer-term results, including 1-year ipsilateral stroke or death and 3-year stroke-related outcomes, were summarized narratively rather than pooled broadly. This was particularly relevant because 1 study reported mid-term follow-up of 1.6 ± 1.5 years rather than a prespecified 1-year interval. Statistical Analysis Study results were meta-analyzed using a random-effects model to generate pooled risk ratios (RRs) with corresponding 95% confidence intervals (CIs). Statistical heterogeneity was assessed using the I² statistic and Cochran Q test, with corresponding P values reported. A 2-sided P value < 0.05 was considered statistically significant. Separate analyses were performed for TCAR versus CEA and for TCAR versus TF-CAS. Statistical analyses were performed using R, version 4.4.0 (R Foundation for Statistical Computing). Leave-one-out sensitivity analyses were performed for the main pooled perioperative outcomes by sequentially omitting 1 study at a time and recalculating the pooled estimate and heterogeneity. Results The initial search of PubMed and Web of Science Core Collection yielded 258 citations. After exclusion of 50 records published more than 10 years before the final search date and removal of 26 duplicates, 182 unique citations remained for screening. Following title and abstract review, 154 records were excluded, and 28 articles underwent full-text assessment. Of these, 23 were excluded after full-text review, and the reasons for exclusion are shown in Fig. 1. A total of 5 comparative studies were ultimately included (Fig. 1). PRISMA flow diagram of study selection CharacterisTics of Studies and Patients All included studies were comparative observational investigations. Most were based on large registry or clinical database analyses, with additional evidence contributed by a single-center retrospective cohort. Comparisons included TCAR versus CEA, TCAR versus TF-CAS, or both. Reported outcomes were centered on perioperative or 30-day stroke, death, stroke or death, and myocardial infarction, together with selected stroke or death outcomes assessed at approximately 1 year; detailed outcome definitions were accepted as reported in the original studies. Some studies also provided longer follow-up stroke-related outcomes, although differences in the definition of asymptomatic status and in follow-up duration limited direct comparability across endpoints. Esquetini-Vernon et al. [22] described a single-center retrospective comparative cohort from a US stroke center that included 293 procedures and directly identifiable asymptomatic patients undergoing CEA, TF-CAS, or TCAR, with assessment of 30-day stroke, death, myocardial infarction, and mid-term outcomes. Kibrik et al. [20], Turner et al. [21], Malas et al. [18], and Columbo et al. [8] were registry-based comparative cohort studies derived from the VQI. Kibrik et al. included 28,571 procedures and compared CEA, TF-CAS, and TCAR in octogenarians, focusing on 30-day stroke or death, death, and myocardial infarction using multivariable logistic regression. Turner et al. [21] analyzed 7158 TCAR procedures and 18,023 TF-CAS procedures, with asymptomatic patients separately identifiable, and evaluated in-hospital stroke, death, and stroke or death stratified by symptom status using multivariable logistic regression. Malas et al. [18] included 41,548 CEA procedures, 5725 TCAR procedures, and 6064 TF-CAS procedures, and used propensity score matching and Cox regression to assess 30-day stroke, death, and myocardial infarction, as well as 1-year ipsilateral stroke or death. Columbo et al. [8] studied 21,234 TCAR procedures, 82,737 CEA procedures, and 14,595 TF-CAS procedures, and evaluated perioperative and 1-year stroke or death using logistic regression, Cox regression, and instrumental variable analysis (Table 1). Table 1Study CharacteristicsStudyCountryData sourceStudy typePatients, No.Asymptomatic subgroup separately identifiableComparisonOutcomes assessedAdjustment method Esquetini-Vernon et al., 2026 [22]USSingle-center stroke center cohortRetrospective comparative cohort293 proceduresYesCEA vs. TF-CAS vs. TCAR30-day stroke, death, and MI; mid-term outcomesNo formal PSM or multivariable adjustment reported Kibrik et al., 2022 [20]US/CanadaVQIRegistry-based comparative cohort28,571YesCEA vs. TF-CAS vs. TCAR30-day stroke/death, death, and MIMultivariable logistic regression Turner et al., 2022 [21]USVQIRegistry-based comparative cohortTCAR 7158; TF-CAS 18,023YesTCAR vs. TF-CASIn-hospital stroke, death, and stroke/death, stratified by symptom statusMultivariable logistic regression Malas et al., 2022 [18]USVQIRegistry-based comparative cohortCEA 41,548; TCAR 5725; TF-CAS 6064PartlyTCAR vs. CEA; TCAR vs. TF-CAS30-day stroke, death, and MI; 1-year ipsilateral stroke/deathPropensity score matching; Cox regression COLUMBO et al., 2022 [8]USVQIRegistry-based comparative cohortTCAR 21,234; CEA 82,737; TF-CAS 14,595PartlyTCAR vs. CEA vs. TF-CASPerioperative and 1-year stroke/deathLogistic regression; Cox regression; instrumental variable analysis Study Characteristics RisK of Bias Assessment Risk of bias across the included observational studies was assessed using the Newcastle-Ottawa Scale. The evaluated domains comprised representativeness of the exposed cohort, selection of the nonexposed cohort, ascertainment of exposure, absence of the outcome at study entry, control of important confounders, baseline comparability, outcome assessment, and adequacy of follow-up. Each domain was classified as having either low or high risk of bias. Studies receiving 7 to 9 stars on the Newcastle-Ottawa Scale were considered to have low risk of bias, those receiving 4 to 6 stars were considered to have moderate to high risk of bias, and those receiving 0 to 3 stars were considered to have high risk of bias. For overall interpretation, studies rated as low risk of bias in at least 7 of the 8 assessed domains were considered to have a relatively low overall risk of bias, whereas those with low-risk ratings in fewer domains were considered more susceptible to bias. On this basis, Kibrik et al. [20], Turner et al. [21], Malas et al. [18], and Columbo et al. [8] were judged to have a relatively low overall risk of bias, whereas Esquetini-Vernon et al. [22] was considered to have a higher risk of bias, primarily because no formal adjustment for important confounders or baseline imbalance was reported (Table 2). Table 2Newcastle-Ottawa Scale Scores for Observation StudiesStudyExposed cohort representativenessSelection of nonexposed cohortExposure ascertainmentOutcome does not present at startControlled for important confoundersBaseline comparabilityOutcome assessmentFollow-up long enough Esquetini-Vernon et al., 2026 [22]Low risk of biasLow risk of biasLow risk of biasLow risk of biasHigh risk of biasHigh risk of biasLow risk of biasLow risk of bias Kibrik et al., 2022 [20]Low risk of biasLow risk of biasLow risk of biasLow risk of biasLow risk of biasLow risk of biasLow risk of biasLow risk of bias Turner et al., 2022 [21]Low risk of biasLow risk of biasLow risk of biasLow risk of biasLow risk of biasLow risk of biasLow risk of biasHigh risk of bias Malas et al., 2022 [18]Low risk of biasLow risk of biasLow risk of biasLow risk of biasLow risk of biasLow risk of biasLow risk of biasLow risk of bias COLUMBO et al., 2022 [8]Low risk of biasLow risk of biasLow risk of biasLow risk of biasLow risk of biasLow risk of biasLow risk of biasLow risk of bias Newcastle-Ottawa Scale Scores for Observation Studies Main Results For TCAR versus CEA, pooled analysis showed no significant differences in 30-day stroke (risk ratio (RR), 1.25; 95% CI, 0.98–1.59; I²=82%), death (RR, 0.94; 95% CI, 0.64–1.38; I²=57%), or stroke or death (RR, 1.10; 95% CI, 0.95–1.26; I²=75%) (Fig. 2). In contrast, 30-day myocardial infarction was less frequent after TCAR (RR, 0.68; 95% CI, 0.48–0.96; I²=71%). At approximately 1 year, pooled analysis showed no significant difference between TCAR and CEA in stroke (RR, 1.27; 95% CI, 0.70–2.29; I²=0%) or death (RR, 1.14; 95% CI, 0.97–1.35; I²=0%), whereas the pooled estimate for stroke or death favored CEA (RR, 1.25; 95% CI, 1.15–1.36; I²=31%). These pooled findings were broadly consistent with the propensity score–matched VQI analysis by Malas et al. [18], in which 30-day stroke, death, and stroke or death did not differ significantly between TCAR and CEA, while stroke/death/myocardial infarction favored TCAR (2.30% vs. 3.25%) and myocardial infarction occurred less frequently after TCAR (0.55% vs. 1.12%). At 1 year, Malas et al. [18] reported similar ipsilateral stroke or death between TCAR and CEA in the matched cohort (6.49% vs. 5.68%; hazard ratio (HR), 1.14; 95% CI, 0.95–1.37), whereas Columbo et al. [8] reported 1-year stroke or death rates of 4.9% after TCAR and 3.8% after CEA among asymptomatic patients. Esquetini-Vernon et al. [22], by contrast, reported mid-term rather than strictly 1-year follow-up. TCAR verus CEA For TCAR versus TF-CAS, the pooled estimates consistently favored TCAR for the principal perioperative neurologic outcomes (Fig. 3). TCAR was associated with lower risks of 30-day stroke (RR, 0.60; 95% CI, 0.48–0.73; I²=0%), death (RR, 0.55; 95% CI, 0.38–0.81; I²=20%), and stroke or death (RR, 0.60; 95% CI, 0.52–0.70; I²=6%). No significant difference was observed for 30-day myocardial infarction (RR, 0.91; 95% CI, 0.65–1.29; I²=44%). At approximately 1 year, pooled analysis showed no significant difference in stroke (RR, 0.75; 95% CI, 0.46–1.23; I²=26%), whereas death (RR, 1.09; 95% CI, 1.00–1.20; I²=66%) and stroke or death (RR, 0.79; 95% CI, 0.71–0.87; I²=45%) differed between groups, with the pooled stroke or death estimate favoring TCAR. These findings were consistent with the matched VQI analysis by Malas et al. [18], in which TCAR was associated with a lower perioperative stroke or death rate than TF-CAS (1.83% vs. 2.55%), whereas 1-year ipsilateral stroke or death did not differ significantly overall (6.07% vs. 7.07%; HR, 0.85; 95% CI, 0.71–1.01). Notably, asymptomatic patients in that study had more favorable 1-year outcomes with TCAR than with TF-CAS (HR, 0.78; 95% CI, 0.62–0.98). Longer-term evidence also favored TCAR. In the VISION registry-claims analysis by Columbo et al. [8], asymptomatic patients undergoing TCAR had lower 3-year risks of stroke (5.1% vs. 9.2%) and stroke or death (22.6% vs. 31.4%) than those undergoing TF-CAS; compared with TCAR, TF-CAS was associated with a higher adjusted hazard of stroke (HR, 1.69; 95% CI, 1.25–2.28). TCAR verus TF-CAS NarraTive Findings From Additional Studies Esquetini-Vernon et al. [22] reported no significant differences among CEA, TCAR, and TF-CAS in a single-center cohort of asymptomatic patients with respect to the 30-day composite outcome of ipsilateral stroke, transient ischemic attack, myocardial infarction, or death. No myocardial infarction or transient ischemic attack events were observed. At mid-term follow-up (mean, 1.6 ± 1.5 years), composite and individual outcome rates also did not differ significantly among the groups. Turner et al. [21] examined TCAR and TF-CAS according to symptom status and showed that patients with a remote history of transient ischemic attack or stroke had higher in-hospital event rates than truly asymptomatic patients. These findings indicate that patients with remote neurologic symptoms should not be considered equivalent to patients who are asymptomatic without prior events when interpreting subgroup comparisons across studies [21]. Sensitivity Analyses Leave-one-out sensitivity analyses were conducted for the main pooled perioperative outcomes. For TCAR versus TF-CAS, omission of any single study did not materially affect the pooled estimate for 30-day stroke or death, and the direction and statistical significance of the association remained stable. For TCAR versus CEA, exclusion of individual studies did not materially change the pooled estimates for 30-day stroke or death or 30-day death. Nevertheless, omission of Kibrik et al. [20] markedly reduced heterogeneity, suggesting that this octogenarian-only cohort contributed importantly to between-study heterogeneity (Supplement 2). Discussion The present analysis suggests that TCAR and TF-CAS are not performing similarly in asymptomatic carotid stenosis. Across the pooled perioperative comparisons, TCAR was associated with lower risks of stroke, death, and stroke or death than TF-CAS. The pattern was less favorable when TCAR was compared with CEA. In that comparison, pooled 30-day neurologic outcomes were broadly similar, although myocardial infarction occurred less frequently after TCAR. These observations should be interpreted with caution, given that the available evidence is derived almost entirely from nonrandomized, registry-based studies. Among the 2 main procedural comparisons, the contrast between TCAR and TF-CAS was the more consistent. This is clinically plausible. By avoiding aortic arch traversal and using flow reversal for neuroprotection, TCAR may reduce embolic exposure relative to TF-CAS. The pooled analyses were aligned with that procedural rationale, showing lower 30-day risks of stroke, death, and stroke or death after TCAR, with little heterogeneity across those outcomes. The direction of effect was also consistent with the matched VQI analysis by Malas et al. [18] and with the VISION comparative effectiveness study by Columbo et al. [8], in which TCAR was associated with lower longer-term stroke risk than TF-CAS among asymptomatic patients. The comparison with CEA was more nuanced. CEA remains the benchmark revascularization procedure in most current treatment frameworks, and the available comparative data do not show a clear perioperative neurologic advantage for TCAR in asymptomatic disease. In the pooled 30-day analyses, stroke, death, and stroke or death did not differ significantly between TCAR and CEA, whereas myocardial infarction was less frequent after TCAR. This overall pattern is in line with the matched analysis by Malas et al., which found no significant difference in 30-day stroke, death, or stroke/death between the 2 procedures, but did report lower stroke/death/myocardial infarction and lower myocardial infarction after TCAR. Columbo et al. similarly showed that perioperative stroke or death was slightly higher after TCAR than after CEA in the overall cohort, while adjusted and instrumental variable analyses at 1 year did not show a statistically significant difference. The heterogeneity in the TCAR-versus-CEA analyses is not unexpected. The included studies differed in case mix, study design, symptom classification, follow-up interval, and adjustment strategy. Kibrik et al. is particularly relevant in this respect, because it was restricted to octogenarians and showed that asymptomatic patients aged 80 years or older had a higher adjusted perioperative stroke or death risk after TCAR than after CEA, while myocardial infarction was less frequent after TCAR. That study is clinically informative, but it also differs materially from the broader registry cohorts included in the pooled analyses. The leave-one-out sensitivity analyses support this interpretation: exclusion of Kibrik 2022 substantially reduced heterogeneity in the principal TCAR-versus-CEA perioperative outcomes, even though the pooled estimates themselves did not materially change. Variation in how “asymptomatic” disease was defined adds another layer of uncertainty. Turner et al. showed that patients with a remote history of transient ischemic attack or stroke had worse in-hospital outcomes than patients without prior neurologic symptoms and should not be treated as equivalent to a truly asymptomatic population. Esquetini-Vernon et al. used a 6-month symptom-free interval to define asymptomatic status and reported mid-term follow-up of 1.6 ± 1.5 years rather than a fixed 1-year time point. These are not trivial differences, because pooled estimates can appear more precise than the underlying clinical definitions actually allow. Several limitations should be considered when interpreting these findings. Most importantly, the evidence base is observational and therefore remains susceptible to confounding by indication, treatment-selection bias, and residual imbalance despite adjustment. In addition, several of the largest and most informative studies were drawn from related VQI-derived data environments, raising the possibility of partial cohort overlap. Outcome definitions, comparator structures, and follow-up intervals were also not fully uniform across studies. Some pooled outcomes were based on only a small number of studies, which limits precision and makes those estimates more exploratory than definitive. Finally, although the TCAR-versus-TF-CAS perioperative findings remained stable in sensitivity analysis, the TCAR-versus-CEA comparisons were more heterogeneous and require more cautious interpretation. At the same time, the review addresses a specific gap in the literature by focusing on asymptomatic carotid stenosis rather than combining asymptomatic and symptomatic cohorts. That distinction matters clinically, and it is not applied consistently in prior TCAR studies. Requiring extractable comparative data for the asymptomatic subgroup allowed the quantitative synthesis to remain closer to the clinical question of interest. The combination of outcome-specific meta-analysis, narrative integration of nonpoolable comparative evidence, and leave-one-out sensitivity analysis also provides a more informative picture than either a purely narrative review or an unrestricted pooled analysis. Further comparative work is still needed, especially studies that reduce treatment-selection bias and report reconstructable event counts for asymptomatic subgroups. Greater consistency in the definition of asymptomatic disease would also improve interpretability, particularly for patients with remote neurologic symptoms. Longer-term follow-up remains important as well, because the perioperative advantage of TCAR over TF-CAS does not automatically establish a durable long-term benefit in stroke prevention. Conclusion In asymptomatic carotid stenosis, the currently available evidence suggests that TCAR compares favorably with TF-CAS, particularly with respect to perioperative stroke, death, and stroke or death, with supportive longer-term evidence for lower stroke risk. When compared with CEA, TCAR does not demonstrate a clear perioperative neurologic advantage, although myocardial infarction may occur less frequently after TCAR. These findings should be interpreted cautiously because the evidence base is nonrandomized, derived largely from registry-based observational studies, and subject to heterogeneity in symptom definition, follow-up duration, and possible cohort overlap. At present, TCAR is best regarded as a promising and potentially preferable endovascular option for selected asymptomatic patients rather than as a definitively established replacement for CEA. Supplementary Information Below is the link to the electronic supplementary material. Supplementary Material 1 (DOCX 20.0 KB) Supplementary Material 2 (DOCX 2.04 MB) Author Contributions SP.L. andD.C. designed the study. JM.H. and XW.W. analyzed the data. ZJ.L. and XL.G. verified the correctness of the data. ZC.J. and QF.H. was major contributors to the manuscript. ZC.J. and T.L. conducted a literature search. All authors contributed to the article, read and approved the final manuscript. Funding This work was supported by the following: Dalian Municipal Guidance Program for the Life and Health Sector; the 2025 In-Hospital Independent Research Project of the Provincial Key Specialty Program, Dalian Central Hospital of Dalian University of Technology (No. 2025SZ021 and 2025SZ022); the Dalian Medical Science Research Program(No. 2311002); the National Natural Science Foundation of China (No. 82504180). Data Availability All data analyzed in this study were derived from previously published articles. Declarations Ethics Statement Ethical approval was not required because this study was a systematic review and meta-analysis of previously published studies. Competing Interests The authors declare no competing interests. Footnotes Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. Zhi-Cheng Jiang and Qi-Fei Hu contributed equally to this work. Contributor Information Dong Chen, Email: neuro-chen@dlut.edu.cn. Shu-Peng Li, Email: neuro_lsp@126.com. References Associated Data Supplementary Materials Supplementary Material 1 (DOCX 20.0 KB) Supplementary Material 2 (DOCX 2.04 MB) Data Availability Statement All data analyzed in this study were derived from previously published articles.