Comprehensive Meta-Analysis of Futile Recanalization in Acute Ischemic Stroke Patients Undergoing Endovascular Thrombectomy: Prevalence, Factors, and Clinical Outcomes

Background: Futile recanalization (FR) continues to raise concern despite the success of endovascular thrombectomy (EVT) in acute ischemic stroke (AIS). Understanding the prevalence of FR and identifying associated factors are crucial for refining patient prognoses and optimizing management strategies. Objectives: This study aims to comprehensively assess the pooled prevalence of FR, explore the diverse factors connected with FR, and establish the association of FR with long-term clinical outcomes among AIS patients undergoing EVT. Materials and Methods: Incorporating studies focusing on FR following EVT in AIS patients, we conducted a random-effect meta-analysis to assess the pooled prevalence and its association with various clinical and imaging risk factors linked to FR. Summary estimates were compiled and study heterogeneity was explored. Results: Our comprehensive meta-analysis, involving 11,700 AIS patients undergoing EVT, revealed a significant pooled prevalence of FR at 51%, with a range of 48% to 54% (Effect Size [ES]: 51%; 95% Confidence Interval [CI]: 48–54%; z = 47.66; p < 0.001). Numerous clinical factors demonstrated robust correlations with FR, including atrial fibrillation (Odds Ratio [OR]: 1.39, 95% CI 1.22 1.59; p < 0.001), hypertension (OR 1.65, 95% CI 1.41 1.92; p < 0.001), diabetes mellitus (OR 1.71, 95% CI 1.47 1.99; p < 0.001), previous stroke or transient ischemic attack (OR 1.298, 95% CI 1.06 1.59; p = 0.012), prior anticoagulant usage (OR 1.33, 95% CI 1.08 1.63; p = 0.007), cardioembolic strokes (OR 1.34, 95% CI 1.10 1.63; p = 0.003), and general anesthesia (OR 1.53, 95% CI 1.35 1.74; p < 0.001). Conversely, FR exhibited reduced likelihoods of smoking (OR 0.66, 95% CI 0.57 0.77; p < 0.001), good collaterals (OR 0.33, 95% CI 0.23 0.49; p < 0.001), male sex (OR 0.87, 95% CI 0.77 0.97; p = 0.016), and intravenous thrombolysis (IVT) (OR 0.75, 95% CI 0.66 0.86; p < 0.001). FR was strongly associated with increasing age (standardized mean difference [SMD] 0.49, 95% CI 0.42 0.56; p < 0.0001), baseline systolic blood pressure (SMD 0.20, 95% CI 0.13 0.27; p < 0.001), baseline National Institute of Health Stroke Severity Score (SMD 0.75, 95% CI: 0.65 0.86; p < 0.001), onset-to-treatment time (SMD 0.217, 95% CI 0.13 0.30; p < 0.001), onset-to-recanalization time (SMD 0.38, 95% CI 0.19; 0.57; p < 0.001), and baseline blood glucose (SMD 0.31, 95% CI 0.22 0.41; p < 0.001), while displaying a negative association with reduced baseline Alberta Stroke Program Early CT Score (ASPECTS) (SMD −0.37, 95% CI −0.46 −0.27; p < 0.001). Regarding clinical outcomes, FR was significantly associated with increased odds of symptomatic intracranial hemorrhages (OR 7.37, 95% CI 4.89 11.12; p < 0.001), hemorrhagic transformations (OR 2.98, 95% CI 2.37 3.75; p < 0.001), and 90-day mortality (OR 19.24, 95% CI 1.57 235.18; p = 0.021). Conclusions: The substantial prevalence of FR, standing at approximately 51%, warrants clinical consideration. These findings underscore the complexity of FR in AIS patients and highlight the importance of tailoring management strategies based on individual risk factors and clinical profiles.


Background
Strokes are a global health concern and rank among the leading causes of mortality and disability worldwide [1].The advent of reperfusion therapy has brought about a revolutionary shift in the management of acute ischemic stroke (AIS), providing substantial benefits to those affected [2].Nevertheless, despite advancements like endovascular thrombectomy (EVT), a significant proportion of patients continue to experience lessthan-optimal functional outcomes, even after achieving complete recanalization [3].This enduring disparity presents an ongoing challenge to the delivery of effective patient care, emphasizing the critical necessity of identifying cases of futile recanalization (FR) [4,5].FR, defined as functional dependence despite successful reperfusion, is a phenomenon that occurs with relative frequency among AIS patients who undergo EVT [6].Beyond its prognostic relevance, recognizing cases of FR holds immense potential for tailoring reperfusion strategies to specific subsets of AIS patients.The prevalence of FR among EVT-treated AIS patients varies across studies [7], with a comprehensive pooled prevalence still eluding researchers [3,7].It is imperative to identify the comorbidities or risk factors associated with FR, as previous evidence has demonstrated unfavorable clinical outcomes post FR [8].Furthermore, delving into the intricate relationships between FR and other clinically and radiologically significant biomarkers and outcomes [9], such as symptomatic intracranial hemorrhages (sICH), hemorrhagic transformations (HT) [6], and indicators like the Alberta Stroke Program Early CT Score (ASPECTS) [10], can significantly enhance our ability to predict outcomes in EVT-treated stroke cases.While some insights into potential factors linked to FR have been gained, our comprehension of post-FR prognosis remains incomplete [7] Within this context, the present study seeks to comprehensively assess post-FR outcomes, aiming to provide invaluable clinical guidance and insights for patients in relevant scenarios.
The primary objectives of this study center around an exploration of FR in the context of AIS patients undergoing EVT.To achieve this, the study aims to address the following pivotal questions: (a) What is the estimated pooled prevalence of FR among AIS patients undergoing EVT?(b) Which specific predictive indicators exhibit significant correlations with the occurrence of FR? (c) How does the occurrence of FR impact various clinical outcomes, and what is the significance level or strength of this relationship?

Literature Search and Study Selection
We conducted a comprehensive search for studies in the PubMed, Embase, and Cochrane Central Registry of Controlled Trials databases covering the period from January 2005 to May 2023.Our search strategy included the terms: "stroke", "ischemic attack", "cerebrovascular disorders", "cerebrovascular accident", "brain ischemia", "brain infarction", "thrombectomy", "endovascular procedures", "reperfusion therapy", "recanalization", "FR", "failed recanalization", "complete recanalization", or "partial recanalization".A detailed version of the search strategy is available in the Online Supplementary Information (Search Strategy).We meticulously examined the reference lists of pertinent articles, sys-tematic reviews, and meta-analyses to identify additional relevant studies.The systematic flow of the search, study inclusion, and the various subgroup analyses performed within the meta-analysis are visually represented using the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) flowchart (Figure 1).Our reporting strictly adhered to the PRISMA 2020 checklist (Supplemental Table S1) and the Meta-analysis Of Observational Studies in Epidemiology (MOOSE) checklist (Supplemental Table S2), all of which are detailed in the Supplemental Information.
Life 2023, 13, x FOR PEER REVIEW 3 of 32 "cerebrovascular disorders", "cerebrovascular accident", "brain ischemia", "brain infarction", "thrombectomy", "endovascular procedures", "reperfusion therapy", "recanalization", "FR", "failed recanalization", "complete recanalization", or "partial recanalization".A detailed version of the search strategy is available in the Online Supplementary Information (Search Strategy).We meticulously examined the reference lists of pertinent articles, systematic reviews, and meta-analyses to identify additional relevant studies.The systematic flow of the search, study inclusion, and the various subgroup analyses performed within the meta-analysis are visually represented using the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) flowchart (Figure 1).Our reporting strictly adhered to the PRISMA 2020 checklist (Supplemental Table S1) and the Meta-analysis Of Observational Studies in Epidemiology (MOOSE) checklist (Supplemental Table S2), all of which are detailed in the Supplemental Information.

Inclusion and Exclusion Criteria
To be considered eligible for inclusion, studies needed to satisfy the following criteria: (a) inclusion of AIS patients who underwent reperfusion therapy (IVT and EVT, or EVT alone); (b) participants aged 18 years or older; (c) availability of comparative data between patients with FR and those with non-FR, along with relevant post-futile-recanalization data; and (d) studies designed with appropriate methodology, including a sufficient sample size of at least 20 patients in each group.Exclusion criteria encompassed: (1) studies not written in English; (2) studies conducted on animals; (3) duplicated publications; (4) unavailability of full-text articles; (5) systematic reviews, meta-analyses, or narrative reviews; and ( 6) studies lacking relevant data on FR.FR was defined as poor functional outcome in AIS patients undergoing EVT, despite successful recanalization.The definitions of poor outcome and successful recanalization or reperfusion varied slightly across different studies (Table 1).

Data Extraction
Initially, all article titles and abstracts were reviewed using Endnote (Clarivate Analytics, London, UK) to exclude articles that did not meet the eligibility criteria.The remaining articles underwent a comprehensive examination to determine their suitability for inclusion in the systematic review or meta-analysis, as per the defined eligibility criteria.Data extraction was conducted using a dedicated data extraction sheet, capturing the following information from each study: 1.
Baseline study demographics: author, country, publication year, registry/trial name, study design, study design, and number of centers; 2.
Definition and criteria of various parameters: definition of poor outcome, successful recanalization, stroke etiology criteria, collateral status criteria, criteria for symptomatic intracerebral hemorrhage (sICH), and definition of mortality; 4.
Patient demographics: age and sex; 5.

Methodological Quality Assessment of Included Studies
The methodological quality assessment of the included studies was conducted using the modified Jadad analysis (MJA) [49] and was completed independently by the primary researcher (Supplemental Table S3).The risk of biases in results due to funding was also evaluated, based on the declaration of funding sources and conflicts of interest extracted from each individual study (Supplemental Table S4).

Statistical Analyses
The statistical analyses in this study were conducted using STATA v. 13.0 (StataCorp, College Station, TX, USA).Baseline characteristics of the included cohort in this metaanalysis were extracted from all incorporated studies.When suitable, means and standard deviations (SDs) were estimated from medians and interquartile ranges (IQRs) using Wan et al.'s [50] method.The pooled prevalence of FR among patients with AIS undergoing EVT was assessed using the "metaprop" package in STATA, performing a random-effects meta-analysis of proportions derived from individual studies, 95% confidence intervals (95% CI) were obtained using the "cimethod (exact)" and "ftt" commands.To explore factors linked to FR and its impact on clinical outcomes, a random-effects meta-analysis designed by DerSimonian and Laird (DL) was employed.The random-effects model was applied across all subgroup analyses, encompassing studies on reperfusion therapy type (EVT or a combination of EVT and IVT), stroke territory (anterior, posterior, and mixed [anterior/posterior]), and study design (retrospective, prospective or mixed [studies with data collected both retrospectively as well as prospectively]).For odds ratios (ORs), 95% confidence intervals (95% CIs), percentage weights, and inter-study heterogeneity within our meta-analysis, forest plots were generated (see Supplemental Figures S1-S6).The sensitivity analysis was performed using the "metaninf" package in STATA to examine changes in the pooled odds ratios resulting from the exclusion of individual studies.Heterogeneity among studies was assessed using I 2 statistics and p-values (I 2 < 30% = low, 30-50% = moderate, 50-75% = substantial, 75-100% = severe).The potential presence of publication bias was assessed using Egger's test and funnel plots (through the use of "metabias" and "metafunnel" STATA packages).An asymmetry on either side of the funnel plot indicated the presence of publication bias, which was further corroborated by the p-value from Egger's test.Summary effects and measures of heterogeneity for both prevalence and association studies were tabulated.Additionally, we took into account Cochran's Q test p-values and estimated between-study variances using Tau-squared.All analyses conducted in this study adhered to a significance level of p < 0.05.

Description of Included Studies
A total of 1430 studies were initially identified by manual and electronic database searches.After removing duplicates, a total of 1015 records remained.Each abstract was meticulously reviewed, leading to the exclusion of 928 records.Among the remaining 86 articles, 49 studies were subsequently excluded for various reasons.Specifically, 19 of these studies did not report the targeted outcomes or lacked sufficient data, while 2 exhibited inappropriate study designs.Additionally, 10 studies were excluded due to overlapping cohorts, 9 had inadequate control groups, 1 had a limited sample size, and 6 lacked a clear definition of FR.Ultimately, a final selection of 39 studies, encompassing a total of 11,700 patients, was included in this meta-analysis.Out of 11,700 patients, FR was observed in 5766 patients.For instances involving reports from the same database or registry, priority was given to studies with the largest or most recent sample size.Of these studies, 3 centered around patients primarily receiving EVT, with or without IVT, while 36 studies focused on patients who primarily underwent EVT with IVT.The mean age of all included studies was 65.6 (n = 11,700).Comprehensive clinical characteristics, associated factors, and outcomes of the studies featured in the meta-analysis are presented in Tables 1-3.Summary effects and heterogeneity related to the estimated pooled prevalence of FR are provided in Table 4.Further insight into the association between discrete predictive indicators, clinical outcomes and FR is presented in Table 5 and Supplemental Table S6.To address the link between FR and continuous predictive indicators, Table 6 and Supplemental Table S7 display the corresponding summary effects and heterogeneity.Additionally, the supplemental materials contribute valuable information.It is important to note that variations in the definitions of FR, sICH, collateral status, and ASPECTS score exist across the studies.To comprehensively assess the studies, the manuscript includes an evaluation of the methodological quality and funding bias in Supplemental Tables S3 and S4.Finally, the assessment of publication bias, conducted using Egger's test followed by sensitivity analysis, is summarized in Supplemental Tables S5 and S6 and Figures S7-S12.

Prevalence of FR in Prospective and Retrospective Studies
One study [37] with a cohort size of 362 reported on the prevalence of FR in AIS patients undergoing EVT through a prospective and retrospective study design.However, a meta-analysis was unable to be performed due to an insufficient number of studies (Table 4).

Prevalence of FR in the Middle East
Two studies [31,42], with a cohort size of 144 reported on the prevalence of FR in AIS patients undergoing EVT in the Middle East.However, a meta-analysis was unable to be performed due to an insufficient number of studies (Table 4).

Predictive Indicators of Futile Recanalization
Tables 5 and 6 summarize the associations between various factors and the likelihood of FR in patients with AIS.For more detailed information on these associations, please refer to the provided text and supplemental figures (Supplemental Figures S1-S6).Supplemental Figures S7 and S8 provide the Egger's plot for assessing publication bias.

Discussion
Our meta-analysis has revealed a pooled prevalence estimate of 51% for FR among AIS patients undergoing EVT.This study is distinct in providing pooled prevalence estimates on FR following EVT for AIS patients, presenting the largest sample size reported to date.Additionally, we have also identified clinical risk factors significantly linked to FR, including age, AF, HTN, DM, history of stroke and/or transient ischemic attack, and smoking.Furthermore, FR is correlated with an elevated risk of severe adverse outcomes, encompassing sICH, HT, and mortality.
We have determined a pooled prevalence of 51% for FR among patients with AIS who undergo EVT.This finding contrasts with earlier meta-analyses, which reported prevalence rates ranging from 32.4% to 56.7% [3,7].Establishing this pooled prevalence of FR is pivotal in recognizing and conveying the potential risks associated with undergoing EVT for AIS patients.The notable heterogeneity observed across the studies could be attributed to procedural disparities among treatment centers, FR in the timing of patient presentations leading to differences in OTT and OTR.Furthermore, the divergence in hospital settings and statuses, particularly between tertiary hospitals and other centers, could contribute to the observed heterogeneity.Tertiary hospitals, likely catering to patients with more severe stroke symptoms, might yield outcomes distinct from centers treating milder symptoms [52,53].The occurrence of FR might be explained by the no-reflow phenomenon.This phenomenon arises from leukocyte-endothelial interactions that depend on adhesive molecules, leading to the aggregation of red blood cells and the formation of microthrombi [6].Other factors contributing to this occurrence include early re-occlusion of arteries, hemorrhagic transformation, and compromised collateral circulation [6,54].
Our comprehensive meta-analysis has revealed a spectrum of factors associated with FR following AIS providing critical insights into the multifaceted nature of this phenomenon.These factors encompass a wide range of characteristics, including patient demographics, clinical variables, treatment modalities, and even laboratory markers.Notably, male gender, smoking, the presence of good collaterals, AF, hypertension, diabetes mellitus, prior stroke or TIA, prior use of anticoagulants, cardioembolic etiology, general anesthesia administration, and adjunct IVT are associated with an increased risk of FR.Furthermore, increasing age, systolic blood pressure, NIHSS at admission, onset-to-treatment time, onset-to-reperfusion time, baseline blood glucose, and reduced ASPECTS were correlated with FR [3].These predictors can help inform clinical decision making and highlight areas for further research.Male sex demonstrated a decreased odd of FR, suggesting that female AIS patients may face a slightly higher risk of FR.AF was significantly associated with increased odds of FR, underscoring the importance of managing AF in AIS patients to improve recanalization outcomes.Hypertension significantly increased the odds of FR, highlighting the need for aggressive blood pressure control in these cases.Conversely, smoking was significantly associated with decreased odds of FR, suggesting a potential protective effect.Furthermore, adjunct intravenous thrombolysis significantly decreased the odds of FR, emphasizing the importance of considering IVT in AIS management.However, the study revealed several important nuances.For instance, certain factors, like alcohol and hyperlipidemia, showed no significant association with FR, suggesting that their impact on recanalization outcomes may be limited.These findings are in line with previous meta-analyses that have identified various factors linked to FR, including age, admission NIHSS score, ASPECTS score, HTN, admission SBP, AF, and the usage of intravenous tissue plasminogen activator (IV tPA), OTT, and OTR [3,7].However, our analysis adds to the existing literature by providing a more comprehensive overview of the factors associated with FR, encompassing a broader range of variables.
It is important to acknowledge that the included studies varied in terms of patient populations, methodologies, and reporting standards.Factors such as race, education levels, body mass index (BMI), wake-up stroke, statin usage, and comorbidities like congestive heart failure and intracranial atherosclerotic stenosis were inconsistently reported or underrepresented [14,20,22,24,25,30,36,47,48,51].Laboratory markers such as leukocyte status [12,32,45], high-sensitivity c-reactive protein (hs-CRP) levels [26,38,42], and various blood cell counts were also underreported [12,16,20,28,38,42,44].Imaging characteristics, stroke location, etiology, and procedural details demonstrated significant heterogeneity across studies.These variations underscore the complexity of FR and emphasize the need for standardized reporting and further research to better understand the interplay of these factors in clinical practice.Additionally, the study detected publication bias in several analyses, underscoring the need for further research and cautious interpretation of these findings.Overall, these results indicate that risk factors associated with FR following AIS are diverse and multifactorial, encompassing patient characteristics, clinical variables, and treatment-related factors.While this meta-analysis sheds light on many of these factors, it also highlights the need for more comprehensive and standardized research to enhance our understanding of FR and improve patient outcomes.Additionally, considering the multifaceted nature of FR, a personalized approach to stroke management, taking into account these various factors, may be necessary to reduce the incidence of FR and optimize stroke care.

Limitations
Our meta-analysis exhibits certain limitations arising from variations in the quality of the included studies, which in turn impacted our precision in data extraction and analysis.Initially, we encountered a range of study designs, spanning from retrospective designs to prospective approaches and randomized controlled trials (RCTs).Furthermore, the research objectives differed; some studies aimed to contrast the outcomes of distinct treatment regimens (such as comparing EVT solely to EVT and IVT), while others focused on identifying potential adverse consequences of EVT.Additionally, a considerable portion of the studies was retrospective in nature, consequently constraining their overall design.
Additionally, it is worth acknowledging the potential impact of the 2015 American Heart Association (AHA)/American Stroke Association (ASA) guideline changes [55], which could introduce variations in results between studies conducted before and after that year.Looking ahead, we propose further investigation into the prevalence of FR among AIS patients, with a specific focus on regional disparities.These forthcoming studies should also delve into the various associated factors with FR, particularly those that have been identified as underreported, thus precluding a comprehensive meta-analysis.

Figure 1 .
Figure 1.PRISMA Flowchart: Inclusion of Studies in the Meta-Analysis.Illustration depicting the flow of study selection according to the PRISMA guidelines, leading to the inclusion of studies in the meta-analysis.Abbreviations: N: number of studies; n: cohort size; PP = pooled prevalence; AF:

Table 1 .
Clinical Characteristics of Studies Included in Meta-Analysis.

Table 3 .
Continuous Predictive Markers of Futile Recanalization Included in Meta-Analysis.
Abbreviations: FR: futile recanalization; n: number of patients; SBP: systolic blood pressure; NIHSS: National Institute of Health Stroke Severity; ASPECTS: Alberta Stroke Program Early CT Score; OTT: onset-to-treatment time; OTR: onset-to-recanalization time; BG: blood glucose; N: number of studies; n: number of patients.

Table 4 .
Meta-Analysis Results for Estimated Pooled Prevalence of Futile Recanalization: Summary Effects and Heterogeneity.
Abbreviations: N: number of studies; n: number of patients; CI: confidence interval.

Table 5 .
Meta-Analysis Results for Discrete Predictive Markers and Outcomes Associated with Futile Recanalization: Summary Effects and Heterogeneity.
Cochran's Q test; H: relative excess in Cochran's Q over its degrees of freedom; I 2 : proportion of total variation in effect estimate due to between study heterogeneity (based on Cochran's Q test); τ 2 : between-study variance to test comparisons of heterogeneity among subgroups; *: values of I≤ are percentages; ¶ : heterogeneity values were calculated from data with 95% CIs based on gamma (random-effects) distribution for Q; Φ : heterogeneity variance estimates (τ 2 ≤) were derived from the DerSimonian and Laird method.

Table 6 .
Meta-Analysis Results for Continuous Predictive Markers Associated with Futile Recanalization: Summary Effects and Heterogeneity.
Abbreviations: SBP: systolic blood pressure; NIHSS: National Institute of Health Stroke Severity; ASPECTS: Alberta Stroke Program Early CT Score; OTT: onset-to-treatment time; OTR: onset-to-recanalization time; BG: blood glucose; N: number of studies; n: number of patients; OR: odds ratio; CI: confidence interval; REDL: DerSimonian and Laird random-effects method; Q: heterogeneity measures were calculated from data with 95% confidence intervals (95% CI), based on noncentral X 2 (common effect) distribution for Cochran's Q test; H: relative excess in Cochran's Q over its degrees of freedom; I 2 : proportion of total variation in effect estimate due to between study heterogeneity (based on Cochran's Q test); τ 2 : between-study variance to test comparisons of heterogeneity among subgroups; *: values of I≤ are percentages; ¶ : heterogeneity values were calculated from data with 95% CIs based on gamma (random-effects) distribution for Q; Φ : heterogeneity variance estimates (τ 2 ≤) were derived from the DerSimonian and Laird method.