Highlights
• The efficacy and safety of endovascular treatment (EVT) in medium vessel occlusion remain uncertain.
• Pooled analysis of randomized trials showed no significant 90 day functional improvement with EVT over best medical treatment.
• EVT was associated with higher risks of symptomatic intracerebral hemorrhage and serious adverse events.
Background and purpose
Endovascular treatment (EVT), a highly effective treatment for ischemic stroke due to large vessel occlusion (LVO) in the internal carotid artery, M1 segment of the middle cerebral artery (MCA) and basilar artery, can be considered in patients up to 24 hours after stroke onset. Reference Papanagiotou, Bonatti and Fiehler1 The meta-analysis of seven randomized controlled trials (RCTs) from the HERMES collaboration demonstrated that in patients with proximal or dominant M2 segment MCA occlusions, functional independence (modified Rankin score (mRS) 0–2) at 90 days significantly improved in the EVT group compared to best medical treatment (BMT) alone. Reference Menon, Hill and Dávalos2
However, a substantial proportion of patients with acute ischemic stroke (AIS) due to medium vessel occlusion (MeVO) fail to achieve favorable functional outcomes with the BMT currently available. Furthermore, approximately one-third of these patients are not functionally independent 90 days after the index stroke. Reference Ospel, Menon and Demchuk3 The prevailing European and American guidelines do not offer any recommendations or contraindications with EVT for patients with medium or distal vessel occlusion (MDVO).
Intravenous thrombolytic agents, such as alteplase and tenecteplase, have been observed to demonstrate a heightened propensity for recanalizing arteries as compared to cases of LVO. This phenomenon can be attributed to the characteristic nature of MeVO, which typically gives rise to smaller thrombi that are less fibrin-rich compared to those observed in LVO. This renders the clot more susceptible to enzymatic degradation by thrombolytic agents such as alteplase or tenecteplase. Reference Bala, Singh and Ignacio4
However, a multicenter prospective study demonstrated that early recanalization following intravenous thrombolysis (IVT) in the distal M1 MCA subgroup was achieved in less than 50% of cases. Reference Menon, Al-Ajlan and Najm5 A paucity of definitive data from prospective clinical trials exists on the efficacy and safety of EVT for the treatment of MeVO. The present systematic review and meta-analyses outline the current evidence from RCTs conducted to evaluate the efficacy and safety of EVT for AIS among patients with MeVO. To consolidate the existing evidence and enhance the robustness of conclusions, we performed a trial sequential analysis (TSA). This approach allowed us to assess whether the available evidence is robust enough to inform clinical practice or if further research is warranted.
Materials and methods
Search strategy
The PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) statement was employed for the purpose of reporting systematic reviews and meta-analyses in the present study. Reference Page, McKenzie and Bossuyt6 The study protocol was registered with the International Prospective Register of Systematic Reviews (PROSPERO) under the registration identification number CRD420250653970.
Inclusion criteria
The present systematic review incorporated RCTs of patients diagnosed with AIS due to confirmed symptomatic and endovascularly treatable MDVO. The selection of patients was based on neurovascular noninvasive imaging, namely, CT angiography (CTA) or magnetic resonance angiography, with the caveat that the patients in question presented within 24 hours of their last known well and were treated with or without IVT. Exclusion criteria encompassed case reports, case series, observational studies, studies without primary efficacy outcome data, studies without primary safety outcome data and studies that did not compare EVT with no EVT in AIS due to MeVO.
Data extraction
A comprehensive literature search was independently conducted by two authors (F.M. and B.P.) using five electronic databases: PubMed, Scopus, EMBASE, MEDLINE and Google Scholar. This search was conducted from database inception until March 2025. Our search followed the PICOS (Population, Intervention, Comparator, Outcome, Study Design) framework as detailed below:
Population
Patients with AIS were identified using the following MeSH terms and keywords: “Acute stroke” OR “Ischemic stroke” OR “Cerebrovascular accident” OR “Cerebral infarction.” These terms were chosen to comprehensively extract all relevant literature related to ischemic strokes.
Intervention
To focus on studies evaluating EVT, we used “Thrombectomy” OR “Endovascular therapy” OR “Mechanical aspiration” OR “Endovascular technique” OR “Endovascular procedure” OR “Stent retriever.” These terms reflect the range of EVT methods used in current clinical practice.
Focus of study
To limit the review of MeVO/MDVO, we included “Medium Vessel Occlusion” OR “Distal Vessel Occlusion” OR “M2 Occlusion” OR “M3 Occlusion” OR “M4 Occlusion” OR “A1 Occlusion” OR “A2 Occlusion” OR “A3 Occlusion” OR “P1 Occlusion” OR “P2 Occlusion” OR “P3 Occlusion.” These terms were selected to isolate studies evaluating non-proximal occlusions, which are often underrepresented in EVT trials.
Comparator
Comparison group was defined using the terms “Best medical treatment alone” OR “Intravascular thrombolysis,” to reflect standard non-interventional therapies used in these patients.
Outcomes
Outcomes of interest were searched using “Excellent functional outcome” OR “Best functional outcome” OR “Symptomatic intracranial hemorrhage” OR “Deaths,” reflecting key safety and efficacy endpoints commonly reported in EVT trials.
Study design
To maintain high methodological quality, we filtered for “Randomized Controlled Trial” OR “RCT” and applied filters to include only RCTs.
Screening and data extraction process
The study screening and selection processes were managed using Covidence software (https://www.covidence.org/), which supported automatic deduplication and title/abstract screening, as well as full-text review, thereby enabling a standardized workflow. The title and abstract were screened independently by two reviewers (F.M. and B.P.). Full-text review was performed on potentially eligible articles based on predefined inclusion and exclusion criteria. Any disagreements were resolved through consensus or adjudication by a third reviewer (R.B.). An Excel-based data extraction form with predefined fields was used to systematically collect relevant data across studies, ensuring consistency and minimizing extraction-related heterogeneity. Extracted variables included study design, sample size, primary and secondary outcomes and safety outcomes in both the intervention and comparator groups.
The final outcome analysis was conducted using the available data from the main manuscript that had already been published. Missing, unclear or outlier data were excluded from the final analysis. However, individual patient-level data were unavailable from the original studies and were therefore not included in the analysis.
Outcomes
The primary efficacy outcome extracted from the studies was defined as the relative risk of achieving an excellent functional outcome, which was categorized as a mRS score of 0–1 at 90 days. The secondary outcome encompassed functional independence, measured by an mRS score of 0–2 at 90 days. The safety outcomes encompassed symptomatic intracranial hemorrhage (sICH), death from any cause at 90 days and any serious adverse events (SAEs) within 90 days post-intervention.
Assessment of risk of bias
Following the selection of relevant studies, the data were extracted into MS Excel by two authors independently (F.M. and B.P.). The information extracted comprised the first author, the year of publication, the number of patients included in each study, mono or multi-centric study, the baseline characteristics of patients included in each study, mRS at 90 days, sICH, death from any cause and any SAEs. Any discrepancy between the two authors was resolved in consultation with a third author (R.B.). The Cochrane Risk of Bias (ROB2) tool was utilized to evaluate the risk of bias in RCTs.
Statistical analysis
Statistical analyses were conducted using the R Project for Statistical Computing (https://www.rproject.org/). A common effects model (fixed effects model) was employed due to minimal heterogeneity (I2 = 0%, τ2 = 0, p = 0.77), indicating negligible variation across studies. Risk ratios (RR) with corresponding 95% confidence intervals (CI) were utilized to generate the forest plot. Statistical heterogeneity was assessed using the Higgins I2 statistic, with values of I2 = 0% suggesting no detectable between-study variability. The classification of study-level estimates was determined by the I2 value, with estimates categorized as heterogeneous if the I2 value exceeded 50%. An I2 value between 50% and 75% was considered indicative of substantial heterogeneity, while a value greater than 75% was regarded as considerable heterogeneity. Since this meta-analysis included a limited number of eligible studies, resulting in a relatively small number of overall randomized participants, it was necessary to assess whether the analysis had sufficient statistical power to reliably evaluate the intervention effects. Consequently, for the primary efficacy outcome, a TSA was also conducted, in which trials were included in chronological order and analyzed as interim analyses relative to the required number of randomized participants. TSA accounted for the required information size (RIS) and established 168 monitoring boundaries to control for type I (false positive, alpha boundary) and type II (false negative, beta boundary) errors. A TSA-adjusted 95% CI and an adjusted threshold for statistical significance, lower than the conventional p < 0.05, were reported if the RIS had not been reached. If the line representing the cumulative z-score fails to cross the alpha boundary, we conclude that the existing studies do not provide sufficient evidence to show a significant difference relative to RIS; if the z line enters the beta boundaries, contingent upon achieving RIS, we can declare futility. We also assessed certainty of the evidence using the GRADE pro GDT tool.
Results
Study selection
A total of 1115 studies were identified, of which 312 duplicates were removed. 803 studies underwent title and abstract screening, and 799 studies were found irrelevant. Four studies underwent full-text screening (ESCAPE-MeVO, Reference Goyal, Ospel and Ganesh7 DISTAL Reference Psychogios, Brehm and Ribo8 DUSK study Reference Mohammaden, de Souza Viana and Abdelhamid9 and DISCOUNT study Reference Clarençon, Durand-Zaleski and Premat10 ). The DUSK study was a retrospective analysis of prospectively collected data, and the DISCOUNT study was excluded as it was a protocol. A total of two randomized control trials (DISTAL and ESCAPE Mevo) comprising 1073 patients met the predefined inclusion and exclusion criteria. Of these patients, 526 were in the EVT arm, and 547 were in the BMT arm. The details are depicted in the PRISMA flow chart (Figure 1).
Preferred Reporting Items for Systematic Reviews and Meta-Analyses flow chart.
Study characteristics
The baseline characteristics of the patients included in the two studies are shown in Table 1. No significant differences in baseline characteristics were observed between the EVT and BMT arms in both studies. The median National Institute of Health Stroke Scale (NIHSS) at admission was found to be marginally higher in the ESCAPE-MeVO trial group compared to the DISTAL trial. The ESCAPE-MeVO trial made a specific exclusion of occlusions of the M1 segment, A1 segment and P1 segment, as well as M4 occlusions. This is in contrast to the DISTAL trial, which included a small proportion of M4 occlusions and certain cases of A1 and P1 occlusions. The prevalence of M2 occlusions was marginally higher in DISTAL EVT, while ESCAPE-MeVO was observed to have a higher proportion of M3 occlusions. The DISTAL trial had a higher number of patients who received IVT compared to the ESCAPE-MeVO trial (355 vs. 305 patients, respectively).
Showing baseline characteristics of patients included in both trials
MeVO = medium vessel occlusion; EVT = endovascular treatment; BMT = best medical treatment; no = number; IQR = interquartile range; TIA = transient ischemic attack; AF = atrial fibrillation; NIHSS = National Institute of Health Stroke Scale; CTA = CT angiography; eTICI = Expanded Thrombolysis in Cerebral Infarction.
Risk of bias assessment across studies
Both the trials included in the systematic review have a low risk of bias in the randomization domain. Concerns were raised regarding the potential for bias due to deviations from the intended interventions (i.e., adherence to the intervention). This was due to the participants and their caregivers or intervention providers being aware of the interventions. The missing outcome data were addressed through the implementation of imputation methods (Table 2). One trial utilized multiple imputation by chained equations, while the other employed a last score carried forward approach, imputing a score of 4 in instances where both day 30 and day 90 scores were missing but the patient was alive at 90 days. These methodologies were employed to minimize bias and ensure the integrity of the data. In both trials, the blinded outcome assessors were successful in minimizing detection bias. The primary outcome analysis was conducted in accordance with a pre-specified plan, thereby ensuring consistency. Overall, the open-label design introduces some bias, but randomization and blinded outcome assessment help mitigate its impact by ensuring balanced groups and objective outcome evaluation (Figure 2).
Traffic plot illustrating the risk of bias across different domains for each study.
Risk of bias assessment in different domains
EVT = endovascular thrombectomy; BMT = best medical treatment; NIHSS = National Institute of Health Stroke Scale; ITT = intention to treat.
Outcomes
Excellent functional outcome (mRS 0–1) at 90 days
The primary efficacy outcome, defined as excellent functional outcome (mRS 0–1 at 90 days), was achieved in 200 out of 526 patients (38%) in the EVT plus BMT arm and 219 out of 543 patients (40.3%) in the BMT arm (Table 3). The pooled RR, calculated using the common effects model, was 0.95 (95% CI: 0.81–1.10; 95% TSA-adjusted CI: 0.79–1.13), indicating no significant difference between the two treatment arms. The I2 test for heterogeneity was 0%, demonstrating consistency across studies (Figure 3A). The TSA showed that with an achieved information size of 1069 (96%), the line representing the cumulative z-score failed to cross the alpha boundary, implying that the existing studies do not provide sufficient evidence to show a significant difference between EVT plus BMT arm versus BMT alone; additionally, the cumulative z-score line has already entered the beta boundaries in the futility zone (Figure 4).
Forest plot of the efficacy outcomes: (A) Forest plot for modified Rankin score (mRS) 0–1; (B) forest plot for mRS 0–2. (A) Forest plot showing excellent functional outcome (mRS 0–1) at 90 days. Squares represent the risk ratios (RR) for each individual study, with square size proportional to study weight. Horizontal lines indicate 95% confidence intervals (CI). The diamond at the bottom represents the pooled RR and 95% CI under a fixed-effect model. A value of RR = 1.0 suggests no difference between groups. I2 = 0% indicates no observed heterogeneity. (B) Forest plot showing functional independence (mRS 0–2) at 90 days. Visual elements are as described in Figure 3A. The pooled RR and CI are shown as a diamond. I2 = 10% indicates low heterogeneity among the included studies.
Trial sequential analysis (TSA) evaluating the cumulative evidence from included studies on a pooled relative risk effect. The x-axis represents the cumulative information size (number of participants/events, in percentage of the required information size), while the y-axis shows the cumulative Z-score for the meta-analysis. RR = risk ratios; AIS = acute ischemic stroke.
Outcomes of both trials
MeVO = medium vessel occlusion; EVT = endovascular treatment; BMT = best medical treatment; mRS = modified Rankin score; sICH = symptomatic intracranial hemorrhage.
Functional independence (mRS 0–2) at 90 days
The secondary efficacy outcome, defined as functional independence (mRS 0–2 at 90 days post-intervention), was achieved in 291 out of 526 patients (55.32%) in the EVT plus BMT arm and 308 out of 543 patients (57%) in the BMT arm (Table 3). The pooled RR using the common effects model was 0.98 (95% CI: 0.88, 1.09), indicating no significant difference between the two treatment arms (Figure 3B). The heterogeneity for this outcome was low (I2 = 10%), indicating minimal variability between studies, suggesting that the results were consistent across trials, and the pooled estimate is reliable for interpretation (Figure 3B).
Safety outcomes
Symptomatic intracranial hemorrhage (sICH) within 24 hours
The primary safety outcome, defined as sICH within 24 hours after randomization, was reported according to each trial’s pre-specified criteria: ESCAPE-MeVO used the Heidelberg bleeding classification, while DISTAL applied the modified Safe Implementation of Thrombolysis in Stroke–Monitoring Study (SITS-MOST) criteria. Despite this difference in sICH definitions, the primary safety outcome occurred in 30 out of 526 patients (5.7%) in the EVT plus BMT arm and 13 out of 546 patients (2.4%) in the BMT arm (Table 3). The pooled RR, calculated using the common effects model, was 2.39 (95% CI: 1.26–4.53), indicating a significantly higher risk of sICH in the EVT plus BMT arm. Additionally, the I2 test for heterogeneity was 0%, demonstrating consistency across studies (Figure 5A).
Safety outcomes: (A) Forest plot of symptomatic intracranial hemorrhage (sICH). (B) Forest plot of all deaths. (C) Forest Plot of any serious adverse events (SAE). 5(A) Forest plot showing the incidence of sICH. 5(B) Forest plot showing incidence of deaths. 5(C) Forest plot showing incidence of SAE.
Deaths from any cause at 90 days
The secondary safety outcome, defined as death from any cause at 90 days, occurred in 76 out of 526 patients (14.4%) in the EVT plus BMT arm and 61 out of 546 patients (11.2%) in the BMT arm (Table 3). The pooled RR, calculated using the common effects model, was 1.29 (95% CI: 0.94–1.76), indicating no significant difference in mortality between the two treatment arms. Additionally, the I2 test for heterogeneity was 16%, suggesting low variability between studies and moderate consistency in findings (Figure 5B).
Serious adverse effects within 90 days
The secondary safety outcome, defined as SAEs within 90 days, occurred in 201 out of 526 patients (38.2%) in the EVT plus BMT arm and 158 out of 546 patients (28.9%) in the BMT arm (Table 3). The pooled RR, calculated using the common effects model, was 1.32 (95% CI: 1.11–1.56), indicating a significantly higher incidence of SAEs in the EVT plus BMT arm. Additionally, the I2 test for heterogeneity was 0%, suggesting no significant variability between studies and high consistency in findings (Figure 5C).
Discussion
Our analysis suggested that among patients with AIS due to MeVO who presented within 24 hours of last being seen normal and had evidence of salvageable tissue on baseline imaging, EVT plus BMT did not significantly improve the probability of achieving an excellent functional outcome at 90 days compared to BMT alone. The findings of this review also suggested that there is no statistically significant improvement in functional independence when EVT is used in conjunction with BMT, as opposed to BMT alone, in those patients. Given the small number of trials included, the TSA findings should be interpreted with caution. The analysis primarily serves to highlight the current insufficiency of data to reach firm conclusions and to underscore the need for further high-quality RCTs to reduce the risk of type I and II errors in future meta-analyses. We also reported that the EVT plus BMT arm had a higher risk of sICH and SAE, while mortality at 90 days showed no significant difference. A low level of heterogeneity while pooling the results indicated consistency across the studies.
We had excluded the DISCOUNT trial (NCT05030142) because the limited results (trial stopped following first interim analysis) available from this trial as a conference abstract did not have data 269 on the primary outcome (mRS 0–1). It included AIS patients with MeVO presenting within 24 hours of last being seen well, with compatible advanced imaging and an NIHSS ≥ 5 or disabling aphasia. The trial randomized 163 patients with occlusions distal to the mid-insula for the MCA and all anterior (A1–A3) and posterior cerebral (P1–P3) artery occlusions. The primary outcome, mRS 0–2 at 90 days, occurred in 60% of patients treated with EVT plus BMT versus 77% in the BMT-alone group (OR: 0.42, 95% CI: 0.20–0.88, after multiple imputations). Reference Clarençon, Durand-Zaleski and Premat10
The results of our meta-analysis of RCTs are significantly different from the previous post hoc analyses of randomized and nonrandomized studies showing a benefit of EVT in MeVO. The HERMES collaboration meta-analysis employed individual patient-level data from multiple major RCTs of EVT in LVO stroke, thus facilitating detailed subgroup analyses including patients with M2 segment MCA occlusions. This demonstrated that EVT in patients with M2 segment MCA occlusions significantly improved functional outcomes (mRS 0–2: 58.2% vs. 39.7%; OR 2.39, 95% 281 CI 1.08–5.28), with efficacy comparable to that seen in proximal occlusions. This difference could be due to multiple factors. Reference Menon, Hill and Dávalos2
First, the reperfusion rates in these trials were significantly lower than LVOs. For example, in the DISTAL trial, successful reperfusion was achieved in 71.7% of patients in the EVT plus BMT group, while in the ESCAPE-MeVO trial, it was observed in 75.1% of cases, likely influenced by technical challenges. Reference Goyal, Ospel and Ganesh7,Reference Psychogios, Brehm and Ribo8 Despite the technical differences between the MeVO-eTICI and eTICI scales, previous LVO trials have reported successful reperfusion rates of 85% or higher. Reference Sarraj, Hassan and Abraham11,Reference Fischer, Kaesmacher and Strbian12 This low incidence of successful reperfusion rates may have been a contributing factor to the neutral findings of EVT observed in the review. SAEs and sICH were more prevalent in the intervention group than in the usual care group. Reference Goyal, Ospel and Ganesh7 These may have contributed cumulatively to the neutral findings in the primary outcome.
Second, in both trials, the median workflow time for EVT was longer than in previous LVO trials, which may have reduced the procedure’s benefit by limiting the amount of salvageable tissue. For example, in the DISTAL trial, the median time from imaging to arterial puncture was 70 minutes, thus exceeding the target of 60 minutes, despite efforts to expedite EVT delivery. This delay was primarily due to a longer interval between imaging and randomization, which was attributed to challenges in detecting distal and medium vessel occlusions, as well as the time required to obtain patient consent. The workflow in the ESCAPE-MeVO trial also revealed a longer median time from onset to recanalization of 359 minutes, in comparison to the 249 minutes observed in the ESCAPE trial. Reference Fischer, Kaesmacher and Strbian12 The time delay in the ESCAPE-MeVO trial was partly attributed to the time required to arrange general anesthesia and technical challenges in achieving reperfusion in distal arteries.
Additionally, distal arteries are end arteries without collaterals, emphasizing the critical importance of timely treatment in MeVO strokes, as they have less compensatory capacity compared to LVO strokes. In the ESCAPE-MeVO trial, a minority of patients with MeVO (39 out of 255) in the EVT arm experienced spontaneous recanalization between noninvasive imaging and the first intracranial angiographic assessment, thereby eliminating the need for thrombectomy despite undergoing angiography. In the majority of cases (32 out of 39), this recanalization was attributed to intravascular thrombolysis, thereby highlighting that spontaneous thrombus resolution can occur in MeVO following thrombolysis. Reference Goyal, Ospel and Ganesh7
The smaller vessel diameters, distal location and intricate branching patterns of MeVOs pose significant challenges to the efficacy of currently employed devices. Reference Uniken Venema, Dankbaar, van der Lugt, Dippel and van der Worp13 These devices have been primarily designed for LVOs, and consequently, their application in the context of MeVO strokes may result in suboptimal perfusion outcomes. Moreover, these anatomical features increase technical difficulty during EVT, including limited catheter reach, vessel tortuosity and higher risk of complications such as vessel perforation or dissection. During stent retriever withdrawal, the straightening of tortuous vessels can generate shear stress at the origin of small perforating branches, potentially leading to vessel wall injury and subarachnoid hemorrhage. In the future, device manufacturers are likely to develop longer devices with low cross-sectional diameter specifically tailored to address the anatomical challenges of MeVO strokes. Reference Ospel, Nguyen and Jadhav14
Advancements in EVT devices and techniques should also parallel improved patient selection. In the ESCAPE-MeVO trial, successful reperfusion (adapted mTICI 2b–3) was achieved in only 75% of patients. Achieving complete recanalization on the first pass is crucial, as the first-pass effect is linked to better outcomes. Reference Goyal, Ospel and Ganesh7 Additionally, subgroup analysis from the DISTAL trial suggested fewer hemorrhagic complications with aspiration catheters compared to stent retrievers or a combination of both. It is also imperative to optimize workflow efficiency, improve imaging to treatment time and streamline the decision-making process in order to mitigate these delays and improve EVT outcomes in MeVO patients. Improved patient selection is essential for optimizing outcomes. First, establishing a standardized definition for the M2 segment and subsegment occlusions would create a more homogeneous patient group. Fasen and colleagues reported that up to 35% of M2 occlusions may be missed on single-phase CTA, underscoring a significant limitation in current acute-phase imaging protocols for detecting MeVOs. This diagnostic gap represents a missed therapeutic window, as unidentified MeVOs may exclude patients from timely administration of EVT or IVT.
Furthermore, emerging evidence suggests that patients with MeVOs, whether or not eligible for immediate reperfusion therapy, require closer neurological monitoring, given their higher risk of early deterioration compared to those without angiographically detectable occlusions. The use of multiphase CTA is increasingly recognized as essential for the accurate detection and management of MeVOs. Reference Fasen, Heijboer, Hulsmans and Kwee15 Second, raising the NIHSS threshold might help identify patients most likely to benefit from EVT. For example, subgroup analysis from the DISTAL trial showed a trend toward better outcomes with EVT as NIHSS increased. This aligns with a population-based study on M2 occlusions, which identified an NIHSS cut-off of 9 points as predictive of poor outcomes, with 78% of patients scoring >9 having unfavorable results compared to 22% of those scoring <9. Reference Rai, Domico and Buseman16 The trials assessed outcomes primarily using the mRS, which is predominantly weighted toward motor function. However, this may underrepresent the true disability experienced by patients or the benefit of intervention, with posterior cerebral artery MeVOs, where visual and cognitive impairments are more prominent. In such cases, an outcome measure that integrates visual and neurocognitive domains may provide a more comprehensive evaluation of functional impact. Reference Ospel, Nguyen and Jadhav14
The principal limitation of this study is the paucity of RCTs published on this research question to date. Moreover, the lack of access to individual patient data restricts the ability to perform subgroup analyses, which could have identified specific patient populations more likely to benefit from EVT in MeVOs. Our meta-analysis included only a limited number of patients with dominant M2 occlusions, as robust nonrandomized data have already demonstrated the safety and benefit of EVT in this subgroup. Consequently, equipoise was lacking, and these patients were preferentially treated outside of trial settings, thereby limiting their enrollment in recent RCTs.
It is important to highlight that several ongoing international RCTs: DISTALS (Distal Ischemic Stroke Treatment With Adjustable Low-Profile Stentriever), 17 STEP (StrokeNet Thrombectomy Endovascular Platform) 18 and ORIENTAL-MeVO (Evaluation of Endovascular Treatment in Acute Intracranial Distal Medium Vessel Occlusion Stroke) 19 are evaluating EVT for MDVOs, employing MeVO-specific devices and refined patient selection strategies. The outcomes of these trials are expected to clarify the safety and efficacy of EVT in MeVO-related stroke and further define its clinical role, particularly in light of the neutral results from DISTAL and ESCAPE-MeVO. Collectively, these studies are likely to be vital in shaping future evidence-based guidelines for this stroke subtype.
Conclusions
EVT progress for managing ischemic stroke has been phenomenal and remains a promising intervention for stroke. Although the use of EVT in MeVOs is a further progress in this direction, yet its current role requires further investigation to determine optimal strategies of intervention techniques, improving patient selection, outcomes and mitigating risks associated with the procedure.
Acknowledgments
We acknowledge the authors of the included studies for their original work. This review was conducted following the PRISMA guidelines and registered with PROSPERO under the registration identification number CRD420250653970.
Author contributions
F.M.: Drafting/revision of the manuscript for content, including medical writing for content, major role in the acquisition of data, analysis or interpretation of data. B.P.: Drafting/revision of the manuscript for content, including medical writing for content, analysis or interpretation of data. P.H.: Drafting/revision of the manuscript for content, including medical writing for content, major role in the acquisition of data, analysis or interpretation of data. R.B.: Drafting/revision of the manuscript for content, including medical writing for content, major role in the acquisition of data, study concept or design, analysis or interpretation of data. All authors reviewed and approved the final manuscript.
Funding statement
No funding sources to disclose.
Competing interests
No competing interest relevant to this work.
Ethical approval
Ethical approval was not required for this study, as it is a systematic review and meta-analysis of previously published data.
