Thrombectomy for Distal Medium Vessel Occlusion: A Meta‐Analysis of Randomized Controlled Trials

Abstract

Background

Approximately 25% to 40% of all acute ischemic strokes result from distal medium vessel occlusions. However, the effectiveness and safety of thrombectomy for distal medium vessel occlusions remain uncertain.

Methods

A systematic search of PubMed, Cochrane, and Embase was conducted for randomized controlled trials published until February 2025. The primary outcomes were functional outcomes at 90 days (excellent, good, and favorable). The secondary outcomes included symptomatic intracranial hemorrhage (ICH), any ICH, and mortality at 90 days.

Results

Three studies involving 1246 patients with stroke with distal medium vessel occlusion were included (614 patients receiving thrombectomy; others with medical management). At 3 months, the odds ratio for an excellent functional outcome was 0.92 (95%, 0.72–1.17), for a good functional outcome was 0.87 (95% CI, 0.70–1.09), and for a favorable functional outcome was 0.84 (95% CI, 0.64–1.10), indicating no significant difference between the 2 groups. However, thrombectomy was associated with a higher risk of hemorrhage, with an odds ratio of 2.18 (95% CI, 1.24–3.83) for symptomatic ICH and 1.96 (95% CI, 1.54–2.50) for any ICH. Mortality at 3 months was similar between the groups, with an odds ratio of 1.20 (95% CI, 0.85–1.70).

Conclusions

Thrombectomy did not significantly improve functional outcomes or reduce mortality but increased symptomatic ICH and any ICH. Further recent randomized controlled trials are needed to define subgroups that may benefit.

Registration

PROSPERO CRD420251073344.

NONSTANDARD ABBREVIATIONS AND ACRONYMS

ACA

anterior cerebral artery

AIS

acute ischemic stroke

BCG

balloon guide catheters

DMVO

distal medium vessel occlusion

EVT

endovascular therapy

ICH

intracranial hemorrhage

MCA

middle cerebral artery

MM

medical management

NIHSS

National Institutes of Health Stroke Scale

PCA

posterior cerebral artery

RCT

randomized controlled trial

sICH

symptomatic intracranial hemorrhage

SR

stent retrievers

Clinical Perspective.

What Is New?

• This meta‐analysis of randomized controlled trials is the first to synthesize evidence on thrombectomy for distal medium vessel occlusion strokes.

• Thrombectomy did not improve functional outcomes or reduce mortality compared with medical management but was associated with increased risks of symptomatic and any intracranial hemorrhage.

What Are the Clinical Implications?

• Routine use of thrombectomy for distal medium vessel occlusion should not be recommended based on current evidence, given the absence of functional benefit and increased hemorrhagic risk.

• Careful patient selection remains crucial; ongoing, and future randomized trials may help define subgroups that could derive benefit from thrombectomy.

Mechanical thrombectomy has demonstrated effectiveness for proximal large vessel occlusion within 24 hours of symptom onset. However, landmark trials have excluded patients with distal medium vessel occlusion (DMVO), which accounts for 25% to 40% of acute ischemic strokes (AIS). ¹ Current guidelines assign a Class IIb recommendation for endovascular therapy in DMVO, reflecting limited evidence. ² The DUSK trial ³ demonstrated that ≈25% of patients receiving medical management (MM) experienced functional disability or mortality yet failed to show superior outcomes with thrombectomy compared with MM.

A recent randomized controlled trial (RCT) evaluated thrombectomy in patients with AIS within 12 hours of symptom onset with occlusions in the M2/M3 segments of the middle cerebral artery (MCA), the A2/A3 segments of the anterior cerebral arteries (ACA), or the P2/P3 segments of the posterior cerebral arteries (PCA). ⁴ Endovascular therapy (EVT) did not reduce disability and was associated with increased symptomatic intracranial hemorrhage (sICH) and mortality at 90 days compared with usual care. The interim analysis of the DISCOUNT trial ⁵ yielded similar results. However, these studies were limited by small subgroup sizes across vascular territories and 1 study’s premature termination following planned interim analysis. Therefore, this meta‐analysis aims to evaluate the effectiveness of thrombectomy in patients with stroke with DMVO.

METHODS

This systematic review of recent RCTs was conducted in accordance with Preferred Reporting Items for Systematic Reviews and Meta‐Analyses guidelines (Table S1). ⁶ The study protocol was registered at the Open Science Framework (protocol available at https://osf.io/jca2v) and PROSPERO (CRD420251073344). Institutional review board approval was waived because this research only involved analysis of existing published data. Data are available upon reasonable request to the corresponding author.

Eligibility Criteria

We included recent RCTs that recruited patients with stroke with DMVO receiving EVT within 24 hours. DMVO was defined as cerebral artery occlusion in the M2/M3 segments of the MCA, the A1/A2/A3 segments of the ACA, or the P1/P2/P3 segments of the PCA. EVT includes thrombectomy using stent retriever (SR), direct aspiration thrombectomy, combined technique, intraarterial lysis, percutaneous transluminal angioplasty, or rescue stenting.

Search Strategy

We systematically searched MEDLINE (via PubMed), Embase, and Cochrane CENTRAL from inception to February 12, 2025, without publication restrictions. No language filters were applied during the search. Non‐English articles were assessed using English abstracts or translated full texts when necessary. The search strategy, developed by stroke specialists (HJJ and LLY), combined Medical Subject Headings and keywords related to “ischemic stroke,” “medium/distal vessel,” and “thrombectomy” (Table S2). Reference lists from included studies or systematic reviews were manually examined to identify additional eligible trials. To locate relevant unpublished material, we expanded our search to include clinical trial registries (eg, ClinicalTrials.gov) and academic conferences.

Study Selection

Three reviewers (HJJ, WSW, or LYY) independently screened titles and abstracts of identified references, followed by full‐text assessment of potentially eligible RCTs. Disagreements were resolved by consensus or third‐reviewer consultation (PHC or CHL). Nonrandomized studies and those without control groups were excluded. For overlapping cohorts, only the most recent studies were included.

Data Collection

Data extraction was performed independently by 3 reviewers (HJJ, WSW, or LYY) using a standardized form, with discrepancies resolved through consensus or consultation with additional reviewers (PHC or CHL). We extracted study characteristics, methodology, participant demographics, intervention details, and outcomes. The corresponding authors were contacted to obtain aggregate data. No data imputation was performed for missing values.

Risk of Bias

The methodological quality of included trials was assessed using the Cochrane Risk of Bias Tool (version 2) by 2 reviewers (HJJ, WSW) without conflicts of interest, and disagreements were adjudicated by a third reviewer (PHC or CHL). While the inherent inability to blind interventional procedures was acknowledged in the assessment, all other domains were evaluated according to standard criteria.

Outcome Measurement

The primary outcomes were as follows: excellent functional outcome, defined as modified Rankin Scale (mRS) score 0–1 at 90 days; good functional outcome (modified Rankin Scale score 0–2) at 90 days; and favorable functional outcome (modified Rankin Scale score 0–3) at 90 days. Secondary outcomes included sICH (ICH with neurological deterioration), any ICH, and mortality at 90 days.

Data Synthesis

Following Cochrane methodology, we used both fixed‐effect and random‐effects (DerSimonian‐Laird) approaches for our statistical evaluation; however, given the small number of included studies, we adopted a fixed‐effect model as the primary analysis due to the instability and imprecision of τ² estimates with limited data. ⁷ , ⁸ , ⁹ For binary end points, we generated odds ratios (OR) with corresponding 95% CIs. We primarily assessed heterogeneity using the I ² metric and Cochrane’s Q statistic (with P <0.10 indicating significance). Additionally, we used the synergy index, calculated as the OR between studies with corresponding 95% CI, to explore pairwise heterogeneity, especially when most data came from a few dominant trials. ¹⁰ Publication bias assessment involved visual examination of funnel plots supplemented by Egger’s regression test. All computations were executed in R environment utilizing ‘metafor’ and ‘meta’ packages, with alpha set at 0.05. ¹¹ , ¹² , ¹³

Sensitivity Analysis, Subgroup Analysis, Meta‐Regression, and Trial Sequential Analysis

We conducted sensitivity analyses excluding the unpublished trial and subgroup analyses stratified by overall risk of bias level. Meta‐regression analyses were conducted to investigate sources of heterogeneity by evaluating the influence of study‐level covariates: mean age, thrombolysis rate, mean National Institutes of Health Stroke Scale (NIHSS) score, and time from last known well to randomization. Trial sequential analysis was performed using the Copenhagen Trial Unit’s software (v0.9.5.10 Beta; www.ctu.dk/tsa) with a relative risk reduction derived from the pooled effect estimates. We applied the O’Brien‐Fleming boundary, assuming 80% power and a 2‐sided α of 0.05. ¹⁴ , ¹⁵

Confidence in the Cumulative Evidence

Through application of the Grading of Recommendations Assessment, Development and Evaluation methodology, we critically appraised the methodological quality and cumulative strength of evidence comparing endovascular intervention with MM across various outcome measures. ¹⁶

RESULTS

Search Strategy

Our comprehensive search strategy initially retrieved 120 citations, with preliminary screening eliminating 5 duplicate records (Figure S1). Subsequent title, abstract, and full‐text evaluations progressively narrowed the corpus, with 102 studies systematically excluded based on predefined exclusion criteria. Thirteen potentially relevant articles underwent meticulous review, with 1 further eliminated due to insufficient relevance. Complementary searching through alternative sources (including websites, conference proceedings, and citation tracking) identified 10 additional potential studies. After rigorous methodological filtering to exclude protocols, nonrandomized investigations, and conference posters, a definitive cohort of 3 RCTs emerged for comprehensive analysis. The final selected 3 studies encompassed 1246 participants, with 614 patients receiving thrombectomy and the remaining cohort assigned to MM. Detailed trial characteristics are comprehensively documented in Table. Patients’ ages ranged from 72.0 to 77.5 years, with median NIHSS scores of 6 to 8. The thrombolysis administration rates varied between 56.5% and 75.0%. The predominant occlusion site was the MCA, accounting for 70.0% to 85.1%, followed by the PCA and ACA.

Table 1.

Characteristics of Included Studies

Author, year	Study design	Country	No. of patients	Age (y)	Female (%)	NIHSS	Hypertension (%)	Diabetes (%)	Dyslipidemia (%)	AF (%)	Previous stroke (%)	Heart disease (%)	Thrombolysis (%)	Occlusion site, MCA (%)	Occlusion site, ACA (%)	Occlusion site, PCA (%)	Successful recanalization (%)	Overall RoB
Goyal, 4 2025	RCT	International multicenter	EVT: 255 MM: 275	74 (63–82) 76 (65–83)	46.3 46.4	8 (6–11) 7 (5–11)	72.2 77.4	23.1 27.0	23.1 27.0	31.8 27.7	23.9 20.4	24.3 29.6	56.5 60.2	85.1 81.8	9.4 4.4	4.7 11.6	75.1	Low
Psychogios, 17 2025	RCT	International multicenter	EVT: 277 MM: 276	77 (68–83) 77.5 (68–84)	42.8 45.2	6 (5–9) 6 (5–9)	66.1 64.7	24.0 24.3	24.0 24.3	25.1 25.4	22.5 21.7	26.2 30.1	62.0 68.8	70.0 70.3	7.2 4.0	20.0 23.4	71.7	Some concern
Clarençon, 5 2025	RCT	France multicenter	EVT: 82 MM: 81	75 (11)* 72 (13)*	46 48	8 (6–11) 8 (6–12)	NA	NA	NA	NA	NA	NA	65.4 75.0	75.0 76.5	7.5 7.4	17.5 16.0	NA	Some concern

Successful Recanalization = a modified Thrombolysis in Cerebral Infarction score of ≥2b. ACA indicates anterior cerebral artery; AF, atrial fibrillation; EVT, endovascular therapy; MCA, middle cerebral artery; MM, medical management; NA, not available; NIHSS, National Institutes of Health Stroke Scale; PCA, posterior cerebral artery; RCT, randomized controlled trial; and RoB, risk of bias.

^*Mean±SD.

Primary Outcomes

In 2 trials (n=1083), there was no significant difference in the rate of achieving excellent functional independence between thrombectomy and MM (37.6% versus 39.7%; OR, 0.92 [95% CI, 0.72–1.17]; I ²=0%, P=0.81; Figure 1A). ⁴ , ¹⁷ Similarly, across 3 trials (n=1246), no difference was observed in the rate of good functional independence (54.7% versus 58.1%; OR, 0.87 [95% CI, 0.70–1.09]; I ² = 63.6%, P=0.06; Figure 1B). ⁴ , ⁵ , ¹⁷ Furthermore, in 2 trials (n=1083), no significant difference was noted in favorable functional independence (72.2% versus 75.7%; OR, 0.84 [95% CI, 0.64–1.10]; I ²=0%, P=0.48; Figure 1C). ⁴ , ¹⁷

Figure 1. Forest plots and pooled estimations among patients with stroke with distal medium vessel occlusion postendovascular therapy.

A, Excellent functional outcome at 3 months. B, Good functional outcome at 3 months. C, Favorable functional outcome at 3 months. EVT indicates endovascular therapy; MM, medical management; and OR, odds ratio.

Secondary Outcomes

For sICH, 3 trials (n=1246) showed a significantly higher rate with thrombectomy compared with MM (6.4% versus 3.0%; OR, 2.18 [95% CI, 1.24–3.83]; I ²=0%, P=0.60; Figure 2A). ⁴ , ⁵ , ¹⁷ For any ICH, 3 trials (n=1246) also demonstrated a higher rate with thrombectomy (38.9% versus 24.7%; OR, 1.96 [95% CI, 1.54–2.50]; I ²=61.1%, P=0.08; Figure 2B). ⁴ , ⁵ , ¹⁷ Regarding 90‐day mortality, no significant difference was observed between the 2 groups (12.7% versus 10.8%; OR, 1.20 [95% CI, 0.85–1.70]; I ²=59.3%, P=0.09; Figure 2C). ⁴ , ⁵ , ¹⁷

Figure 2. Forest plots and pooled estimations among patients with stroke with distal medium vessel occlusion postendovascular therapy.

A, Symptomatic intracranial hemorrhage. B, Any intracranial hemorrhage. C, Mortality at 3 months. EVT indicates endovascular therapy; ICH, intracranial hemorrhage; MM, medical management; OR, odds ratio; and Symp., symptomatic.

Heterogeneity

Synergy indices are summarized in Table S3. No substantial heterogeneity was observed between the ESCAPE‐MeVO ⁴ and DISTAL ¹⁷ trials across all outcomes. However, significant pairwise differences were observed when compared with the DISCOUNT trial. ⁵ For good functional outcome, the synergy index was 2.40 (95% CI, 1.18–4.89) between the DISTAL ¹⁷ and DISCOUNT ⁵ trials. For any ICH, synergy indices were 2.24 (95% CI, 1.07–4.73) and 2.24 (95% CI, 1.06–4.77) when comparing the DISCOUNT ⁵ with ESCAPE‐MeVO ⁴ and DISTAL ¹⁷ trials, respectively. A difference in mortality was also found between the ESCAPE‐MeVO ⁴ and DISCOUNT ⁵ trials, with a synergy index of 6.50 (95% CI, 1.16–36.54).

Sensitivity Analysis, Subgroup Analysis, Meta‐Regression, and Trial Sequential Analysis

In sensitivity analysis, heterogeneity was reduced for good functional outcomes, any ICH, and mortality (Figure S2). The overall trend remained consistent, with only slight increases in the pooled estimates for sICH and any ICH. In the subgroup analysis, heterogeneity was not reduced when stratified by risk of bias. These findings suggest that the DISCOUNT trial ⁵ may have contributed disproportionately to between‐study variation (Figure S3). Meta‐regression analysis examined the relationship between 4 variables (mean age, rate of thrombolysis, mean NIHSS score, and mean time from onset to randomization) and all outcomes. No significant correlations were found between these variables and the outcomes (Table S4). However, the DISCOUNT trial ⁵ did not provide data on excellent functional outcomes, favorable outcomes, or the time from onset to randomization. As a result, the effect size for these parameters could not be estimated. Trial sequential analysis results showed that z‐statistic lines for functional outcomes and 90‐day mortality did not cross monitoring boundaries. For sICH, the boundaries were crossed, and the required information size was reached for any ICH. Detailed plots are provided in Figure S4.

Summary of Findings and Recommendations

Only the ESCAPE‐MeVO trial ⁴ demonstrated low overall risk of bias, while the other trials had some concerns in the quality assessment (Figure S5), despite lack of double blinding. No publication bias was detected for any outcomes (Figure S6). The quality of evidence per Grading of Recommendations, Assessments, Developments, and Evaluations criteria was as follows: moderate for excellent functional outcome at 90 days; low for good functional outcome; moderate for favorable functional outcome; high for sICH; moderate for any ICH; and low for mortality at 3 months (Table S5).

DISCUSSION

In this meta‐analysis of RCTs involving patients with AIS due to DMVO, EVT did not significantly enhance functional outcomes or lower mortality compared with usual care but was linked to elevated risks of sICH and any ICH. Approximately 25% of patients receiving the best medical therapy still had substantial disability at 90 days, consistent with previous cohort studies.

Recent meta‐analyses have shown that SR provide higher reperfusion rates than aspiration catheter alone (86.1% versus 77.5%; OR, 1.80 [95% CI, 1.17–2.78]), while combined SR and aspiration catheter techniques improve functional independence (OR, 1.33 [95% CI, 1.06–1.67]) and reduce mortality (OR, 0.69 [95% CI, 0.50–0.94]) compared with aspiration catheter alone. ¹⁸ However, our findings indicate that SR‐based thrombectomy lacked similar efficacy. The ESCAPE‐MeVO trial ⁴ used Solitaire X devices as the first attempt, while the DISTAL trial ¹⁷ used a combined approach. Unlike proximal large vessels, distal medium cerebral vessels are smaller, more tortuous, and have thinner walls, increasing the risk of device–vessel mismatch and procedural injury. Newer smaller devices, such as the Tigertriever 13 (Rapid Medical, Yoqneam, Israel) and RED 43 aspiration catheter (Penumbra, Alameda, CA), have demonstrated promising recanalization rates (≈83%) and favorable safety in DMVOs. ¹ , ¹⁹ , ²⁰ However, high‐quality data remain limited, and identifying the optimal device strategy in this setting remains a key research priority. The ongoing DISTALS trial (ClinicalTrials.gov; identifiers: NCT05152524) using the Tigertriever 13 may offer further insights. Future studies should ensure device‐specific outcome reporting to clarify comparative effectiveness.

The subgroup analysis by occlusion location did not demonstrate consistent effectiveness of EVT across all vessels, with current evidence largely driven by MCA occlusions. Due to small sample sizes, particularly in the ACA and PCA, the confidence intervals were widened, and the efficacy remains uncertain. However, specific technical challenges should be carefully considered to avoid procedural complications. In ACA occlusions, the acute angle (<90 degrees) between the A1 segment of the ACA and the C7 segment of the internal carotid artery can reduce device accessibility. Additionally, increased vessel tortuosity raises the risk of hemorrhage, especially when the SR is withdrawn without the support of a distal access catheter. ²¹ In MCA occlusions, the high frequency of early bifurcation leads to inconsistent definitions and variability in inclusion and exclusion criteria. ²² Furthermore, differences in the functional significance of the superior and inferior M2 branches may affect EVT effectiveness, due to the distinct brain regions they supply. For isolated PCA occlusions, the PLATO study indicated that EVT did not improve functional independence and was associated with a higher risk of sICH and mortality. ²³ Similar outcomes were observed in a meta‐analysis, raising concerns about the effectiveness and safety of EVT in this subgroup. ²⁴

In the DISTAL trial, ¹⁷ balloon guide catheters (BGC) were used in ≈38% of cases to address distal embolization and tandem lesions. BGC can reduce antegrade flow and distal embolization. Post‐hoc analyses suggest that BGC use in large vessel occlusions of the anterior circulation improves recanalization rates, first‐pass effect, procedural time, and functional outcomes. ²⁵ , ²⁶ , ²⁷ , ²⁸ However, an RCT suggests that BGC may not restore functional independence due to prolonged puncture‐to‐reperfusion times, reduced collateral flow, and increased risk of carotid artery vasospasm, which could confound results. ²⁹ Similarly, Orscelik et al ³⁰ reported that BGC use in 123 of 231 patients was associated with lower modified Thrombolysis in Cerebral Infarction scores, decreased first‐pass effect rates, and longer procedure times, indicating that BGCs may not enhance functional outcomes. Additionally, variations in inner lumen diameters can affect compatibility with thrombectomy devices, further reducing navigation and accessibility.

Because the occlusion site occurred distal to a large vessel occlusion, patients typically exhibited milder clinical deficits, with NIHSS scores ≈6 to 8. The question of whether EVT should be performed in patients with stroke with mild symptoms and proximal large vessel occlusions remains debated. A meta‐analysis of 14 studies suggests that EVT is safe and feasible but may not reduce disability. ³¹ However, in the subgroup analysis of the DISTAL trial, ¹⁷ there was a trend toward better improvement in the mRS for patients with severe neurological deficits (NIHSS >9), aligning with current guidelines for patients with anterior circulation occlusions. Similarly, the post‐hoc analysis of the PLATO study indicated that baseline stroke severity is an important modifier for outcomes in isolated PCA occlusion. ³² The ongoing ORIENTAL MeVo (ClinicalTrials.gov; identifiers: NCT06146790) specifically targets patients with higher stroke severity (NIHSS ≥6), aiming to clarify the benefit of EVT in this subgroup.

Compared with the HERMES trials, ³³ patients in this analysis were older, had a lower rate of thrombolysis, and experienced a longer time from last known well to randomization. Evidence for thrombolysis in patients with minor stroke was also limited. Coutts et al ³⁴ demonstrated that despite higher recanalization rates, thrombolysis failed to improve functional outcomes compared with standard care, with numerically increased sICH. Procedure delays may be attributed to subtle symptomatology, more nuanced decision‐making processes, and delays in arranging general anesthesia. ³⁵ Although Ospel et al ³⁶ found that early successful recanalization was strongly associated with better outcomes, recanalization times were not provided for further analysis. Notably, subgroup analysis revealed that EVT was favored in the ESCAPE‐MeVO trial ⁴ (median time from last known well to randomization <260 minutes), whereas MM was favored in the DISTAL trial ¹⁷ (median time <360 minutes), suggesting that longer time to treatment may lead to divergent outcomes. These factors may flaw the outcomes. Fortunately, the promising results from the ANGEL‐TNK and PEARL trials ³⁷ , ³⁸ suggest that intra‐arterial thrombolysis followed by thrombectomy could help restore distal perfusion and potentially improve outcomes. Ongoing trials such as DUSK (ClinicalTrials.gov; identifiers: NCT05983757; comparing 0 to 6 hours versus 6 to 12 hours) and FRONTIER‐AP (Australian New Zealand Clinical Trials Registry; identifiers: ACTRN12621001746820p; 0 to 9 hours) are designed to evaluate the effectiveness of EVT across different time windows.

Perfusion imaging is commonly used for reperfusion therapy in the late time window, as demonstrated in trials such as DIFFUSE and DAWN. ³⁹ , ⁴⁰ However, it is time‐consuming and exposes patients to more contrast compared with computed tomography angiography (CTA) alone. A retrospective study involving 13 774 patients with AIS showed that the use of computed tomography perfusion (CTP) within the first 6 hours increased EVT treatment rates from 8.5% to 20.1% compared with CTA alone, with most of the findings attributed to DMVO rather than proximal large vessel occlusions. The median time to reperfusion therapy was slightly faster in CTP‐triaged patients compared with those receiving CTA alone. ⁴¹ Furthermore, these neutral results highlight the need for improved patient selection in DMVO. Higher cerebral blood volume index and hypoperfused tissue volumes were associated with excellent thrombectomy outcomes. ⁴² , ⁴³ In our meta‐analysis, we observed a low incidence of contrast‐associated acute kidney injury, with no significant correlation between contrast volume and kidney injury rate (mean volume: 184.7 mL; range: 71.2–251.5 mL). ⁴⁴ Overall, perfusion imaging remains a crucial tool for accurate diagnosis and identifying salvageable tissue without adding significant risk.

Strength and Limitations

To our knowledge, this is the first meta‐analysis of RCTs examining the effectiveness and benefits of thrombectomy for patients with stroke with DMVO. Trial sequential analysis was conducted to minimize type I and type II errors and provided cumulative information. Nevertheless, certain constraints must be considered when evaluating the findings of this investigation. Subgroups of patients may respond differently to thrombectomy, which could be better explored through individual patient data meta‐analysis. To begin with, the definition of anatomy varied, although a recent international consensus statement has proposed a systematic classification for the distal, middle‐size cerebral vasculature. ⁴⁵ Future studies should adopt these standardized definitions to enhance consistency and comparability. Furthermore, most studies focused on MCA occlusion, leaving small sample sizes for ACA and PCA occlusions. The ESCAPE‐MeVO trial ⁴ did not include ACA‐A1 and PCA‐P1, potentially precluding meaningful conclusions for these territories and introducing selection bias. Furthermore, the DISCOUNT trials, ⁵ based on conference data, appeared to show a divergent trend and contributed to heterogeneity, as reflected in the synergy index and sensitivity analyses, so caution is advised in interpreting its findings. In addition, the inclusion of only 3 trials, with most patients extracted in just 2 studies (ESCAPE‐MeVO ⁴ and DISTALS trials ¹⁷ ) and a small sample size in the DISCOUNT trial, ⁵ compromises the external validity and reliability of our pooled estimates. Moreover, the mean time from last known well to randomization was <6 hours, leaving the effectiveness of the EVT in the late time window unclear. Additionally, most trials predominantly involved North American and European populations, which may limit applicability to other populations. Also, because <10 trials were included, there remained limited power to detect publication bias. Furthermore, technical heterogeneity persisted across the included studies. Insufficient details regarding thrombectomy strategies, the permitted number of thrombectomy attempts, and devices precluded device‐specific outcome analysis. Another key point is that post‐thrombectomy care protocols, including rehabilitation, antihypertensive therapy, and antiplatelet management, lack individualization and long‐term follow‐up data. Finally, 4 RCTs are currently underway (Table S6). The DISTALS trial aims to evaluate the effectiveness of an adjustable low‐profile SR for the treatment of distal ischemic stroke. The DUSK trial enrolled patients within 12 hours of symptom onset and compared outcomes between the 0 to 6 hour and 6 to 12 hour time windows. The ORIENTAL MeVo trial was designed to investigate the effectiveness of EVT in patients with DMVO presenting with severe neurological symptoms (admission NIHSS ≥6). The FRONTIER‐AP trial focused exclusively on patients with anterior circulation stroke. These trials are expected to address current limitations and inform future clinical practice.

CONCLUSIONS

Thrombectomy was not found to significantly regain neurological functional status or reduce mortality, while it was associated with higher rates of sICH and any ICH. Further randomized trials are warranted to identify subgroups that may benefit.

Sources of Funding

This study was supported by the Tri‐Service General Hospital: TSGH‐D‐114168.

Disclosures

None.

Supporting information

Tables S1–S6

Figures S1–S6