Endovascular Treatment for Medium and Distal Vessel Occlusion Stroke: Are We Missing the Point?

Abstract

BACKGROUND:

Recent randomized controlled trials of endovascular treatment (EVT) for distal and medium vessel occlusion (DMVO) stroke did not demonstrate clinical benefit over best medical therapy and suggested potential safety concerns. We aimed to compare baseline characteristics of patients with DMVO stroke enrolled in these trials with those treated with EVT in routine clinical practice (RCP).

METHODS:

We conducted a retrospective cohort study of a prospectively recorded dual-center database including all consecutive patients with an acute ischemic stroke due to isolated primary DMVO (distal or nondominant M2, M3, A1, A2, A3, P1, P2) between January 2022 and July 2024 who fulfilled trials inclusion/exclusion criteria. Both centers participated in the DMVO trials during the study period. Patients were categorized in 2 groups: those who underwent EVT under RCP (RCP group) versus those randomized in a trial for DMVO stroke (clinical trial, computed tomography group). The 2 groups were evaluated and compared based on demographic, clinical, imaging, and treatment characteristics.

RESULTS:

Among 182 patients (median age, 79 years; interquartile range [IQR], 69–86; 46.2% female), 53 (29.1%) were included in a trial, and 129 (70.9%) were treated in RCP. Compared with RCP patients, those in the computed tomography group presented lower baseline stroke severity (median NIHSS, 5 [IQR, 4–7] versus 10 [IQR,7–15]; P<0.001), smaller admission hypoperfusion volumes (27 mL [IQR, 16–42] versus 48 mL [IQR, 24–70]; P<0.001) and a trend toward more vascular risk factors (2, [IQR, 1–3] versus 1 [1–2] in RCP patients; P=0.062) was observed. Angiographic outcomes among patients undergoing EVT trended toward higher successful reperfusion in the RCP group (87.6% versus 72.0%; P=0.051).

CONCLUSIONS:

In our study, conducted at 2 centers participating in DMVO clinical trials, randomized patients presented milder clinical severity and smaller hypoperfusion areas compared with those treated in daily practice, suggesting a potential selection bias. Our findings highlight the importance of consecutive recruitment in future DMVO trials to clarify the role of EVT.

CLINICAL PERSPECTIVE.

What Is New?

• Patients who underwent endovascular treatment as part of a clinical trial showed lower National Institutes of Health Stroke Scale scores, smaller hypoperfusion volumes on admission, and the rate of successful recanalization was numerically lower compared with those treated in routine clinical practice.

What Are the Clinical Implications?

• Patients with an acute ischemic stroke due to distal/medium vessel occlusion may not be automatically excluded from endovascular treatment, and, instead, individual clinical and imaging features may be taken into account.

Endovascular thrombectomy (EVT) has become the standard of care for the treatment of acute ischemic stroke due to large vessel occlusion involving anterior circulation and basilar artery.^1,2 However, 24% to 40% of acute ischemic stroke cases involve distal medium vessel occlusions (DMVO), for which the role of EVT remains uncertain.³

Recently, 2 randomized controlled trials (DISTAL [Endovascular Therapy Plus Best Medical Treatment (BMT) Versus BMT Alone for Medium Vessel Occlusion Stroke] and ESCAPE-MeVO [Endovascular Treatment to Improve Outcomes for Medium Vessel Occlusions]) have failed to demonstrate superiority of EVT over best medical management for the treatment of DMVO stroke and signaled potential safety concerns, showing higher rates of intracranial hemorrhage and mortality in the EVT arm.^4,5 In addition, an interim analysis of the DISCOUNT trial recommended its premature discontinuation due to safety concerns and futility of the procedure.

These neutral results have prompted critical evaluation within the scientific community, aiming to elucidate potential methodological limitations and clinical variables that may have contributed to these findings, particularly given the poor natural history of DMVO stroke: nearly half of patients do not achieve excellent recovery, and up to one-third remain functionally dependent at 90 days despite best medical care.⁶ Most importantly, included patients presented a relatively low baseline severity stroke (median National Institutes of Health Stroke Scale [NIHSS], 6–8) compared with previous registries,^7–9 elder age than previous stroke trials,^1,10 and lower-than-expected recanalization that may dilute the effect of thrombectomy. Moreover, available thrombectomy devices were originally designed for proximal large vessels, raising concerns about technical efficacy and safety in smaller, more tortuous distal arteries.

The aim of our study was to compare the baseline clinical and imaging characteristics of patients with DMVO stroke treated with EVT in routine clinical practice with those enrolled in DISTAL and ESCAPE-MeVO during the same time period.

Methods

Anonymized data supporting this study’s findings are available for any qualified investigator on reasonable request to the corresponding author. Data were prospectively recorded in an institutional database approved by the local ethics committees. Because of the retrospective nature of the study, the need for written informed consent was waived.

This study followed the STROBE statements (Strengthening the Reporting of Observational Studies in Epidemiology).¹¹

Study Design and Population

We conducted a retrospective comparative study using a prospectively recorded, dual-center database. Both centers participated in randomized clinical trials focused on DMVO stroke, with 1 involved in the DISTAL trial and the other in the ESCAPE-MeVO trial. Both centers have 24/7 clinical, interventional, and research capabilities, with no restrictions on patient enrollment, and perform between 200 and 400 thrombectomies per year. We included all consecutive patients with an acute ischemic stroke due to isolated primary DMVO, defined as distal or nondominant M2, M3, A1, A2, A3, P1, P2 on admission CT angiography, who underwent EVT and were randomized into a DMVO clinical trial from January 2022 to July 2024. Those patients with DMVO who were not randomized and did not undergo EVT were excluded. This time period was selected to coincide with the enrollment period of both randomized controlled trials (RCTs).

The cohort was divided into 2 groups according to whether patients were enrolled in a distal occlusion clinical trial (CT group) or were treated according to the routine clinical practice (RCP) at the treating center (RCP group). All included patients met the eligibility criteria for the DISTAL or ESCAPE-MeVO clinical trials (Supplemental Material; Tables 1 and 2).

Table 1.

Baseline Characteristics and Demographics

Table 2.

Angiographic Data, Overall and by Subgroups Regarding Whether They Were Included in a RCT or Not

Clinical and Radiological Parameters

Baseline characteristics comparison was defined as the primary objective. Recorded variables in both groups included age, sex, baseline degree of disability measured by modified Rankin Scale, presence of vascular risk factors that included hypertension, diabetes, dyslipidemia and smoking habit, stroke severity assessed by NIHSS on admission, admission Alberta Stroke Program Early Computed Tomography Score, occlusion location based on initial CT angiography, and CT perfusion parameters automatically analyzed by RAPID software.

Treatment-related differences were defined as the secondary objective. We also registered intravenous thrombolysis rate administered before EVT, workflow times (onset to imaging, onset to puncture, and puncture to recanalization times), and endovascular treatment characteristics. Angiographic outcomes were defined according to the expanded Thrombolysis in Cerebral Infarction grading system at each center.

Functional outcome at 90 days was described but was not included in the analysis.

Statistical Analysis

Kolmogorov–Smirnov and Shapiro–Wilk tests were used to ensure normality of continuous variables. Categorical variables were presented as absolute values and percentages, and continuous variables as median (interquartile range [IQR]) or means (±SD) as indicated. Statistical significance for intergroup differences was assessed by the Pearson χ² test or Fisher exact test for categorical variables and by the Mann-Whitney U test or Student t test as appropriate to continuous variables. There was no missing data.

A P<0.05 was considered statistically significant. All analyses were performed using the IBM SPSS Statistics (version 25) software and R (version 4.3.0, R Foundation for Statistical Computing, Vienna, Austria).

Results

From January 2022 to July 2024, a total of 182 patients with an acute ischemic stroke due to isolated DMVO were included in the study. Fifty-three patients (29.1%) were enrolled in DISTAL or ESCAPE-MeVO trials (CT group), of whom 25 (47.2%) received EVT. The remaining 129 (70.9%) patients received EVT according to routine clinical practice at the participating centers (RCP group).

Clinical Characteristics

The overall median age was 79 years (IQR, 69–86), 84 patients (46.2%) were female, and the median baseline modified Rankin Scale was 1 (IQR, 0–2). There were no differences between groups in terms of age, sex, or premorbid functional status. However, patients in the CT group presented a trend toward more vascular risk factors (2, [IQR 1–3] versus 1 [1–2] in RCP patients; P=0.062). The CT group had a significantly lower baseline NIHSS score (5 [4–7] versus 10 [7–15] in RCP; P<0.001). A nonsignificant trend toward a longer onset-to-imaging interval was observed in the CT group (median, 431 minutes [IQR, 150–630] versus 265 minutes [IQR, 100–606]; P=0.149). Table 1 summarizes the baseline characteristics, demographics, and neuroimaging parameters.

Admission Neuroimaging Parameters

Overall, the occlusion was located in 126 (69.2%) patients in the MCA (M2 in 107 [58.8%], M3 in 19 [10.4%]); in 24 (13.2%) patients in the ACA (A2 in 16 [8.8%], A3 in 8 [4.4%]), and in 32 [17.6%] patients in the PCA (P1 in 13 [7.1%], P2 in 19 [10.4%]). No between-group differences were observed concerning occlusion location (Table 1).

The median Alberta Stroke Program Early Computed Tomography Score on admission was 10 (IQR, 9–10), with no between-group significant differences. On CT perfusion, the CT group revealed significantly lower hypoperfusion volume at admission (median Tmax <6s in the CT group, 27 mL [16–42] versus 48 mL [24–70] in the RCP group; P<0.001).

Angiographic and Procedural Data

Among patients who underwent EVT, thrombectomy technique selection was similar between groups (combined technique: RCP, 101 [80.8%] versus CT 17 [68%]; direct aspiration: RCP 24 [19.2%] versus 2 [8%]; P=0.527) and successful reperfusion (expanded Thrombolysis in Cerebral Infarction, 2b–3) was numerically lower in the CT group compared with the RCP group (72.0% versus 87.6%; P=0.051). Rates of first-pass effect and complete reperfusion (expanded Thrombolysis in Cerebral Infarction, 2c–3) did not differ significantly between groups. Multiple passes were performed in only 28% of patients in the CT group, versus 41.9% in the RCP group (P=0.29).

Safety outcomes were similar across groups, with no significant differences in the incidence of vessel perforation, arterial dissection, or symptomatic intracranial hemorrhage despite a numerical imbalance against the RCP group. Table 2 summarizes the angiographic and procedural data of patients who underwent EVT.

Discussion

Our study shows that the majority of patients presenting with a DMVO stroke were treated with mechanical thrombectomy according to routine clinical practice (70.9%) and were not enrolled in any of the ongoing RCTs during the study period. Importantly, patients recruited into clinical trials presented with lower stroke severity, smaller perfusion deficits, and a trend to higher burden of vascular comorbidities compared with those treated outside of trials. These findings suggest a lack of equipoise and a potential selection bias, whereby trial enrollment favored patients with milder deficits, while those with more severe presentations were preferentially managed with EVT in routine care, and raise a concern about the generalizability of clinical trial results for all patients with DMVO.

Our data are consistent with published clinical trials, as the mean baseline NIHSS scores were 6 (5–9) and 8 (5–11) in the DISTAL and ESCAPE-MeVO trials, respectively.^4,5 By contrast, nonrandomized observational studies of EVT in DMVO commonly report higher baseline NIHSS scores, ranging from 8 to 14,^6,9,12–15 and demonstrate signal benefit in selected patients.^16,17 The underrepresentation of higher-risk patients in DMVO trials may have diluted potential treatment effects and contributed to neutral results. This phenomenon resembles the experience with early large vessel occlusion thrombectomy trials (IMS-III [Interventional Management of Stroke III], MR RESCUE [Mechanical Retrieval and Recanalization of Stroke Clots Using Embolectomy], SYNTHESIS [Local Versus Systemic Thrombolysis for Acute Ischemic Stroke])¹⁸ and basilar occlusion thrombectomy trials,^19,20 which were initially negative but later overcame by trials with more appropriate patient selection, consecutive randomization, and contemporary devices.¹

Several studies have suggested that EVT may be more beneficial for patients presenting with higher baseline clinical deficits (NIHSS >6) and larger stroke volumes on initial imaging (Tmax >6>60 mL). Similarly, a greater treatment effect was observed in patients treated within 180 minutes from symptom onset in the ESCAPE-MeVO trial. However, as observed in our study, these patients are underrepresented in both clinical trials. It is reasonable to assume that in patients with milder symptoms and a smaller penumbra area, demonstrating a net clinical benefit of EVT is more challenging and requires careful evaluation of RCT protocols.

Clinical trial data suggest that EVT of distal occlusions may carry a higher risk of complications, such as arterial dissection or vessel perforation, compared with large vessel occlusions, owing to the more distal location of the target lesion, the greater technical difficulty of catheter navigation, and the increased vascular tortuosity.²¹ Dedicated devices optimized for DMVO navigation and clot retrieval in smaller and more tortuous vessels may be needed to maximize benefit. However, the safety profile of EVT in our study was comparable across groups, with no significant increase in periprocedural complications or symptomatic intracranial hemorrhage.

Moreover, regarding angiographic and technical characteristics of EVT, we have observed a trend in fewer number of passes and lower rate of successful reperfusion in the CT group (87.6% opposed to 72% in trial patients). This may reflect that the neurointerventional approach varies according to whether patients are managed within the framework of a clinical trial or routine care. A potential reason might be that more distal occlusions are characterized by having longer and more tortuous endovascular access, as evidenced by smaller hypoperfusion volumes on CT perfusion maps.

Our findings highlight that the scientific community may have engaged in a reverse cherry-picking phenomenon, preferentially including patients with milder symptoms in clinical trials, while treating those with more severe presentations in routine clinical practice under the assumption that endovascular therapy would be beneficial. The neutral results of recent DMVO trials should not be interpreted as definitive evidence of futility, but rather as a reflection of a lack of consecutive randomization and trial design limitations. Ensuring consecutive recruitment within the framework of clinical equipoise is essential to avoid selection bias and ensure the external validity of the findings. Moreover, future study designs should prioritize: (1) enrichment strategies targeting patients with higher NIHSS scores, larger perfusion lesions, or unfavorable collateral profiles,(2) shorter onset-to-treatment intervals, and (3) the use of dedicated or optimized devices for distal thrombectomy.

The main strength of our study was the inclusion of consecutive patients treated in real-world daily clinical practice. In addition, this study was performed in 2 different centers with patient inclusion in 2 different RCTs for DMVO strokes, which reinforces the external validity of the obtained results.

The main limitations of our study include a retrospective analysis and the difference in sample size across groups. An additional methodological concern is the lack of imaging evaluation by a central core lab imaging analysis, which could have introduced certain variability in the assessment of both the Alberta Stroke Program Early Computed Tomography Score and final expanded Thrombolysis in Cerebral Infarction scores. Given that the number of participating centers is limited to 2, a potential concern is the generalizability of the findings, especially considering the larger number of centers involved in the DMVO RCTs. Finally, the inclusion of patients with DMVO excluded from trials and for endovascular therapy would increase the rigor of the article.

Conclusions

In our study, conducted at 2 centers participating in DMVO clinical trials during the inclusion period, randomized patients presented milder clinical severity and smaller hypoperfusion areas compared with those treated in routine clinical practice. These findings suggest a potential selection bias that may lead to an underestimation of the true benefit of EVT and underscore the importance of consecutive patient recruitment.

ARTICLE INFORMATION

Sources of Funding

None.

Disclosures

Dr Requena reports compensation from iVascular for consultant services. Dr Diana reports compensation from Medtronic and Balt for consultant services. Dr Ribo reports compensation from Rapid Pulse for consultant services; stock holdings in Nora; compensation from Sensome for data and safety monitoring services; compensation from Philips for consultant services; stock holdings in Anaconda Biomed; compensation from Vesalio for consultant services; compensation from Cerenovus for consultant services; compensation from Medtronic MiniMed Inc for consultant services; compensation from AptaTargets for consultant services; compensation from Stryker Corporation for consultant services; and stock holdings inMethinks. Dr Ribo serves on the editorial board of Stroke: Vascular and Interventional Neurology. Editorial board members are not involved in the handling or final disposition of submissions. The other authors report no conflicts.

Supplemental Material

Tables S1–S2

STROBE Checklist