Review of Current Large Core Volume Stroke Thrombectomy Clinical Trials: Controversies and Progress

Abstract

Large clinical trials have helped establish the benefit of endovascular treatment (EVT) in patients with acute ischemic stroke with large vessel occlusion and small infarct core volume as determined by scores ≥6 on the Alberta Stroke Program Early CT Scores. Several small studies have suggested that patients with large infarct core volume (LICV) may also benefit from EVT. Currently, at least 6 randomized clinical trials are examining the benefit of extending EVT to this population of patients with acute ischemic stroke. These trials were independently conceived and have significant differences in their inclusion criteria. Understanding these inclusion criteria and other differences in trial design is pivotal for the field to interpret the upcoming results of these trials. In this review, the designs of the 6 trials are summarized and compared. Specific differences are described, including (1) the rationale for EVT treatment in patients with LICV, (2) how to define LICV and the imaging modality used to identify LICV, (3) inclusion of an Alberta Stroke Program Early CT Score 0 to 5 versus 3 to 5, (4) use of the mismatch between blood flow and the size of infarct as an inclusion criterion, and (5) inclusion of early window and/or late window patients. The potential impact of these trial results on current guidelines for acute ischemic stroke is discussed. Differences in trial design as well as inclusion and exclusion criteria may influence trial outcomes. The implications of these trial results will likely be enhanced by a pooled analysis.

Nonstandard Abbreviations and Acronyms

AIS

acute ischemic stroke

ASPECTS

Alberta Stroke Program Early CT Score

DWI

diffusion‐weighted imaging

EVT

endovascular treatment

LICV

large infarct core volume

LVO

large vessel occlusion

MM

medical management

mRS

modified Rankin Scale

MT

mechanical thrombectomy

sICH

symptomatic intrac hemorrhage

Large clinical trials on patients with early and late window stroke have helped to establish the indications for endovascular treatment (EVT) in patients with acute ischemic stroke (AIS) with large vessel occlusion (LVO). This includes patients with an Alberta Stroke Program Early CT Score (ASPECTS) ≥6 within 6 hours and patients meeting DAWN (DWI or CTP Assessment With Clinical Mismatch in the Triage of Wake‐Up and Late Presenting Strokes Undergoing Neurointervention With Trevo) and DEFUSE 3 (The Endovascular Therapy Following Imaging Evaluation for Ischemic Stroke 3) criteria in 6 to 16 or 24 hours. ¹ , ²

Since then, many clinical trials have been focused on expanding the indications for patients with AIS with LVO. Currently, at least 6 clinical trials are examining the benefit of extending EVT to patients with AIS and LVO with large infarct core volume (LICV). These 6 trials are summarized in the Table 1. Mainly, these LICV trials enroll patients with ASPECTS <6 along with otherwise varying inclusion criteria. Although the differences among all 6 trials are not significant, debates on these differences exist. ³ , ⁴ , ⁵ The merits of different inclusion and exclusion criteria were discussed among the lead investigators in the Large Core Thrombectomy forum at a recent Chinese Stroke Association annual meeting. A discussion on various aspects of LICV trial design and implications is presented here to update the scientific community. We invited the principal investigators of some of these large core thrombectomy randomized clinical trials (RCTs) to discuss and complete this review.

Table 1.

Summary of the Current 6 “Large Core” Trials

Trial	TENSION	LASTE	TESLA	RESCUE‐Japan LIMIT	SELECT‐2	ANGEL‐ASPECT
NCT number/links	NCT03094715	NCT03811769	NCT03805308	NCT03702413	NCT03876457	NCT04551664
Participating Country(ies)	Europe, Canada	Europe, United States	United States	Japan	United States, Canada, Europe	China
Major inclusion criteria
Major imaging inclusion criteria	NCCT or DWI ASPECTS 3–5	NCCT or DWI ASPECTS 0–5 (−5 for > 80 y)	NCCT or ASPECTS 2–5	CT ASPECTS 3–5 or DWI ASPECTS 3–5	1. ASPECTS ≥6 and CTP core ≥50 cc 2. ASPECTS 3–5 and core ≥50 cc 3. ASPECTS 3–5 and core <50 cc	1. ASPECTS 3–5 2. ASPECTS >5 (>6 h) and core 70–100 cc 3. ASPECTS <3 and core 70–100 cc
Mismatch (clinical or radiographic)	No	Yes (clinical mismatch)	No	No	No	No
NIHSS score	<26	>5	>6	≥6	≥6	6–30
Age, y	>18	≥18	18–85	>18	18–85	18–80
Time window	<12 h LSW	<6.5 h LKW	Random <24 h	Random <6 h LKW, 6–24 FLAIR (–)	Treat <24 h (0–12 vs 6–24)	Random <24 h
Occlusion site	Terminal ICA and MCA M1	Intracranial ICA, MCA M1 or M1–M2	Terminal ICA and M1	Terminal ICA and M1	ICA or MCA M1, tandem	Terminal ICA and/or MCA M1 Tandem occlusion included
Required time limit from randomization to puncture	No	Yes Randomization to arterial access 30 min	No	Yes Randomization to arterial access 60 min	No	Yes Randomization to arterial access 60 min
Bridging therapy permitted	Yes	Yes	Yes	Yes	Yes	Yes
Intervention model	EVT, no specified device	EVT, no specified device	FDA‐approved EVT devices	EVT, no specified device	EVT with SR Device: Trevo, Solitaire, and EmboTrap	EVT, IA‐thrombolysis, angioplasty Device: Solitaire, EMBOTRAP, Reco SR, Penumbra aspiration catheter, or CFDA‐approved EVT devices
Major exclusion criteria	Mass effect on CT Vascular disease prevent MT (eg, aortic dissection or aneurysm, no arterial transfemoral access)	Suspicion of aortic dissection, excessive tortuosity of cervical vessels on vascular imaging Multiple occlusion Cervical tandem lesion that requires stent placement	Refractory hypertension Mass effect on CT Tandem lesion Difficult endovascular access on vascular images	Mass effect on CT Clinical evidence of chronic occlusion High risk of hemorrhage (platelet <40,000 μL−1, aPTT >50 s or PT‐INR >3.0)	Mass effect on CT Inability to undergo CTA and/or CTP tPA 3–4.5 h With special situation* ICA dissection or aortic dissection Multiple occlusions	Refractory hypertension Mass effect on CT Multiple occlusions INR >1.7 or aPTT >35 s; platelet count <100×109 L−1
Primary outcome	mRS score shift analysis	mRS score at 90 and 180 d	Utility‐weighted 90‐d mRS score	mRS score 0–3 at 90 d	Shift on 90‐d mRS score	mRS score at 90 d
Actual study start date	July 20, 2018	April 7, 2019	July 16, 2019	November 2018	October 11, 2019	September 28, 2020

Source: https://clinicaltrials.gov. ANGEL‐ASPECT, Study of EVT in Acute Anterior Circulation LVO Patients with a largE infarCT core; Core: rCBF <30% on CT perfusion or ADC <620. ADC indicates apparent diffusion coefficient; aPTT, activated partial thromboplastin time; ASPECTS, Alberta Stroke Program Early CT Score; CFDA, China Food and Drug Administration; CT, computed tomography; CTA, computed tomography angiography; CTP, computed tomography perfusion; DWI, diffusion‐weighted imaging; EVT, endovascular treatment; FDA, US Food and Drug Administration; FLAIR, fluid‐attenuated inversion recovery; IA, intra‐arterial; ICA, internal carotid artery; INR, international normalized ratio;LASTE, Large Stroke Therapy Evalusation; LKW, last known well; LSW, last seen well; LVO, large vessel occlusion; MCA, middle cerebral artery; mRS, modified Rankin Scale; MT, mechanical thrombectomy; NCCT, noncontrast CT; NCT, National Clinical Trial; NIHSS, National Institutes of Health Stroke Scale;PT‐INR, Pro‐thrombin Time‐International Normalized Ratio; RESCU‐Japan LIMIT, Randomized Controlled Trial of Endovascular Therapy for Acute Large Vessel Occlusion With Large Ischemic Core; rCBF, relative cerebral blood flow;SELECT‐2, Thrombectomy for Emergent Salvage of Large Anterior Circulation Ischemic Stroke; SR, stent retriever; TESLA, Thrombectomy for Emergent Salvage of Large Anterior Circulation Ischemic Stroke and tPA, tissue plasminogen activator.

^* (1) Age >80 y, (2) current anticoagulant use, (3) history of diabetes and prior stroke, (4) NIHSS >25, and (5) ischemic involvement of more than one‐third of MCA territory.

Rationale for EVT Treatment for Patients with LICV

Several retrospective studies, prospective studies, and meta‐analyses suggest that patients with LICV may benefit from EVT. The THRACE (Thrombectomie des Artères Cerebrales) trial ⁶ was the only major early randomized trial to enroll patients with ASPECTS <6. A subgroup analysis of the THRACE trial on patients with large baseline diffusion‐weighted imaging (DWI) stroke lesions showed that among 53 participants with a DWI volume of >70 mL, 12 (22.6%) patients in the EVT group had good clinical outcomes (modified Rankin Scale [mRS] scores ≤2 at 3 months). ⁷ The prospective German Stroke Registry–Endovascular Treatment also showed that 22% of 152 patients with thrombectomy with ASPECTS <6 achieved independence with mRS scores 0 to 2 at 90 days. ⁸ The HERMES (Highly Effective Reperfusion Evaluated in Multiple Endovascular Stroke Trials) collaboration pooled the data from the 6 trials. It showed a benefit of EVT over control was observed in 126 patients with ASPECTS 0 to 4 ⁹ and a computed tomography perfusion (CTP) or DWI magnetic resonance imaging (MRI) core volume of ≥70 mL. ¹⁰ Functional improvements, measured with an mRS score of 0 to 2 at 90 days, were achieved in 25% of patients with ASPECTS 0 to 4 with EVT versus 14% in controls. ⁹

In the SELECT (Optimizing Patient's Selection for Endovascular Treatment in Acute Ischemic Stroke) trial, ¹¹ the prespecified secondary analysis of 105 patients (of whom 62 received EVT) with ASPECTS ≤5 or CTP‐determined ischemic core volume ≥50 mL showed that functional independence was achieved in 31% in the EVT group versus 14% in the control group. Incidence rates of death, neurologic decline, and symptomatic intracranial hemorrhage (sICH) were similar in both groups. EVT was also associated with less infarct growth (44 versus 98 mL; P=0.006) and smaller final infarct volume (97 versus 190 mL; P=0.001) than medical management (MM).

A recent meta‐analysis ¹² including 17 studies and 1378 patients with ASPECTS 0 to 6 (1194 EVT, 184 mechanical thrombectomy [MT]) found that an mRS score 0 to 2 was achieved in 30.1% of cases after EVT and in 3.2% after MM (odds ratio [OR], 4.76; P=0.01). The marked lower rate in the MM group compared with previous RCTs (HERMES, 14%; SELECT,14%) is likely attributed to the imbalance of the patient population in these retrospective studies: more advanced age in the MT group (75 years [range, 50–94 years] versus 68.7 years [range, 43–91 years]), higher National Institutes of Health Stroke Scale scores (19 [range, 5–40] versus 18 [range, 9–28]), lower intravenous thrombolysis (47.8% versus 56.8%), and longer symptom onset to admission time (130 minutes [range, 37–642 minutes] versus 115 minutes [range, 10–380 minutes]). Successful recanalization (Thrombolysis in Cerebral Infarction grade 2b–3) gave higher odds of mRS scores 0 to 2 than unsuccessful reperfusion (OR, 5.2; P=0.001). Another pooled random‐effect meta‐analysis ¹³ including 12 studies of patients with large core volumes by either definition (ASPECTS <6 or ischemic core volume ≥50 mL) or both demonstrated increased functional independence (mRS score 0–2) rates with EVT (25% versus 7%; pooled OR, 4.39 [95% CI, 2.53–7.64]) and decreased mortality rates (23% versus 33%; pooled OR, 0.53 [95% CI, 0.40–0.71]).

In a matched case‐control study ¹⁴ of 56 patients (28 pairs) with internal carotid artery, M1, and M2 occlusion and CTP‐determined infarct core >50 mL, EVT led to higher rates of functional independence (90‐day mRS score 0–2, 25% versus 0%; P=0.04) and smaller final infarct volumes (87 versus 242 mL; P<0.001). One control (4%) and 2 treatment patients (7%) developed a parenchymal hematoma type 2 (P>0.99). The rates of hemicraniectomy (7% versus 21%; P=0.10) and 90‐day mortality (29% versus 48%; P=0.75) were lower in the EVT arm. Sensitivity analysis for patients with a baseline ischemic core volume >70 mL (12 pairs) revealed a significant reduction in final infarct volumes (110 versus 319 mL; P<0.001) but only a nonsignificant improvement in the overall distribution of mRS scores favoring the treatment group (P=0.18).

Interestingly, 1 observational cohort study ¹⁵ included a consecutive sample of 170 patients with anterior circulation stroke and initial ASPECTS ≤5 (99 EVT, 71 MM). Clinical outcome after failed or incomplete EVT (Thrombolysis in Cerebral Infarction grade 0–2b) was significantly better compared with patients with MM only (median mRS score 5, interquartile range 4–6 versus 5–6; P=0.03). Failed EVT (Thrombolysis in Cerebral Infarction grade 0–2a) was not associated with a worse outcome than MM.

Given the poor prognosis of patients with LICV, the difference in clinical outcome between the EVT and control groups may not reach statistical significance. Other surrogates of successful recanalization such as the functional improvement outcome (mRS score 0–3), less infarct edema, ¹⁶ , ¹⁷ and less need for decompressive craniectomy ¹⁴ , ¹⁸ , ¹⁹ , ²⁰ may need to be used to evaluate the effect of EVT.

Although multiple studies and meta‐analyses suggest a potential benefit of EVT in LICV strokes, the small sample size and heterogeneity of response preclude firm conclusions. Large, multicenter RCTs are needed to clarify the benefit of EVT and direct treatment guidelines.

Define LICV and Image Modality to Identify Large Infarct Core

Clearly defined inclusion criteria are critical for any clinical trial of EVT in patients with AIS and LVO. Currently, some debate exists surrounding the definition of large infarct core and other inclusion criteria. ³ , ⁴ , ⁵

ASPECTS is a widely accepted tool used to assess infarct volume. American Heart Association/American Stroke Association guidelines define ASPECTS ≥6 as a class I indication for EVT treatment for the patient within 6 hours of stroke onset. In general, ASPECTS <6 is regarded as a “large core infarct.” However, multiple studies have shown low interrater agreement with ASPECTS. ²¹ , ²² An inaccurate ASPECTS can misassign patients between the EVT and control groups, weakening any trial conclusions.

Many studies use CTP‐determined ischemic core volume to define “large core.” CTP‐determined volumes of ischemic stroke and the surrounding tissue at risk (penumbra) correlate well with acute DWI lesion volume. ²³ , ²⁴ , ²⁵ CTP is more readily available than MRI, and results are available more quickly. The correlation between the CTP‐determined core volume and ASPECTS is not well established. One study showed the optimum ASPECTS cutoff to detect a DWI lesion ≥70 mL was <7 (sensitivity, 0.74; specificity, 0.86; Youden J=0.60), and the optimum CTP‐determined core volume cutoff was ≥50 mL (sensitivity, 0.86; specificity, 0.97; Youden J=0.84). ²⁵

There is some debate about whether “large core” should be defined as 50 mL versus 70 mL on CTP. ³ , ⁴ , ⁵ Most studies and previous trials have defined a large core as a CTP or DWI MRI volume of 70 mL. CTP tends to overestimate infarct core in the early time window and in patients with very low ASPECTS. ²⁶ , ²⁷ , ²⁸ , ²⁹ , ³⁰ Using 50 mL as the cutoff for “large core” may exaggerate this error. This error could be mitigated by decreasing the threshold of infarct core of relative cerebral blood flow on CTP from <30% to <20% in the early stage. On the other hand, core lesions between 50 and 70 mL are not considered by all to be “large core.” Only the SELECT‐2 trial included patients with 50 to 70 mL cores of the 6 ongoing trials.

Whether ASPECTS or CTP/DWI MRI or both should be used for defining large core is not clear. The subgroup analysis in the meta‐analysis comparing outcomes between these 2 definitions did not find significant heterogeneity in the results when LICV was defined based on ASPECTS or ischemic core volume of CTP. ¹³ Although 1 study ²⁵ found a good correlation between ASPECTS and CTP/DWI volume, others found them to be discordant. ¹¹ , ¹³ , ³¹ The correlation between ASPECTS and CTP core is moderate and somewhat stronger in patients with later time windows after stroke onset. ²¹ The trials, including the computed tomography (CT) ASPECTS score and CTP core volume, will be able to evaluate the heterogeneity of treatment effect after thrombectomy in patients with discordant imaging profiles and identify the optimal imaging modality to select patients for EVT.

DWI hyperintensity in an acute stroke does not necessarily represent irreversibly infarcted tissue. Several studies have demonstrated that substantial proportions of DWI lesions are salvageable. Yoo et al evaluated the factors associated with DWI reversal after EVT. ³² In 404 patients who met the study criteria, DWI reversal was found in 63 patients (15.5%) after EVT. Complete reperfusion and shorter imaging time to recanalization were independently associated with DWI reversal among patients with AIS who received EVT. DWI reversal group had a baseline volume of 30.1±36.7 mL, and the change in DWI volume was −12.3±18.7 mL. A systemic review of 18 studies ³³ showed significant variability in the reversal rates of DWI lesions (0%–83%), with a surprisingly high mean rate (24% of pooled patients). Some studies ³⁴ , ³⁵ found that reversibility correlated with the spatial relationship between DWI and perfusion‐weighted imaging lesions. The regions most likely to show reversal did not have superimposed perfusion‐weighted imaging deficits and had only modest reductions in apparent diffusion coefficient values. ³⁶ The good response to recanalization in patients with LICV may be partly explained by the reversibility of larger DWI lesions. Therefore, even patients with very low ASPECTS (0–2) may have a chance of recovery. DWI is more sensitive than CT for the detection of AIS. CT had the tendency to underestimate extension of early ischemic change than DWI. So, when compared with the DWI and noncontrast CT ASPECTS score for patients with hyperacute stroke, DW ASPECTS scored approximately 1 point lower than CT ASPECTS. ³⁷

Studies also have shown that clinical recovery may be affected by recanalization time. One study of 132 patients has shown that CTP thresholds associated with final infarct volume primarily depend on the CTP‐to‐reperfusion time rather than the time from stroke onset to CTP. ³⁸ Another study of 132 patients from the HERMES collaboration showed a relationship between optimal perfusion thresholds and imaging‐to‐reperfusion time, ³⁹ suggesting the importance of stroke onset‐to‐reperfusion time. These studies highlight the need to account for and possibly standardize imaging to reperfusion time in the large core trials.

Aspects 0 to 5 Versus 3 to 5

Previous studies ⁴⁰ , ⁴¹ , ⁴² identified ASPECTS thresholds of <2 or 3, below which treatment appears futile. A meta‐analysis ⁹ showed that ASPECT 0 to 2 favored MM instead of EVT. Other studies showed a lack of benefit if CTP‐determined core volume exceeded 100 and 150 mL. ¹⁰ , ¹¹ A recent meta‐analysis of 17 studies and 1378 patients reported that an mRS score 0 to 2 was achieved by 37.7%, 33.3%, 22.1%, and 17.1% of patients with ASPECTS 6, 5, 4, and 0 to 4, respectively. ¹² The studies by Mourand et al ²⁰ and Inoue et al ⁴³ showed favorable outcomes between 16% and 20% of patients with ASPECTS 0 to 3 after EVT.

The upper limit of large core volume at which EVT is unlikely to provide any benefit remains uncertain in patients with ASPECTS 0 to 2. Most reports are from nonrandomized, single‐arm, retrospective studies, and the results of MM at different ASPECTS are not available for comparison. Choosing a patient population with ASPECTS 3 to 5 for clinical trials may yield more positive results.

In a study of 232 patients with middle cerebral artery infarcts, patients with DWI ASPECTS 0 to 2 account for 43.4% (62/143) of all patients with ASPECTS 0 to 5 and 26.7% (62/232) of all patients with ASPECTS. ⁴⁴ In another study of 337 patients with stroke onset time >6 hours, CT ASPECTS 0 to 2 comprised 30.2% of all patients with ASPECTS <6 and 11.6% of all patients with ASPECTS. ³⁰ These studies demonstrated that there is a significant number of patients with ASPECTS 0 to 2 in both the early and late windows. CTP or DWI may help identify patients with a reasonable “large core” volume who may benefit from EVT. Among all 6 trials, ASPECTS used for inclusion were 0 to 5 in LASTE, 2 to 5 in TESLA, and 3 to 5 in the other 4 trials. SELECT‐2 and ANGEL‐ASPECT used CTP in addition to ASPECTS to identify patients with ASPECTS 0 to 2 with a reasonable volume of a large core infarct.

Is Penumbral Imaging Necessary for Patient Selection?

The penumbra around a cerebral infarct represents potentially salvageable tissue. Multiple studies ¹⁰ , ⁴⁵ , ⁴⁶ have demonstrated that the favorable clinical response of EVT correlates with the mismatch (penumbra) volume. The larger the penumbra, the better the results. One study showed that 80% of patients with AIS have penumbra regardless of the volume of infarct core, ⁴⁵ but the prevalence of penumbra in the patients with low ASPECTS is low. Per 1 study, the prevalence of mismatch declines with decreasing ASPECTS (6.4%/point). ⁴⁷ The clinical core mismatch is present in 77%, 65%, and 13% in the ASPECTS groups of 9 to 10, 6 to 8, and 0 to 5, respectively.

In the HERMES study, a penumbral imaging meta‐analysis showed that in patients who underwent CTP in RCTs testing MT, the vast majority of them had a mismatch on baseline imaging as a result of a probable selection. ¹⁰ The study by Seners et al also found that about half of the patients had a mismatch for large core LVO within 6 hours from last seen well and only patients with a mismatch appeared to benefit from MT. In addition, MT+Best medical management (BMM) was associated with increased rates of parenchymal hemorrhage over BMM alone regardless of the baseline perfusion imaging profile, calling into question the rationale for MT in patients with large‐core LVO without mismatch. ⁴⁸

Study enrollment based on a large CTP alone could randomly assign patients with ASPECTS 6 to 10 into the non‐EVT cohort. Thus, potentially withholding the guideline‐based standard therapy (American Heart Association/American Stroke Association, European Stroke Organisation/European Society for Minimally Invasive Neurological Therapy guidelines). ¹ , ² On the other hand, difficulty in defining ASPECTS in RCTs with Non‐contrast CT only could lead to withholding standard therapy in patients with erroneously low ASPECTS readings who would have been correctly classified using a more standardized CTP diagnosis.

CTP seems useful in predicting functional outcome but cannot reliably identify patients who will not benefit from intraarterial therapy. ⁴⁹ Given the low prevalence of mismatch in the group with ASPECT 0 to 5, incorporation of CTP into the inclusion criteria may not have many benefits however it may risk delaying revascularization. The addition of CTP into the inclusion criteria mainly serves to potentially offset the inter‐rater variability of using CT ASPECTS alone. Clinical core mismatch may sufficiently replace penumbral imaging however of the 6 trials, only LASTE included clinical core mismatch in the inclusion criteria.

The collateral status is also a predictor of a good outcome. One study showed that reperfusion in patients with ASPECTS ≤5 led to better clinical outcomes among patients with good collaterals compared with those with no collaterals. ⁵⁰ In patients with ASPECTS 0 to 5, 1 study ⁵⁰ showed good collaterals in 27%, whereas another ⁵¹ in 44.8%. The final clinical outcome is more favorable in patients with better collaterals. Interestingly, collaterals were associated with lower core volumes but not higher penumbra volumes. ⁵² This suggests a major role of collaterals in preventing early tissue loss and suggests limited significance as markers of salvageable tissue. None of the 6 trials include collateral status as part of the inclusion criteria. Secondary subgroup analysis may be applied to determine the role of the collateral status in selecting patients for EVT in the population with ASPECTS 0 to 5.

Should Late Time Window Patients be Enrolled?

Subgroup analyses in 1 meta‐analysis ¹³ found that patients with LICV did not show a significant difference in outcomes among studies reporting <6, <12, and <24 hours for stroke‐to‐EVT time windows. This may be because most patients with LICV presented in the early time window, reducing the power to detect the difference between early and late windows. The SELECT trial, ¹¹ with 105 patients with large core infarct, did show a progressive decline in functional outcomes for each hour of treatment delay, with a low likelihood of benefit from EVT after 12 hours. This differed from the findings in patients with small core infarct, in whom good outcomes have been documented up to 24 hours after onset. A meta‐analysis ¹² by Cagnazzo et al that included 17 studies demonstrated that a shorter time from onset to reperfusion was associated with a higher probability of functional independence after EVT in patients with ASPECTS 0 to 6. The reduced benfit of EVT over time is likely related to a progressive decreasing mismatch (1.6% per hour) and decreasing ASPECTS (6.4% per point). ⁴⁷ Patients in the hyperacute phase of stroke showed increased ASPECTS lesion growth from imaging to recanalization, suggesting a benefit of faster recanalization in these patients. ⁵³

Among the 6 trials, TENSION, LASTE, and RESCUE‐Japan LIMIT used early window (12 hours) inclusion criteria (with up to 24 hours in RESCUE‐Japan LIMIT if MRI fluid‐attenuated inversion recovery negative). TESLA, SELECT‐2, and ANGEL‐ASPECT enrolled patients up to 24 hours. The inclusion of later time windows may reduce the aggregate difference between EVT and MM outcomes but may provide valuable data for the subgroup analysis of the late time window population.Pooled individual patient data from these trials could increase the power of subgroup analysis in patients with late‐window LICV.

What to Expect and How to Interpret Differences in Trial Results

There are 2 fundamental questions to be answered by these “large core” trials, with a slight difference in their future application: First, will EVT benefit a patient with “large core” infarct and more precisely, which subgroup of patients with “large core” infarct? Second, will EVT result in significant sICH with a large core infarct and therefore is it harmful? The sICH and procedure‐associated complications will likely affect the trial results significantly.

There are conflicting data on the incidence of sICH with EVT. In the HERMES study, ⁹ the rate of sICH was 4 times higher with EVT than with controls (19% versus 5%) in patients with ASPECTS 0 to 4, however, other studies did not confirm a significant difference in sICH between patients with EVT and patients with MM. One meta‐analysis ¹² showed lower and another ¹³ showed higher rates of sICH with EVT. Both did not reach statistical significance. Interestingly, one study ¹⁵ has shown even failed EVT (Thrombolysis in Cerebral Infarction grade 0–2a) was not associated with a worse outcome than MM. Common EVT‐associated complications include vasospasm (3%–10.3%), emboli (3.4%–5.7%), vessel injury (2%–6.7%), subarachnoid hemorrhage (2.8%–2.9%), intracerebral hemorrhage (3.5%–5.9%), and access complications (0.4%–1.9%). ⁵⁴ , ⁵⁵ , ⁵⁶ , ⁵⁷ , ⁵⁸ Many of these complications, such as vasospasm, emboli to the distal territory, and focal dissection/occlusion at intervened vessels, are much less detrimental given that the patient already has extended nonsalvageable infarct.

There are several scenarios for the possible result of the trials. First, the trial result is “positive”; there is a statistically significant benefit of EVT on patients with large core infarct with no increased risk of sICH. This would be the ideal result. In this case, the indications for the EVT will be much expanded. Given that this result indicates that almost all patients with stroke, regardless of whether the infarct core is “large” or “small,” will likely benefit from EVT, and the advanced image modality to select the patient for EVT may become unnecessary. This will provide evidence and support for the direct thrombectomy approach/strategy to shorten the time from onset to recanalization. This would potentially improve the clinical outcome of the stroke population with LVO by bypassing the time‐consuming process of image modality selection. Second, the trial result is “negative” (no benefit of EVT with increased risk of sICH). This would be the worst result. The current recommendations would remain unchanged. Third, the trial result is “negative,” with no benefit of EVT, but no increased risk of sICH either. This result is indeed the “bad” news for the individual patient with large infarct core, but not necessarily bad for the stroke population. More recent studies ⁵⁹ , ⁶⁰ showed that more sophisticated imaging workups for either early or late windows are not correlated with better clinical outcomes than without these imaging selections. Currently, the sophisticated pre‐EVT imaging workup mainly excludes this group of patients with large core infarct. If the trial result showed no harm for this group of patients, we could skip the imaging selection modality for all clinically eligible patients for direct angiosuite intervention. Although there is no benefit on the individual patient level, the large potential benefit will be on the patient population level.

We acknowledge that the direct thrombectomy approach/strategy remains contraversial. People may highlight the ethical challenge of exposing a subset of patients to an unnecessary endovascular procedure carrying the risk of sICH and other complications. One recourse after negative LIV trial results would be to conduct additional clinical trials with further refined imaging selection criteria to find a subgroup of patients with LIV who are more likely to benefit from MT or the factors more likely causing increased sICH.

Each of the LIV RCTs has its own unique selection criteria as discussed in the above sections and summarized in Table 1. Variability in clinical trial criteria include imaging modality (CT versus MRI ASPECTS), time window (early versus late), and use of CTP versus no CTP. The result of each individual trial consequently may only apply to that specific trial patient population. For example, the first published low ASPECTS thrombectomy trial, RESCUE‐Japan LIMIT, ⁶¹ showed that patients with large cerebral infarctions had better functional outcomes with endovascular therapy than with medical care alone, but had more intracranial hemorrhages. The trial population was enriched for mismatch based on MRI DWI/fluid‐attentuated inversion recovery signal (no early fluid‐attentuated recovery signal change) ratio. Moreover, the primary outcome was mRS score 0 to 3 as opposed to the more commonly used mRS score 0 to 2. Moreover, >25% of the patients with LVIC received intravenous thrombolysis, which is known to be associated with increased intracerebral hemorrhage. ⁶² Hence, the “positive” result seen in the RESCUE‐Japan LIMIT trial may not be applicable across all populations of patients with LIV and should not lead to the early termination of other ongoing trials with different selection criteria. The results from these other trials will provide more solid evidence to define the specific patient populations with a large core who will benefit from MT.

Conclusions

Differences in design and inclusion/exclusion criteria may affect the result of EVT in LICV trials. These differences may affect or complement the interpretation of trial results. The power of these trials may be enhanced by a pooled analysis of multiple trials.

Sources of Funding

None.