← Back
NeuroTrials.ai
Neurology Clinical Trial Database

MR CLEAN-LATE

Endovascular treatment versus no endovascular treatment after 6-24 h in patients with ischaemic stroke and collateral flow on CT angiography (MR CLEAN-LATE) in the Netherlands: a multicentre, open-label, blinded-endpoint, randomised, controlled, phase 3 trial

Year of Publication: 2023

Authors: Olthuis SGH, Pirson FAV, Pinckaers FME, ..., et al.; on behalf of the MR CLEAN-LATE investigators

Journal: The Lancet

Citation: Lancet 2023; 401: 1371-80

Link: https://doi.org/10.1016/S0140-6736(23)00575-5


Clinical Question

Can collateral flow on CT angiography alone be used to select patients with anterior circulation large-vessel occlusion for endovascular treatment in the 6-24 hour late window?

Bottom Line

In patients with anterior circulation ischemic stroke presenting 6-24 h from symptom onset or last seen well who were not eligible for treatment under DAWN/DEFUSE-3 criteria, endovascular treatment selected by the presence of collateral flow on CTA improved functional outcomes at 90 days, supporting a more pragmatic and inclusive selection strategy that does not require perfusion imaging.

Major Points

  • Endovascular treatment produced a favorable shift in 90-day mRS (adjusted common OR 1.67, 95% CI 1.20-2.32)
  • Median mRS was 3 (IQR 2-5) with endovascular treatment vs 4 (IQR 2-6) with best medical treatment alone
  • Symptomatic intracranial hemorrhage was significantly more common with endovascular treatment (7% vs 2%; adjusted OR 4.59, 95% CI 1.49-14.10)
  • All-cause 90-day mortality was numerically lower but not statistically different (24% vs 30%; adjusted OR 0.72, 95% CI 0.44-1.18)
  • Selection using CTA collateral flow alone is a viable, pragmatic alternative to perfusion imaging for late-window thrombectomy eligibility
  • Findings support extending late-window thrombectomy indications beyond current DAWN/DEFUSE-3 criteria, particularly where perfusion imaging is limited

Design

Study Type: Multicentre, open-label, blinded-endpoint, randomised, controlled, phase 3 trial (PROBE design)

Randomization: 1

Blinding: Open-label with blinded endpoint assessment (PROBE design)

Allocation: 1:1 web-based randomisation, block sizes 8-20, stratified by centre

Enrollment Period: Feb 2, 2018 to Jan 27, 2022

Follow-up Duration: 90 days

Centers: 18

Countries: Netherlands

Sample Size: 535

Analyzed: 502

Analysis: Modified intention-to-treat: all randomly assigned patients who provided deferred consent or died before consent could be obtained. Adjusted for predefined confounders. Treatment effect estimated with ordinal logistic regression, reported as adjusted common odds ratio with 95% CI.

Registration: ISRCTN19922220


Inclusion Criteria

  • Age ≥18 years
  • Ischemic stroke
  • Presentation in the late window (6-24 h from symptom onset or last seen well)
  • Anterior circulation large-vessel occlusion
  • Collateral flow on CTA (grades 1, 2, or 3)
  • NIHSS score ≥2
  • Not eligible for late-window endovascular treatment under Dutch national guidelines (based on DAWN/DEFUSE-3 clinical and perfusion imaging criteria)

Exclusion Criteria

  • Intracranial hemorrhage on baseline imaging
  • Pre-stroke dependency (mRS ≥3)
  • Ischemic stroke within the previous 6 weeks with persistent neurological symptoms
  • Clinical evidence of hemorrhagic diathesis
  • Clearly demarcated hypodensity of more than one-third of the middle cerebral artery territory consistent with current symptoms
  • Participation in other medical or surgical intervention trials (with exceptions listed in protocol)
  • Eligibility for late-window endovascular treatment under national guidelines (treated per those guidelines and excluded from enrollment)

Baseline Characteristics

0:

  • Characteristic: Age, years (median, IQR)
  • Intervention: 74 (64-80)
  • Control: 74 (64-81)

1:

  • Characteristic: Female
  • Intervention: 148 (58%)
  • Control: 113 (46%)

2:

  • Characteristic: Male
  • Intervention: 107 (42%)
  • Control: 134 (54%)

3:

  • Characteristic: NIHSS score (median, IQR)
  • Intervention: 10 (6-17)
  • Control: 10 (6-18)

4:

  • Characteristic: Prior ischaemic stroke
  • Intervention: 51/254 (20%)
  • Control: 40/246 (16%)

5:

  • Characteristic: Atrial fibrillation
  • Intervention: 51/254 (20%)
  • Control: 53/246 (22%)

6:

  • Characteristic: Diabetes
  • Intervention: 35/254 (14%)
  • Control: 39/246 (16%)

7:

  • Characteristic: Hypertension
  • Intervention: 142/254 (56%)
  • Control: 118/245 (48%)

8:

  • Characteristic: ASPECTS (median, IQR)
  • Intervention: 9 (7-10)
  • Control: 8 (7-9)

9:

  • Characteristic: Tandem lesion
  • Intervention: 51/248 (21%)
  • Control: 57/236 (24%)

10:

  • Characteristic: Collateral grade 0 (absent)
  • Intervention: 11/251 (4%)
  • Control: 15/246 (6%)

11:

  • Characteristic: Collateral grade 1 (≤50%, >0%)
  • Intervention: 67/251 (27%)
  • Control: 73/246 (30%)

12:

  • Characteristic: Collateral grade 2 (>50%, <100%)
  • Intervention: 109/251 (43%)
  • Control: 96/246 (39%)

13:

  • Characteristic: Collateral grade 3 (100%)
  • Intervention: 64/251 (25%)
  • Control: 62/246 (25%)

14:

  • Characteristic: Time from onset/last seen well to randomisation, min (median, IQR)
  • Intervention: 709 (542-910)
  • Control: 682 (532-887)

15:

  • Characteristic: IV thrombolytics
  • Intervention: 12 (5%)
  • Control: 19 (8%)

16:

  • Characteristic: Witnessed stroke onset
  • Intervention: 35 (14%)
  • Control: 21 (9%)

Arms

FieldEndovascular treatment + best medical treatmentControl
N255247
InterventionEndovascular thrombectomy plus best medical treatmentBest medical treatment alone (no endovascular treatment)
DurationSingle procedure with 90-day follow-up90-day follow-up

Outcomes

OutcomeTypeControlInterventionHR / OR / RRP-value
Modified Rankin Scale (mRS) score distribution (ordinal shift analysis)PrimaryMedian mRS 4 (IQR 2-6)Median mRS 3 (IQR 2-5)1.67
mRS 0-1 vs 2-6 at 90 daysSecondary39 (16%)54 (21%)Adjusted OR 1.63
mRS 0-2 vs 3-6 at 90 daysSecondary84 (34%)100 (39%)Adjusted OR 1.54
mRS 0-3 vs 4-6 at 90 daysSecondary103 (42%)129 (51%)Adjusted OR 1.73
NIHSS score at 24 h (median, IQR)Secondary10 (4 to 18)7 (3 to 16)β-coefficient -0.18; difference -16%
NIHSS score at 5-7 days or discharge (median, IQR)Secondary7 (3 to 14)3 (1 to 10)β-coefficient -0.31; difference -27%
EQ-5D-5L utility score at 90 days (median, IQR)Secondary0.28 (0.00 to 0.83)0.53 (0.00 to 0.88)β-coefficient 0.08
Barthel Index at 90 days (median, IQR)Secondary95 (55 to 100)100 (81 to 100)β-coefficient 8.2
Recanalisation (mAOL ≥2) at 24 h on CTA/MRASecondary84/160 (53%)154/190 (81%)Adjusted OR 3.14
Follow-up infarct volume, mL (median, IQR)Secondary43 (13 to 121)28 (6 to 87)β-coefficient -0.45; difference -36%
All-cause mortality at 90 daysSafety74/247 (30%)62/255 (24%)Adjusted OR 0.72Not significant
Symptomatic intracranial hemorrhageSafety4/247 (2%)17/255 (7%)Adjusted OR 4.59
Any intracranial haemorrhageSafety67/188 (36%)119/222 (54%)Adjusted OR 1.88
Haemorrhagic infarction type 1Safety35/188 (19%)31/222 (14%)Adjusted OR 0.59
Haemorrhagic infarction type 2Safety19/188 (10%)26/222 (12%)Adjusted OR 1.10
Parenchymal haematoma type 1Safety8/188 (4%)29/222 (13%)Adjusted OR 2.31
Parenchymal haematoma type 2Safety3/188 (2%)16/222 (7%)Adjusted OR 2.17
Embolisation in new territory on DSASafety0/24646/245 (19%)NA
Infarction in new territory within 7 daysSafety3 (1%)4 (2%)Adjusted OR 1.08
Femoral-artery pseudoaneurysmSafety1 (<1%)3 (1%)NA
Groin haematomaSafety09 (4%)NA

Criticisms

  • Open-label design (though endpoint assessment was blinded) may introduce bias in management decisions
  • Conducted entirely in the Netherlands, limiting generalisability to other health systems and populations
  • Increased risk of symptomatic intracranial hemorrhage with endovascular treatment must be weighed against functional benefit
  • Enrollment excluded patients eligible under DAWN/DEFUSE-3 criteria, so results apply to a complementary (not overlapping) population

Funding

Collaboration for New Treatments of Acute Stroke (CONTRAST) consortium, Dutch Heart Foundation, Stryker, Medtronic, Cerenovus, Top Sector Life Sciences & Health, and the Netherlands Brain Foundation

Based on: MR CLEAN-LATE (The Lancet, 2023)

Authors: Olthuis SGH, Pirson FAV, Pinckaers FME, ..., et al.; on behalf of the MR CLEAN-LATE investigators

Citation: Lancet 2023; 401: 1371-80

Content summarized and formatted by NeuroTrials.ai.