Residency · Residency · Neurology
Mechanical Thrombectomy: Patient Selection and Evidence
Overview
Mechanical thrombectomy is now the standard of care for acute ischemic stroke caused by large vessel occlusion in the anterior circulation. Five landmark randomized controlled trials published in 2015 established overwhelming benefit, and subsequent trials have expanded eligibility to include patients presenting in late time windows (up to 24 hours) and those with large infarct cores. With a number needed to treat as low as 2 to 3 for reduced disability, thrombectomy stands as one of the most effective interventions in all of medicine.
Landmark Thrombectomy Trials (2015)
MR CLEAN (Netherlands)
MR CLEAN was the first positive trial, evaluating intra-arterial treatment within 6 hours of anterior circulation large vessel occlusion. It demonstrated an absolute 13.5% increase in functional independence (mRS 0-2) and used predominantly first-generation stent retrievers. Notably, no perfusion imaging was required for patient selection.
ESCAPE (Canada)
ESCAPE permitted thrombectomy up to 12 hours from onset and required moderate-to-good collaterals on multiphase CTA. The trial emphasized rapid workflow, with an ideal onset-to-groin-puncture time of under 4 hours, and excluded patients with large established infarcts (ASPECTS below 6). The number needed to treat was 4 for reduced disability.
EXTEND-IA (Australia)
EXTEND-IA used CT perfusion-based selection within 6 hours, requiring a target mismatch between core and penumbra. The results were dramatic, with a 71% reperfusion rate and significant improvement in early neurological recovery, demonstrating the power of perfusion imaging to select patients who stand to benefit most.
SWIFT PRIME (International)
SWIFT PRIME evaluated thrombectomy within 6 hours using the Solitaire stent retriever, required confirmed large vessel occlusion on CTA, and excluded patients with large cores on CT perfusion. Functional independence was achieved in 60% of thrombectomy patients versus 35% in the control group.
REVASCAT (Spain)
REVASCAT treated patients within 8 hours and required an ASPECTS of 7 or greater. It confirmed the benefit of thrombectomy with an absolute 15.5% increase in functional independence and reinforced the generalizability of results across different health systems.
| Trial | Year | Time Window | Selection Criteria | Key Result |
|---|---|---|---|---|
| MR CLEAN | 2015 | 0–6 h | CTA-confirmed LVO | +13.5% functional independence |
| ESCAPE | 2015 | 0–12 h | Good collaterals, ASPECTS ≥ 6 | NNT 4 |
| EXTEND-IA | 2015 | 0–6 h | CT perfusion mismatch | 71% reperfusion rate |
| SWIFT PRIME | 2015 | 0–6 h | CTA + CT perfusion | 60% vs 35% independence |
| REVASCAT | 2015 | 0–8 h | ASPECTS ≥ 7 | +15.5% functional independence |
| DAWN | 2018 | 6–24 h | Clinical-imaging mismatch | Tissue-based paradigm |
| DEFUSE-3 | 2018 | 6–16 h | Target mismatch on CTP/MRI | 45% vs 17% independence |
| SELECT2 | 2023 | 0–24 h | Large core (ASPECTS 3–5) | Net benefit despite higher sICH |
| ANGEL-ASPECT | 2023 | 0–24 h | Large core (ASPECTS 3–5) | Improved mRS distribution |
| RESCUE-Japan LIMIT | 2022 | 0–24 h | Large core (ASPECTS 3–5) | NNT 4 |
Pooled Meta-Analysis (HERMES)
The HERMES collaboration pooled individual patient data from all five trials and found a strong benefit across all prespecified subgroups, including age, NIHSS, and occlusion site. Benefit was observed regardless of whether IV tPA had been administered first, and although earlier treatment produced greater benefit, meaningful improvement persisted up to 7.3 hours. The overall number needed to treat for reduced disability on the mRS was 2.6.
Extended Time Window Trials
DAWN (2018)
DAWN enrolled patients 6 to 24 hours from last known well, selecting them based on a clinical-imaging mismatch in which a high NIHSS was disproportionate to a small ischemic core on CTP or DWI. Three mismatch strata based on age and NIHSS were defined. The adjusted difference in utility-weighted mRS strongly favored thrombectomy, establishing the tissue-based paradigm for late-window stroke treatment.
DEFUSE-3 (2018)
DEFUSE-3 studied the 6-to-16-hour window using target mismatch criteria on CTP or MRI: a core under 70 mL, mismatch ratio exceeding 1.8, and penumbra greater than 15 mL. Functional independence was achieved in 45% of thrombectomy patients versus 17% in the control arm, and the trial was stopped early for efficacy.
Large Core Infarct Trials
SELECT2 (2023)
SELECT2 enrolled patients with large vessel occlusion and large ischemic cores (ASPECTS 3-5 or core volumes of 50 to 100 mL). Thrombectomy significantly improved functional outcomes despite the large infarct burden. Although rates of symptomatic ICH were higher, the net benefit was maintained.
ANGEL-ASPECT (2023)
Conducted in a Chinese population, ANGEL-ASPECT enrolled patients with ASPECTS 3-5 or core volumes of 70 to 100 mL and allowed thrombectomy within 24 hours using perfusion-based selection. The trial demonstrated a significant improvement in the distribution of mRS scores.
RESCUE-Japan LIMIT (2022)
RESCUE-Japan LIMIT studied large core patients (ASPECTS 3-5) treated within 6 hours, or within 24 hours with perfusion-based selection. Thrombectomy showed significant benefit with a number needed to treat of 4. Collectively, these three trials expanded eligibility to patients who were previously considered too far advanced for intervention.
Patient Selection Criteria
Imaging Requirements
Non-contrast CT head is obtained first to rule out hemorrhage and assess ASPECTS. CT angiography confirms the large vessel occlusion location, whether in the ICA, M1, or proximal M2 segment. CT perfusion assesses the ischemic core and penumbra, which is particularly important for extended time window and large core decision-making. The ASPECTS is a 10-point score assessing early ischemic changes in the MCA territory; a score of 6 or above is the traditional threshold for standard thrombectomy, though the large core trials now include scores of 3 to 5.
Target Vessels
The primary targets for thrombectomy include the internal carotid artery terminus and the middle cerebral artery M1 segment, which is the most common occlusion site. Growing evidence supports treatment of proximal M2 branches, though they have been less rigorously studied in RCTs. For the posterior circulation, the ATTENTION and BAOCHE trials now support thrombectomy for basilar artery occlusion. More distal occlusions in the M3, A2, and P2 segments remain an active area of investigation.
Clinical Criteria
Patients should have pre-stroke functional independence (traditionally mRS 0-1, though some benefit has been shown for mRS 0-2). The NIHSS is typically 6 or above, although no strict cutoff exists in guidelines. Age is not a limiting factor, as the HERMES subgroup analysis demonstrated benefit in patients over 80 years.
Thrombectomy Technique
Stent retriever thrombectomy involves deploying the device (such as Solitaire or Trevo) across the thrombus, allowing it to integrate with the clot, and then withdrawing it along with the thrombus. Direct aspiration, known as the ADAPT technique, uses a large-bore aspiration catheter placed at the face of the clot with vacuum applied to extract it. The combined approach, sometimes called the "Solumbra" technique, uses aspiration plus a stent retriever simultaneously. Reperfusion is graded on the modified TICI scale, where TICI 2b/3 (more than 50% reperfusion) is the minimum target and TICI 2c/3 (near-complete to complete reperfusion) is associated with the best outcomes. Achieving complete reperfusion in the first pass, known as the first-pass effect, is associated with superior clinical results.
Workflow and Time Metrics
At a comprehensive stroke center, the door-to-groin-puncture target is under 90 minutes, or under 120 minutes when transfer is involved. Groin puncture to reperfusion should be under 60 minutes. Some centers have implemented direct-to-angio-suite protocols that bypass the emergency department in select cases. The debate between drip-and-ship (giving thrombolysis at the presenting hospital and transferring) and mothership (direct transport to a thrombectomy-capable center) models continues, though the simplicity of a tenecteplase bolus before transfer has strengthened the drip-and-ship approach.
Posterior Circulation Thrombectomy
Basilar Artery Occlusion
The ATTENTION trial in 2022 demonstrated significant benefit of thrombectomy within 12 hours for basilar occlusion, with 46% achieving a good outcome versus 23% in the control group. BAOCHE in 2022 confirmed this benefit within 24 hours using favorable imaging selection. Although the BASICS trial in 2021 did not show benefit, it had methodological limitations including prolonged enrollment and a heterogeneous population. Current practice favors thrombectomy for basilar artery occlusion based on ATTENTION and BAOCHE.
<image>A visual summary of the five landmark 2015 thrombectomy trials (MR CLEAN, ESCAPE, EXTEND-IA, SWIFT PRIME, REVASCAT) presented as a forest plot showing the odds ratio for functional independence (mRS 0-2) for thrombectomy vs control in each trial and the pooled HERMES meta-analysis estimate. Each trial is labeled with its key selection criteria (CTA only, collateral assessment, perfusion imaging, ASPECTS threshold) and time window. The pooled estimate shows a clear, statistically significant benefit with the overall NNT annotated.</image>
<image>An ASPECTS scoring diagram on a non-contrast CT head showing two axial slices (ganglionic level and supraganglionic level) with the 10 ASPECTS regions labeled and outlined (C, L, IC, I, M1-M6). One example shows a normal ASPECTS 10 scan, and another shows early ischemic changes in 4 regions yielding an ASPECTS of 6. A color-coded scale indicates the traditional threshold (>=6 for standard thrombectomy, 3-5 for large core trials, <3 generally excluded).</image>
<image>A procedural illustration of mechanical thrombectomy techniques. Panel A shows stent retriever thrombectomy: a microcatheter is navigated past the clot, the stent retriever is deployed and integrates with the thrombus, then the device is withdrawn under flow arrest with a balloon guide catheter, removing the clot. Panel B shows direct aspiration (ADAPT): a large-bore aspiration catheter is advanced to the face of the clot, suction is applied, and the clot is extracted. Panel C shows the modified TICI reperfusion grading scale (0, 1, 2a, 2b, 2c, 3) with corresponding angiographic images illustrating each grade from no reperfusion to complete reperfusion.</image>
Clinical Pearls
Thrombectomy has a number needed to treat of 2 to 3 for reduced disability, placing it among the most effective treatments in all of medicine. IV thrombolysis should never delay thrombectomy; the bolus should be given and the patient moved directly to the angiography suite. ASPECTS is a pragmatic bedside tool but suffers from significant interrater variability, and automated software platforms such as RAPID and Viz.ai improve scoring consistency. Large core infarcts (ASPECTS 3-5) are no longer an absolute contraindication to thrombectomy, as demonstrated by SELECT2, ANGEL-ASPECT, and RESCUE-Japan LIMIT. Age alone should not exclude patients, given the clear benefit shown in patients over 80 in the HERMES meta-analysis. Posterior circulation large vessel occlusion, specifically basilar artery thrombosis, should be treated with thrombectomy based on the ATTENTION and BAOCHE trials. First-pass complete reperfusion (TICI 2c/3) is associated with the best clinical outcomes, underscoring that operator technique and experience matter.
References
- Berkhemer OA, et al. A randomized trial of intraarterial treatment for acute ischemic stroke (MR CLEAN). N Engl J Med. 2015;372(1):11-20.
- Goyal M, et al. Randomized assessment of rapid endovascular treatment of ischemic stroke (ESCAPE). N Engl J Med. 2015;372(11):1019-1030.
- Campbell BC, et al. Endovascular therapy for ischemic stroke with perfusion-imaging selection (EXTEND-IA). N Engl J Med. 2015;372(11):1009-1018.
- Goyal M, et al. Endovascular thrombectomy after large-vessel ischaemic stroke: a meta-analysis of individual patient data from five randomised trials (HERMES). Lancet. 2016;387(10029):1723-1731.
- Sarraj A, et al. Endovascular thrombectomy for large ischemic stroke across ischemic core volumes (SELECT2). N Engl J Med. 2023;388(14):1259-1271.
- Tao C, et al. Endovascular treatment for acute basilar-artery occlusion (ATTENTION). N Engl J Med. 2022;387(15):1361-1372.
