Mechanical thrombectomy is recommended for patients with acute ischemic stroke caused by a large vessel occlusion, but only when specific clinical, imaging, and timing criteria are met. The standard window is within 6 hours of symptom onset, though carefully selected patients can be treated up to 24 hours later. Getting the selection right matters enormously: in the right patient, thrombectomy dramatically improves functional outcomes and reduces mortality, while in the wrong patient, it can cause hemorrhage or death with no benefit.
Core Eligibility Within 6 Hours
The foundational criteria for thrombectomy within the first 6 hours center on three things: stroke severity, the amount of brain already damaged, and where the clot is located. Current guidelines recommend thrombectomy for patients with an NIHSS score of 6 or higher, an ASPECTS score between 3 and 10, and a confirmed occlusion of the internal carotid artery or the first segment of the middle cerebral artery. The patient should also have been functionally independent before the stroke, defined as a pre-stroke modified Rankin Scale score of 0 to 1.
Patients with mild strokes (NIHSS below 6) who have a proximal large vessel occlusion represent a gray area. Thrombectomy is the standard of care at NIHSS 6 and above, but treatment in the minor stroke subgroup is still being explored. These patients can deteriorate rapidly if the clot doesn’t resolve, so clinical judgment plays a role even when they fall below the formal threshold.
Which Arteries Qualify
Strong trial evidence supports thrombectomy for occlusions of the internal carotid artery (both intracranial and cervical segments, including tandem occlusions) and the M1 and M2 segments of the middle cerebral artery. M2 occlusions were included in several landmark trials, though in smaller numbers, and are now considered a class I indication alongside ICA and M1 occlusions.
Distal medium vessel occlusions, such as M3 segments or smaller branches, are a different story. Three recent randomized trials tested thrombectomy for these smaller clots and found no clinical benefit. Worse, the thrombectomy groups showed higher rates of symptomatic brain bleeding and even increased mortality. Recanalization rates for these smaller vessels were notably lower than in large vessel trials, likely because the devices are less effective in narrow, tortuous arteries. For now, thrombectomy in distal vessels is not broadly recommended and should only be considered in highly selected cases with disabling deficits and clear perfusion mismatch.
Extending the Window to 24 Hours
Two landmark trials, DAWN and DEFUSE-3, established that thrombectomy can work well beyond 6 hours if the right patients are identified through advanced imaging. The key concept is “mismatch”: a large area of brain tissue that is starving for blood but not yet dead. If the area of irreversible damage (the core) is small relative to the total area at risk, there is still brain worth saving.
To qualify in this extended window, patients need a confirmed ICA or proximal middle cerebral artery occlusion, an ischemic core volume of less than 70 ml, a mismatch ratio of 1.8 or greater (meaning the total threatened tissue is at least 1.8 times the size of the core), and at least 15 ml of potentially salvageable brain tissue. In the DAWN trial specifically, patients were selected based on a disproportionately severe clinical deficit compared to the size of the infarct on imaging. This approach applies to patients who wake up with stroke symptoms or present late, as long as imaging confirms there is viable tissue to rescue.
How Imaging Defines the Decision
Advanced imaging is what makes patient selection possible, especially beyond the 6-hour window. CT perfusion and MRI perfusion scans generate maps that distinguish dead tissue from tissue that is struggling but alive.
The ischemic core, representing brain tissue that is almost certainly already dead, is identified as areas where relative cerebral blood flow has dropped below 30% compared to the opposite side of the brain. The total area of critically low blood flow, which includes both the core and the salvageable penumbra, is defined by a processing delay (called Tmax) greater than 6 seconds. The difference between these two volumes is the penumbra: the tissue thrombectomy aims to save. Automated software calculates these volumes in minutes, giving the treatment team objective numbers to guide the decision.
Within the first 6 hours, a non-contrast CT scan scored using ASPECTS is often sufficient. ASPECTS ranges from 0 to 10, with 10 being a normal scan and lower scores indicating more extensive early damage. Most trial evidence supports thrombectomy at ASPECTS 6 and above, but newer data has expanded this range.
Large Core Infarcts
Until recently, patients with large areas of already-damaged brain were excluded from thrombectomy on the assumption that the damage was too extensive to justify the risks. The SELECT2 trial challenged this, enrolling patients with ASPECTS scores of 3 to 5 or core volumes of 50 ml or more. These patients had occlusions of the internal carotid artery or M1 segment and were treated within 24 hours of symptom onset.
The trial found that even in this population, thrombectomy improved outcomes compared to medical therapy alone. However, these patients need to have been functionally independent before the stroke (pre-stroke mRS of 0 or 1) and must have no evidence of intracranial hemorrhage on imaging. The benefit in large-core patients is real but more modest, and the risk of complications is higher, so careful case-by-case evaluation is essential.
Blood Pressure Requirements
Blood pressure must be controlled before the procedure can begin. The threshold is below 185/110 mmHg for patients who have received or will receive clot-dissolving medication. If blood pressure cannot be brought below this level, thrombectomy is generally not performed. During and after the procedure, the target tightens slightly to below 180/105 mmHg. Evidence suggests that patients who arrive with even lower systolic pressures tend to do better, though aggressive lowering in someone with a large vessel occlusion requires balancing the risk of worsening the ischemia.
Contraindications
Several conditions rule out thrombectomy or make it too risky. Intracranial hemorrhage on imaging is an absolute contraindication, since opening a blocked vessel into an area that is already bleeding would be catastrophic. A large completed infarct with minimal salvageable penumbra means there is little brain left to save and a high risk of converting dead tissue into a hemorrhage. Small vessel occlusions are not amenable to mechanical retrieval. Uncorrectable coagulopathies significantly raise bleeding risk during the procedure. Persistently elevated blood pressure that cannot be brought below 185/110 mmHg is also a contraindication, as is a high likelihood of reocclusion from factors like extreme platelet counts, pre-existing vessel narrowing, or residual clot material.
Functional Independence Before the Stroke
Nearly all major thrombectomy trials required patients to have been functionally independent before their stroke, typically defined as a modified Rankin Scale score of 0 or 1. This means the person could carry out daily activities without assistance. Patients who were already significantly dependent (mRS of 3 or higher) before the stroke are routinely excluded from standard protocols. The reasoning is straightforward: people who were already disabled face higher rates of institutionalization, mortality, and complications after the procedure, and the potential for meaningful recovery is lower. That said, the mRS is a blunt tool, and some patients with moderate pre-existing disability may still benefit. This remains an area where clinical teams weigh individual circumstances rather than applying a rigid cutoff.