Skip to content

Acute ischemic stroke

TL;DR — Acute ischemic stroke is a race to restore perfusion without treating a hemorrhage or exposing irreversibly infarcted tissue to disproportionate risk. Intravenous alteplase improves disability outcomes when given within 4.5 hours, with benefit falling continuously as treatment is delayed; MRI or perfusion mismatch can identify selected patients with unknown onset or up to 9 hours from onset (Emberson 2014, PMID 25106063; Thomalla 2018, PMID 29766770; Ma 2019, PMID 31067369). For anterior-circulation large-vessel occlusion, thrombectomy produces a large ordinal disability benefit (common OR 2.49, number needed to treat about 2.6 for one-step mRS improvement), extends to selected patients at 6–24 hours, and now includes many patients with a large established core (Goyal 2016, PMID 26898852; Nogueira 2018, PMID 29129157; Sarraj 2023, PMID 36762865). Tenecteplase 0.25 mg/kg is a bolus alternative with non-inferior functional outcomes to alteplase and higher early reperfusion before thrombectomy in one trial (Campbell 2018, PMID 29694815; Menon 2022, PMID 35779553). The operational principle is parallelism: stabilize physiology, establish last-known-well and baseline function, exclude hemorrhage, identify occlusion, and start eligible reperfusion without allowing advanced imaging or transfer to create avoidable delay.

Immediate diagnostic sequence

The initial question is not “is this definitely a stroke?” but “is there disabling focal deficit, hemorrhage, a treatable arterial occlusion, or a dangerous mimic?” Non-contrast CT rapidly separates hemorrhage from a scan that may be normal early in ischemia; CTA from arch to vertex identifies large-vessel occlusion, tandem disease, and relevant vascular anatomy. MRI diffusion is more sensitive for small, posterior-fossa, and very early infarction but must not delay otherwise-indicated reperfusion; imaging performance is synthesized in classification and diagnostic workup.

Parallel action Decision enabled Delay-sensitive failure
Confirm last-known-well, witnessed onset, and baseline function Standard vs tissue-based treatment window Treating awakening time as onset; missing a witnessed earlier baseline
NIHSS plus explicit disabling-deficit assessment Severity, thrombectomy triage, thrombolysis judgment NIHSS underweights gait, hand function, vision and posterior-circulation deficits
Point-of-care glucose Detect hypoglycemic mimic; quantify hyperglycemia Waiting for a full chemistry panel when no specific contraindication exists
Non-contrast CT Exclude hemorrhage; estimate early ischemic change Delaying thrombolysis for MRI in a conventional-window candidate
CTA head/neck Detect LVO and activate thrombectomy pathway Waiting for creatinine in a patient without a known severe contrast contraindication
CT perfusion or MRI mismatch when relevant Late/unknown-window selection; core estimate Applying late-window thresholds to an early-window patient and losing time
Medication/history check Anticoagulants, recent surgery/bleeding, prior disability Treating an electronic medication list as proof of adherence

Clinical severity and imaging are complementary. A low NIHSS can coexist with a disabling aphasia, hemianopia, distal occlusion, or basilar disease; conversely, a high NIHSS can result from seizure with post-ictal deficit, hypoglycemia, migraine, or functional symptoms. Reperfusion decisions therefore require a deficit-by-deficit assessment, not an NIHSS cutoff alone.

Intravenous thrombolysis: what the trials establish

The NINDS trial established benefit within 3 hours: despite no neurological improvement at 24 hours, alteplase increased the odds of minimal or no disability at 3 months by at least 30% across four outcome scales; symptomatic intracerebral hemorrhage within 36 hours occurred in 6.4% versus 0.6%, without a significant mortality difference (NINDS rt-PA Stroke Study Group 1995, PMID 7477192). ECASS III extended treatment to 3–4.5 hours: mRS 0–1 occurred in 52.4% with alteplase and 45.2% with placebo (OR 1.34, 95% CI 1.02–1.76), while symptomatic ICH by the trial definition occurred in 2.4% versus 0.2% (Hacke 2008, PMID 18815396).

IST-3 broadened enrollment to older and less-selected patients within 6 hours. Its primary dichotomous outcome was neutral, but ordinal analysis favored alteplase; fatal or non-fatal symptomatic ICH within 7 days was 7% versus 1%, emphasizing that early hazard and later functional benefit coexist (Sandercock 2012, PMID 22632908). Individual-participant meta-analysis resolved the time interaction: alteplase improved the odds of a good outcome when started within 4.5 hours, benefit declined with delay, and proportional benefit did not disappear solely because of age over 80 or severe stroke (Emberson 2014, PMID 25106063).

Population / strategy Main result Safety signal Interpretation
Alteplase ≤3 h ≥30% relative increase in odds of minimal/no disability sICH 6.4% vs 0.6% Large time-dependent benefit despite early bleeding hazard (PMID 7477192)
Alteplase 3–4.5 h mRS 0–1: 52.4% vs 45.2%; OR 1.34 sICH 2.4% vs 0.2% Benefit persists but is smaller later (PMID 18815396)
Alteplase ≤6 h, broad eligibility Primary dichotomy neutral; ordinal shift favorable fatal/non-fatal sICH 7% vs 1% Do not infer equal benefit throughout 0–6 h (PMID 22632908)
MRI DWI-positive/FLAIR-negative unknown onset favorable mRS 53.3% vs 41.8%; adjusted OR 1.61 death 4.1% vs 1.2%; sICH 2.0% vs 0.4% Tissue clock can substitute for an unavailable wall clock in selected patients (PMID 29766770)
Perfusion mismatch 4.5–9 h or wake-up excellent outcome 35.4% vs 29.5%; adjusted RR 1.44 sICH 6.2% vs 0.9% Modest absolute benefit with selected imaging phenotype (PMID 31067369)

Minor stroke is not synonymous with non-disabling stroke

PRISMS enrolled patients with NIHSS 0–5 whose deficits were judged non-disabling. The trial stopped early after 313 of a planned 948 participants; favorable outcome was 78.2% with alteplase and 81.5% with aspirin, while symptomatic ICH occurred in 3.2% and 0%, respectively (Khatri 2018, PMID 29998337). Because it was underpowered and excluded disabling minor deficits, PRISMS supports avoiding alteplase for clearly non-disabling presentations; it does not justify withholding therapy for aphasia, hemianopia, dominant-hand weakness, or gait failure merely because NIHSS is low.

Tenecteplase

Tenecteplase is administered as a single bolus, simplifying transfer and avoiding a one-hour infusion. In EXTEND-IA TNK, 0.25 mg/kg before thrombectomy achieved substantial reperfusion before the procedure in 22% versus 10% with alteplase (incidence ratio 2.2, 95% CI 1.1–4.4) and improved ordinal mRS (common OR 1.7, 95% CI 1.0–2.8) (Campbell 2018, PMID 29694815). The pragmatic Canadian AcT trial found 0.25 mg/kg tenecteplase non-inferior to alteplase for mRS 0–1 at 90–120 days: 36.9% versus 34.8%, with symptomatic ICH 3.4% versus 3.2% (Menon 2022, PMID 35779553). TRACE-2 independently found non-inferiority in thrombolysis-eligible patients not planned for thrombectomy: mRS 0–1 in 62% versus 58%, risk ratio 1.07 (95% CI 0.98–1.16) (Wang 2023, PMID 36774935). These data support 0.25 mg/kg; they should not be extrapolated to higher tenecteplase doses.

Mechanical thrombectomy

Five pivotal 2015 trials aligned vessel imaging, modern devices, and fast workflow. MR CLEAN showed an adjusted common OR of 1.67 for a better 90-day mRS with intra-arterial treatment (Berkhemer 2015, PMID 25517348). ESCAPE selected patients with a small core and moderate-to-good collaterals and increased functional independence from 29.3% to 53.0%, while reducing mortality from 19.0% to 10.4% (Goyal 2015, PMID 25671798). EXTEND-IA used perfusion selection and increased reperfusion at 24 hours from 37% to 89% and independence from 40% to 71% (Campbell 2015, PMID 25671797). SWIFT PRIME increased functional independence from 35% to 60% (Saver 2015, PMID 25882376), and REVASCAT improved the ordinal mRS distribution (adjusted common OR 1.7, 95% CI 1.05–2.8) (Jovin 2015, PMID 25882510).

The HERMES individual-participant meta-analysis combined 1,287 patients: thrombectomy yielded a common OR 2.49 (95% CI 1.76–3.53) for lower disability, with a number needed to treat of 2.6 for one patient to improve by at least one mRS category. Benefit was present across age, sex, severity, alteplase eligibility, and geography; 90-day mortality and symptomatic ICH did not differ significantly (Goyal 2016, PMID 26898852). Treatment speed remained decisive: every hour of delay to arterial puncture was associated with less functional benefit, and benefit became non-significant after about 7.3 hours in the early-window trial population (Saver 2016, PMID 27673305).

Late window: physiology, not permission to wait

DAWN selected patients 6–24 hours from last known well whose clinical deficit was disproportionate to infarct volume. Utility-weighted mRS at 90 days was 5.5 versus 3.4, and functional independence was 49% versus 13%; symptomatic ICH did not differ significantly (6% vs 3%) (Nogueira 2018, PMID 29129157). DEFUSE 3 selected a core under 70 mL, mismatch ratio at least 1.8, and penumbra at least 15 mL at 6–16 hours; independence was 45% versus 17% and mortality 14% versus 26% (Albers 2018, PMID 29364767). These are selection trials, not evidence that time no longer matters.

Large established core

Recent trials overturned the assumption that a large core makes thrombectomy futile, while also showing that absolute prognosis often remains poor.

Trial / selection Window Primary signal Important tradeoff
RESCUE-Japan LIMIT, ASPECTS 3–5 ≤6 h, or ≤24 h without early FLAIR change mRS 0–3: 31.0% vs 12.7%; RR 2.43 Any ICH 58.0% vs 31.4% (Yoshimura 2022, PMID 35138767)
ANGEL-ASPECT, ASPECTS 3–5 or core 70–100 mL ≤24 h generalized OR 1.37 for better mRS Any ICH 49.1% vs 17.3%; sICH 6.1% vs 2.7% (Huo 2023, PMID 36762852)
SELECT2, ASPECTS 3–5 or core ≥50 mL ≤24 h generalized OR 1.51; functional independence 20% vs 7% Vascular complications occurred; sICH uncommon in both arms (Sarraj 2023, PMID 36762865)
TENSION, ASPECTS 3–5, mainly non-perfusion selection ≤12 h adjusted common OR 2.58; mortality 40% vs 51% Supports simpler imaging selection where perfusion is unavailable (Bendszus 2023, PMID 37837989)

The clinically relevant question has shifted from “large core: yes or no?” to how age, core volume and location, edema risk, pre-stroke function, technical access, and the patient's valuation of survival with dependency modify net benefit.

Basilar-artery occlusion

BASICS was neutral overall: favorable outcome (mRS 0–3) at 90 days occurred in 44.2% with endovascular therapy and 37.7% with medical therapy (RR 1.18, 95% CI 0.92–1.50), with sICH 4.5% versus 0.7% (Langezaal 2021, PMID 34010530). Two later Chinese trials were positive in more selected, generally severe presentations. ATTENTION found mRS 0–3 in 46% versus 23% within 12 hours (adjusted rate ratio 2.06), with mortality 37% versus 55% (Tao 2022, PMID 36239644). BAOCHE found mRS 0–3 in 46% versus 24% at 6–24 hours (adjusted rate ratio 1.81), with sICH 6% versus 1% (Jovin 2022, PMID 36239645). Differences in eligibility, crossover, stroke severity, background thrombolysis, and population limit simple pooling.

Thrombolysis before thrombectomy

For an alteplase-eligible patient presenting directly to a thrombectomy-capable center, randomized trials ask whether thrombectomy alone preserves outcome while avoiding hemorrhage and delay. DIRECT-MT met its prespecified non-inferiority boundary for thrombectomy alone but with a wide margin and predominantly Chinese population (Yang 2020, PMID 32374959). MR CLEAN-NO IV did not show superiority or non-inferiority of either strategy, leaving the treatments clinically non-equivalent on that trial's criteria (LeCouffe 2021, PMID 34758251). SKIP also failed to demonstrate non-inferiority for thrombectomy alone (Suzuki 2021, PMID 33464334), while DIRECT-SAFE failed its non-inferiority test in an international population (Mitchell 2022, PMID 35810757). The evidence therefore does not support delaying thrombectomy for infusion completion, but neither does it support routinely omitting otherwise-indicated IV thrombolysis.

Supportive care changes outcome

Reperfusion occurs within a physiological system. Routine low-dose oxygen in non-hypoxic stroke did not improve 90-day death or disability in 8,003 participants (Roffe 2017, PMID 28973619). Intensive glucose control to 80–130 mg/dL versus standard 80–179 mg/dL did not improve 90-day function and produced more severe hypoglycemia (2.6% vs 0%) in SHINE (Johnston 2019, PMID 31334795). After alteplase, targeting systolic blood pressure 130–140 mm Hg rather than the guideline threshold below 180 mm Hg reduced any intracranial hemorrhage but did not improve 90-day functional status in ENCHANTED (Anderson 2019, PMID 30739745).

Very early, high-dose mobilization is not automatically beneficial. AVERT randomized mobilization within 24 hours and found fewer favorable 3-month outcomes with the intensive protocol (46% vs 50%; adjusted OR 0.73), despite no significant mortality difference (AVERT Trial Collaboration 2015, PMID 25892679). Organized stroke-unit care, swallowing assessment before oral intake, fever evaluation, venous-thromboembolism prevention, pressure-area care, and early individualized rehabilitation remain distinct from an aggressive mobilization dose.

If no thrombolytic is given and hemorrhage has been excluded, the large IST and CAST factorial trials found a small early net benefit from aspirin, reducing early recurrent ischemic stroke without a net excess of death or hemorrhagic stroke (IST Collaborative Group 1997, PMID 9174558; CAST Collaborative Group 1997, PMID 9186381). Antiplatelets are generally withheld for the first 24 hours after IV thrombolysis pending follow-up imaging; mechanism-specific secondary prevention then replaces generic acute treatment (secondary prevention).

Deterioration after presentation

Change Time-critical possibilities Immediate reassessment
New headache, vomiting, hypertension, reduced consciousness Hemorrhagic transformation, reperfusion hemorrhage, edema Stop antithrombotic/thrombolytic infusion if running; urgent CT; coagulation and fibrinogen
Worsening focal deficit Re-occlusion, thrombus migration, incomplete reperfusion, edema, seizure Repeat NIHSS, glucose, CT/CTA; discuss rescue treatment
Declining consciousness after large hemispheric infarct Malignant edema and herniation CT, airway assessment, neurosurgical escalation
Fluctuating brainstem signs Basilar thrombosis progression or re-occlusion Urgent vascular imaging and thrombectomy-center review
Fever, hypoxia, tachypnea Aspiration, infection, pulmonary embolism Oxygenation, chest evaluation, swallow status

For malignant middle-cerebral-artery infarction in patients 18–60 years treated within 48 hours, pooled randomized data found decompressive surgery reduced 1-year mortality from 71% to 22% and increased mRS ≤3 from 21% to 43%; survival benefit includes survivors with substantial disability, making early values-based discussion essential (Vahedi 2007, PMID 17303527).

Evidence boundaries and equity

Trial eligibility is not the same as biological eligibility. The pivotal thrombectomy trials underrepresented severe pre-stroke disability, frailty, very distal occlusions, pregnancy, children, and settings without rapid advanced imaging. Late-window and large-core trials often used expert centers with high reperfusion rates, so their effect sizes may not transport unchanged to systems with long transfer or anesthesia delays. Conversely, requiring perfusion imaging where it is unavailable can exclude patients whom non-perfusion large-core trials suggest may benefit (Bendszus 2023, PMID 37837989).

Open questions

  • Which patients with low NIHSS but proven LVO benefit from immediate thrombectomy rather than rescue after deterioration? PRISMS addressed non-disabling thrombolysis, not the occlusion-specific thrombectomy question (Khatri 2018, PMID 29998337).
  • What is the net benefit of thrombectomy at the extremes of large core, advanced age, or pre-stroke dependency? The large-core trials moved the boundary but retained important exclusions (Yoshimura 2022, PMID 35138767; Sarraj 2023, PMID 36762865).
  • Can tenecteplase's early reperfusion advantage be translated into better patient-centered outcome across transfer networks? EXTEND-IA TNK showed reperfusion superiority, whereas AcT and TRACE-2 established overall non-inferiority (Campbell 2018, PMID 29694815; Menon 2022, PMID 35779553; Wang 2023, PMID 36774935).
  • Which basilar-occlusion phenotypes account for the different results of BASICS, ATTENTION, and BAOCHE (Langezaal 2021, PMID 34010530; Tao 2022, PMID 36239644; Jovin 2022, PMID 36239645)?
  • Can prehospital triage improve time to reperfusion without delaying thrombolysis for the larger group without LVO? The steep time-response relationship defines the harm that any routing strategy must overcome (Saver 2016, PMID 27673305).

References

  1. National Institute of Neurological Disorders and Stroke rt-PA Stroke Study Group. Tissue plasminogen activator for acute ischemic stroke. N Engl J Med. 1995;333:1581-1587. PMID 7477192.
  2. Hacke W, et al. Thrombolysis with alteplase 3 to 4.5 hours after acute ischemic stroke. N Engl J Med. 2008;359:1317-1329. PMID 18815396.
  3. Sandercock P, et al. The benefits and harms of intravenous thrombolysis with recombinant tissue plasminogen activator within 6 h of acute ischaemic stroke (IST-3). Lancet. 2012;379:2352-2363. PMID 22632908.
  4. Emberson J, et al. Effect of treatment delay, age, and stroke severity on the effects of intravenous thrombolysis with alteplase for acute ischaemic stroke. Lancet. 2014;384:1929-1935. PMID 25106063.
  5. Khatri P, et al. Effect of intravenous alteplase vs aspirin on functional outcome for patients with acute ischemic stroke and minor nondisabling neurologic deficits. JAMA. 2018;320:156-166. PMID 29998337.
  6. Thomalla G, et al. MRI-guided thrombolysis for stroke with unknown time of onset. N Engl J Med. 2018;379:611-622. PMID 29766770.
  7. Ma H, et al. Thrombolysis guided by perfusion imaging up to 9 hours after onset of stroke. N Engl J Med. 2019;380:1795-1803. PMID 31067369.
  8. Campbell BCV, et al. Tenecteplase versus alteplase before thrombectomy for ischemic stroke. N Engl J Med. 2018;378:1573-1582. PMID 29694815.
  9. Menon BK, et al. Intravenous tenecteplase compared with alteplase for acute ischaemic stroke in Canada (AcT). Lancet. 2022;400:161-169. PMID 35779553.
  10. Wang Y, et al. Tenecteplase versus alteplase in acute ischaemic cerebrovascular events (TRACE-2). Lancet. 2023;401:645-654. PMID 36774935.
  11. Berkhemer OA, et al. A randomized trial of intraarterial treatment for acute ischemic stroke. N Engl J Med. 2015;372:11-20. PMID 25517348.
  12. Goyal M, et al. Randomized assessment of rapid endovascular treatment of ischemic stroke. N Engl J Med. 2015;372:1019-1030. PMID 25671798.
  13. Campbell BC, et al. Endovascular therapy for ischemic stroke with perfusion-imaging selection. N Engl J Med. 2015;372:1009-1018. PMID 25671797.
  14. Saver JL, et al. Stent-retriever thrombectomy after intravenous t-PA vs. t-PA alone in stroke. N Engl J Med. 2015;372:2285-2295. PMID 25882376.
  15. Jovin TG, et al. Thrombectomy within 8 hours after symptom onset in ischemic stroke. N Engl J Med. 2015;372:2296-2306. PMID 25882510.
  16. Goyal M, et al. Endovascular thrombectomy after large-vessel ischaemic stroke. Lancet. 2016;387:1723-1731. PMID 26898852.
  17. Saver JL, et al. Time to treatment with endovascular thrombectomy and outcomes from ischemic stroke. JAMA. 2016;316:1279-1288. PMID 27673305.
  18. Nogueira RG, et al. Thrombectomy 6 to 24 hours after stroke with a mismatch between deficit and infarct. N Engl J Med. 2018;378:11-21. PMID 29129157.
  19. Albers GW, et al. Thrombectomy for stroke at 6 to 16 hours with selection by perfusion imaging. N Engl J Med. 2018;378:708-718. PMID 29364767.
  20. Yoshimura S, et al. Endovascular therapy for acute stroke with a large ischemic region. N Engl J Med. 2022;386:1303-1313. PMID 35138767.
  21. Huo X, et al. Trial of endovascular therapy for acute ischemic stroke with large infarct. N Engl J Med. 2023;388:1272-1283. PMID 36762852.
  22. Sarraj A, et al. Trial of endovascular thrombectomy for large ischemic strokes. N Engl J Med. 2023;388:1259-1271. PMID 36762865.
  23. Bendszus M, et al. Endovascular thrombectomy for acute ischaemic stroke with established large infarct. Lancet. 2023;402:1753-1763. PMID 37837989.
  24. Langezaal LCM, et al. Endovascular therapy for stroke due to basilar-artery occlusion. N Engl J Med. 2021;384:1910-1920. PMID 34010530.
  25. Tao C, et al. Trial of endovascular treatment of acute basilar-artery occlusion. N Engl J Med. 2022;387:1361-1372. PMID 36239644.
  26. Jovin TG, et al. Trial of thrombectomy 6 to 24 hours after stroke due to basilar-artery occlusion. N Engl J Med. 2022;387:1373-1384. PMID 36239645.
  27. Yang P, et al. Endovascular thrombectomy with or without intravenous alteplase in acute stroke. N Engl J Med. 2020;382:1981-1993. PMID 32374959.
  28. LeCouffe NE, et al. A randomized trial of intravenous alteplase before endovascular treatment for stroke. N Engl J Med. 2021;385:1833-1844. PMID 34758251.
  29. Suzuki K, et al. Effect of mechanical thrombectomy without vs with intravenous thrombolysis on functional outcome among patients with acute ischemic stroke. JAMA. 2021;325:244-253. PMID 33464334.
  30. Mitchell PJ, et al. Endovascular thrombectomy versus standard bridging thrombolytic with endovascular thrombectomy within 4.5 h of stroke onset. Lancet. 2022;400:116-125. PMID 35810757.
  31. Roffe C, et al. Effect of routine low-dose oxygen supplementation on death and disability in adults with acute stroke. JAMA. 2017;318:1125-1135. PMID 28973619.
  32. Johnston KC, et al. Intensive vs standard treatment of hyperglycemia and functional outcome in patients with acute ischemic stroke. JAMA. 2019;322:326-335. PMID 31334795.
  33. Anderson CS, et al. Intensive blood pressure reduction with intravenous thrombolysis therapy for acute ischaemic stroke (ENCHANTED). Lancet. 2019;393:877-888. PMID 30739745.
  34. AVERT Trial Collaboration group. Efficacy and safety of very early mobilisation within 24 h of stroke onset (AVERT): a randomised controlled trial. Lancet. 2015;386:46-55. PMID 25892679.
  35. International Stroke Trial Collaborative Group. The International Stroke Trial (IST). Lancet. 1997;349:1569-1581. PMID 9174558.
  36. CAST Collaborative Group. CAST: randomised placebo-controlled trial of early aspirin use in 20,000 patients with acute ischaemic stroke. Lancet. 1997;349:1641-1649. PMID 9186381.
  37. Vahedi K, et al. Early decompressive surgery in malignant infarction of the middle cerebral artery. Lancet Neurol. 2007;6:215-222. PMID 17303527.