Ischemic Stroke

Details of the Concept

Acute ischemic stroke (AIS) results from sudden focal brain ischemia due to thromboembolism or in situ thrombosis of intracranial or extracranial vessels, causing neurological deficits corresponding to the territory of ischemia. Time-sensitive reperfusion — intravenous thrombolysis (IVT) and/or endovascular thrombectomy (EVT) — is the cornerstone of acute treatment. The 2026 AHA/ASA guideline marks a paradigm shift: tenecteplase replaces alteplase as the preferred thrombolytic (COR 1A), EVT indications are substantially expanded (ASPECTS 3–5 now COR 1A; basilar occlusion within 24h COR 1A), and post-EVT BP <140 mmHg within 72h is established as COR 3:Harm.

Key Facts

Classification

• NIHSS: Stroke severity scale (0–42); minor stroke NIHSS 0–5; severe ≥15
• ASPECTS: Alberta Stroke Program Early CT Score (0–10); assesses MCA territory ischemia on CT; ≥6 = favourable imaging; 3–5 = extensive early ischemia; 0–2 = near-complete infarction
• TOAST etiology: Large artery atherosclerosis (LAA), cardioembolism (CE), small vessel occlusion (lacunar), other determined, cryptogenic
• Cryptogenic stroke is a diagnosis of exclusion and accounts for about 20% of ischaemic strokes after a full workup — more in younger patients, where one-third of strokes are cryptogenic. Embolic stroke of undetermined source (ESUS), introduced in 2014, standardises the definition for strokes with an embolic appearance on imaging and no mechanism after workup; it may be caused by occult paroxysmal AF, non-stenotic atherosclerosis, or a patent foramen ovale. (sources/pfo-jama-2025, rating: high)
• PFO-associated stroke is now a named entity, not a cryptogenic subgroup. An international working group proposed in 2020 that it be removed from the cryptogenic category — an ischaemic stroke with a cortical, large white-matter or retinal infarct, a PFO, and no other identified likely cause. The redefinition makes the causal question explicit, because the PFO must be shown to be pathogenic rather than incidental. See concepts/Patent-Foramen-Ovale.

Cryptogenic Stroke and the Patent Foramen Ovale

• A quarter of all adults have a PFO and about half of patients ≤60 with an ESUS do — so roughly half the PFOs found in this population are incidental, and the diagnostic task is discrimination, not detection. (sources/pfo-jama-2025, rating: high)
• Attributable fraction depends on age and vascular risk: the PFO is the likely cause in 80% (75–84%) of cryptogenic strokes under 55 and 52% (34–66%) in older patients. Overall, paradoxical embolism accounts for ~5% of all ischaemic strokes and ~10% under 55.
• Causal likelihood is scored by RoPE and PASCAL and it predicts the size of the treatment effect — see concepts/PFO-Closure-Selection.
• Closure plus antiplatelet therapy is recommended in selected patients aged 18–60 after PFO-associated stroke, with a strong recommendation on high-quality evidence (sources/pfo-eso-2024, rating: very high; sources/pfo-jscai-2022, rating: very high). Pooled trial data give an annualised stroke rate of 0.47% versus 1.09% (adjusted HR 0.41).
• Do not close a PFO without excluding AF first. Occult paroxysmal AF is the likelier mechanism in older patients with vascular risk factors, and closure itself provokes AF; SCAI recommends against routine closure in patients with a history of AF who have had a stroke. See concepts/Poststroke-AF-Monitoring.
• Closure is not the end of secondary prevention: four years after the device, stroke risk remains ~6-fold the general population (2.5% vs 0.4%) and tracks hypertension, diabetes, dyslipidaemia and smoking. (sources/pfo-jama-2025)
• TIA: Transient ischemic attack — symptoms resolve; treated as equivalent to minor AIS for DAPT purposes (sources/ais-aha-2026, rating: very high)

Prehospital and EMS

• Mobile Stroke Units (MSU) — COR 1 A: Recommended over conventional EMS for thrombolytic-eligible patients; reduces onset-to-treatment times; COR 1A for areas with MSU programs (sources/ais-aha-2026, rating: very high)
• Direct EMS destination: Transport to closest EVT-capable hospital preferred; in well-coordinated systems with short transport times, bypass to comprehensive stroke centre is acceptable (sources/ais-aha-2026, rating: very high)
• Pre-hospital severity scales (LAMS, CPSS, sNIHSS) guide destination selection for suspected LVO

Intravenous Thrombolysis (IVT)

• Tenecteplase 0.25 mg/kg (max 25 mg) — COR 1 A: Single IV bolus; non-inferior to alteplase 0.9 mg/kg in multiple RCTs (EXTEND-IA TNK, ATTEST-2, NOR-TEST 2A, TASTE, AcT); now the preferred agent (sources/ais-aha-2026, rating: very high)
• Tenecteplase 0.4 mg/kg — COR 3: No Benefit: Higher dose not recommended
• Alteplase 0.9 mg/kg (max 90 mg) — COR 1 A: Remains standard when tenecteplase unavailable
• Bridging IVT before EVT: Do not skip IVT to hasten EVT — COR 1; IVT improves clot access and may reduce thrombus burden
• Extended window 4.5–9h with perfusion imaging mismatch — COR 2a: EXTEND, ECASS-4, WAKE-UP trials; DWI-FLAIR mismatch guides wake-up stroke selection (sources/ais-aha-2026, rating: very high)
• Non-disabling minor stroke — COR 3: No Benefit: NIHSS 0–5 without disabling deficits; IVT not recommended; DAPT preferred
• Pre-IVT BP requirement: SBP <185 mmHg / DBP <110 mmHg (use IV labetalol, nicardipine, or clevidipine)
• Contraindications: Prior intracranial haemorrhage, platelets <100,000, anticoagulation with elevated INR or Xa/thrombin inhibitor; SBP ≥185/DBP ≥110 unless treatable

Endovascular Thrombectomy (EVT) — Adults

• Basilar artery occlusion — COR 1 A: EVT within 24h, NIHSS ≥10, PC-ASPECTS ≥6; ATTENTION (HR 2.06 for functional independence) and BAOCHE trials — major upgrade from 2018 guideline (sources/ais-aha-2026, rating: very high)
• Pediatric EVT: Age ≥6y within 6h — COR 2a; Age ≥6y within 6–24h — COR 2a; Age 28d–6y within 24h — COR 2b

Blood Pressure Management

• Pre-IVT: Reduce SBP to <185 mmHg / DBP to <110 mmHg; IV antihypertensives required if above threshold (sources/ais-aha-2026, rating: very high)
• Post-IVT (24h): Maintain <180/105 mmHg; targeting SBP <140 mmHg post-IVT — COR 3: No Benefit (ENCHANTED trial)
• Post-EVT successful reperfusion (mTICI 2b/2c/3) — SBP <140 mmHg within 72h — COR 3: HARM: ENCHANTED2/MT trial — intensive lowering to <140 mmHg after successful EVT increased death and disability vs. <180 mmHg; optimal target 140–180 mmHg for 72h post-reperfusion (sources/ais-aha-2026, rating: very high)
• No reperfusion after EVT or no reperfusion therapy: Permissive hypertension (SBP up to 220 mmHg) acceptable in first 24h; avoid aggressive lowering in absence of organ damage
• See entities/Hypertension for full BP management framework

Blood Glucose

• Treat hyperglycemia >180 mg/dL: Target 140–180 mg/dL with insulin — COR 1 (sources/ais-aha-2026, rating: very high)
• Intensive glucose control (80–130 mg/dL) — COR 3: No Benefit: SHINE trial (n=1,151) — no functional benefit; increased hypoglycemia; moderate control preferred
• Hypoglycemia (<60 mg/dL): Treat immediately; glucose <60 is a contraindication to IVT

Antiplatelet Therapy

• DAPT for minor AIS / high-risk TIA (onset <24h) — COR 1 A:
  • Aspirin 325 mg + clopidogrel 300–600 mg loading, then aspirin 75–100 mg + clopidogrel 75 mg × 21 days → single antiplatelet
  • CHANCE (n=5,170) and POINT (n=4,881) trials (sources/ais-aha-2026, rating: very high)
• Ticagrelor + aspirin × 30 days for LAA etiology — COR 2a: THALES trial (HR 0.83); higher bleeding; LAA mechanism only (sources/ais-aha-2026, rating: very high)
• Aspirin within 24–48h of AIS (non-IVT/EVT patients) — COR 1 A: IST and CAST trials — modest consistent benefit
• Aspirin NOT within 24h of IVT/EVT: Risk of hemorrhagic transformation
• See concepts/DAPT-Strategies for full antiplatelet framework

Early Anticoagulation After AF-Related Stroke

• ELAN (n=2,013) is the primary randomised evidence for early initiation. Open-label trial at 103 sites; severity defined by infarct size rather than NIHSS. Early arm: DOAC within 48 h for minor or moderate stroke, day 6–7 for major stroke. Later arm: day 3–4 (minor), day 6–7 (moderate), day 12–14 (major). Primary 30-day composite of recurrent stroke, systemic embolism, major extracranial bleeding, symptomatic ICH or vascular death: 2.9% versus 4.1%; adjusted risk difference −1.18 percentage points (95% CI −2.84 to 0.47); OR 0.70 (0.44–1.14). (sources/elan-doac-timing-nejm-2023, rating: very high)
• The components moved the right way, and sICH did not increase. Recurrent ischaemic stroke 1.4% versus 2.5% (OR 0.57, 0.29–1.07); systemic embolism 0.4% versus 0.9%; symptomatic ICH 0.2% in both arms (OR 1.02, 0.16–6.59); major extracranial bleeding 0.3% versus 0.5%; vascular death 1.1% versus 1.0%. At 90 days the composite was 3.7% versus 5.6% (OR 0.65, 0.42–0.99), but this was not a prespecified secondary outcome and is exploratory. (sources/elan-doac-timing-nejm-2023)
• ELAN tested no hypothesis. It was powered for precision, not for non-inferiority; the authors' reading is that early initiation "can be supported" because the CI extends from a 2.8-point reduction to only a 0.5-point increase, with a 98% probability that early treatment adds no more than 0.5 percentage points of risk. The current COR 2a recommendation for early OAC reflects that evidentiary status. (sources/elan-doac-timing-nejm-2023; sources/ais-aha-2026, rating: very high)
• Timing by severity — what the evidence actually supports.

(sources/elan-doac-timing-nejm-2023)

• Severe stroke remains the weakest ground. The guideline advice to delay ≥14 days after a large infarct with NIHSS >15 is consensus-based; ELAN's "early" major-stroke arm began at day 6–7 and excluded parenchymal haematoma, so neither the guideline delay nor a day 3 start has randomised support. (sources/ais-aha-2026; sources/elan-doac-timing-nejm-2023)
• Heparin bridging — COR 3: Harm: No benefit; increases ICH; not recommended while awaiting OAC
• See entities/Atrial-Fibrillation for full AF anticoagulation framework

DVT Prevention

• Intermittent pneumatic compression (IPC) — COR 1 A: CLOTS 3 trial — reduces proximal DVT (4.6% vs. 7.7%); recommended in all immobilised AIS patients (sources/ais-aha-2026, rating: very high)
• Elastic compression stockings — COR 3: HARM: CLOTS 1 trial — no benefit, increased skin breakdown and complications; explicitly contraindicated in AIS
• Subcutaneous heparin/LMWH for DVT prophylaxis: may be used after 24h in stable non-reperfused patients without hemorrhagic transformation

Dysphagia

• Swallowing screen before any oral intake — COR 1: All AIS patients; bedside water-swallow test or formal SLP evaluation (sources/ais-aha-2026, rating: very high)
• Pharyngeal electrical stimulation (PES) — COR 2a: NEW in 2026 guideline; PHAST trial — modest dysphagia improvement; delivered via nasogastric tube electrode
• Nasogastric tube preferred over early PEG (first 2–3 weeks); FOOD trial showed no survival advantage with early PEG

Rehabilitation

• Early mobilization within 24–72h (low-to-moderate intensity) — COR 1: Short sitting/standing sessions to prevent complications (sources/ais-aha-2026, rating: very high)
• High-dose very early mobilization <24h — COR 3: HARM: AVERT trial (n=2,104) — intensive early mobilization within 24h associated with worse 3-month functional outcomes (OR 0.59 for favourable outcome)
• SSRIs for motor recovery — COR 3: No Benefit: FOCUS, AFFINITY, EFFECTS (combined n>3,000) — fluoxetine 20 mg/day × 6 months did not improve functional outcome; increases fracture risk

Brain Swelling and Decompressive Surgery

• Decompressive hemicraniectomy ≤60 years (malignant MCA infarction) — COR 1 A: DESTINY, DECIMAL, HAMLET trials — NNT ≈2 for survival; mortality reduced from ~78% to ~29%; 43% achieve mRS ≤3 (sources/ais-aha-2026, rating: very high)
• Decompressive hemicraniectomy >60 years — COR 2b: DESTINY II — mortality reduced (33% → 70%) but most survivors have severe disability (mRS 4–5); requires individualized goals-of-care discussion (sources/ais-aha-2026, rating: very high)
• IV glibenclamide for cerebral edema — COR 3: No Benefit: CHARM trial (phase 2b, n=94) — failed to reduce MRI-measured brain swelling
• Osmotic therapy (mannitol/hypertonic saline): reasonable for acute herniation as bridge to decompression; no RCT outcome data

Contradictions / Open Questions

• "Cryptogenic" and "ESUS" select different populations, and the stroke literature uses them interchangeably. Cryptogenic is a diagnosis of exclusion after any workup; ESUS requires an embolic imaging pattern and no mechanism. A patient can be cryptogenic but not ESUS (lacunar-appearing infarct with no cause found) or ESUS but not cryptogenic (two possible mechanisms identified). Trials of PFO closure enrolled "cryptogenic" stroke, and the relevance of their results to an ESUS-labelled patient is assumed rather than demonstrated. (sources/pfo-jama-2025, rating: high)
• The decision to close a PFO rests on how thoroughly AF was excluded, and nobody has defined how much monitoring is enough. AHA/ASA 2021 recommends ≥24 hours; a consortium of eight European societies proposes 6 months in patients at high AF risk before deciding on closure. A stroke attributed to a PFO after 24 hours of monitoring in a 65-year-old with hypertension may simply be an AF stroke with a coincidental PFO. (sources/pfo-jama-2025)
• IVT vs. direct EVT in LVO: SELECT2, DIRECT-MT, DEVT, MR CLEAN NO IV trials suggest direct EVT may be non-inferior to bridging thrombolysis in LVO; 2026 guideline still recommends bridging IVT — ongoing debate, especially in centres with rapid door-to-groin times
• Post-EVT BP optimal range: COR 3:Harm established for SBP <140 mmHg, but the upper limit above which harm begins has not been established by RCT; 140–180 mmHg is operationally recommended without a specific target
• ASPECTS threshold for EVT: ASPECTS 3–5 is now COR 1A based on SELECT2/TESLA/TENSION, but these trials enrolled selected patients; real-world ASPECTS 3–5 populations may have worse outcomes than trial populations
• ELAN's early-arm safety in severe stroke is inferred, not tested. The trial excluded parenchymal haematoma at randomisation, enrolled a population with median NIHSS 3, and did not centrally adjudicate infarct severity; the "early" major-stroke arm started at day 6–7, so earlier initiation in large infarcts is untested. Its confidence interval straddles zero because the trial tested no formal hypothesis. (sources/elan-doac-timing-nejm-2023)
• Basilar occlusion PC-ASPECTS threshold: ATTENTION enrolled predominantly Asian patients; generalizability to non-Asian populations and validity of PC-ASPECTS ≥6 threshold require further study
• Pediatric IVT and EVT: Both receive lower evidence-level recommendations (COR 2a–2b); no dedicated large RCTs exist for pediatric AIS reperfusion

Connections

• Related to concepts/Patent-Foramen-Ovale — the commonest structural cause found in cryptogenic stroke
• Related to concepts/PFO-Closure-Selection — who should be closed, and why the answer is probabilistic
• Related to concepts/Poststroke-AF-Monitoring — excluding the alternative mechanism before treating the PFO
• Related to entities/Hypertension — BP targets pre/post-IVT and post-EVT; COR 3:Harm for <140 mmHg within 72h of reperfusion
• Related to entities/Atrial-Fibrillation — early OAC timing after AF-related AIS (ELAN trial)
• Related to concepts/DAPT-Strategies — DAPT for minor AIS/TIA (COR 1A, 21-day regimen)
• Related to concepts/ASCVD-Risk-Assessment — secondary stroke prevention

Sources

• sources/ais-aha-2026
• sources/elan-doac-timing-nejm-2023
• sources/pfo-jama-2025
• sources/pfo-eso-2024
• sources/pfo-jscai-2022