19.1 IV Reperfusion Therapy (03.C.3)

19.1 IV Reperfusion Therapy

Quick Answer: Eligible adults within 0–4.5 hours of last known well receive tenecteplase 0.25 mg/kg IV (maximum 25 mg, single bolus) or alteplase 0.9 mg/kg IV (maximum 90 mg: 10% bolus, remainder over 60 minutes). Pretreatment blood pressure ≤185/110 mm Hg; during treatment and for 24 hours keep <180/105 mm Hg. Stop the drug, scan, and give cryoprecipitate (target fibrinogen ≥150 mg/dL) for symptomatic intracranial hemorrhage. Wake-up stroke with DWI-positive, FLAIR-negative mismatch can still be a lytic candidate.

Minutes of untreated large-vessel ischemia destroy cortex. On the ABIM Neurocritical Care examination, stems punish a 100 kg tenecteplase dose of 90 mg, a lytic given at 198/118 mm Hg, a DOAC swallowed that morning treated as “unknown INR so proceed,” and a wake-up hemiparesis dismissed because the clock from last seen normal is 10 hours. Independent OpenExamPrep teaching here covers intravenous reperfusion for acute ischemic stroke as listed among Neurocritical care diseases in the ABPN Content Specifications. This guide is not an ABIM or ABPN product. Pharmacology of alteplase, tenecteplase, antiplatelets, and anticoagulants is also taught in the reperfusion-drugs chapter; this section is eligibility, dosing arithmetic, blood pressure, exclusions, hemorrhage rescue, and unknown-onset imaging.

Why the first 4.5 hours still matter

Intravenous thrombolysis (IVT) recanalizes some occlusions and improves 90-day disability in patients with a disabling deficit after hemorrhage is excluded on noncontrast CT (NCCT). The National Institute of Neurological Disorders and Stroke (NINDS) alteplase trial established benefit in 0–3 hours. ECASS III extended alteplase to 3–4.5 hours in selected patients. Benefit is time-dependent: earlier needle time means more modified Rankin Scale (mRS) 0–1 outcomes and less hemorrhage. Do not wait for creatinine, a complete lipid panel, or an MRI that will not change a 4.5-hour decision. Do not withhold a lytic merely to “get the patient to thrombectomy faster” if the patient is IVT-eligible — bridging IVT plus endovascular thrombectomy (EVT) remains the default, not a skip-lytic strategy.

Mild, nondisabling deficits (isolated sensory change, isolated dysarthria that the patient and you agree will not change function) did not benefit from lytics in dedicated trials. Dual antiplatelet therapy is the usual path for that phenotype. Disabling aphasia, hemianopia that wrecks driving, or a hand that cannot hold a cup is not “too mild.”

Alteplase and tenecteplase: doses you must calculate

Alteplase (recombinant tissue plasminogen activator) dose is 0.9 mg/kg, maximum 90 mg. Give 10% as an IV bolus over about 1 minute; infuse the remaining 90% over 60 minutes. If the infusion infiltrates or is interrupted, you have a pharmacokinetics problem and a documentation problem — tenecteplase’s single bolus exists partly to avoid that failure mode.

Tenecteplase is a genetically modified tPA with longer half-life and greater fibrin specificity. The stroke dose is 0.25 mg/kg as a single IV bolus, maximum 25 mg. The Alteplase compared to Tenecteplase (AcT) trial (Menon and colleagues, Lancet 2022) showed noninferiority of this dose versus alteplase for 90-day mRS 0–1. The 2026 AHA/ASA Guideline for the Early Management of Patients With Acute Ischemic Stroke gives a Class 1 recommendation for either tenecteplase 0.25 mg/kg (max 25 mg) or alteplase 0.9 mg/kg in eligible adults within 4.5 hours of symptom onset or last known well. Many systems now lead with tenecteplase because one bolus is simpler in the emergency department and in the angiography suite. That practical advantage is not a license to use the 0.4 mg/kg dose studied in NOR-TEST 2, which increased symptomatic intracranial hemorrhage (sICH) without a benefit worth the bleed.

Worked arithmetic:

• 80 kg alteplase: 0.9 × 80 = 72 mg total. Bolus 7.2 mg. Infusion 64.8 mg over 60 minutes.
• 110 kg alteplase: 0.9 × 110 = 99 mg → cap at 90 mg. Bolus 9 mg. Infusion 81 mg over 60 minutes.
• 80 kg tenecteplase: 0.25 × 80 = 20 mg once.
• 110 kg tenecteplase: 0.25 × 110 = 27.5 mg → cap at 25 mg once.

A 100 kg patient does not receive 90 mg of tenecteplase. That is the alteplase ceiling misapplied to the wrong drug.

Blood pressure before and after the lytic

Pivotal alteplase protocols required systolic BP <185 mm Hg and diastolic BP <110 mm Hg before treatment. After the drug is in, the target is <180/105 mm Hg during administration and for 24 hours. AcT excluded patients whose hypertension could not be brought under <180/105 mm Hg. Lower is not automatically better in the untreated large-vessel occlusion still waiting for the groin stick; overshoot hypotension can enlarge the infarct. Use easily titrated IV agents (labetalol, nicardipine, clevidipine — whatever your unit actually stocks) rather than a one-time hydralazine prayer. If you cannot reach ≤185/110 mm Hg without crashing perfusion, the patient is not a lytic candidate on BP grounds; EVT may still proceed if an LVO is present.

Absolute and relative exclusions

Absolute stops are findings that make hemorrhage risk unacceptable or make the diagnosis not ischemic stroke:

• Intracranial hemorrhage on NCCT, or a clinical picture of subarachnoid hemorrhage even if the CT is early.
• Active internal bleeding.
• Recent intracranial or intraspinal surgery, or recent serious head trauma (classically within 3 months).
• INR >1.7 or PT >15 seconds on warfarin; platelet count <100,000/µL when a coagulopathy is suspected (do not delay the bolus in a previously healthy patient just to wait for platelets that are almost never the blocker).
• Treatment-dose low-molecular-weight heparin within 24 hours, or unfractionated heparin within 48 hours with an aPTT above the upper limit of normal.
• Direct oral anticoagulant (DOAC) taken recently: the 2019-era rule that still appears on examinations is last dose within 48 hours (or unknown timing) without a documented normal specific assay (dilute thrombin time / ecarin for dabigatran; calibrated anti-Xa for apixaban/rivaroxaban/edoxaban). Observational 2026-era series report that selected DOAC-exposed patients given IVT did not show a large sICH signal, but that is not a blanket “apixaban at 06:00, lytic at 08:00” policy. If the last DOAC dose was this morning, the safe exam answer is withhold IVT; EVT can still be offered for LVO.

Relative exclusions require judgment and a documented risk discussion:

• Major surgery or serious trauma within 14 days (bleeding into the wound).
• Gastrointestinal or genitourinary hemorrhage within 21 days.
• Prior intracranial hemorrhage — remote, small, and explained (for example a 10-year-old traumatic epidural) is not the same as recurrent lobar hemorrhages suggesting amyloid angiopathy.
• Seizure at onset with a residual deficit that might be Todd paresis; treat if you have vascular imaging or early ischemic change supporting stroke.
• Pregnancy, arterial puncture at a noncompressible site within 7 days, known unruptured aneurysm or vascular malformation — individualize.
• Glucose extremes: historically <50 mg/dL or >400 mg/dL. The 2026 AHA/ASA text defines severe hypo- and hyperglycemia around those same poles (<50 and >400 mg/dL), tells you to correct glucose and reassess, and still supports IVT if a disabling deficit persists after correction. A glucose of 32 mg/dL that resolves with dextrose is a mimic, not a lytic indication. A glucose of 32 mg/dL that you correct, with persistent aphasia and a hyperdense MCA, is still a stroke.

Do not delay IVT to count cerebral microbleeds on an MRI you do not need. A known burden up to about 10 microbleeds is not an automatic stop; a florid amyloid pattern with countless lobar bleeds is a different conversation.

Symptomatic intracranial hemorrhage after the lytic

sICH is a clinical decline plus hemorrhage on imaging, classically a ≥4-point NIHSS rise in NINDS-style definitions. It is not a trivial petechial blush you only see because you ordered a 24-hour CT. When the nurse reports sudden vomiting, a new blown pupil, or a patient who stopped following after a previously improving exam, stop alteplase immediately (tenecteplase is already in as a bolus — you cannot retrieve it, but you still stop any remaining push and you still reverse the fibrinolytic state).

Send CBC, PT/INR, aPTT, fibrinogen, and type and cross. Obtain emergent NCCT. The 2026 AHA/ASA adult table for bleeding within 24 hours after alteplase or tenecteplase is operational, not theoretical:

• Cryoprecipitate (contains fibrinogen and factor VIII): 10 units over 10–30 minutes, aiming for fibrinogen ≥150 mg/dL. A rule of thumb in that table: 10 units raise fibrinogen by about 50 mg/dL.
• If cryoprecipitate is delayed or refused, tranexamic acid 1000 mg IV over 10 minutes, or ε-aminocaproic acid 4–5 g over 1 hour then 1 g/h until bleeding is controlled.
• Treat BP, ICP, cerebral perfusion pressure, temperature, and glucose as you would any ICH. Call hematology and neurosurgery when evacuation or ICP drainage is on the table.

Do not give platelets “because tPA affects platelets” as a first move. Do not start a heparin drip to “protect the remaining clot.” Orolingual angioedema is a different emergency: stop the infusion, hold ACE inhibitors, protect the airway (fiberoptic if the floor of mouth is rising), and use steroid plus antihistamine plus epinephrine pathways as in the AHA table; icatibant or C1-inhibitor appears in refractory ACE-inhibitor–like swelling.

Wake-up and unknown onset: the tissue clock

Last known well is not the same as wake-up time. A patient last seen normal at 22:00 who is found aphasic at 07:00 is 9 hours on the wall clock and may still have tissue that is only 1–2 hours old if the stroke happened near dawn. The WAKE-UP trial (Thomalla and colleagues, NEJM 2018) enrolled unknown-onset patients who could be treated within 4.5 hours of symptom recognition and who had MRI mismatch: an acute lesion on diffusion-weighted imaging (DWI) without marked FLAIR hyperintensity in the same region — a biologic signature that the infarct is likely <4.5 hours old. Patients with DWI lesions larger than one-third of the MCA territory were excluded. IV alteplase improved mRS 0–1 at 90 days (53.3% versus 41.8%). Conceptually, DWI-positive / FLAIR-negative means “still a lytic clock.” DWI and FLAIR both bright means the tissue has been ischemic long enough to accumulate vasogenic water — treat that as a late infarct, not an ECASS III clone.

Perfusion mismatch (CTP or MR PWI: small core, large penumbra) selected patients in EXTEND for alteplase in a 4.5–9 hour window and appears in later extended-window work. Do not invent a 9-hour lytic for an unmarked NCCT without mismatch. Do not refuse EVT in a wake-up LVO merely because you are arguing about FLAIR; late-window thrombectomy has its own imaging rules in the next section.

Worked stems

An 80 kg man at 2.5 hours with NIHSS 12, NCCT without blood, BP 176/98, INR 1.1: alteplase 72 mg (7.2 mg bolus + 64.8 mg infusion) or tenecteplase 20 mg once. Either is a 2026 Class 1 choice.

A 90 kg woman on rivaroxaban last taken 6 hours ago, NIHSS 18, left M1 occlusion: no IVT on standard DOAC timing. Activate EVT. Do not wait 48 hours “for the DOAC to wear off” while the MCA dies.

A 70 kg man found at 08:00, last seen normal at 23:00, NIHSS 8, MRI DWI bright in a small MCA patch, FLAIR dark: this is WAKE-UP physiology. If otherwise eligible, IVT is the evidence-based discussion, not automatic exclusion because last known well is 9 hours.

A patient drops from NIHSS 6 to 18 forty minutes into an alteplase infusion, vomits, and the left pupil enlarges: stop the infusion, stat NCCT, send fibrinogen, give cryoprecipitate targeting ≥150 mg/dL, and manage BP/ICP. Do not complete the remaining 50 mL “because the protocol bag is still hanging.”

Exam traps

Capping tenecteplase at 90 mg. Using 0.4 mg/kg tenecteplase. Lytic at BP 200/120. Treating INR 2.1. Skipping glucose. Calling every wake-up stroke ineligible without asking for DWI–FLAIR or perfusion. Observing “to see if the lytic works” before calling the interventionalist. Independent practice items at /practice/abim-neurocritical-care drill these numbers; they are a study bank, not the computer-based examination administered by ABPN.