Stroke TrialsTeaching session
⌂ Home

Sixty-eight trials · door to discharge

Every order we write
came out of a trial.

This is all of them — arranged in the order you actually make decisions, from the door of resus to the discharge letter. Each one opens into its full record: objective, summary, intervention, arms, outcomes.

The colours throughout are the modified Rankin Scale, 0 through 6. Green is independence. Rust is death. Every trial card is edged with the colour of what it achieved.

Dr A. Aburaba — Academic Chief Resident, Neurology

MODIFIED RANKIN SCALE
0 — no symptoms6 — death

Before we start

Four questions and you have read it

Every record in this session is laid out to answer these in order. You never need to memorise a study design again.

Who was in it?

Read Intervention and Arms. If your patient would not have been enrolled, the result is not about your patient. This settles most ward arguments on its own.

Against what?

A drug that beats placebo can still lose to what you already do. Modern medical therapy is a punishing comparator — ask CREST-2.

What was measured?

Function, or a picture? First line of Outcomes. A smaller clot on the scan is not a patient who walks. You will see that trap four times today.

What did it cost?

Every record lists the safety outcome next to the efficacy one. Know the bill before you quote the benefit.

Almost every winning trial here won by doing something earlier, more completely, or in a better-chosen patient. Very few won by inventing a molecule. Most of the improvement still available to us is in how we organise ourselves.

Warm-up · drag the handle

Everything in the next chapter turns on this number

0 h6 h12 h18 h24 h
0h 00
Every neurone still alive.
M1 OCCLUSION CORE · PENUMBRA
A window is a countdown, not a permission slip. Inside every green lane, earlier is still better. Nobody has ever shown that waiting helps.
Chapter one

Open the vessel

Thirty years of pushing two questions: how late, and how sick.

Chapter 1 · dissolving the clot · open any card for the record

Thrombolysis

Chapter 1 · the ward version

What that means at 3am

Do

  • Tenecteplase 0.25 mg/kg (max 25 mg) or alteplase 0.9 mg/kg inside 4.5 hours. One bolus is faster and easier to get right.
  • Treat on disability, not on the NIHSS number. An isolated aphasia or hemianopia is disabling at NIHSS 2.
  • Past 4.5 hours, let imaging decide — perfusion mismatch, or DWI positive with FLAIR negative.
  • Lysis first, then theatre when the patient qualifies for both.

Don't

  • Don't wait for bloods in a patient with no reason to be coagulopathic.
  • Don't lyse a truly non-disabling deficit — dual antiplatelets do the job with less risk.
  • Don't use low-dose alteplase to feel safer; non-inferiority was never proven.
  • Don't fear the mimic. Symptomatic bleeding in mimics is very low. Missing a stroke is not.
Every 15 minutes saved on door-to-needle puts roughly one more patient per hundred back on their feet unaided. Time is the cheapest treatment effect we own.

Chapter 1 · pulling the clot out · open any card for the record

Thrombectomy

Chapter 1 · the ward version

Who do I phone about, and how fast

Phone now

Any ICA, M1 or basilar occlusion with a disabling deficit, within 24 hours. Do not pre-decide that they are too late, too old or too sick — that call belongs to the person holding the images.

Discuss carefully

Large established core, ASPECTS 3–5. There is benefit, but most survivors stay dependent. Consent for that reality, not for a rescue.

Do not extrapolate

M2 distal, M3, A2, P2. Two randomised trials in 2025 found no benefit and more complications. Smaller is not simply smaller.

The job is not finished when the vessel opens. That is the moment blood pressure becomes the intervention — and three trials have shown that pushing it down does harm.
Chapter two

Keep it open

The drugs we prescribe most often, and get wrong most often.

Chapter 2 · thirty years learning one lesson · open any card for the record

Antiplatelets and dual therapy

Chapter 2 · print this one

The DAPT card

Who gets it

Minor ischaemic strokeNIHSS ≤ 3 (≤ 5 in newer data)
or high-risk TIAABCD² ≥ 4
Start within24 h — up to 72 h per INSPIRES
Loading doseAspirin 300 mg + clopidogrel 300 mg
ThenAspirin 75 mg + clopidogrel 75 mg
For how long21 days, then one agent for life
AlternativeTicagrelor + aspirin, 30 days
CYP2C19 loss of functionUse ticagrelor

Who does not

  • Large or disabling strokes. The trials enrolled minor strokes only.
  • Anyone in atrial fibrillation. They need an anticoagulant, not two antiplatelets.
  • Anyone past three weeks. The benefit has gone; only the bleeding is left.
The commonest error in our specialty is DAPT started correctly and never stopped. MATCH, CHARISMA and SPS3 all found the same thing. Write the stop date on the discharge summary.
Chapter three

The pressure

Four different right answers in the same week. This is where we do most of our accidental harm.

Chapter 3 · know which patient you are standing next to

Blood pressure

Infarct, not reperfused

Leave it alone unless above roughly 220/120. The penumbra is living on that pressure.

Infarct, after lysis

Below 180/105. Lower shrinks the bleeding on the scan and does nothing for the patient.

Infarct, after thrombectomy

Do not chase a low number. Three trials found intensive lowering after reperfusion causes harm.

Haemorrhage

Towards 140 in 1–2 hours, smoothly, and hold it. Not below 120.

Chapter four — the long one

When it bleeds

One trial asked the question properly. Everything in the twelve years since has been an answer to it.

LOBAR STICH II TESTED THIS DEEP GREY EXCLUDED VENTRICLES EXCLUDED

Spotlight · Lancet 2013

STICH II — do we take the clot out?

STICH (2005) was neutral overall but left a hint: shallow lobar clots might do better with surgery, deep ones might do worse. STICH II tested that hint properly, and it is still the cleanest surgical question ever asked in haemorrhage.

Enrolled601 patients · 78 centres · 27 countries
ConsciousGCS eye ≥ 2, motor ≥ 5
The clotLobar, 10–100 mL, ≤ 1 cm from surface
VentriclesClean — no intraventricular blood
TimingRandomised < 48 h, surgery < 12 h after
The operationOpen craniotomy
Notice who is missing. STICH II excluded exactly the patients we most want to operate on — the deep bleed, the drowsy patient, the ventricles full of blood. Design first, result second.

Bad outcome at six months

Each square is one patient in a hundred

Early surgery — 59 Good outcome — 41
Conservative — 62 Good outcome — 38

OR 0.86 (0.62–1.20), p = 0.367. Three and a half squares. Not significant.
Six-month death 18 % vs 24 % — OR 0.71 (0.48–1.06), a trend that never quite went away.

And why it may have been blunted: 21 % of the conservative group had surgery anyway. The honest description is that STICH II compared operate now with operate on one in five, later.

Spotlight · what came after · pick a year

Twelve years answering one question

Operate

Cerebellar clot over 3 cm with brainstem compression, hydrocephalus or a falling GCS — go now, and take the clot out rather than just placing a drain. Lobar 30–80 mL inside 24 hours if you have a minimally invasive service.

Drain

Ventricles full with hydrocephalus — external drain. Thrombolysing the ventricles lowers mortality but does not restore function. Tell families in those words.

Still unsolved

The deep hypertensive bleed — the commonest one we admit and the one with nothing proven. Craniectomy is a rescue on weak evidence. Say so out loud.

Chapter 4 · the rest of the toolkit

Everything else we do for a bleed

Read the pattern. Three separate drugs made the bleed smaller on the scan and none made the patient better. The only thing that moved function was a bundle — pressure, glucose, temperature, reversal, all inside the hour.
Chapter five

Stop the next one

Find the cause first. The drug follows from it, never the other way round.

Chapter 5 · when the clot came from the heart

Fibrillation, ESUS and the hole in the heart

The takeaway is a negative one. Empirical anticoagulation for a cryptogenic stroke does not work — two large trials, then a third for atrial cardiopathy. So the work is finding the fibrillation, not guessing at it.

Chapter 5 · the artery in the neck and the one in the head

Large artery disease

Two rules that never change. A symptomatic tight carotid is a two-week emergency, not an outpatient referral. And an intracranial stenosis is a medical problem — SAMMPRIS stopped early because stenting caused the strokes it was meant to prevent.

Chapter 5 · the unglamorous part that saves the most people

Lipids, pressure and the ward routine

If you get one change made in our service this year, make it a written bundle for the first hour — pressure, glucose, temperature, reversal, imaging. It will not be published under your name and it is the highest-yield thing in this session.

Reference · leave this open for questions

All records

Closing

Five things to take to the ward

  • Time is still the whole game. Imaging buys us hours we used to refuse, but earlier is always better inside any window.
  • Dual antiplatelets are a three-week drug. Write the stop date on the discharge summary.
  • Blood pressure has four right answers in one week. Know which patient you are standing next to.
  • STICH II said no to routine craniotomy. ENRICH said yes to early minimally invasive surgery for shallow lobar clots. The deep bleed is unsolved, and we should say so.
  • A better scan is not a better patient. Ask for the functional outcome every single time.
Thank you for sitting through it. The records slide stays open — search it during the questions.

Selected references

1 Prabhakaran S, Gonzalez NR, Zachrison KS, et al. 2026 guideline for the early management of patients with acute ischemic stroke. Stroke. 2026. doi:10.1161/STR.0000000000000513.

2 Mendelow AD, Gregson BA, Rowan EN, et al. Early surgery versus initial conservative treatment in spontaneous supratentorial lobar intracerebral haematomas (STICH II). Lancet. 2013;382(9890):397–408.

3 Pradilla G, Ratcliff JJ, Hall AJ, et al. Trial of early minimally invasive removal of intracerebral hemorrhage (ENRICH). N Engl J Med. 2024;390(14):1277–89.

4 Broderick JP, Grotta JC, Naidech AM, et al. Recombinant factor VIIa within 2 h of intracerebral haemorrhage (FASTEST). Lancet. 2026;407(10530):773–83.

5 Trident Research Group. Three low-dose antihypertensive agents in a single pill after intracerebral hemorrhage. N Engl J Med. 2026;394(16):1571–82.

6 Wang Y, Wang Y, Zhao X, et al. Clopidogrel with aspirin in acute minor stroke or transient ischemic attack (CHANCE). N Engl J Med. 2013;369(1):11–19.

7 Johnston SC, Easton JD, Farrant M, et al. Clopidogrel and aspirin in acute ischemic stroke and high-risk TIA (POINT). N Engl J Med. 2018;379(3):215–25.

8 Goyal M, Menon BK, van Zwam WH, et al. Endovascular thrombectomy after large-vessel ischaemic stroke (HERMES). Lancet. 2016;387(10029):1723–31.

9 Ma L, Hu X, Song L, et al. Third intensive care bundle with blood pressure reduction in acute cerebral haemorrhage (INTERACT3). Lancet. 2023;402(10395):27–40.

10 Sharma M, Dong Q, Hirano T, et al. Asundexian for secondary stroke prevention (OCEANIC-STROKE). N Engl J Med. 2026;394:1467–79.

11 Brott TG, Howard G, Lal BK, et al. Medical management and revascularization for asymptomatic carotid stenosis (CREST-2). N Engl J Med. 2026;394(3):219–31.

12 Ziai WC, Shah VA. Intracerebral hemorrhage. Continuum (Minneap Minn). 2026;32(3):836–68.

Records were assembled from the published reports and have not been checked against a live citation database. Confirm figures in PubMed before circulating.

Dr A. Aburaba — Academic Chief Resident, Neurology