The Code Stroke Workflow
- A code stroke is a race against the clock — brain tissue dies fast, so the whole workflow is built for speed.
- The first job of imaging is almost philosophical: rule out bleeding. Treatment for a clot and treatment for a bleed are opposites, so you must know which one you're dealing with before you do anything.
- The non-contrast head CT is the gatekeeper: fast, everywhere, and great at spotting blood.
- Vessel imaging (CTA) and perfusion then answer two more questions: is there a blocked artery I can fix, and is there still brain worth saving.
- Radiology isn't a bystander here — the imaging IS the decision. Read fast, read right.
Somewhere a phone is ringing, an overhead page is going off, and a stranger's brain is quietly suffocating. That's a code stroke. Someone showed up with a face that won't smile evenly, an arm that won't lift, or words that come out scrambled, and now an entire hospital is sprinting in their direction. Our corner of that sprint is the scanner — and the clock on the wall is the actual villain of the story.
Why the clock is the whole point
When an artery to the brain clogs, the tissue downstream stops getting oxygen. Brain doesn't tolerate that well. There's a grim slogan in stroke care — "time is brain" — and it is not poetry, it's physiology. Every minute of ischemic stroke chews through neurons that don't grow back.
So the entire workflow is engineered around one goal: shave minutes. Treatments like clot-busting drugs and mechanical clot retrieval work best the sooner they're given, and after certain time windows they stop helping and start risking harm. Imaging is the bottleneck everyone is waiting on, which is a thrilling and slightly terrifying place to sit.
Step one: is it a bleed or a clot?
Here's the fork in the road that makes the whole thing tense. A stroke comes in two flavors: a blocked vessel (ischemic) or a burst vessel (hemorrhagic). They look similar at the bedside — same droopy face, same garbled speech — but their treatments are mortal enemies.
Give a clot-buster to someone whose problem is a clot: good. Give that same clot-buster to someone who's actively bleeding into their brain: catastrophic. It's like reaching for the fire extinguisher and grabbing the flamethrower. So before anyone treats anything, imaging has to answer the first question cleanly: blood, yes or no?
That's the job of the non-contrast head CT — and it's why this study is the gatekeeper of the entire code.
Step two: the non-contrast head CT
CT earns its starring role here for unglamorous reasons: it's fast, it's available at basically every hospital at 3 a.m., and fresh blood lights up bright white on it. We do this scan first, and we do it without contrast, because contrast can muddy the one thing we most need to see — intracranial hemorrhage.
If there's blood, the pathway changes completely and the clot-busters stay in the drawer. If there's no blood, we move on — but we keep our eyes open, because early ischemic stroke is sneaky. In the first hours, the CT can look almost normal, with only whispers of trouble: a subtle loss of the usual gray-white contrast, or a faintly too-bright artery where a clot is hiding. These early stroke signs are exactly the kind of thing a tired eye skates right past.
A normal-looking head CT does NOT rule out a stroke. Early ischemia is often invisible on the first scan — the CT's main early job is excluding blood, not confirming a clot. Trust the clinical picture and keep moving through the workflow.
Step three: find the blocked artery (CTA)
If we've cleared blood and the patient looks like a big stroke, the next question is mechanical: where is the clog, and is it a big one we can physically pull out? This is where CT angiography comes in — a contrast-enhanced scan that maps the brain's plumbing.
A blockage in one of the large, proximal arteries (a "large vessel occlusion") is the kind that interventional teams can reach with a catheter and retrieve. CTA shows it as an abrupt cutoff — a highway that just... ends. Spotting it fast can route the patient straight to the angio suite. (This does mean giving iodinated contrast, which in the heat of a code is usually a fast risk-benefit call rather than a leisurely one.)
Step four: how much brain is still worth saving (perfusion)
Imagine a wildfire. Part of the forest is already ash — that's brain that's gone, the core. Around it is forest that's smoldering but not yet burned — starved of blood but still alive if you restore flow fast. That at-risk-but-salvageable ring is the penumbra, and it's the entire reason we bother.
Perfusion imaging measures blood flow across the brain to estimate how big the dead core is versus how big the salvageable penumbra is. A small core with a large penumbra means there's a lot to rescue — a strong argument to intervene, sometimes even later in the time window. A huge core means the horse has largely left the barn, and aggressive treatment may do more harm than good.
Putting it together
Each study answers one clean question, and they stack:
| Step | Study | The question it answers |
|---|---|---|
| 1 | Non-contrast head CT | Is there blood? (If yes, stop and rethink.) |
| 2 | (same CT) | Any early ischemic changes? |
| 3 | CT angiography | Is there a big, retrievable blocked artery? |
| 4 | CT perfusion | Is there still brain worth saving? |
The patient is on the scanner table while the clock ticks and a roomful of people wait on your read. Speed matters, but so does not crying wolf — confidently calling a stroke mimic like a seizure or low blood sugar a stroke can send someone down the wrong path entirely.
The one thing to remember
A code stroke is a sequence of yes/no gates, and imaging holds every key. First prove there's no bleed. Then find the clot. Then decide if there's brain left to rescue. Do it fast, do it in order, and remember that on this particular study, reading quickly and reading correctly are the same act of saving someone's brain.