Non-contrast computed tomography (CT) of the head is the test to prioritize when a patient arrives in the emergency department with signs of a stroke, because it rapidly separates ischemic stroke from intracerebral hemorrhage and clears the patient for time-critical thrombolysis. It is the first test in essentially every acute stroke pathway, and it should be obtained within minutes of arrival, before any other imaging or laboratory result delays a treatment decision.

The logic is not that CT is the best test for stroke. It is the fastest test that answers the one question that changes immediate management: is there blood in the brain? Everything else — which artery is blocked, how much brain is salvageable, whether a large vessel occlusion is present — comes second, because no treatment can proceed until hemorrhage is excluded.

Mechanism: Why Hemorrhage Must Be Excluded First

Acute stroke splits into two broad categories with opposite initial treatments. Roughly 87% of strokes are ischemic, caused by a clot or embolus occluding an artery, and about 13% are hemorrhagic, caused by a ruptured vessel bleeding into brain tissue (per American Heart Association/American Stroke Association statistics). Ischemic stroke is treated with intravenous thrombolysis and, for large vessel occlusions, mechanical thrombectomy. Hemorrhagic stroke is treated by blood pressure control, reversal of anticoagulation, and neurosurgical evaluation. Giving a thrombolytic to a patient with intracranial hemorrhage worsens bleeding and can be fatal.

Non-contrast CT detects acute intracranial hemorrhage with very high sensitivity once blood has been present for a short period, and it does so in a scan that takes seconds to acquire. The scanner measures X-ray attenuation: fresh blood is denser than brain parenchyma, so it appears bright white against gray tissue. That physical contrast is the entire basis for the first decision in stroke care.

A non-contrast CT also does something subtler. It can reveal early ischemic changes — loss of gray-white differentiation, sulcal effacement, a hyperdense vessel sign — that hint at a clot before any contrast is given. These signs are insensitive early, but when present they support the ischemic diagnosis. The scan is therefore not purely a hemorrhage rule-out; it is a first look at the whole picture.

Plain language: the first scan answers "is this a bleed or a blockage?" because that single answer decides whether the patient gets clot-busting drug or blood-pressure and reversal treatment.

Quantified Context: Time, Sensitivity, and Dose

The numbers explain why CT wins the priority contest. Intravenous thrombolysis is generally most effective when given within 4.5 hours of symptom onset, and mechanical thrombectomy windows extend to 24 hours in selected patients with salvageable tissue. Every minute of delay costs brain: pooled analyses of large registries associate each 15-minute delay in thrombolysis with worse functional outcomes, and every hour of untreated large vessel occlusion is estimated to cost on the order of 120 million neurons. These figures are the reason stroke pathways are built around speed rather than test elegance.

Non-contrast CT acquisition itself takes seconds to a couple of minutes. Door-to-CT times of under 25 minutes are achievable in organized stroke centers, and guidelines emphasize rapid imaging so that thrombolysis decisions are not delayed. Compare that with MRI, which can take 20 to 45 minutes for a full stroke protocol and requires screening for implanted devices and patient tolerance. MRI is more sensitive for early ischemia, but its time cost is a real cost when the treatment clock is running.

The counterintuitive detail: a normal non-contrast CT does not exclude ischemic stroke. Early ischemia is often invisible on CT within the first hours, and a normal scan is entirely compatible with a treatable clot. The scan's job is to rule out hemorrhage and permit treatment, not to confirm infarction.

Comparison: Which Test Answers Which Question

Test Primary question answered Typical speed Key limitation
Non-contrast head CT Hemorrhage present or absent Seconds to a few minutes Insensitive to early ischemia
CT angiography (CTA) Large vessel occlusion? Minutes, added to CT Requires contrast; renal and allergy considerations
CT perfusion Salvageable tissue vs. core infarct? Minutes, added to CT Requires contrast and software; not universal
MRI (DWI) Acute infarction, even small or posterior 20–45 minutes for protocol Slower; device and tolerance screening
Carotid ultrasound Carotid stenosis (not acute phase) Variable Not an emergency triage test
Lumbar puncture Subarachnoid hemorrhage when CT equivocal After CT, hours into workup Invasive; reserved for specific scenarios

The practical reading of this table: non-contrast CT is the triage test, and CTA usually follows immediately in the same session to answer the thrombectomy question. MRI is the problem-solver when the diagnosis remains uncertain, the patient is outside the thrombolysis window, or posterior circulation stroke is suspected.

Timeline: The First Hours of a Stroke Workup

Time from arrival Action Purpose
0–10 minutes Triage, ABCs, glucose check, stroke scale, IV access Confirm stroke mimic exclusion begins; glucose is a mandatory check
0–15 minutes Non-contrast head CT ordered and patient transported Exclude hemorrhage
10–25 minutes CT acquired and read Hemorrhage rule-out; early ischemic signs
15–30 minutes CTA (and CT perfusion where available) Detect large vessel occlusion; assess salvageable tissue
25–60 minutes Thrombolysis decision and administration if eligible Reperfusion within the 4.5-hour window
30–90 minutes Thrombectomy team activation and transfer if needed Mechanical reperfusion for large vessel occlusion
60 minutes onward MRI if diagnosis unclear or posterior circulation suspected Resolve diagnostic uncertainty

These intervals are targets, not guarantees, and they shift with local resources. The ordering principle does not: non-contrast CT precedes every treatment decision.

Edge Cases: When the First Scan Is Not Enough

Several scenarios bend the rule without breaking it. Posterior circulation stroke — basilar artery occlusion, cerebellar infarction — is notoriously missed on non-contrast CT because the brainstem and cerebellum are small structures adjacent to dense bone. A patient with vertigo, diplopia, ataxia, or crossed deficits and a normal CT needs MRI or CTA urgently. The CT still goes first to exclude hemorrhage, but it cannot be the last test.

Subarachnoid hemorrhage from a ruptured aneurysm can be subtle. Non-contrast CT is highly sensitive within the first hours but sensitivity declines over days, so a negative CT in a patient with sudden severe headache may still require lumbar puncture or CTA. Wake-up stroke and stroke of unknown onset time need imaging-based selection — MRI or CT perfusion — to identify patients who might still benefit from reperfusion.

Large vessel occlusion changes the priority sequence in a specific way. Once hemorrhage is excluded, CTA should be obtained without delay because thrombectomy eligibility depends on it. In centers with CT perfusion, the combination of non-contrast CT, CTA, and perfusion imaging is often acquired in a single visit to the scanner, so the "first test" and the "next test" are practically one event.

Red-flag warnings:

A normal non-contrast CT never excludes ischemic stroke; do not stop the workup on a negative scan. Sudden severe headache with a negative CT still warrants evaluation for subarachnoid hemorrhage. Posterior circulation symptoms with a normal CT require MRI or CTA before discharge. Do not delay CT for laboratory results other than glucose; treatment decisions should not wait on routine panels. Anticoagulant use raises hemorrhage risk and makes hemorrhage exclusion even more urgent, not less. A patient outside the thrombolysis window still needs imaging, because thrombectomy windows extend to 24 hours in selected cases.

Misconceptions: What the Scan Does and Does Not Do

The most common misconception is that MRI should come first because it is "more sensitive." Sensitivity is not the deciding property in the first minutes; speed and specificity for hemorrhage are. MRI is the better diagnostic test for infarction, and it is the right second-line test, but it is the wrong first-line gatekeeper in a time-limited pathway.

A second misconception is that CT "diagnoses stroke." It diagnoses hemorrhage reliably and suggests ischemia indirectly. The clinical diagnosis of stroke rests on history, examination, and the stroke scale, with imaging used to classify and to guide treatment. A patient with a convincing syndrome and a normal CT is still a stroke patient until proven otherwise.

A third misconception concerns contrast. CTA and CT perfusion require iodinated contrast, which carries small risks in renal impairment and allergy. Those risks are real but are weighed against the benefit of identifying a thrombectomy-eligible occlusion. The non-contrast scan itself requires no contrast, which is another reason it can be first without hesitation.

The professional caveat worth stating plainly: imaging protocols are local. Some centers use MRI-first pathways for selected patients, and mobile stroke units bring CT to the patient. The priority of non-contrast CT reflects the dominant, guideline-supported pathway, not a universal law.

Practical Actions: What to Expect and What to Ask

For clinicians, the sequence is straightforward. Obtain non-contrast head CT as the first imaging study, read it for hemorrhage and early ischemic change, and proceed immediately to CTA when large vessel occlusion is possible. Check glucose early because hypoglycemia is a common stroke mimic. Do not let routine laboratory results delay imaging or thrombolysis decisions.

For patients and families, the practical actions are about speed and information. Note the time symptoms began or when the patient was last known well, because every treatment window is measured from that moment. Tell the team about anticoagulant use, recent surgery, and implanted devices. Ask whether the first scan has been done and whether a vessel imaging study is planned. Do not wait for symptoms to resolve before seeking care.

For emergency departments, the operational thresholds matter. Target door-to-CT within 20 to 25 minutes in stroke alerts, and door-to-needle within 60 minutes for thrombolysis. Build the pathway so that CT, CTA, and perfusion are available in one scanner visit. Audit the intervals, because the mechanism of benefit is time.

A system-level caution: access varies. Not every hospital has CT perfusion software, 24-hour MRI, or thrombectomy capability. Transfer decisions should be made early when a large vessel occlusion is suspected, because the receiving center's capabilities change the plan.

Limitations: What This Priority Does Not Settle

Non-contrast CT cannot tell you the stroke subtype with precision, cannot quantify the penumbra, and cannot reliably detect small or posterior circulation infarcts. It answers the hemorrhage question and little more in the first minutes. The choice of CT first is a triage decision made under uncertainty, and it should be revisited as soon as CTA, perfusion, or MRI results arrive.

The evidence base also has boundaries. Imaging technology, software, and protocol recommendations change, and local pathways differ. The general principle — exclude hemorrhage fast, then characterize the occlusion — is stable, but the specific intervals and test combinations are institutional and evolve over time.

This article is general educational information, not a diagnosis or treatment plan. Stroke is a medical emergency; call local emergency services immediately if someone shows signs of stroke, and consult a qualified local clinician for any personal medical decision.