The Ischemic Stroke Workup: Finding the Mechanism
The ischemic stroke workup: standard imaging, cardiac, and lab evaluation, the cryptogenic add-ons (prolonged monitoring, PFO), and the young-stroke panel.
Every ischemic stroke gets the same core workup — vessels, brain, heart, rhythm, and blood — and when that is unrevealing you escalate to prolonged rhythm monitoring, a shunt hunt, and, in the young, a search for dissection and hypercoagulability.
- → Standard panel for everyone: head-and-neck vessel imaging (CTA or MRA), brain MRI with DWI, echocardiogram with bubble study, 24–72 h of telemetry, and basic labs (CBC, BMP, A1c, lipids, troponin, coagulation studies).
- → Cryptogenic after that means escalate, not stop: prolonged outpatient monitoring (patch or implantable loop recorder) markedly raises atrial-fibrillation detection, and a PFO with no other source may warrant closure in selected younger patients.
- → Stroke under roughly 55 earns a wider net — thrombophilia and vasculitis labs, dedicated dissection imaging, drug screen, and pregnancy or postpartum considerations.
The purpose of the ischemic stroke workup is not to fill a checklist — it is to find the mechanism, because mechanism drives secondary prevention. A cardioembolic stroke from atrial fibrillation, a symptomatic carotid stenosis, a small-vessel lacune, and a paradoxical embolus through a patent foramen ovale (PFO) lead to four very different prevention plans. The workup below is organized the way you should think at the bedside: a core panel that every patient gets, then a tiered escalation when that panel comes back clean, and a parallel young-stroke track when the patient is too young for garden-variety atherosclerosis. Institutional protocols and guideline emphasis vary; treat the specific tests and intervals here as the common ground, not as a rigid order set.
The core workup every ischemic stroke gets
Five domains, every patient, regardless of how obvious the cause looks on admission:
- Vessels, head and neck. CT angiography or MR angiography from the arch to the vertex. You are looking for large-vessel occlusion, symptomatic extracranial or intracranial stenosis, dissection, and aneurysm. Do not let a clean intracranial study distract you from a tandem cervical lesion.
- Brain parenchyma. MRI with diffusion-weighted imaging (DWI). DWI confirms infarction, dates it roughly, and — critically — the pattern points at mechanism: multiple lesions in different vascular territories suggest a proximal embolic (often cardiac) source, while a single deep lacune suggests small-vessel disease.
- Heart, structure. Transthoracic echocardiogram (TTE) for thrombus, valvular disease, and wall-motion abnormality. Add an agitated-saline bubble study to look for a right-to-left shunt (PFO or pulmonary AVM). Transesophageal echo is reserved for higher-yield questions such as aortic arch atheroma, suspected endocarditis, or a strongly suspected shunt that TTE cannot resolve.
- Heart, rhythm. Continuous telemetry for 24–72 hours inpatient. The single most consequential finding you can make is occult atrial fibrillation, because it flips prevention from antiplatelet to anticoagulation.
- Blood. CBC, basic metabolic panel, hemoglobin A1c, fasting lipid panel, troponin, and coagulation studies (PT/INR and PTT). These define vascular risk factors, screen for the occasional coagulopathy, and flag concurrent myocardial injury.
Bedside pearl
Read the DWI for distribution, not just for the lesion. Scattered infarcts across multiple territories are an embolic signature and should push your monitoring and echo workup harder, even if the admission rhythm is sinus.
When the core workup is clean: the cryptogenic escalation
Roughly a quarter to a third of ischemic strokes remain cryptogenic after the standard panel. Cryptogenic is a prompt to escalate, not a diagnosis to settle for. The highest-yield next step is prolonged rhythm monitoring, because paroxysmal atrial fibrillation hides easily in a 24–72 hour window. In the CRYSTAL-AF trial, an implantable loop recorder detected atrial fibrillation in about 8.9% of cryptogenic-stroke patients at 6 months versus 1.4% with conventional follow-up — a roughly six-fold increase — and the gap widened with longer monitoring.1 In practice this means an external patch monitor for 14–30 days or an implantable loop recorder for selected patients, chosen by pretest suspicion and how long you are willing to watch.
The other major cryptogenic question is the PFO. A bubble study identifies the shunt; the harder judgment is whether it is the culprit. Closure is most defensible in younger patients with a large shunt or associated atrial septal aneurysm and no competing stroke source. The CLOSE trial found no recurrent stroke in the closure arm versus a meaningful rate on antiplatelet therapy alone in patients with a large shunt or septal aneurysm,3 and REDUCE showed PFO closure plus antiplatelet therapy lowered recurrent ischemic stroke compared with antiplatelet therapy alone, at the cost of more device-related atrial fibrillation.4 The 2021 AHA/ASA secondary prevention guideline frames PFO closure as a shared decision for selected patients, typically younger than 60 with an otherwise cryptogenic embolic-appearing infarct.2
Pitfall
A PFO is common in the general population, so finding one does not prove it caused the stroke. Closing a PFO in an older patient whose real mechanism is undetected atrial fibrillation or aortic atheroma leaves the true source untreated. Exclude competing causes — especially occult AF — before attributing the event to the shunt.
The young-stroke track (roughly under 55)
When a patient is too young for typical atherosclerotic disease, run a parallel investigation in addition to the core panel rather than waiting for it to come back negative:
- Dissection imaging. Cervical artery dissection is a leading cause of stroke in the young. Standard CTA or MRA may miss it; dedicated vessel-wall MRI better demonstrates the intramural hematoma and the eccentric wall changes.
- Thrombophilia / hypercoagulable panel. Reasonable in younger patients, particularly with a suspected paradoxical or venous source, prior thrombosis, or a suggestive family history. Many assays are unreliable in the acute, anticoagulated, or pregnant state, so timing and interpretation matter.
- Vasculitis and inflammatory labs. Inflammatory markers and a targeted autoimmune workup when the history, multifocal imaging, or angiographic appearance raises CNS or systemic vasculitis.
- Toxicology. A urine drug screen — sympathomimetics such as cocaine and amphetamines are an under-recognized stroke trigger in young patients.
- Pregnancy and postpartum. Check pregnancy status and consider peripartum-specific mechanisms, including reversible cerebral vasoconstriction syndrome, eclampsia-related disease, and cerebral venous sinus thrombosis.
Turning the workup into a prevention plan
Each domain maps to an action. A symptomatic carotid stenosis points toward revascularization plus intensive medical therapy. Atrial fibrillation — whether found on admission telemetry or on a loop recorder months later — generally moves the patient to anticoagulation. Small-vessel lacunar disease is managed with antiplatelet therapy and aggressive risk-factor control. A culprit PFO in the right patient may be closed. The 2021 AHA/ASA guideline is the practical reference that ties these mechanism-specific decisions together and is worth keeping at hand when the workup resolves.2 The corollary is that a truly cryptogenic stroke is a reason to keep monitoring, not to default permanently to aspirin and move on — the mechanism you have not found yet may still be findable, and may change the drug.
This page is educational and does not replace individualized clinical judgment; test selection, monitoring duration, and intervention thresholds vary by institution and by guideline version, and should be applied to the specific patient in front of you.
Frequently asked questions.
What is the minimum workup for every ischemic stroke?
Head-and-neck vessel imaging (CTA or MRA), brain MRI with diffusion-weighted imaging, an echocardiogram (TTE with a bubble study to look for a shunt), 24–72 hours of cardiac telemetry, and basic labs including CBC, BMP, hemoglobin A1c, lipids, troponin, and coagulation studies. This core panel is run on every patient regardless of how obvious the cause appears.
What does "cryptogenic stroke" actually mean?
It means no clear mechanism is identified after the standard workup — no culprit large-vessel stenosis, no atrial fibrillation on initial monitoring, no obvious cardiac source. It is a prompt to escalate (prolonged rhythm monitoring, a shunt evaluation, and in younger patients a hypercoagulable and dissection workup), not a final diagnosis.
Why is prolonged cardiac monitoring recommended after a cryptogenic stroke?
Paroxysmal atrial fibrillation is intermittent and easily missed in a 24–72 hour inpatient window. In the CRYSTAL-AF trial an implantable loop recorder detected atrial fibrillation in about 8.9% of cryptogenic-stroke patients by 6 months versus 1.4% with usual follow-up. Finding AF changes prevention from antiplatelet therapy to anticoagulation, which is why extended monitoring with a patch (14–30 days) or a loop recorder is pursued.
Does finding a PFO mean it caused the stroke?
Not necessarily. A patent foramen ovale is common in the general population, so it is often an incidental finding rather than the culprit. Attribution is strongest in a younger patient with a large shunt or atrial septal aneurysm, an embolic-appearing infarct, and no competing source such as occult atrial fibrillation or aortic atheroma. Closure is a shared decision in selected patients, supported by trials such as CLOSE and REDUCE.
When should I add a young-stroke workup?
Roughly under age 55, or whenever the patient is too young or too vascularly healthy for typical atherosclerosis. Add dedicated dissection imaging (vessel-wall MRI), a thrombophilia panel, vasculitis and inflammatory labs, a urine drug screen, and pregnancy or postpartum considerations — run in parallel with the standard panel rather than only after it returns negative.
Is transesophageal echo needed for every stroke?
No. Transthoracic echocardiography with a bubble study is the routine first test. Transesophageal echo is reserved for specific higher-yield questions — aortic arch atheroma, suspected endocarditis, or a strongly suspected right-to-left shunt that the transthoracic study cannot adequately characterize.
References.
- Sanna T, Diener HC, Passman RS, et al. Cryptogenic stroke and underlying atrial fibrillation (CRYSTAL AF). N Engl J Med. 2014;370:2478–2486. PubMed
- Kleindorfer DO, Towfighi A, Chaturvedi S, et al. 2021 Guideline for the Prevention of Stroke in Patients With Stroke and Transient Ischemic Attack: A Guideline From the American Heart Association/American Stroke Association. Stroke. 2021;52:e364–e467. PubMed
- Mas JL, Derumeaux G, Guillon B, et al. Patent foramen ovale closure or anticoagulation vs. antiplatelets after stroke (CLOSE). N Engl J Med. 2017;377:1011–1021. PubMed
- Søndergaard L, Kasner SE, Rhodes JF, et al. Patent foramen ovale closure or antiplatelet therapy for cryptogenic stroke (Gore REDUCE). N Engl J Med. 2017;377:1033–1042. PubMed
More clinical tools
Keep the stroke service moving.
- NIHSS Pocket Guide A bedside guide to the NIH Stroke Scale - all 15 items, the scoring conventions that trip people up, severity bands, serial use, and the posterior-circulation blind spot.
- Modified Rankin Scale (mRS) The modified Rankin Scale explained: all seven grades 0-6, the independence and walking boundaries, baseline-to-90-day anchoring, and the mRS 0-2 trial cut-point.
- ASPECTS ASPECTS explained: the 10-region MCA score on non-contrast CT, the regions most often missed, historical and large-core thrombectomy thresholds, and automated scoring.
- IV Thrombolysis Eligibility IV thrombolysis for acute ischemic stroke: the 4.5-hour and extended windows, tenecteplase and alteplase dosing, key exclusions, and where guidelines diverge.
- All clinical tools The full bedside reference index.
Related reading
From the articles.
- What is a TIA? Why a resolved deficit still needs urgent workup.
- Types of stroke explained The mechanisms the workup is searching for.
Reviewed by