Reviewed by Dr Edgar Tay, interventional cardiologist and structural heart specialist at Asian Heart & Vascular Centre. Last reviewed September 2026.
Reading time: about 8 minutes.
TL;DR
About one stroke in four has no identified cause after a full investigation, and is then called cryptogenic. In younger patients a patent foramen ovale (PFO), a small unsealed flap between the upper chambers of the heart, is one of the explanations worth looking for, because it allows a clot from the venous circulation to bypass the lungs and reach the brain. Finding a PFO does not by itself explain the stroke, since roughly a quarter of adults have one [1]. The work is deciding whether it is the cause or a coincidence, and international guidance sets out how [2, 3].
A stroke that does not fit
A stroke usually comes with an explanation. A narrowed carotid artery, an irregular rhythm that threw a clot, long-standing high blood pressure, diabetes, high cholesterol, smoking. The investigation looks for those, finds one, and treatment follows from what was found.
Sometimes it finds nothing. The arteries are clear on imaging. The rhythm monitoring is normal. The cholesterol is unremarkable and the blood pressure has always been fine. The patient is often in their thirties or forties, fit, with no family history that helps.
That is a cryptogenic stroke, meaning a stroke of unknown origin. It is an unsatisfying label, and it is also a specific one. It does not mean the investigation was incomplete. It means a thorough investigation was done and the cause was not identified.
All of these are ischaemic strokes, caused by an artery being blocked rather than by bleeding. Where the blockage is a clot that has travelled from somewhere else, the term used is thromboembolic. Where that clot formed inside the heart, most often because of an irregular rhythm, it is described as cardioembolic, and that mechanism is set out separately on the cardioembolic stroke page. A cryptogenic stroke is what is left when none of them has been demonstrated.
For patients this is often harder than a clear diagnosis. There is nothing obvious to correct, and no clear answer to the question of whether it will happen again. That question is exactly why the search continues past the standard workup.
What has to be excluded first
Before a stroke can reasonably be called cryptogenic, several things need ruling out, and the order matters.
- Large-artery disease. Imaging of the arteries in the neck and head, looking for narrowing or a dissection.
- Small-vessel disease. The pattern on brain imaging often distinguishes this.
- Atrial fibrillation. This is the most important one, and the most easily missed. An irregular rhythm that comes and goes will not appear on a single ECG. Detection needs prolonged monitoring, sometimes over weeks, occasionally with an implanted monitor. Atrial fibrillation is a commoner cause of stroke than a PFO and is treated completely differently, with anticoagulation rather than a procedure.
- Clotting disorders. Blood tests for inherited and acquired thrombophilia.
- Less common causes. Vasculitis, infection, and in some settings a cardiac source other than a PFO.
A PFO assessed before atrial fibrillation has been excluded is a premature conclusion. This is the single most common sequencing error, and it produces the wrong treatment. A patient closed for a PFO who actually has intermittent atrial fibrillation remains at risk, because the real mechanism was never addressed.
How a PFO can cause a stroke
The mechanism is called paradoxical embolism, and it is worth understanding because it explains why the circumstances of the stroke matter as much as the anatomy.
Blood returning from the body carries small clots from time to time. Normally this is harmless. Venous blood passes through the right side of the heart into the lungs, where the fine pulmonary capillaries filter out anything of a size that could cause trouble. Filtered blood then returns to the left side and is pumped to the brain and the rest of the body.
A PFO offers a short cut across that filter. Under normal conditions the flap sits closed, held shut because pressure in the left atrium exceeds the right. When right-sided pressure briefly rises above left-sided pressure, the flap can open. Straining, coughing, heavy lifting, bearing down, and the ascent phase of a dive all do this.
If a clot is in the venous circulation at the moment the flap opens, it can cross to the left side, bypass the lungs entirely, and travel to the brain. That is a paradoxical embolism: a clot of venous origin causing an arterial event.
Three consequences follow.
The source clot usually forms in the legs or pelvis. Which is why a cryptogenic stroke investigation sometimes includes looking for deep vein thrombosis, even without leg symptoms.
Circumstances carry evidence. A stroke that began during or shortly after a long flight, heavy straining or a dive fits the mechanism better than one that began at rest.
Anatomy modifies the likelihood. A larger opening, one that shunts blood without any straining at all, or one accompanied by a mobile atrial septal aneurysm, makes the mechanism more plausible.
Coincidence or cause: how the judgement is made
Because roughly one adult in four has a PFO [1], and because most of them will never have a stroke, the presence of a PFO in a stroke patient is not evidence that it caused the stroke. In a considerable number of patients the PFO is a bystander.
Clinicians therefore assess causal plausibility rather than mere presence. One tool used is the RoPE score, a ten-point system based on age and vascular risk factors, in which a higher score indicates a greater probability that the PFO is causally related rather than incidental [2].
The logic is worth spelling out because it is counter-intuitive. A young patient with no vascular risk factors scores highly, and that high score means the PFO is more likely to be the cause. Not because youth is dangerous, but because in a 35-year-old with clear arteries, normal blood pressure and no diabetes, there is very little else that could plausibly have caused a stroke. The PFO is the remaining explanation. Conversely a 68-year-old with hypertension, diabetes and a smoking history scores low, because in that patient the stroke has several plausible causes and the PFO is probably incidental.
The anatomy of the shunt and the clinical circumstances are then weighed alongside the score.
What treatment follows
Where the PFO is judged to be the likely cause, closure may be offered. Three randomised trials, CLOSE, RESPECT and REDUCE, compared closure plus medical therapy against medical therapy alone in patients with cryptogenic stroke and a PFO, and were published together in 2017 [4, 5]. In selected patients, meaning younger patients with anatomically significant shunts and no alternative explanation, closure reduced recurrent stroke more than medical therapy alone. Five-year outcomes have since been reported [6]. Current European and American guidance sets out the selection process, and recommends that the decision be reached jointly by cardiology and neurology rather than by either specialty alone [2, 3, 7].
The procedural detail is set out on the PFO device closure page, and the wider question of whether any given hole in the heart needs treating is covered in our guide to whether a PFO needs closing.
Where the PFO is judged incidental, treatment is medical. Antiplatelet therapy is standard, with anticoagulation in specific circumstances. This is a legitimate treatment rather than a lesser option, and for a patient with a low RoPE score it is the correct one.
Where atrial fibrillation is found after all, the plan changes entirely to anticoagulation, and the PFO becomes irrelevant to the decision.
Transient ischaemic attack, migraine and the other associations
A transient ischaemic attack (TIA) produces stroke-like symptoms that resolve completely, usually within an hour and by definition within twenty-four hours, without permanent damage on imaging. It is caused by the same mechanisms as a stroke and is investigated the same way. It is not a lesser event. It is a warning that carries a genuine short-term risk of a completed stroke, and it warrants urgent assessment rather than a wait-and-see approach. A TIA with no identified cause raises the same PFO question as a cryptogenic stroke.
Migraine with aura is associated with PFO more often than chance would predict, and patients frequently ask whether closing a PFO will help the headaches. On current evidence it does not reliably do so. The randomised trials, including PREMIUM, did not meet their primary endpoints, although some secondary measures such as monthly headache days improved [8], and a later pooled analysis examined the combined trial data [9]. Migraine alone is not an indication for closure. Where a patient has both migraine with aura and a cryptogenic stroke, the stroke drives the decision.
Platypnoea orthodeoxia syndrome is an uncommon situation in which blood oxygen falls when a patient sits or stands upright and improves on lying flat, caused by shunting across a PFO in certain anatomical configurations. It is rare, it is treatable, and it is one of the non-stroke situations in which closure is considered [7].
Cryptogenic stroke in the Singapore context
Stroke is a leading cause of adult disability in Singapore, and it is not confined to older patients. Cases in adults under fifty are seen regularly, and it is in that group that a cryptogenic classification is most likely, because the conventional risk factors have not had time to accumulate.
Three things about the local picture are worth knowing.
The investigation available here is thorough, which is why the label means something. Brain and vessel imaging, extended rhythm monitoring, echocardiography with bubble contrast, transcranial Doppler and thrombophilia screening are all accessible. When a stroke is called cryptogenic after that sequence, it reflects a genuinely unidentified cause rather than a limited workup.
Regional travel is part of the clinical history and is easy to overlook. Long-haul and regional flights are routine from Singapore, and prolonged immobility is a recognised trigger for venous clot formation. A stroke that began during or shortly after travel is worth flagging without being prompted, because it supports the paradoxical embolism mechanism in a way that no test result can.
Diving is a genuine consideration rather than an unusual one. Recreational diving is common across the region and accessible from Singapore, and the pressure changes of ascent are among the situations that open a PFO flap. A history of diving, and particularly any previous episode of decompression sickness, belongs in the assessment.
For patients who travel to Singapore for assessment from Indonesia, the Philippines or elsewhere in the region, the practical question is continuity. Rhythm monitoring runs over weeks, and any treatment decision carries a follow-up schedule and a period on medication afterwards. Both need to work in the country where you actually live, and that is a conversation for the first consultation rather than the last.
What happens after the assessment
What to expect
Expect it to take time. The rate-limiting step is usually prolonged rhythm monitoring, which cannot be rushed without undermining its purpose. Expect more than one specialty to be involved. And expect a discussion rather than a single recommendation, because for many patients both closure and medical therapy are defensible and the choice depends on how the individual weighs a procedure against long-term medication.
Expect, too, the possibility that the answer stays uncertain. Some strokes remain unexplained even after a PFO has been assessed and judged incidental. The plan is then medical, secondary prevention is optimised, and that is a considered outcome rather than a failure of investigation.
What to bring, and what to ask
Bring the discharge summary, the brain imaging report, the echocardiogram report, and any rhythm monitoring results including the duration of monitoring. The duration matters, because a 24-hour monitor and a 30-day monitor are not equivalent for detecting intermittent atrial fibrillation.
Questions worth asking:
- Has atrial fibrillation been excluded, and over how long a monitoring period?
- Is my PFO the likely cause of this stroke, or an incidental finding?
- What is my RoPE score, and what does it imply?
- If we treat this medically instead, what does that mean long term?
- Who else is involved in this decision?
Who does this at Asian Heart & Vascular Centre
Cryptogenic stroke assessment spans two sub-specialties. Structural heart intervention, led by Dr Edgar Tay, covers PFO assessment and transcatheter closure. Rhythm assessment, which is where atrial fibrillation is confirmed or excluded, sits with the electrophysiology team. Where the cause is genuinely unclear at the outset both are involved, which is the arrangement current guidance recommends [2].
Diagnostic work including bubble contrast studies and transcranial Doppler is performed in-house.
If you have had a stroke or a transient ischaemic attack and been told the cause was not found, that answer is not necessarily final. A consultation at Asian Heart & Vascular Centre can establish what has already been excluded, what has not, and whether a patent foramen ovale is a plausible explanation in your case.
References
- Hagen PT, Scholz DG, Edwards WD. Incidence and size of patent foramen ovale during the first 10 decades of life: an autopsy study of 965 normal hearts. Mayo Clinic Proceedings. 1984;59(1):17-20.
- Pristipino C, Sievert H, D’Ascenzo F, et al. European position paper on the management of patients with patent foramen ovale. General approach and left circulation thromboembolism. European Heart Journal. 2019;40(38):3182-3195.
- Caso V, Turc G, Abdul-Rahim AH, et al. European Stroke Organisation (ESO) guidelines on the diagnosis and management of patent foramen ovale (PFO) after stroke. European Stroke Journal. 2024. doi:10.1177/23969873241247978
- Søndergaard L, Kasner SE, Rhodes JF, et al. Patent foramen ovale closure or antiplatelet therapy for cryptogenic stroke (REDUCE). New England Journal of Medicine. 2017;377:1033-1042. doi:10.1056/NEJMoa1707404
- Saver JL, Carroll JD, Thaler DE, et al. Long-term outcomes of patent foramen ovale closure or medical therapy after stroke (RESPECT). New England Journal of Medicine. 2017;377:1022-1032. doi:10.1056/NEJMoa1610057
- Kasner SE, Rhodes JF, Andersen G, et al. Five-year outcomes of PFO closure or antiplatelet therapy for cryptogenic stroke. New England Journal of Medicine. 2021. doi:10.1056/NEJMc2033779
- Pristipino C, Germonpré P, Toni D, et al. European position paper on the management of patients with patent foramen ovale. Part II: decompression sickness, migraine, arterial deoxygenation syndromes and select high-risk clinical conditions. EuroIntervention. 2021.
- Tobis JM, Charles A, Silberstein SD, et al. Percutaneous closure of patent foramen ovale in patients with migraine: the PREMIUM trial. Journal of the American College of Cardiology. 2017;70(22):2766-2774. doi:10.1016/j.jacc.2017.09.1105
- Mojadidi MK, Kumar P, Mahmoud AN, et al. Pooled analysis of PFO occluder device trials in patients with PFO and migraine. Journal of the American College of Cardiology. 2021;77(6):667-676. doi:10.1016/j.jacc.2020.11.068
Frequently Asked Questions
What is a cryptogenic stroke?
A stroke with no identified cause after a full investigation. It does not mean the investigation was incomplete. It means large-artery disease, small-vessel disease, atrial fibrillation and clotting disorders were all looked for and none explained the event.
Can a hole in the heart cause a stroke?
It can, through a mechanism called paradoxical embolism. A clot in the venous circulation normally gets filtered by the lungs. If the PFO flap opens, during straining, coughing, lifting or the ascent from a dive, a clot can cross to the arterial side and reach the brain.
Why do young people have strokes?
Because the usual explanations often do not apply. With clear arteries, normal blood pressure and no diabetes, there is little conventional risk to find. That is why a cryptogenic classification is most likely in younger patients, and why a PFO becomes a plausible explanation in that group.
Does finding a PFO mean it caused my stroke?
No. Roughly a quarter of adults have a PFO and most never have a stroke, so its presence is not evidence of cause. In a considerable number of patients it is a bystander. Clinicians assess causal plausibility rather than mere presence.
What is the RoPE score?
A ten-point score based on age and vascular risk factors. A higher score indicates a greater probability that the PFO is causally related to the stroke rather than incidental. A young patient with no risk factors scores highly; an older patient with several scores low.
What tests are done after an unexplained stroke?
Imaging of the arteries in the neck and head, brain imaging, prolonged rhythm monitoring to look for atrial fibrillation, and blood tests for clotting disorders. Echocardiography with bubble contrast assesses for a PFO. Monitoring duration matters, since intermittent atrial fibrillation is missed by a single ECG.
What is the difference between a TIA and a stroke?
A transient ischaemic attack produces stroke-like symptoms that resolve completely, usually within an hour and by definition within twenty-four hours, with no permanent damage on imaging. It is not a lesser event. It carries a real short-term risk of a completed stroke and needs urgent assessment.
Does closing a PFO prevent another stroke?
In selected patients, yes. Three randomised trials published in 2017, CLOSE, RESPECT and REDUCE, found that closure reduced recurrent stroke more than medical therapy alone in younger patients with anatomically significant shunts and no alternative explanation. Selection is what makes the difference.
Can a PFO cause a TIA?
Yes. A transient ischaemic attack is caused by the same mechanisms as a stroke and is investigated the same way, including the search for a patent foramen ovale. A TIA with no identified cause raises the same PFO question as a cryptogenic stroke, and warrants the same assessment.
Which specialist should I see after an unexplained stroke?
Assessment spans two sub-specialties. Structural heart intervention covers PFO assessment and closure, and electrophysiology confirms or excludes atrial fibrillation as the true cause. Where the cause is unclear at the outset both teams are involved, which is the arrangement current guidance recommends for these decisions.