What are the indications for percutaneous patent foramen ovale (PFO) closure?

Indications for Patent Foramen Ovale (PFO) Closure

PFO closure should be considered in carefully selected patients aged 18-60 years with cryptogenic stroke who have high-risk PFO features (large shunt or atrial septal aneurysm) and a RoPE score >8, after excluding other stroke etiologies including atrial fibrillation, carotid disease, and left atrial thrombus. 1, 2

Primary Indication: Cryptogenic Stroke in Young Patients

The strongest evidence supports PFO closure for secondary stroke prevention in patients ≤60 years old with cryptogenic embolic stroke and high-risk PFO characteristics. 3, 2

Patient Selection Criteria

You must confirm ALL of the following before considering closure:

• Age 18-60 years (trials excluded patients >60, creating an evidence gap in older populations) 1, 4
• Confirmed cryptogenic stroke after comprehensive workup excluding:
  • Atrial fibrillation (requires minimum 30 days of cardiac monitoring) 4, 2
  • Significant carotid stenosis 4
  • Left atrial thrombus or appendage clot 5
  • Severe thoracic aortic atherosclerosis 1
  • Lacunar infarcts (small deep infarcts suggest small vessel disease, not paradoxical embolism) 1
• Cortical infarct pattern on imaging consistent with embolic mechanism 4, 2
• No indication for long-term anticoagulation (closure benefit is negated if anticoagulation required) 1, 4

High-Risk PFO Features That Favor Closure

The PASCAL classification system identifies patients most likely to benefit: 2

• Large shunt (>25 microbubbles on contrast echocardiography) 1
• Atrial septal aneurysm (increases stroke risk dramatically: OR 15.59 in patients ≤55 years) 5, 1
• RoPE score >8 (scores of 7,8, and 9-10 correspond to 72%, 84%, and 88% probability that PFO caused the stroke) 5, 6
• Absence of vascular risk factors (younger patients without hypertension, diabetes, hyperlipidemia) 2

Expected Outcomes

In PASCAL "probable" patients (young, high-risk PFO features, no vascular risk factors), closure reduces recurrent stroke by 90% at 2 years (absolute risk reduction 2.1%, NNT=48). 2

Pooled trial data shows: 3, 7

• Recurrent stroke/TIA: 0.47% per year with closure vs 1.09% with medical therapy (HR 0.41)
• NNT to prevent one stroke: 37 patients 7
• All-cause mortality reduction: OR 0.49 3

Absolute Contraindications to Closure

Do NOT close PFO in patients with: 1, 4

• Age >60 years (insufficient trial data, competing stroke etiologies more likely) 1, 4
• Requirement for long-term anticoagulation (for any indication) 1, 4
• Lacunar stroke pattern (suggests small vessel disease, not paradoxical embolism) 1
• Low-risk PFO (small shunt, no atrial septal aneurysm, low RoPE score <6) 1, 2
• Identified alternative stroke mechanism (atrial fibrillation, carotid disease, etc.) 4

Relative Contraindications and Uncertain Indications

Migraine with Aura

There is insufficient evidence to recommend PFO closure for migraine with aura. 5, 1 Despite the association between PFO and migraine, closure should not be performed for this indication alone.

Decompression Sickness and High-Risk Occupations

Consider closure only for multiple recurrent events in high-volume divers, compressed-air tunnel workers, high-altitude aviators, or astronauts who must continue their high-risk occupation. 5, 1 This should occur only in centers maintaining closure registries or participating in trials.

Peripheral (Non-Cerebral) Paradoxical Embolism

There are no evidence-based recommendations for PFO closure in peripheral embolism (myocardial infarction, renal infarction, limb ischemia), as the causal connection lacks supporting data. 5, 1

Right-Sided Cardiac Disease with Elevated Pressures

In patients with right-sided cardiac disease causing elevated right atrial pressures and right-to-left shunting, closure may be considered to control cyanosis, but this requires highly individualized assessment. 5, 1 The primary treatment should target the underlying right-sided disease.

PFO with Deep Vein Thrombosis

When both PFO and documented DVT/PE are present, anticoagulation is the primary treatment. 5 PFO closure might be considered depending on recurrent DVT risk, but anticoagulation takes precedence. 5

Procedural Risks to Discuss

Patients must be counseled about device-related complications: 1, 2

• Atrial fibrillation: 4.6-6.6% (NNH=49, meaning for every 37 strokes prevented, you cause AF in 49 patients) 1, 7
• Serious device-related adverse events: 1.4-5.9% 1
• Late complications: Pericardial effusion, device erosion, thrombus formation on device 1
• Procedural success rate: 98.9% 1

Medical Therapy Alternative

For patients who do not meet closure criteria or decline the procedure, single antiplatelet therapy (aspirin 81-325 mg daily or clopidogrel 75 mg daily) is the guideline-concordant approach. 5, 4

• Antiplatelet therapy is preferred over anticoagulation in the absence of another indication for anticoagulation 5
• Dual antiplatelet therapy is not recommended for PFO-associated stroke 4

Critical Pitfalls to Avoid

• Do not assume PFO is causal without excluding other mechanisms, especially in older patients with vascular risk factors 4
• Do not close small PFOs without high-risk features (uncertain benefit, definite procedural risk) 1, 4
• Do not pursue closure if patient requires anticoagulation for any reason 1, 4
• Do not close PFO in "unselected" patients with TIA or cryptogenic stroke (closure is not beneficial without proper risk stratification) 5
• Ensure adequate atrial fibrillation screening (minimum 30 days monitoring) before attributing stroke to PFO 4, 2

Guideline Strength and Evolution

Current American guidelines carry only Class IIb recommendation (insufficient data for definitive recommendations), reflecting uncertainty despite positive trial results. 5, 1 However, Canadian guidelines have upgraded to Level A for carefully-selected patients. 1 The Netherlands Society of Cardiology specifically states closure should be considered only in patients with RoPE score >8 and at least one clinical risk factor. 5

The as-treated analysis (excluding protocol violations) showed stronger benefit than intention-to-treat analysis (3.6% vs 5.8% recurrent events, OR 0.62), suggesting closure is effective when actually performed in appropriate patients. 5, 1