From Wiki Journal Club
Jump to: navigation, search
Furlan AJ, et al. "Closure or medical therapy for cryptogenic stroke with patent foramen ovale". The New England Journal of Medicine. 2012. 366(11):991-999.
PubMedFull textPDF

Clinical Question

In patients with cryptogenic stroke or TIA with a patent foramen ovale (PFO), does percutaneous PFO closure reduce the risk of a recurrent event?

Bottom Line

Although many patients with cryptogenic stroke/TIA have a PFO, closure of the PFO does not reduce the 30-day rate of recurrent stroke/TIA, all-cause mortality, or death from neurologic causes.

Major Points

Up to 50% of young patients with cryptogenic stroke have PFOs at autopsy. Thus, the evaluation of cryptogenic stroke or TIA often includes echocardiography to identify a right-to-left intracardiac shunt. However, PFOs and other atrial septal abnormalities are discovered incidentally among some 25% of the general population.[1] Based upon the French PFO-ASA and PICSS trials, aspirin is the mainstay of medical therapy for individuals with cryptogenic stroke/TIA and PFO. Surgical closure of the defect, however, had been a subject of controversy, particularly when evidence was sparse and PFO closure carried significant morbidity. With the advent of minimally invasive percutaneous PFO closure, however, investigators considered that the benefits of PFO closure may outweigh its risks among individuals with cryptogenic stroke/TIA.

CLOSURE randomized 909 patients ≤60 years old with cryptogenic stroke to receive either percutaneous PFO closure or medical therapy. The device group received aspirin plus clopidogrel, while the medical therapy group received aspirin, warfarin, or both. At a mean follow-up of 2 years, there was no difference between groups in the primary composite endpoint of stroke/TIA, 30-day mortality, or death from neurologic causes. PFO closure was achieved in 87% of patients randomized to device therapy, which is comparable to other trials of the STARFlex closure device, but inferior to other devices.[2] Vascular complications and atrial fibrillation were more common in the device group.

While the trial was well designed, there may have been problems with the particular PFO closure device selected. For example, the STARFlex device may be more prone to thrombus formation, atrial fibrillation, and residual shunt compared to other PFO closure devices, and therefore have a higher risk of recurrent stroke/TIA.[3] In addition, the makers of the STARFlex device went out of business in 2011, rendering the device unavailable; thus it is difficult, if not impossible, to extrapolate CLOSURE's results to other closure devices.[4]


  • AHA/ASA (2011)[5]
    • For patients with an ischemic stroke or TIA and a PFO, antiplatelet therapy is reasonable (class IIa, level B)
    • There are insufficient data to establish whether anticoagulation is equivalent or superior to aspirin for secondary stroke prevention in patients with PFO (class IIb, level B)
    • There are insufficient data to make a recommendation regarding PFO closure in patients with stroke and PFO (class IIb, level C)


  • Multicenter, open-label, parallel group, randomized, superiority trial
  • N=909
    • Percutaneous PFO closure (n=447)
    • Medical therapy (n=462)
  • Setting: 87 sites in the US and Canada
  • Enrollment: 2003-2008
  • Follow-up: 2 years
  • Analysis: Intention-to-treat
  • Primary endpoint: Composite of stroke or TIA at 2 years, all-cause mortality at 30 days, and death from neurologic causes at 31 days to 2 years


Inclusion Criteria

  • Age 18-60 years
  • Ischemic stroke or TIA in prior 6 months
  • PFO on TTE with bubble study

Exclusion Criteria

  • Stroke etiology clearly not PFO

Baseline Characteristics

Derived from the closure group, which was similar to the medical group.


  • Age: 46.3 years
  • Males: 52.1%
  • White: 89.0%

Medical data:

  • Inclusion event:
    • Cryptogenic stroke: 72.6%
    • TIA: 27.4%
  • Successful PFO repair in closure group: 89.4%
  • TTE findings:
    • Moderate or substantial shunt: 55.9%
    • Atrial septal aneurysm ≥10mm: 37.6%
  • Smoker in previous year: 21.5%
  • MAP: 92 mmHg
  • Medical history:
    • HTN: 33.8%
    • HLD: 47.4%
    • CVD family history: 55.3%
    • CHD: 0.4%
    • Ischemic heart disease: 1.3%
    • MI: 1.6%
    • Valvular dysfunction: 11.0%
    • Arrhythmia: 5.8%
    • Catheterization: 5.1%
    • PTCA: 1.3%
    • PVD: 1.1%
    • Gerbezius-Morgagni-Adams-Stokes syndrome: 0.9%
    • PE: 0%
    • Pericarditis: 0.4%
    • Cardiomyopathy: 0.2%


  • Random assignment with stratification by site and atrial septal aneurysm to one of two groups:
    • Closure group: percutaneous closure of PFO with STARFlex device, followed by clopidogrel 75 mg daily for 6 months and aspirin 81-325 mg daily for 2 years
    • Medical group: warfarin (titrated to INR 2-3), aspirin 325 mg daily, or both


Primary Outcome

Composite of stroke or TIA at 2 years, all-cause mortality at 30 days, and death from neurologic causes at 31 days to 2 years
5.5% vs. 6.8% (HR 0.78; 95% CI 0.45-1.35; P=0.37)

Secondary Outcomes

Major bleeding
2.6% vs. 1.1% (P=0.11)
All-cause mortality
<0.1% vs. 0.1% (P=NS)
2.9% vs. 3.1% (HR 0.90; 95% CI 0.41-1.98; P=0.79)
3.1% vs. 4.1% (HR 0.75; 95% CI 0.36-1.55; P=0.44)

Subgroup Analysis

There was no difference in the subgroups for the primary outcome. Pre-defined subgroups included sex, atrial septal aneurysm, shunt size, type of inclusion event (stroke vs. TIA), or baseline medications.

Adverse Events

Atrial fibrillation
5.7% vs. 0.7% (P<0.001)
Periprocedural major vascular complications
3.2% vs. 0%


  • Trial was underpowered to detect small differences in the event rate (the reduced sample size reduced from competition from off-trial PFO closure and thus slow enrollment)[2]
  • No standardized definition of cryptogenic stroke[4]
  • Many strokes deemed cryptogenic occurred in patients with significant vascular risk factors, and may have not been solely due to paradoxical embolism
  • STARFlex device is no longer available, so difficult to extrapolate CLOSURE's results to other devices[4]
  • There may have been a significant patient self-selection bias, whereby patients with a higher risk of recurrent stroke may opt out of clinical trials, and thus are not represented in this trial[3]
  • 2-year follow up may be inadequate for statistically significant differences in the primary outcome to emerge


NMT Medical

Further Reading

  1. 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 Clin Proc. 1984;59:17–20.
  2. 2.0 2.1 Johnston SC. "Patent Foramen Ovale Closure — Closing the Door Except for Trials." NEJM. 2012; 366:1048-1050
  3. 3.0 3.1 Tobis J, Shenoda M. "Percutaneous Treatment of Patent Foramen Ovale and Atrial Septal Defects." Journal of the American College of Cardiology. 2012. 60(18);1722-1732.
  4. 4.0 4.1 4.2 "Correspondence: Device Closure for Stroke with Patent Foramen Ovale." N Engl J Med 2012. 366:2322-2324
  5. Furie KL, et al. "Guidelines for the prevention of stroke in patients with stroke or transient ischemic attack." Stroke. 2011;42(1):227-76.