Closing a PFO After Unexplained Stroke: What the Trials Support

For a person under 60 whose ischemic stroke has no other explanation after a careful workup, three randomized trials support closing a patent foramen ovale (PFO) to lower the risk of another stroke. RESPECT, CLOSE, and REDUCE each found fewer recurrent strokes with closure than with medication alone. The benefit is real but modest in absolute terms, and it applies to a specific, carefully selected group. Tools like the RoPE score help estimate whether a PFO was likely the culprit, which is central to deciding whether closure makes sense. This article is educational and is not medical advice.

What a PFO is and why it matters after stroke

A patent foramen ovale is a small flap-like opening between the heart's two upper chambers that fails to seal after birth. It is common, present in roughly a quarter of adults, and in most people it never causes a problem. The concern arises when a clot from the venous side of the circulation slips through that opening into the arterial side and travels to the brain, a process called paradoxical embolism.

The difficulty is that a PFO is common enough that finding one in a stroke patient does not prove it caused the stroke. Many strokes in younger adults are labeled cryptogenic, meaning the standard evaluation, including imaging, heart-rhythm monitoring, and vascular studies, turns up no clear cause. When such a workup is negative and a PFO is present, the PFO becomes a plausible but unproven explanation. That uncertainty is the reason patient selection matters so much.

What the three trials actually showed

The RESPECT trial, published in the New England Journal of Medicine in 2013, randomized 980 patients aged 18 to 60 with a cryptogenic stroke and a PFO to device closure or medical therapy. Its result is often misread. In the primary intention-to-treat analysis, closure did not reach statistical significance, with a hazard ratio of 0.49 and a confidence interval that crossed 1 (95% CI 0.22 to 1.11, P=0.08). Prespecified per-protocol and as-treated analyses did favor closure. Extended follow-up reported in 2017, out to a median of roughly six years, showed a significant reduction in recurrent ischemic stroke with closure. So the honest reading is that the original headline endpoint was negative, and confidence grew with longer follow-up and secondary analyses.

The CLOSE and REDUCE trials, both published in 2017, were designed to test closure more cleanly. CLOSE enrolled patients with what it defined as high-risk anatomy, meaning an associated atrial septal aneurysm or a large right-to-left shunt. No patient in its closure group had a recurrent stroke during follow-up, while strokes did occur in the antiplatelet-only group, a striking separation. REDUCE randomized patients to closure plus antiplatelet therapy versus antiplatelet therapy alone and found recurrent ischemic stroke in 1.4 percent of the closure group versus 5.4 percent of the medication group. Taken together, the three trials moved the field from equipoise to a reasonably consistent signal that closure prevents recurrent stroke in the right patients.

The RoPE score and the selection problem

Because a PFO can be an innocent bystander, clinicians use the Risk of Paradoxical Embolism (RoPE) score to estimate the probability that a given PFO actually caused the stroke. The score runs from 0 to 10 and rewards features such as younger age, absence of vascular risk factors like hypertension and diabetes, and a cortical stroke on imaging. A low score, in the range of 0 to 3, suggests the PFO is probably incidental, while a high score of 9 or 10 corresponds to an attributable fraction near 90 percent. In practical terms, a higher score points to a PFO more likely to be pathogenic, and analyses have linked higher scores to greater relative benefit from closure.

The 2022 guideline from the Society for Cardiovascular Angiography and Interventions (SCAI) folds this thinking into its recommendations. It notes that a RoPE score of 7 or higher may identify patients likely to gain more from closure. The score is a guide, not a verdict, and it works best alongside a neurologist's judgment that the stroke was genuinely cryptogenic.

What the evidence does not cover

The trials were built on a narrow population, and their conclusions do not stretch cleanly beyond it. Nearly all randomized data come from adults 18 to 60 years old. The SCAI guideline makes a strong recommendation for closure in that age band while offering only a conditional, low-certainty suggestion for patients 60 and older, precisely because the trial evidence thins out there. Patients whose strokes had another identifiable cause, who had prior atrial fibrillation, or whose evaluation was incomplete were not the people these trials enrolled.

Closure is also not risk-free. Procedural complications, device-related issues, and a well-documented increase in atrial fibrillation, often transient and occurring around the time of the procedure, are part of the trade-off. The absolute benefit is meaningful but not large. Because the yearly recurrence risk on medication is low to begin with, many patients must be treated to prevent one additional stroke over several years. That arithmetic is why the SCAI guideline frames the choice around shared decision-making: a patient who weighs procedural risk heavily and values the uncertain incremental benefit less may reasonably decline closure.

The practical takeaway

The strongest case for PFO closure is a younger adult, under 60, with a genuinely unexplained stroke, a high RoPE score, and often high-risk anatomy, evaluated jointly by neurology and cardiology. For that person, RESPECT, CLOSE, and REDUCE support closure as a way to reduce recurrent stroke. Outside that profile, the evidence is weaker, the balance shifts, and the decision belongs to a careful conversation rather than a reflex. Reading these trials well means holding both truths at once: closure helps the right patient, and defining the right patient is the hard part.