Four years after three major studies found insufficient evidence to conclude that patent foramen ovale (PFO) closure lowers the risk of stroke better than medical therapy, three new analyses are concluding that it does.

The newest studies “provide compelling evidence that PFO closure does reduce the risk of recurrent stroke in carefully selected patients where the PFO was the most likely cause of the first event,” Dr. Steven Messe at the University of Pennsylvania told Reuters Health by email. He had previously opposed PFO closure outside a clinical trial.

Still, although the new evidence shows the benefit of the procedure, he and other researchers interviewed by Reuters Health cautioned against overuse of closure in cases where the source of the stroke is not immediately clear.

About 25 percent of the population has a PFO, “so in most patients with a PFO and a stroke, the PFO was incidental and closing it won’t help,” said Dr. David Spence of Western University in Ontario, coauthor of one of the new studies, known as Gore REDUCE.

“Patient selection will be paramount to avoid recommending PFO closure in patients where the PFO was an ‘innocent bystander,'” said Dr. Messe, who was peripherally involved in Gore REDUCE. “Compared to almost any other stroke etiology, PFO appears to be low risk for recurrence, and thus if a competing cause is identified, it is probably the proximate cause, rather than the PFO.”

The three previous studies that showed no significant benefit involved 2,798 patients. They were published in the New England Journal of Medicine, and were known as CLOSURE I (2012), PC (2013) and RESPECT (2013).

But when those results were pooled, closure did show a significant benefit, Dr. Spence told Reuters Health in an email. The individual tests from 2012-2013 “were just under-powered to show benefit individually,” and too many patients were getting the PFO device who weren’t likely to benefit from them.

The three new studies, involving 2,307 participants, were released online September 13, again by the NEJM.

One of the new tests is a follow-up to the RESPECT study, and those researchers found it took time for the benefits of closure to show up. At the 2- to 4-year mark, the benefit was not statistically significant. But it was significant at a median follow-up of 5.9 years, reducing the ischemic stroke rate by 45 percent (P=0.046) compared to standard medical therapy.

“The longer-term follow-up, a median of nearly six years, indicates durable benefit from treatment,” chief author Dr. Jeffrey Saver of the University of California, Los Angeles, told Reuters Health by email.