CLOSURE I: No Overall Benefit, No Reduction in Stroke or TIA With PFO Closure

Shelley Wood

November 15, 2010

November 15, 2010 (Chicago, Illinois) — Full final results from the first randomized controlled trial of patent foramen ovale (PFO) closure for stroke/transient ischemic attack (TIA) to reach completion have found no differences in the primary end point of stroke or TIA at two years, all-cause mortality at 30 days, and neurological mortality between 31 days and two years.

Dr Anthony Furlan

Dr Anthony Furlan (University Hospitals Case Medical Center, Cleveland, OH) presented the results for CLOSURE I here on day two of the American Heart Association 2010 Scientific Sessions.

The detailed results, fleshing out top-line results released early by the company, should, in the words of trial discussant Dr Pierre Amarenco (Bichat Hospital, Paris, France), have a "major impact" on the management of "the million of patients with stroke of unknown cause who also have a PFO."

Dr Pierre Amarenco

Furlan said it's "too soon to say" what impact the results will have on PFO-closure procedures: several trials are still ongoing, and while the devices are investigational in the US, they are market-approved in Europe and elsewhere.

"Ultimately I think CLOSURE I, if the other ongoing trials show similar results, will obviously ratchet down the number of endovascular PFO closures [performed]," he said.

"The challenge now, in my view, for the endovascular community is to refine the selection criteria and not so liberally be closing these holes in cryptogenic-stroke patients."

Closure Details

CLOSURE I enrolled 909 patients randomized equally to PFO closure using the STARFlex closure device (NMT Medical) as well as six months of aspirin and clopidogrel (and an additional 18 months of aspirin) or to best medical therapy--aspirin or warfarin or a combination.

According to Furlan, the study employed three independent core labs and an independent clinical events committee to review all neurological and study end points, to make up for the lack of blinding in the randomization. One critique of NMT Medical's other randomized controlled trial of PFO closure (this one, for migraine), MIST, was that it did not employ an independent core lab; MIST, however, used a sham procedure in the control arm in an effort to blind patients to treatment allocation.

At two years, the composite primary end point, as well as rates of stroke or TIA alone, were no different between groups. An analysis of outcomes according to baseline characteristics, including shunt size or presence/absence of atrial shunts, also found no differences between groups.

Both major vascular complications and atrial fibrillation, mostly periprocedural, were significantly more common in the intervention group, but other safety end points were no different between study arms. Procedural and technical success rates (no or trace residual leaking) were high, with 86.7% percent of PFOs closed at one year.

Efficacy and Safety

End point Device (%) Medical therapy (%) p
Composite end point 5.9 7.7 0.30
Stroke 3.1 3.4 0.77
TIA 4.6 4.6 0.39
Major vascular complications 3.2 0.0 <0.001
Atrial fibrillation 5.7 0.7 <0.001


Investigators also conducted an analysis that looked at the etiologies of recurrent strokes, making an "important" finding, Furlan said.

"When we looked at the causes of recurrent events, strokes or TIA, we could find an explanation [for the event] that had nothing to do with paradoxical embolism in about 80% of the patients," he said. "The important point is that many of these cryptogenic strokes [were] not due to paradoxical embolism."

Indeed, both Furlan and Amarenco made the point that one of the reasons the PFO closure/stroke theory grew so popular was that when young patients experienced stroke of unknown cause, a PFO seemed like a plausible etiology, and clinicians simply stopped looking for another explanation.

Cryptogenic may mean, well, how hard did you look?

"The American Stroke Association has not even defined what workup must be done in every cryptogenic stroke patient," Furlan said. "For example, not every patient has their CV 30-day events monitored. So cryptogenic may mean, well, how hard did you look? We looked very hard within this trial, but in routine practice, the looking tends to stop when you find the hole."

And as Amarenco put it: "It is now clear that the selection of patients for PFO closure should be refined, and we cannot now consider closing PFO in all patients who have stroke of unknown cause. The term cryptogenic strokes is probably unnecessarily cryptic to many physicians and patients . . . too vague."

"Positive" criteria--instead of a lack of criteria--should be used to guide treatments, he added, such as infarction in both hemispheres, in anterior and posterior circulation, or brain infarction and silent renal or splenic infarction. "These positive criteria can increase the likelihood of finding the source of the embolism--we should go this way in future trials, if there are new trials done."

Several trials are still ongoing, the CLOSE and PC trials in France and Switzerland, respectively, plus St Jude Medical's/ AGA's (acquisition deal still in process) RESPECT trial and Gore's REDUCE trial. Gore has issued a media statement reaffirming its commitment to completing the trial. Both AGA and St Jude told heartwire that they cannot comment on the status of the trial until the ink is dry on their agreement.

As for whether there are any patients in whom PFO closure may still be justified, Furlan and Amarenco left the door open.

"On a case-by-case basis, after careful evaluation by a vascular neurologist and cardiologist, we may select some patients with a potential causal relationship between PFO and stroke embolism. However, these cases are likely to be very rare," Amarenco said.

And Furlan pointed out that the original humanitarian device exemption (HDE) in the US, scrapped in 2006, may have had it right in the first place.

"The old HDE . . . which required [patients to have had] recurrent events and failed warfarin and medical therapy, that's a group that deserves further study."


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.