乌鸦传媒

Using a device for patent foramen ovale (PFO) closure did not reduce recurrent events in patients with cryptogenic ischemic stroke, although there was a hint of benefit, a meta-analysis of three recent trials showed.

In the pooled, intention-to-treat analysis, the rate of nonfatal ischemic cerebrovascular events and transient ischemic attack (TIA) was 3.7% with PFO closure and 5.3% with medical therapy, a nonsignificant difference (odds ratio 0.70, 95% CI 0.47-1.05), Andres Pineda, MD, of Mount Sinai Heart Institute in Miami Beach, and colleagues reported online in .

They pointed to an analysis of patients who actually received the assigned treatment, however, to suggest that PFO closure could be effective. In the as-treated analysis, the rate of recurrent events was 3.6% with PFO closure and 5.8% with medical therapy (OR 0.62, 95% CI 0.41-0.94).

"Despite all of the ... limitations of this analysis and the controversy about the topic, [these] pooled data revealed a potential benefit of the PFO closure device versus medical management alone," Pineda and colleagues wrote.

"Furthermore," they wrote, "this scientific information may be more reasonable to be followed when deciding about when, how, or on whom to intervene instead of waiting for the regulatory or health insurance agencies to decide, by arbitrary or perhaps economic parameters, when this intervention should take place, as has been suggested after the RESPECT and PC trials were published."

The presence of , which account for 30% to 40% of ischemic strokes.

Observational studies have in reducing recurrent events after a cryptogenic stroke, but the technique has not been effective in clinical trials. Thus, the effectiveness of PFO closure remains a topic of debate.

Pineda and colleagues performed a meta-analysis of three recent trials of PFO closure:

• CLOSURE I trial using the STARFlex septal closure system
• RESPECT trial using the Amplatzer PFO occluder
• PC trial using the Amplatzer PFO occluder

All three trials failed to establish the superiority of PFO closure over medical therapy.

The pooled analysis included 2,303 patients (mean age 45.7, 52.7% male).

Average follow-up among the three trials ranged from 2 to 4 years. Medical therapy varied among the trials and included any of the following: warfarin, aspirin, clopidogrel, general thienopyridine, and extended-release dipyridamole.

The outcomes considered for the primary endpoint of each trial also varied. For the current analysis, the primary endpoint was a composite of nonfatal ischemic cardiovascular events and TIA.

In the main intention-to-treat analysis, there were no differences between the PFO closure and medical therapy groups on rates of the primary endpoint, either of its components, or bleeding.

There was a greater number of atrial fibrillation cases with PFO closure (32 versus eight) but the difference was not significant (OR 3.29, 95% CI 0.86-12.60). Most of those cases came from the CLOSURE I trial and "therefore our result may have been skewed by the safety profile of the STARFlex device and not of the PFO closure in general," the authors wrote.

They acknowledged that the meta-analysis was limited by the small number of trials included, variations between the studies in inclusion criteria, endpoints, and use of antithrombotic agents and anticoagulation, as well as the inability to extrapolate the results to PFO closure devices that were not included in the three trials.

From the American Heart Association:

Comment