Background: Patent foramen ovale (PFO) is present in 40% of patients with cryptogenic stroke and may be associated with paradoxical emboli to the brain. Therapeutic options include antiplatelet agents, anticoagulation, percutaneous device and surgical closure. We assessed the hypothesis that there are differences in rates of recurrent TIA or stroke between patients in the four treatment groups.
Methods: Patients presenting from January 1997 with cryptogenic stroke or TIA and PFO were followed prospectively until June 2003. Treatment choice was made on an individual case basis. The primary outcome was recurrent stroke. The secondary outcome was a composite of stroke, TIA, and vascular death.
Results: Baseline. Our cohort consisted of 121 patients; 64 (53%) were men. Median age was 43 years. Sixty-nine percent presented with stroke and 31% with TIA. One or more vascular risk factor was present in 40%. Atrial septal aneurysm (ASA) was present in 24%. Treatment consisted of antiplatelet agents (34%), anticoagulation (17%), device (39%) and surgical closure (11%). Follow-up. Recurrent events occurred in 16 patients (9 antiplatelet, 3 anticoagulation, 4 device closure); 7 were strokes, 9 were TIA. Comparing individual treatments there was a trend toward more strokes in the antiplatelet arm (p = 0.072); a significant difference was seen for the composite endpoint (p = 0.012). Comparing closure versus combined medical therapy groups, a significant difference was seen for primary (p = 0.014) and secondary (p = 0.008) outcomes, favoring closure. Age and pre-study event predicted outcome.
Conclusion: Patent foramen ovale closure was associated with fewer recurrent events. Complications of surgical and device closure were self-limited.