The optimal management of patients with cryptogenic stroke (CS) and patent foramen ovale (PFO) remains controversial. We conducted a meta-analysis to assess the effect of PFO closure for secondary prevention of stroke on patients with CS. We searched the literature for randomized control trials assessing the recurrence of stroke after PFO closure compared with medical therapy (antiplatelet and/or anticoagulation). Five randomized control trials with a total of 3,440 patients were included. The mean age was 45.2 ± 9.7 years and follow-up duration ranged from 2.0 to 5.9 years. PFO closure significantly reduced the risk of stroke compared with the medical therapy (2.8% vs 5.8%; relative risk [RR] 0.48, confidence interval [CI] 0.27 to 0.87, p = 0.01, I2 = 56%). The number needed to treat for stroke prevention was 10.5. PFO closure was associated with an increased risk of atrial fibrillation compared with medical therapy (4.2% vs 0.7%; RR 4.55, CI 2.16 to 9.6, p = 0.0001, I2 = 25%). There was no significant difference in all-cause mortality (RR 1.33, CI 0.56 to 3.16, p = 0.52, I2 = 0%), as well as no difference in bleeding risk between the 2 groups (RR 0.94, CI 0.49 to 1.83, p = 0.86, I2 = 29%). In conclusion, our meta-analysis demonstrates that PFO closure is associated with significantly lower risk of recurrent stroke in patients with PFO and CS compared with medical therapy. However, atrial fibrillation was more common among closure patients.
This meta-analysis updates the estimate of risks and benefits for PFO closure following cryptogenic stroke (embolic stroke of undetermined source--ESUS) with recent trial data. PFO closure reduced the risk for stroke recurrence, but at the cost of substantially increasing the rate for subsequent atrial fibrillation. There was no effect on mortality. PFO closure following ESUS remains a difficult decision that requires careful assessment of risks and benefits.
As a hematologist, these are peripherally relevant as we occasionally get asked to comment on stroke reduction, although I suspect this is more relevant to those in cardiology, neurology, and general internal medicine. This study helps provide a stronger quality of evidence on a controversial topic.