Patent Foramen Ovale in Patients With Cerebral Infarction

OBJECTIVES: To determine the frequency of patent foramen ovale (PFO) among various subtypes of cerebral infarction. To determine whether any historical or clinical characteristics predict the presence or absence of PFO in these patients. DESIGN: Comorbidity and infarct subtype study. SETTING: Referral-based study. PATIENTS: One hundred sixteen patients with cerebral infarction consecutively referred for transesophageal echocardiography during a 6-month period. MAIN OUTCOME MEASURES: Infarct subtype classification was made using a clinical and radiographic diagnostic rubric similar to that used by the Stroke Data Bank of the National Institute of Neurological Diseases and Stroke. The frequency of various risk factors and clinical characteristics in patients with and in those without PFO and the frequency of PFO in patients with various infarct subtypes were compared (chi 2 or Fisher exact tests). RESULTS: Patent foramen ovale was detected in 37 patients (32%). Mean age was similar in those with (60 years) and those without (64 years) PFO. Patent foramen ovale was more frequent among men (39%) than women (20%, P = .03). Patients with PFO had a lower frequency of atrial fibrillation, diabetes mellitus, hypertension, and peripheral vascular disease compared with those without PFO. There was no difference in frequency of the following characteristics in patients with PFO compared with those without PFO: pulmonary embolus, chronic obstructive pulmonary disease, pulmonary hypertension, peripheral embolism, prior cerebral infarction, nosocomial cerebral infarction, Valsalva maneuver at the same time of cerebral infarction, recent surgery, or hemorrhagic transformation of cerebral infarction. Patent foramen ovale was found in 22 (40%) of 55 patients with infarcts of uncertain cause and in 15 (25%) of 61 with infarcts of known cause (cardioembolic, 21%; large vessel atherostenosis, 25%; lacune, 40%) (P = .08). When the analysis was restricted to patients who underwent Valsalva maneuver, PFO with right to left or bidirectional shunt was found in 19 (50%) of 38 patients with infarcts of uncertain cause and in 6 (20%) of 30 with infarcts of known cause (P = .01). CONCLUSION: Although PFO was overrepresented in patients with infarcts of uncertain in our and other studies, it has a high frequency among patients with cerebral infarction of all types. The relation between PFO and stroke requires further study.