The Association between the Diameter of a Patent Foramen Ovale and the Risk of Embolic Cerebrovascular Events Herwig W. Schuchlenz, MD, Wolfgang Weihs, MD, Susanne Horner, MD, Franz Quehenberger, PhD PURPOSE: We sought to determine whether the size of a patent foramen ovale affected the risk of embolic cerebrovascu- lar events of unknown origin. PATIENTS AND METHODS: We ascertained the presence and measured the size of patent foramen ovale using multiplane transesophageal echocardiography in 121 consecutive patients younger than 60 years who had transient ischemic attacks or ischemic strokes and in 123 control subjects. None of the pa- tients had left heart, aortic, or carotid sources of embolism, or echocardiographic signs of elevated left or right atrial pressure. We used multivariate logistic regression to determine whether the size of the patent foramen ovale was an independent risk factor for cerebrovascular events. RESULTS: The mean ( SD) diameter of a patent foramen ovale was significantly larger in patients (4  2 mm) than in control subjects (2  1 mm, P 0.0001). A patent foramen ovale greater than 4 mm was associated with an increased risk of transient ischemic attacks [odds ratio (OR) = 3.4; 95% confi- dence interval (CI), 1.0 to 11, P = 0.04], ischemic strokes (OR = 12; 95% CI, 3.3 to 44, P = 0.0001), and, especially, having evi- dence of two or more strokes (OR = 27; 95% CI, 4.7 to 160, P = 0.0002). CONCLUSION: The diameter of a patent foramen ovale is an independent risk factor for ischemic events, especially recurrent strokes. Am J Med. 2000;109:456 – 462. 2000 by Excerpta Medica, Inc. I n approximately 40% of patients, the source of a ce- rebral embolism remains unknown (1,2). Using transthoracic contrast echocardiography, Lechat et al (3) and Webster et al (4) showed that a patent foramen ovale was common in patients with strokes of undeter- mined origin, presumably because the patent foramen ovale allows small peripheral venous emboli to bypass the natural pulmonary capillary filter (paradoxical embo- lism). Other studies have suggested that a patent foramen ovale is an incidental finding in patients with cryptogenic strokes (5,6). This suggestion is consistent with autopsy data showing that a patent foramen ovale is present in 20% to 34% of the normal population (7). Transesophageal echocardiography is better at detect- ing a patent foramen ovale than is transthoracic echocar- diography (8 –10), and a biplane technique is better than a monoplane technique (11). A multiplane transesopha- geal probe allows visualization of the entire area of the fossa ovalis and measurement of the diameter of a patent foramen ovale. We therefore performed a case-control study to test the hypothesis that the extent of cerebral damage caused by paradoxical embolism was related to the diameter of the patent foramen ovale. A secondary hypothesis was that the diameter of a patent foramen ovale was an indepen- dent risk factor for embolic stroke, after adjusting for traditional atherosclerosis risk factors (12) and for atrial septal aneurysm (8). MATERIAL AND METHODS Selection of Cases Between June 1995 and January 1998 we studied patients with unexplained cerebrovascular events who were re- ferred for transesophageal echocardiography. We mea- sured the patent foramen ovale diameter and atrial septal abnormalities in 150 consecutive patients younger than 60 years who had a recent transient ischemic attack, isch- emic stroke, or recurrent cerebral stroke. The diagnosis of a cerebrovascular embolic event was based on the sudden onset of neurologic deficits and the corresponding mor- phologic findings on computed tomography (CT) or magnetic resonance (MR) imaging scans. All patients un- derwent Doppler and Duplex sonography of the carotid and vertebral arteries, transcranial Doppler sonography, and laboratory studies including measurement of anti- cardiolipid antibodies, protein C and protein S levels, re- sistance to activated protein C, and antithrombin III lev- els. Patients whose neurologic workup yielded no identifi- able cause for the cerebral embolic event were included in the study as cases. We excluded those with carotid steno- sis greater than 30%, ulcerated carotid plaques, flow- From the 2. Medizinische Abteilung (HWS, WW), LKH-Universita ¨- tsklinikum Graz, Universita ¨tsklinik fu ¨ r Neurologie (SH), LKH-Univer- sita ¨tsklinikum Graz, Institut fu ¨ r Medizinische Informatik, Statistik und Dokumentation (FQ), Karl-Franzens-Universita ¨t Graz, Graz, Austria. Requests for reprints should be addressed to Herwig Walter Schuchlenz, MD, 2. Medizinische Abteilung, LKH-Universita ¨tsklini- kum Graz, Auenbruggerplatz 15, A-8036 Graz, Austria. Manuscript submitted March 1, 2000, and accepted in revised form June 16, 2000. 456 2000 by Excerpta Medica, Inc. 0002-9343/00/$–see front matter All rights reserved. PII S0002-9343(00)00530-1