PFO and Stroke Review by compaejopo



PFO and Stroke
What Are the Data?
Lawrence R. Wechsler, MD

Abstract: Patent foramen ovale (PFO) is a frequent finding on echocardiography and occurs in over 25% of the population. In young patients with cryptogenic stroke, the frequency is much higher suggesting paradoxical embolization may be responsible for the clinical events. There are conflicting data from studies examining the association between PFO and stroke. The combination of atrial septal aneurysm and PFO, and PFO size and severity of right-to-left shunt may add additional risk but again the data are insufficient for definite conclusions. Available information suggests no difference in subsequent stroke in patients with PFO treated with aspirin or warfarin for secondary prevention. Endovascular closure is technically feasible, but not without the possibility of periprocedural complications. Comparison of medical and endovascular closure of PFOs in patients with stroke is ongoing in 2 major randomized trials. Key Words: PFO, stroke, endovascular closure (Cardiology in Review 2008;16: 53–57)

Antiplatelet therapy, anticoagulation, and closure of the PFO are all options; however, little definitive data exist to guide the choice of treatment.

The foramen ovale forms in the fetus from the incomplete closure of the septum primum and septum secundum. In the fetus this opening acts as a 1-way valve, bypassing the lungs and permitting passage of oxygenated blood to the left side of the heart and the systemic circulation. After birth, the foramen typically closes as right atrial pressures decline and blood passes through the lungs. The foramen ovale usually rapidly fibroses and becomes permanently closed. In a significant proportion of individuals, however, the foramen does not close and remains as a potential passage of embolic material from the venous to the systemic circulation. A PFO is found at autopsy in 27% of individuals, although the incidence decreases with increasing age.1 The prevalence of PFO detected by echocardiography is less certain; however, a population-based study using TEE found a PFO in 25.6% of randomly selected individuals.2 Excessive excursion of the interatrial septum during respiration constitutes an atrial septal aneurysm (ASA). An ASA occurred in 2.2% of the population studied by TEE.2 The majority of patients with ASA also have PFO and the combination of septal abnormalities may be associated with larger right-toleft shunts.3


he relationship between patent foramen ovale (PFO) and stroke is complex and controversial. Paradoxical embolization through a right-to-left shunt was first described over 400 years ago, but until recently was thought to be a relatively uncommon cause of stroke. Contrast echocardiography and transesophageal echocardiography (TEE) revolutionized the diagnosis of PFO, greatly increasing recognition of this abnormality both in patients with and without stroke. Identification of a PFO in a patient with stroke, particularly without another apparent explanation (cryptogenic stroke) raises the issue of paradoxical embolization through the PFO. Evidence linking PFO and stroke is mostly circumstantial and one or more of the classic elements of diagnosis of paradoxical embolization is often lacking. In a specific patient it is difficult to know whether the PFO or some other unseen mechanism was responsible for the stroke. Even assuming the PFO is related to the stroke, optimal treatment is uncertain.

Certain conditions might increase the risk of stroke in women with PFO. An increase in thrombosis during pregnancy and prolonged valsalva during delivery may increase the risk of paradoxical embolism. Hypercoagulable states may also increase the likelihood of thrombosis and stroke associated with PFO. Despite these risks, cryptogenic stroke occurs with similar frequency in men and women.4 Similarly, most studies find no gender difference among stroke patients with and without PFO.5–7 In 1 study of consecutive patients under age 55 with cryptogenic stroke, a greater proportion of patients with atrial septal abnormalities were women compared with those without atrial septal abnormalities (47.8% vs. 38.2%; P 0.03).8 The risk of recurrent stroke was higher in men, but the difference did not reach statistical significance.

From the Department of Neurology, UPMC Stroke Institute, University of Pittsburgh Medical School, Pittsburgh, Pennsylvania. Correspondence: Lawrence R. Wechsler, MD, Department of Neurology, UPMC Stroke Institute, University of Pittsburgh Medical School, 200 Lothrop St., Suite C400, Pittsburgh, PA 15213-2582. E-mail: lwechsler@ Copyright © 2007 by Lippincott Williams & Wilkins ISSN: 1061-5377/08/1601-0053 DOI: 10.1097/CRD.0b013e31815771e4

Cardiology in Review • Volume 16, Number 1, January/February 2008



Cardiology in Review • Volume 16, Number 1, January/February 2008

TABLE 1. Comparison of PFO Frequency in Cryptogenic Stroke vs. Control
Study Lechat et al Webster et al13 Di Tullio et al10 Hausmann et al11 Cabanes et al9

Cryptogenic Stroke (%) 54 50 47 50 56

Control (%) 10 15 4 11 18

P 0.01 0.01 0.01 0.01 0.01

Each element of this scenario may be a relatively unlikely event, making the final result, a clinical stroke, even less likely. The source of thrombus in patients with PFO and suspected paradoxical embolus is uncertain. Deep vein thrombosis is found in only a minority of patients16 and even pelvic magnetic resonance imaging does not demonstrate a source of thrombus in many cases.17 Other possible sources of embolus include thrombus forming in an ASA or within the PFO canal itself.

Before the modern era of echocardiography, reports of paradoxical embolization were rare. Contrast echocardiography greatly increased the identification of PFO and stroke registries began investigating the causes of stroke mostly in hospitalized patients. Patients in whom no source of stroke could be found, or cryptogenic stroke, composed up to 40% of patients in some series.4 Echocardiography became part of the standard evaluation of such patients. In series of patients studied with echocardiography, a PFO was found in 47–56% of young patients with cryptogenic stroke, but in only 4 –18% of controls,9 –13 suggesting the PFO might be a source of paradoxical embolization (Table 1). A meta-analysis of casecontrol studies found that PFO increased the relative risk of stroke by a factor of 1.83 95% confidence interval (CI), 1.25–2.66 . In individuals younger than 55 years old, the relative risk was 3.10 (95% CI, 2.39 – 4.21).14 Not all studies have confirmed this association. In a population-based study in which TEE was performed routinely, PFO was found in 25% of individuals with cryptogenic stroke and in 15% of referral controls. However, in a group of controls randomly selected from the population, PFO occurred in a similar 25%.15 This important finding suggests that referral bias and possibly examiner bias may play a role in the higher frequency of PFO among cryptogenic stroke patients found in previous studies. The examiner may search more diligently for evidence of PFO in patients with cryptogenic stroke due to the lack of any other identified cause. Another population-based study5 also failed to find any association between stroke and PFO based on contrast transthoracic echocardiography. These population studies may be limited in detection of all stroke events and diagnosis of PFO possibly obscuring any association. Despite the weaknesses they suggest further study of this issue is needed. The association of stroke and PFO does not prove causation. Reports of thrombus visualized within the PFO canal confirms that at least in some cases paradoxical embolization does occur. However, whether this mechanism is responsible for stroke in the majority of patients with PFO is uncertain. Several conditions must be met for a paradoxical embolization through a PFO to occur. First, there must be a source of venous thrombus. The thrombus must then embolize and while passing through the heart, it must be subjected to increased right heart pressures allowing the thrombus to pass through the PFO to the systemic circulation. Once in the systemic circulation it must be directed to the brain and then lodge in a sufficiently large artery to cause a clinical stroke.

Several factors have been suggested to increase the risk for stroke or recurrent stroke in patients with PFO. In the meta-analysis by Overell et al younger patients had a greater risk of stroke associated with PFO. The relative risk for stroke in patients under age 55 was 6.00 (95% CI, 3.72–9.68) and 2.26 (95% CI, 0.96 –5.31) for those older than 55.14 This was based on only 3 studies including patients older than 55 and there was considerable heterogeneity in the results across studies. One of the 3 studies did find an association in this older age group. Homma et al6 examined the risk of recurrent stroke associated with PFO in the PFO in Cryptogenic Stroke Study (PICSS), a substudy of the Warfarin Aspirin Recurrent Stroke Study (WARSS). An association between PFO and stroke was found only for individuals greater than 65 years old and there was no increased risk for younger patients. The reason for these disparate results is unclear. Significant differences in study populations, criteria for PFO diagnosis and statistical analysis may all play a role. In all studies, the number of patients in each age group with PFO is small. For example, in the study by Homma et al, only 29 patients in the age 65 and older group had a PFO. Another factor possibly increasing the risk of stroke with PFO is the presence of ASA. Echocardiography-based studies initially identified ASA as a factor increasing the association between PFO and stroke.9,18,19 In a study of 100 consecutive stroke patients under age 55 investigated by TEE, the combination of ASA and PFO was associated with a stroke odds of 33.3 (95% CI, 4.1–270) when compared with controls without stroke.9 In the meta-analysis mentioned previously, ASA alone was associated with both stroke and cryptogenic stroke but the combination of PFO and ASA greatly magnified the risk. In patients with cryptogenic stroke, odds ratio was 3.16 (95% CI, 2.30 – 4.35) for PFO, 3.65 (95% CI, 1.34 –9.97) for ASA, and 23.26 (95% CI, 5.24 –103.2) for PFO/ASA.14 The last result was based on only 2 studies resulting in a very wide CI reflecting the uncertainty of the true odds ratio. The PFO in cryptogenic stroke study, a prospective study of consecutive patients under age 55 with cryptogenic stroke treated with aspirin, reported no increased risk of stroke recurrence in patients with PFO or ASA alone, but a dramatic increase in those with both abnormalities.8 The risk of recurrence after 4 years was 4.2% in patients without either septal abnormality, 2.3% in those with PFO and 15.7% with PFO and ASA (Table 2). A careful look at the numbers, however, reveals several issues that cloud the significance of these findings. First, although the total group included 581 patients, only 27 patients with
© 2007 Lippincott Williams & Wilkins


Cardiology in Review • Volume 16, Number 1, January/February 2008

PFO and Stroke

TABLE 2. Recurrent Stroke on Aspirin in the PFO and Cryptogenic Stroke Study
No. Pts No PFO/ASA PFO PFO ASA 304 216 51 Stroke Rate at 2 Yr 3.7% (1.6–5.8) 1.8% (0.05–3.6) 4.0% (0.0–9.4) Stroke Rate at 4 Yr 4.2% (1.8–6.6) 2.3% (0.3–4.3) 15.2% (1.8–28.6)

TABLE 3. Frequency of Recurrent Stroke on Aspirin or Warfarin in the PICSS Trial
No. Pts PFO PFO 98 152 Warfarin (%) 9.5 8.3 Aspirin (%) 17.9 16.3 RR (95% CI) 0.52 (0.16–1.67) 0.50 (0.19–1.31) P 0.3 0.2

Data from Mas JL, Arquizan C, Lamy C, et al. Recurrent cerebrovascular events associated with patent foramen ovale, atrial septal aneurysm, or both. N Engl J Med. 2001;345:1740 –1746.

Data from Homma S, Sacco RL, Di Tullio MR, et al. Effect of medical treatment in stroke patients with patent foramen ovale: patent foramen ovale in cryptogenic stroke study. Circulation. 2002;105:2625–2631. RR indicates relative ratio.

PFO and ASA were at risk at 4 years. Second, the CI for the recurrent stroke rate at 4 years in the PFO/ASA group was 1.8 –28.6%, overlapping with the stroke rate in the PFO and no septal abnormality groups. Finally, at 2 years there was no increase in risk with a recurrent stroke rate of 3.7% without septal abnormalities and 4.0% in those with PFO/ASA. It is difficult to understand why the PFO/ASA combination would increase stroke risk only between years 2 and 4 after the initial event. The PICSS study also found no increased risk associated with PFO/ASA, although results for cryptogenic stroke patients were not separately reported.7 In the Northern Manhattan Study (NOMAS) of patients examined by contrast transthoracic echocardiography and prospectively followed, PFO/ASA did not increase the risk of subsequent stroke.5 Despite the suggestion from early studies, further evidence is needed to establish an increased potential for stroke in patients with both ASA and PFO. Other potential risk factors include size of PFO, severity of right-to-left shunt, and the presence of right-to-left shunt at rest. The size of the PFO tunnel varies considerably from 1 mm to 15 mm. In many studies, PFO size is estimated from the number of bubble crossing from right-toleft rather than by direct measurement of the PFO size. The size of the right-to-left shunt as measured by transcranial doppler has been associated with stroke risk,20 and patients with PFO and cryptogenic stroke have been found to have larger PFOs than those with identifiable causes of stroke or age-matched controls.7,21 However, in both prospective studies,7,8 no increased risk of recurrent stroke was associated with large PFOs. In 1 study of patients with cryptogenic stroke, a right-to-left shunt at rest was significantly more common in patients with PFO and stroke compared with a control group with PFO and no history of stroke.22 The control group included only 13 patients and treatment of patients with PFO varied according to the discretion of the treating physician. Some studies have linked hypercoagulable abnormalities including prothrombin gene mutation and Factor V Leiden mutation with PFO-associated stroke.23–25

Treatment options for PFO in patients with stroke include antiplatelet therapy, anticoagulants, endovascular closure, or surgical closure. Endovascular treatment has largely replaced surgical closure due to the morbidity associated with the surgical approach. Because of the presumption of venous thromboembolism even in the absence of an identifiable source of throm© 2007 Lippincott Williams & Wilkins

bus, many clinicians use anticoagulants for prevention of stroke recurrence. Little data are available to compare the efficacy of the 2 treatment modalities. The Lausanne study26 prospectively followed patients with PFO and stroke treated with anticoagulants, antiplatelet agents, or surgical closure but no randomization was performed and treatment was at the discretion of the treating physician. There was no effect of treatment on stroke recurrence and the overall stroke recurrence rate was low (1.9% per year). In the PICSS study,7 patients were randomized to adjusted dose warfarin (INR, 1.4 –2.8) or 325 mg/day of aspirin but the randomization occurred in the WARSS trial from which the PICSS cohort was selected. No difference in stroke recurrence between warfarin- and aspirin-treated patients was observed either for all stroke patients with PFO or those with cryptogenic stroke and PFO. In this cohort there was no increased stroke risk associated with either PFO or PFO/ASA (Table 3). It should be noted that the average age of these patients was 59 and there was a high proportion of typical risk factors for atherosclerotic disease. Event rates were also higher in the PICSS study. No data are provided regarding the relative efficacy of warfarin and aspirin in the small subset of patients with both PFO and ASA. Major bleeding was not significantly different in the warfarin- and aspirin-treated patients, but in both studies minor bleeding was increased in patients treated with warfarin. Overall, there are little data to support the use of anticoagulation in patients with PFO and stroke, although those with demonstrated deep vein thrombosis, pulmonary embolism, and possibly those with hypercoagulable states, should receive anticoagulation. Compared with surgical treatment, transcatheter closure of PFO does not require general anesthesia and allows rapid treatment usually resulting in discharge within 24 hours. Reported results of endovascular treatment are limited to case series, which show a high rate of technical success. Despite device closure, recurrent stroke occurs in 0 –5% of patients and major procedural complications are seen in 1.5% with minor complications in 7.9% of treated patients.27 Recent series report periprocedural complication rates of 2– 6%28 –30 (Table 4). Whether the outcomes from device closure compare favorably with medical therapy is unclear. Windecker et al30 reported the results of a retrospective review of 308 patients with cryptogenic stroke and PFO treated with either medical therapy or endovascular closure. Treatment was not randomized and was based on preference of the patient and treating physician. Periprocedural complications occurred in 6% of patients and 17% had residual



Cardiology in Review • Volume 16, Number 1, January/February 2008

TABLE 4. Procedural Complications From Endovascular PFO Closure
Windecker30 Procedures Device embolization (%) Cardiac tamponade (%) Retroperitoneal hematoma (%) Air embolus (%) Access site problems (%) Transient ST elevations (%) AV fistula (%) 150 4 (2.7) — — 3 (2) 2 (1.3) — — Windecker29 78 3 (3.9) 1 (1.3) 1 (1.3) — — 2 (2.6) — Braun28 307 1 (0.3) — — — — 5 (1.6) 1 (0.3)

1. Hagen PT, Scholz DG, Edwards WD. Incidence and size of patent foramen ovale during the first 10 decades of life: an autopsy study of 965 normal hearts. Mayo Clin Proc. 1984;59:17–20. 2. Meissner I, Whisnant JP, Khandheria BK, et al. Prevalence of potential risk factors for stroke assessed by transesophageal echocardiography and carotid ultrasonography: the SPARC study. Stroke prevention: assessment of risk in a community. Mayo Clin Proc. 1999;74:862– 869. 3. Fox ER, Picard MH, Chow CM, et al. Interatrial septal mobility predicts larger shunts across patent foramen ovales: an analysis with transmitral doppler scanning. Am Heart J. 2003;145:730 –736. 4. Sacco RL, Ellenberg JH, Mohr JP, et al. Infarcts of undetermined cause: the NINCDS Stroke Data Bank. Ann Neurol. 1989;25:382–390. 5. Di Tullio MR, Sacco RL, Sciacca RR, et al. Patent foramen ovale and the risk of ischemic stroke in a multiethnic population. J Am Coll Cardiol. 2007;49:797– 802. 6. Homma S, DiTullio MR, Sacco RL, et al. Age as a determinant of adverse events in medically treated cryptogenic stroke patients with patent foramen ovale. Stroke. 2004;35:2145–2149. 7. Homma S, Sacco RL, Di Tullio MR, et al. Effect of medical treatment in stroke patients with patent foramen ovale: patent foramen ovale in cryptogenic stroke study. Circulation. 2002;105:2625–2631. 8. Mas JL, Arquizan C, Lamy C, et al. Recurrent cerebrovascular events associated with patent foramen ovale, atrial septal aneurysm, or both. N Engl J Med. 2001;345:1740 –1746. 9. Cabanes L, Mas JL, Cohen A, et al. Atrial septal aneurysm and patent foramen ovale as risk factors for cryptogenic stroke in patients less than 55 years of age. A study using transesophageal echocardiography. Stroke. 1993;24:1865–1873. 10. Di Tullio M, Sacco RL, Gopal A, et al. Patent foramen ovale as a risk factor for cryptogenic stroke. Ann Intern Med. 1992;117:461– 465. 11. Hausmann D, Mugge A, Becht I, et al. Diagnosis of patent foramen ovale by transesophageal echocardiography and association with cerebral and peripheral embolic events. Am J Cardiol. 1992;70:668 – 672. 12. Lechat P, Mas JL, Lascault G, et al. Prevalence of patent foramen ovale in patients with stroke. N Engl J Med. 1988;318:1148 –1152. 13. Webster MW, Chancellor AM, Smith HJ, et al. Patent foramen ovale in young stroke patients. Lancet. 1988;2:11–12. 14. Overell JR, Bone I, Lees KR. Interatrial septal abnormalities and stroke: a meta-analysis of case-control studies. Neurology. 2000;55:1172–1179. 15. Petty GW, Khandheria BK, Meissner I, et al. Population-based study of the relationship between patent foramen ovale and cerebrovascular ischemic events. Mayo Clin Proc. 2006;81:602– 608. 16. Stollberger C, Slany J, Schuster I, et al. The prevalence of deep venous thrombosis in patients with suspected paradoxical embolism. Ann Intern Med. 1993;119:461– 465. 17. Cramer SC, Rordorf G, Maki JH, et al. Increased pelvic vein thrombi in cryptogenic stroke: results of the paradoxical emboli from large veins in ischemic stroke (pelvis) study. Stroke. 2004;35:46 –50. 18. Lamy C, Giannesini C, Zuber M, et al. Clinical and imaging findings in cryptogenic stroke patients with and without patent foramen ovale: the PFO-ASA Study. Atrial septal aneurysm. Stroke. 2002;33:706 –711. 19. Mas JL, Zuber M. Recurrent cerebrovascular events in patients with patent foramen ovale, atrial septal aneurysm, or both and cryptogenic stroke or transient ischemic attack. French study group on patent foramen ovale and atrial septal aneurysm. Am Heart J. 1995;130:1083–1088. 20. Anzola GP, Zavarize P, Morandi E, et al. Transcranial Doppler and risk of recurrence in patients with stroke and patent foramen ovale. Eur J Neurol. 2003;10:129 –135. 21. Schuchlenz HW, Weihs W, Horner S, et al. The association between the diameter of a patent foramen ovale and the risk of embolic cerebrovascular events. Am J Med. 2000;109:456 – 462. 22. De Castro S, Cartoni D, Fiorelli M, et al. Patent foramen ovale and its embolic implications. Am J Cardiol. 2000;86:51G–52G. 23. Karttunen V, Hiltunen L, Rasi V, et al. Factor V Leiden and prothrombin gene mutation may predispose to paradoxical embolism in subjects with patent foramen ovale. Blood Coagul Fibrinolysis. 2003;14:261–268. 24. Lichy C, Reuner KH, Buggle F, et al. Prothrombin g20210a mutation, but not factor V Leiden, is a risk factor in patients with persistent foramen ovale and otherwise unexplained cerebral ischemia. Cerebrovasc Dis. 2003;16:83– 87.

shunts at follow-up. During follow-up of 4 years, the combined event rate for stroke, transient ischemic attack (TIA), and death showed a nonsignificant reduction in the endovascular group with rates of 8.5% in the endovascular group and 22.2% in the medical treatment group (P 0.05; 95% CI, 0.23–1.01). Event rates for TIA and stroke alone were 7.8% and 22.2% respectively (P 0.08; 95% CI, 0.23–1.11). Most events were TIAs and only 1 major stroke occurred in either group. A significant reduction in stroke and TIA was observed for the subgroup with more than one event at baseline and those with complete PFO closure. Although this is the most detailed information to date comparing endovascular and medical therapy, the retrospective design, single center report, lack of standardized assessment, and nonrandomized treatments limit the reliability of the comparison. No randomized trials of endovascular closure and medical therapy have yet been reported. However, 2 major randomized trials are currently underway. In the CLOSURE trial, over 650 patients under age 60 with cryptogenic stroke and PFO have already been randomized. Regardless of the results, this study and the results of the other major randomized trial, the RESPECT trial, should provide much needed information regarding treatment effects, risk factors for stroke and event rates. Hopefully cardiologists and neurologists will now join together to rapidly bring these trials to completion to define the optimal treatment for this condition.

Paradoxical embolization clearly occurs, but the frequency of this mechanism of stroke in patients with PFO remains unclear. Many patients with PFO and no other identifiable etiology for stroke do not clearly meet criteria for paradoxical embolization in that a source of venous thromboembolism is not present. Population-based studies have failed to confirm the association of PFO and stroke in young patients with cryptogenic stroke observed in case-control studies. Factors such as the presence of ASA, large PFO, or right-to-left shunt at rest may predict a higher stroke risk but studies to date include too few patients or events to reach any definitive conclusions. Limited information suggests no advantage of warfarin over aspirin in preventing recurrent stroke. Whether endovascular closure is superior to medical therapy awaits the results of randomized clinical trials, which are nearing completion.


© 2007 Lippincott Williams & Wilkins

Cardiology in Review • Volume 16, Number 1, January/February 2008

PFO and Stroke

25. Pezzini A, Del Zotto E, Magoni M, et al. Inherited thrombophilic disorders in young adults with ischemic stroke and patent foramen ovale. Stroke. 2003;34:28 –33. 26. Bogousslavsky J, Garazi S, Jeanrenaud X, et al. Stroke recurrence in patients with patent foramen ovale: the Lausanne Study. Lausanne Stroke with Paradoxal Embolism Study Group. Neurology. 1996;46: 1301–1305. 27. Khairy P, O’Donnell CP, Landzberg MJ. Transcatheter closure versus medical therapy of patent foramen ovale and presumed paradoxical thromboemboli: a systematic review. Ann Intern Med. 2003;139:753–760.

28. Braun M, Gliech V, Boscheri A, et al. Transcatheter closure of patent foramen ovale (PFO) in patients with paradoxical embolism. Periprocedural safety and mid-term follow-up results of three different device occluder systems. Eur Heart J. 2004;25:424 – 430. 29. Windecker S, Wahl A, Chatterjee T, et al. Percutaneous closure of patent foramen ovale in patients with paradoxical embolism: long-term risk of recurrent thromboembolic events. Circulation. 2000;101:893– 898. 30. Windecker S, Wahl A, Nedeltchev K, et al. Comparison of medical treatment with percutaneous closure of patent foramen ovale in patients with cryptogenic stroke. J Am Coll Cardiol. 2004;44:750 –758.

© 2007 Lippincott Williams & Wilkins


To top