The Incidence of Patent Foramen Ovale in 1,000 Consecutive Patients*

Transcription

1 The Incidence of Patent Foramen Ovale in 1,000 Consecutive Patients* A Contrast Transesophageal Echocardiography Study Daniel C. Fisher, MD; Edward A. Fisher, MD; jacqueline H. Budd, RN; Stacey E. Rosen, MD; and Martin E. Goldman, MD Study objective: Patent foramen ovale (PFO) is present in 10 to 35% of people and has been reported to be an important risk factor for cardioembolic cerebrovascular accidents (CV As) and transient ischemic attacks (TIAs), especially in younger patients. While contrast transthoracic echocardiography has been used to detect PFO, contrast transesophageal echocardiography (TEE) has a greater sensitivity. Prior studies reported the incidence of PFO in patients presenting with a CV A or TIA. Design: To determine the incidence of PFO in a more general population, we reviewed 1,000 consecutive TEEs performed with contrast and color Doppler for the presence of PFO and other cardioembolic risk factors, including atrial septal aneurysm (ASA), aortic plaque, atrial fibrillation (AFib), and atrial thrombi. While imaging with monoplane or biplane TEE, multiple injections of agitated saline solution were injected during cough or Valsalva maneuver to detect flow through a PFO. Patients: There were 482 male and 518 female patients with mean age of 60 ± 17 years (range 11 to 93 years). Results: Patent foramen ovale was found in 9.2% of all patients and, though seen in all age groups divided by decade, the incidence in patients aged 40 to 49 years was greater than those aged 70 to 79 years (12.96% vs 6.15%, p=0.03). Contrast TEE had a much higher detection rate than color Doppler alone. Importantly, there was no greater incidence of PFO in patients with CV A vs those without CV A, or in male vs female patients. Also, there was a very strong correlation between the presence of ASA and PFO (p<.001). Conclusion: Thus, PFO detected by TEE, frequently seen with ASA, is seen in all age groups and does not in itself present a risk factor for CV A. The association of PFO with peripheral thrombosis and CV A needs further study. (CHEST 1995; 107: ) ASA =atrial septal aneurysm; CV A =cerebrovascular accident; DVT=deep vein thrombosis; NS=not significant; PFO=patent foramen ovale; TEE=transesophageal echocardiography; TIA=transient ischemic attack; TTE=transthoracic echocardiograph Key words: patent foramen ovale; stroke; transesophageal echocardiography _N>proximately 30% of all patients with ischemic stroke or transient ischemic attack (TIA) have a potential cardioembolic source. 1 2 Patent foramen ovale (PFO), a remnant of a flap between the septae primum and secundum allowing fetal right atrial blood to pass into the left atrium, has been implicated as a risk factor for cardioembolic events. A recent study utilizing contrast transthoracic echocardiography (TTE) found the incidence of PFO to be 40% in 60 adults younger than 55 years who had ischemic *From the Mount Sinai Medical Center, New York. Manuscript received August 4, 1994; revision accepted November 7. Reprint requests: Dr. Goldman, Mt. Sinai Medical Center, PO Box 1030, One Gustave Levy Place, New York, NY stroke, compared with 10% in an age-matched control group. 3 The autopsy incidence of PFO by probe patency has been reported to be 27.3%, which declines with age (occurring in 34.3% during the first three decades of life and 20.2% in the ninth and tenth decades). 4 Interestingly, the size of the PFO increased with advancing age from a mean of 3.4 mm in the first decade to 5.8 mm in the tenth decade of life. 4 Cumulative data from clinical investigations in more than 1,000 patients (Table 1) has demonstrated 9.3% incidence of PFO by TTE and 11.2% by transesophageal echocardiography (TEE). 5 To determine the incidence of PFO in a general referral population, and to evaluate the potential for PFO as a cardioembolic source, we performed a retrospective review of 1,000 consecutive patients referred for TEE. Clinical Investigations

2 Table 1-PFO Detection by Echocardiography No. of TTE, No. TEE, No. Source, yr patients (%) (%) Zenker et al, (0) 1 (2.5) Lechat et al, (21.3) Pearson et al, (2) 6 (12) DeCoodt et al, (8) 17 (27) Daniel et al, (0.8) 29 (6.1) Ellis et al, (12.5) Konstadt et al, so 13 (26) Lee et al, so 0(0) 4 (8) Teague and Sharma (26) DiTullio et al, (23) Siostrzonek et al, (6) 30(20) Pearson et al, (5) 13 (16.5) Cujec et al, O(O) 2(3.2) Brickner et al, (29.4) DiTullio et al, (18) Barinagarrementeria et al, (57) Total 1, (9.3) ll5 (ll.2) Study Population METHODS 1,442 1,025 We reviewed 1,000 consecutive nonsurgical TEEs that included 518 female subjects and 482 male subjects, with a mean age of 60 ± 17 years (range, ll to 93 years). Patients were referred to the echocardiography laboratory from both hospital and outpatient settings. The most common clinical indications for the procedures included evaluation to rule out possible cardioembolic sources of stroke (including patients with TIA, stroke, peripheral emboli, atrial fibrillation, or other arrhythmias prior to cardioversion, and those with suspected but not proven neurologic events) (595 patients), valvular heart disease (286), congenital heart disease (117), aortic disease (100), and possible endocarditis (40); some patients had more than one referral diagnosis. Cerebrovascular accident (CV A) / TIA is a referral diagnosis made on the basis of clinical presentation and, when available, is confirmed by neurologic imaging such as CT scan or magnetic resonance imaging. Transesophageal Echocardiography Studies were performed with monoplane and multiplane equipment manufactured by several companies (Acuson, Mountainview, Calif; Advanced Technology Laboratories, Bothell, Wash; and Hewlett-Packard, Andover, Mass). The patient's medical history, allergies, and medications were reviewed. Informed consent was obtained, and the patient fasted 4 to 6 h prior to the procedure. Intravenous access was established and nasal oxygen was supplied. The patient's electrocardiogram and blood pressure were monitored routinely. Endocarditis prophylaxis was administered as per the recommendations of the American Heart Association. 6 Following titration of premedication used in virtually all patients, esophageal intubation and a thorough TEE cardiac examination was performed on all patients 7 Contrast TEE The transesophageal transverse four-chamber and the longitudinal vena cavae views were utilized to locate the thinnest portion of the interatrial septum and the junction between the septum primum and secundum to interrogate for the presence of PFO. Right-sided contrast injections using 10 ml of manually agitated saline solution or dextrose solution were rapidly injected into a peripheral vein to generate echogenic microbubbles that normally enter the right atrium and exit the right ventricle into the pulmonary artery and dissipate in the pulmonary circulation. In the presence of a PFO, these microbubbles may traverse the interatrial septum from the right to left atrium. Passage is aided by transient elevation of right-sided pressure produced by a cough or Valsalva maneuver. Multiple contrast injections were performed (in the horizontal and longitudinal planes that maximized imaging of the interatrial septum and a PFO would it have been present), both with and without maneuvers to transiently elevate right-sided pressures. Each TEE was also assessed for the presence of an atrial septal aneurysm (ASA), aortic plaque, atrial and/or ventricular thrombus, spontaneous echo contrast or "smoke," left-sided valvular vegetations, and prosthetic valve thrombosis. Color flow Doppler was performed to investigate flow patterns and regurgitant lesions. Definitions Patent foramen ovale was defined by the presence of at least three microbubbles in the left-sided chambers within three cardiac cycles after opacification of the right atrium after a peripheral contrast injection (in the absence of pulmonary arteriovenous fistulas). Atrial septal aneurysm was defined as at least a 1.1-cm excursion into either atrium of at least a 1.5-cm segment of the atrial septum 8 Significant aortic plaque was defined as any protruding or mobile irregularity in the lumen of the aorta. Spontaneous echo contrast or "smoke" in the left atrium or appendage, which is due to stasis and red cell rouleau formation, is identified echocardiographically as swirling low-intensity echoes. Major echocardiographic risk factors for stroke included left ventricular, atrial, or appendiceal thrombus, left-sided native or prosthetic valvular vegetation or thrombus, ASA, atrial septal defect, ventricular septal defect, and aortic plaque or dissection. 2 Statistical Analysis Chi-square analysis was performed to evaluate the association between PFO and cardioembolic event, as well as other variables including sex, age, atrial fibrillation, ASA, and spontaneous echo contrast. We specifically compared patients with PFO to those without PFO, and patients with CV A vs those without CV A for differences in number, age, and cardioembolic risk factors. Additionally, we analyzed the 92 patients with PFO for differences in cardioembolic risk factors based on the presence of CV A. Since "smoke" is associated with thromboembolic events, it was categorized with thrombus for statistical purposes. RESULTS Patent foramen ovale was detected in 9.2% of all subjects (Table 2), with equal incidence in male and female subjects (9.54 vs 8.88%, respectively; p=not significant [NSD The mean age of patients with PFO was 58.3 years compared with 60.6 years without PFO (p=ns). When the patients were separated by their age by decades, PFO was seen in all groups (Table 3). Because our patients were referred pri- Table 2-Incidence of PFO in 1,000 TEEs* Mean Age, No. yr CVA, No.(%) ASA, No.(%) PFO (42.4) 14 (15.2) No PFO (38.8) 55 (6.1) p *Comparison of patients with and without patent foramen ovale. CHEST /107/6/ JUNE,

3 Table 3-PFO per Decade % of PFOs in Age, No. of No. of %of PFOs in 92 Patients yr Patients PFOs All Patients With PFOs * * * * *p=0.03. marily by adult cardiologists, only 53 of our patients were younger than 30 years, in whom only five PFOs were seen (9.4%). The percentage of PFO detected by patients per decade was greater in patients 40 to 49 years old (14/ 108, 12.96%) than in patients 70 to 79 years old (15/ 244, 6.15%, p=0.03) (Table 3). Both color-flow Doppler and multiple peripheral contrast injections were used to interrogate for PFO in all1,000 patients. Since the presence of a PFO was defined by contrast injections, all 92 PFOs were detected by contrast, while only 22 were seen by color-flow Doppler. Furthermore, all PFOs detected by color Doppler were confirmed by contrast. Fifty PFOs were detected by monoplane, and 42 were detected with biplane echo. Interatrial septal aneurysms were found in 69 patients (6.9%). The incidence of ASA in patients with PFO was 15.2% compared with 6.1% of patients without PFO (p=0.001) (Table 2). The mean age of the 391 patients referred because of a neurologic event (63.9 ± 16.4 years) was comparable to those patients without an event (58.1 ± 17.3 years). There was an equal incidence of stroke in male and female patients. The incidence of PFO in the patients with a neurologic event was 10.0% compared with 8.7% in patients without a neurologic event (p=ns) (Table 4). Of the 88 stroke patients with spontaneous echo contrast, 54 were in atrial fibrillation (61.4%), compared with 112 patients without stroke with spontaneous echo contrast, 77 of whom were in atrial fibrillation (68.8%) (NS). Of the 92 patients with PFO, 39 had a neurologic event (Table 5). When echocardiograms were evaluated for cardioembolic sources other than PFO, there was a similar incidence of patients with 0, 1, 2, or 3 other potential cardioembolic sources in those with vs those without stroke (Tables 5 and 6). DISCUSSION Previous studies have implicated PFO as a possible cardioembolic source for cerebrovascular events To our knowledge, the current retrospective report is 1506 Table 4-Cardioembolic Risk in 1,000 TEEs* Mean Age, PFO, Plaque, ASA, AFIB, Smoke, No. y r No.(%) No.(%) No. (%) No. (%) No.(%) CVA (10.0) (26.6) (8.2) (20.5) (22.5) NoCVA (8.7) (22.8) (6.1) (23.5) (18.4) *Comparison of those patients with and without CV A or TIA. AFIB=atrial fibrillation. the largest study to examine each subject for the presence of PFO by both color Doppler and contrast TEE. Additionally, we evaluated 1,000 consecutive patients referred for TEE, not just those with suspected cardioembolic events. The incidence of PFO in this study ~ a9.2%, s comparable with findings of other studies of normal populations, 3 9 but lower than studies that selected specific patient subgroups. 4 While there was a significant difference between patients 40 to 49 years old and 70 to 79 years old, it is unlikely that the incidence of PFO decreases with increasing age, and this is possibly due to referral bias. We also found no significant difference in the incidence of PFO in patients with a history of stroke compared to those without stroke, regardless of age, a finding contrary to other studies. 3 Mechanism of PFO In the absence of pulmonary hypertension or right ventricular failure, transient elevation of right ventricular pressures (inducible with cough or Valsalva maneuver) is usually required for the passage of nonprecision microbubbles (average diameter, 20 p.m) from the right to left atrium through a PF0. 10 Normally, left atrial pressure is slightly higher than that on the right. However, during the release phase of the Valsalva maneuver or with a cough, central venous return of blood to the right atrium increases. This transient elevation in right atrial pressure leads to a momentary right-to-left pressure gradient and shunting through a PFO, confirmed by contrast microbubbles traversing from the right to left atrium. However, the detection of microbubbles in the left atrium does not definitively prove a cause-and-effect relationship between a neurologic event and a PFO. One must assume that a fortuitous microthrombus embolized from a venous site, crossed through the PFO, traversed the aorta, entered the cerebral circulation, and lodged to obstruct cerebral blood flow producing a transient or permanent neurologic deficit. While PFO has previously been shown to be a pathway for thrombus,'l a left atrial, ventricular, or appendiceal thrombus, native or prosthetic valve thrombus, or vegetation or carotid plaque has a less Clinical Investigations

4 Table 5-Cardioembolic Risk in 92 Patients With PFO* Aortic LA,LAA Plaque, ASA, Smoke, Valvular, Veg, ASD/ VSD, No. No. % No.(%) No.(%) No.(%) No.(%) No.(%) CVA (23) (20.5) (17.9) (5.1) (2.6) Non-CVA (15.1) (11.3) (7.5) (15.1) (3.8) (p=0.005) *LA, LAA Smoke="smoke" or thrombus visualized in left atrium or left atrial appendage; valvular=native or prosthetic valve problems; VEG=vegetation or fibrin strands; ASD/ VSD=atrial/ventricular septal defect. circuitous route to the cerebral vessels, and therefore has a higher likelihood of precipitating a neurologic event. And although intraoperative TEE has recently documented paradoxical emboli through a PFO as a cause for fat embolism syndrome, this occurred in the setting of extensive pulmonary emboli, and only after a previously intact atrial septum was subjected to marked elevations in pressure due to cor pulmonale.l2 We noted a strong association between PFO and ASA. The incidence of ASA by TTE has been reported to be 0.22% in 36,200 patients. 13 Because of its higher resolution, TEE has increased the detection of ASA (8% of 410 patients). 8 Atrial septal aneurysms are thin-walled aneurysms composed of redundant tissue of the septum primum that bulges at least 1.1 em beyond the plane of the atrial septum into either atrium. As the ASA swings into the left atrium, stretching the separation between the septum primum from secundum, right to left shunting of microthrombi through PFO may be facilitated. Pearson et al 8 reported a significantly higher incidence of ASA in patients with stroke (15%) compared with those without 4%). Thrombi found within the aneurysm may also predispose to cerebral emboli.l 3 Comparison With Other Studies Our incidence of PFO detected by TEE (9.2%) was similar to that of 1,492 patients compiled from 16 studies performed from 1988 to 1992 (9.3% by TTE and 11.2% by TEE; Table 1). DiTullio et ajl 4 studied 146 patients referred with an acute ischemic stroke and found an 18% overall incidence of PFO. However, the incidence of PFO in patients with cryptogenic stroke (no documented source,despite a thorough cerebral, vascular, or cardiac workup) was 42%, compared with 7% in patients who had stroke of determined etiology.l 4 However, as Falk 15 mentioned in the accompanying editorial to that article, their study did not evaluate patients for the presence or absence of venous thrombosis, and therefore a causeand-effect relationship between PFO and cryptogenic stroke was not proved. In a recent study, Stallberger et ap 6 attempted to correlate PFO with underlying deep vein thrombosis (DVT) detected by venography. They found a higher prevalence of DVTs as possible sources of cryptogenic stroke.l 6 The current study differed significantly from previous studies assessing the incidence of PFO because all of our subjects referred for TEE were included, not just those referred solely for the evaluation of a source for a cerebrovascular event. Our overall incidence of PFO was 9.2%; 10% in patients with a neurologic event compared with 8.7% in patients without (NS). We did not separate our patients with stroke or TIA into specific etiologies because the elderly population has a higher incidence of cardiac and large-vessel atherosclerosis that makes it difficult to define a cause-and-effect relationship for one of the multiple possible causes. Limitations All of our patients were referred for TEE and therefore were not a true reflection of a "normal" control population, which may have affected our rate of PFO detection. Sedation administered in virtually all patients may have impeded our patients' ability to perform an adequate Valsalva maneuver, often required to show a right-to-left shunt. However, we attempted to have the patient cough several times during each study. In addition, approximately 50% of TEEs were performed with monoplane echocardiography, which may be less sensitive than biplane imaging in visualizing a PFO, although contrast and color-flow imaging should negate this anatomic limitation. We did not quantify the amount of shunting through the PFO or the separation of septum primum from secundum, which may be valuable in deter- Table 6-Cardioembolic Sources Other Than PFO* CVA (39) Non-CV A( 53) 0, No. (%) 1, No. (%) 2, No. (%) 3, No. (%) 21 (53.8) 28 (52.8) 10 (25.6) 18 (33.9) 7 (17.9) 6 (11.3) 1 (2.5) 1 (1.9) *Atrial fibrillation, left ventricular, atrial or appendiceal thrombus, left-sided native or prosthetic valvular vegetation or thrombus, atrial septal aneurysm, atrial septal defect, ventricular septal defect, and aortic plaque or dissection. CHEST /107/6/ JUNE,

5 mining if the PFO is clinically significant. Homma et al 17 found that patients with cryptogenic stroke had significantly more microbubbles seen in the left atrium and larger PFO dimension compared with patients with identifiable cause of stroke (2.1 ± 1.7 mm vs 0.57 ± 0.78 mm, p<0.01). The use of microbubbles is not a precise method for detecting the presence of PFO. A contrast injection in a peripheral vein provides adequate mixing of blood with microbubbles in the right atrium. Following a vigorous cough or Valsalva maneuver, the entire left side of the heart (including atrium, ventricle, and aorta) can be interrogated for single microbubbles crossing through a patent foramen ovale. Color-flow Doppler is a velocity map that is angle dependent and requires appropriate machine settings, ie, low filter, to detect a tiny jet of color traversing a PFO. Thus, complete opacification of the right atrium with adequate maneuvers to transiently elevate right atrial pressure should have a good sensitivity and specificity for PFO detection. Peripheral venous studies to confirm the presence of DVT were not performed to correlate with the presence of a stroke, and carotid Doppler results were not evaluated in all of the patients with PFO. Also, CT scans were not performed to confirm the diagnosis of CV A, and strokes were therefore not categorized in all patients. Patients were categorized by the presence of left atrial or appendiceal "smoke" and/ or thrombus; however, no distinction was made as to how many patients had "smoke" and how many had thrombus. Additional limitations include the inherent difficulty in classifying which echocardiographic findings qualify as an etiology for stroke and which may be "innocent by-standers:" for instance, a patient with atrial fibrillation with no appendiceal thrombus seen, but who had a PFO, spontaneous contrast in the left atrium, and ascending aortic plaque. The presence of echocardiographic risk factors found by TEE defines a group of patients who may be at higher risk for a cardioembolic event in the future. For example, the presence of echogenic smoke and/or thrombus in the left atrium or atrial appendage identifies patients in atrial fibrillation who require anticoagulation therapy to prevent embolization. Other diagnoses such as PFO and other intracardiac shunts and aortic plaque require another variable such as thrombus formation in a peripheral vein or on top of ulcerated plaques to create a potential thrombus for embolization. CONCLUSIONS Our 9.2% incidence of PFO detected in all subjects was similar to findings in normal subjects in other studies, 3 9 but significantly lower than 27.3% inci- dence found in an autopsy study. 4 Hausman et ap 8 demonstrated better detection of PFO with color and contrast TEE, and an increased prevalence of PFO in young patients with otherwise unexplained ischemic stroke. However, to our knowledge, our study of 1,000 patients represents the largest single report of the use of contrast TEE on a general referral population. There was also a significant association between PFO and the presence of an ASA. RECOMMENDATIONS Transesophageal echocardiography with contrast may detect the cause from a cardioembolic source in patients with neurologic events. While PFO has been correlated with paradoxic emboli, one cannot assume the presence of PFO to be the culprit of all cryptogenic strokes. Currently, aspirin may be appropriate therapy for large-vessel atherosclerotic disease and for nonvalvular atrial fibrillation in patients younger than 75 years. 19 However, anticoagulation with warfarin may be the preferred prophylactic treatment for a PFO in the presence of lower extremity DVT or other predisposing conditions to peripheral thrombus formation, ie, hypercoaguable state, extended immobilization of the lower extremities. 20 Therefore, in patients with cryptogenic stroke in whom a PFO is found, further studies are warranted to confirm a causal relationship. REFERENCES 1 DeRook FA, Comess KA, Albers GW, eta!. Transesophageal echocardiography in the evaluation of stroke. Ann Intern Med 1992; 117: Cerebral Embolism Task Force. Cardiogenic brain embolism. Arch Neuroll989; 46: Lechat P, Mas JL, Lascault G, eta!. Prevalence of patent foramen ovale in patients with stroke. N Eng! J Med 1988; 318: Hagen PT, Scholz DG, Edwards WD. Incidence and size of patent foramen ovale during the first ten decades of life: an autopsy study of 965 normal hearts. Mayo Clin Proc 1984; 59: Movsowitz C, Podolsky LA, Meyerowitz CB, et a!. Patent foramen ovale: a nonfunctional embryological remnant or a potential cause of significant pathology? JAm Soc Echocardiogr 1992;5: Committee on Rheumatic Fever, Endocarditis, and Kawasaki Disease. Prevention of bacterial endocarditis: recommendations by the American Heart Association. JAMA 1990; 264: Fisher EA, Stahl JA, Budd JH, eta!. Transesophageal echocardiography: procedures and clinical application. J Am Coli Cardiol1991; 18: Pearson AC, Nagelhout D, Castello R, eta!. Atrial septal aneurysm and stroke: a transesophageal echocardiographic study. J Am Coli Cardioll991; 18: Lynch JJ, Schuchard GH, Gross CM, et a!. Prevalence of right-to-left atrial shunting in a healthy population: detection by Valsalva maneuver contrast echocardiography. Am J Cardiol1984; 53: Clinical Investigations

6 10 Langholz D, Louie EK, Konstadt SN, et al. Transesophageal echocardiographic demonstration of distinct mechanisms for right to left shunting across a patent foramen ovale in the absence of pulmonary hypertension. J Am Col! Cardiol 1991; 18: Tsao LY, Wilbright W A, Heit JA, eta!. Patent foramen ovale causing stroke in an elderly woman. J Stroke Cerebrovasc Dis 1992; 2: Pel! ACH, Hughes D, Keating J, et al. Brief report: fulminating fat embolism syndrome caused by paradoxical embolism through a patent foramen oval e. N Engl J Med 1993; 329: Hanley PC, Tajik AJ, Ynes JK, et al. Diagnosis and classification of atrial septal aneurysm by two-dimensional echocardiography: report of 80 consecutive cases. JAm Col! Cardiol1985; 6: DiTullio M, Sacco RL, Gopal A, et al. Patent foramen ovale as a risk factor for cryptogenic stroke. Ann Intern Med 1992; 117: Falk RH. PFO or UFO?: the role of a patent foramen ovale in cryptogenic stroke. Am Heart J 1991; 121: Stollberger CS, Slany J, Shuster I, eta!. The prevalence of deep vein thrombosis in patients with suspected paradoxical embolism. Ann Intern Med 1992; 119: Homma S, DiTullio MR, Sacco RL, et al. Characteristics of patent foramen ovale associated with cryptogenic stroke. Stroke 1994; 25: Hausman 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: SP AF Investigators. Preliminary report of the Stroke Prevention in Atrial Fibrillation Study. N Eng! J Med 1 990; 322: SP AF Investigators. The Stroke Prevention in Atrial Fibrillation Study: final results. Circulation 1991; 84: Zenker G, Erbel R, Kramer G, et a!. Transesophageal two-dimensional echocardiography in young patients with cerebral ischemic events. Stroke 1988; 19: Pearson A, Gomez C, Ojile M, et a!. Comparative yield of transesophageal and transthoracic echocardiography in patients with stroke or TIA [abstract]. Circulation 1989; 80: DeCoodt P, Kacenelenbogen R, Heuse D, et al. Detection of patent foramen ovale in stroke by transesophageal contrast echocardiography [abstract]. Circulation 1989; 80: Daniel WG, Angermann C, Engberding R, et a!. Transesophageal echocardiography in patients with cerebral ischemic events and arterial embolism-a European multi-center study. Circulation 1989; 80: Ellis JE, Lance J, Feinteiun S, eta!. Right heart dysfunction, pulmonary embolism and paradoxical embolization during liver transplantation. Anesth Analg 1989; 68: Konstadt SN, Louie EK, BlackS, eta!. Intraoperative detection of patent foramen ovale by transesophageal echocardiography. Stroke 1991; 22: Lee RJ, Bartzokis T, Yeoh TK, eta!. Enhanced detection of intracardiac sources of cerebral emboli by transesophageal echocardiography. Stroke 1991; 74: Teague SM, Sharma MK. Detection of paradoxical cerebral echo contrast embolization by transcranial Doppler ultrasound. Stroke 1991; 22: DiTullio MR, Massaro A, Hoffman M, et a!. Transcranial Doppler with contrast injection in stroke patients with patent foramen ovale [abstract]. Circulation 1991; 84: Siostrzonek P, Zangeneh M, Gossinger H, et al. Comparison of transesophageal and transthoracic contrast echocardiography for detection of a patent foramen ovale. Am J Cardiol 1991; 68: Cujec B, Polasek P, Vol! C, eta!. Transesophageal echocardiography in the detection of potential cardiac sources of embolism in stroke patients. Stroke 1991; 22: Brickner ME, Gray burn PA, Fadel B, eta!. Detection of patent foramen ovale by Doppler color flow mapping in patients undergoing cardiac catheterization. Am J Cardiol1991; 68: Barinagarrementeria F, Diaz F, Vargas J, eta!. Prevalence of patent foramen ovale in young patients with stroke: role of color-flow echocardiography. J Stroke Cerebrovasc Dis 1992; 2:7-11 More Study Opportunities in Cardiovascular Disease CHEST {1995} October 29- November 2, 1995 New York, New York CHEST 1995* offers you rhoughr-provoking sessions on currenr issues in chesr medicine, presenred by world-renowned experrs-rhe same qualiry science found each monrh in Chest. If rhis article on cardiovascular disease was beneficial, you will wanr ro rake advantage of the educational opportunities ar CHEST 1995-where cardiovascular disease will be one of four srudy tracks offered. For more information, call ACCP Product and Registration Services: or fonnerly Annual International Scientific Assembly CHEST I 107 I 6 I JUNE,