Clinical and Imaging Findings in Cryptogenic Stroke Patients With and Without Patent Foramen Ovale
The PFO-ASA Study
Background and Purpose— Patent foramen ovale (PFO) has been identified as a potential risk factor for stroke, but the mechanisms of PFO-associated stroke remain unsettled. The aim of our study was to evaluate possible differences in stroke risk factors and stroke patterns between patients with and without PFO that may give clues to the mechanism of PFO-associated stroke.
Methods— This prospective, multicentric study involved 581 young cryptogenic stroke patients. The presence of PFO and atrial septal aneurysm was assessed by transesophageal echocardiography and reviewed independently by 2 experienced sonographers. Clinical, brain, and vascular imaging findings were reviewed by 2 neurologists and 2 neuroradiologists.
Results— Of the 581 stroke patients, 267 (45.9%) had PFO. Patients with PFO were younger (OR, 0.95; 95% CI, 0.93 to 0.97) and less likely to have traditional risk factors such as hypertension (OR, 0.49; 95% CI, 0.28 to 0.85), hypercholesterolemia (OR, 0.56; 95% CI, 0.34 to 0.93), or current smoking (OR, 0.67; 95% CI, 0.47 to 0.97). Features suggestive of paradoxical embolism, such as Valsalva-provoking activities or deep vein thrombosis, were not more frequent in patients with PFO. Migraine was more common in patients with PFO (27.3%) than in those without PFO (14.0%). PFO (OR, 1.75; 95% CI, 1.08 to 2.82), particularly when associated with atrial septal aneurysm (OR, 2.71; 95% CI, 1.36 to 5.41), was significantly associated with migraine after adjustment for age and sex.
Conclusions— Differences in stroke risk factors and stroke patterns suggest that different stroke mechanisms occur in patients with and without PFO. PFO is significantly and independently associated with migraine, and this association is even stronger in patients with PFO and atrial septal aneurysm.
Patent foramen ovale (PFO) and atrial septal aneurysm (ASA) have been identified as potential risk factors for stroke.1–6⇓⇓⇓⇓⇓ In a recent meta-analysis of case-control studies comparing patients <55 years of age with ischemic stroke with nonstroke control subjects, ORs of stroke were 3.1 (95% CI, 2.3 to 4.2) for PFO and 6.1 (95% CI, 2.5 to 15.2) for ASA.6
Whether the relationship between PFO and stroke is causal and, if so, the precise stroke mechanism are still a matter of debate.6,7⇓ Potential mechanisms include paradoxical embolism from a venous source,7,8⇓ direct embolization from thrombi formed within the PFO or an associated ASA,9–11⇓⇓ and thrombus formation caused by atrial arrhythmias, such as paroxysmal atrial fibrillation.12
This study was part of a prospective, multicentric study (the PFO-ASA study) whose methods have been reported in detail elsewhere.16 Briefly, the PFO-ASA study was designed to assess the absolute and relative risks of recurrent stroke associated with PFO, ASA, or both in young patients with an otherwise unexplained ischemic stroke. The study was conducted in 30 European neurology departments. Patients were consecutively included between May 1, 1996, and December 31, 1998.
Inclusion Criteria and Initial Workup
Inclusion criteria were as follows: age ≥18 and ≤55 years; recent (≤3 months) ischemic stroke (neurological deficit lasting >24 hours); no definite cause of stroke after an extensive and standardized etiological workup, including cerebral CT scan (n=535) or MRI (n=428), routine blood tests, and detailed coagulation study (with protein S, protein C, antithrombin III, and antiphospholipid antibodies); 12-lead ECG and echocardiography (see below); and ≥1 of the following arterial investigations (within 1 month of stroke onset): catheter angiography (n=360; within 2 weeks after stroke onset in 83.2% of cases), magnetic resonance angiography (n=220), or cervical and transcranial ultrasonography (n=495). The decisions to perform additional investigations and to search for latent deep vein thrombosis or pulmonary embolism in patients with PFO were left to the investigator in charge of the patient. In 10 centers, 24-hour ECG recording was performed systematically in all patients (n=173, 84 with PFO) between days 7 and 21 after stroke onset. Twenty-four-hour ECG monitoring was performed after the acute phase of stroke to exclude arrhythmias secondary to stroke. Criteria derived from the Trial of Org 10172 in Acute Stroke Treatment (TOAST) classification17 were used to rule out definite causes of stroke.16
The clinical and imaging data of all patients were collected by the coordinating center and reviewed by 2 neurologists and 2 neuroradiologists who were blinded to the results of transesophageal echocardiography regarding the presence of PFO and ASA. Cases with potential violation of inclusion criteria were submitted to the Validation Committee.
The following information was systematically recorded: (1) baseline characteristics and traditional risk factors of stroke (Table 1); (2) past vascular events, such as stroke, deep venous thrombosis, or pulmonary embolism; (3) history of migraine according to International Headache Society criteria18; (4) stroke severity assessed on a modified Rankin scale19 at the time of inclusion; (5) palpitations preceding or accompanying stroke onset; (6) neurological features reported as suggestive of cardiogenic embolism,20,21⇓ such as abrupt, nonprogressive onset defined as no deficit on waking from sleep, peak deficit within the first 10 minutes, no subsequent deterioration during the first 24 hours, diminished level of consciousness at onset, cortical deficits, including Wernicke’s aphasia, isolated hemianopsia, hemineglect, and apraxia; and (7) features suggesting paradoxical embolism, such as the presence of deep venous thrombosis or pulmonary embolism, Valsalva-provoking activity within the 30 minutes preceding stroke onset (sporting effort, straining at stool, intercourse, lifting a heavy weight, getting up, laughing, and coughing),8 and circumstances predisposing to deep venous thrombosis before stroke onset, such as immobilization, anesthesia, surgery, or pregnancy.22
Brain and Vascular Imaging Findings
The following brain and vascular imaging features were analyzed: (1) stroke arterial territory following previously published templates23–25⇓⇓; (2) imaging features of previous stroke; and (3) neuroimaging data suggestive of cardiogenic embolism,20,26–28⇓⇓⇓ such as hemorrhagic infarct on neuroimaging performed within 2 weeks of stroke onset,29 superficial infarct, infarct larger than one half of the cerebral hemisphere, involvement of specific (Table 2) or multiple arterial territories, and the presence of intracranial occlusion on catheter angiography.
The presence of PFO and ASA was assessed by transesophageal echocardiography with a contrast study performed at rest and during provocative maneuvers (Valsalva and cough test) by use of 5-MHz multiplane (86.4%) or biplane (13.6%) transducers. Examinations were recorded on super VHS videotapes and independently reviewed by 2 experienced sonographers.16
The contrast study was considered positive if ≥3 microbubbles appeared in the left atrium, either spontaneously or after provocative maneuvers, within 3 cardiac cycles after complete opacification of the right atrium. The degree of shunting was defined as small (3 to 9 microbubbles), moderate (10 to 30 microbubbles), or large (>30 microbubbles). ASA was diagnosed when the atrial septum exhibited an excursion into the left or right atrium or both of ≥11 mm.
The 2 sonographers disagreed on the presence or absence of PFO in 13.9% of cases, the presence or absence of ASA in 6.6%, the degree of shunt in 26.5%, and the size of ASA in 8.4% of cases. Disagreements were resolved by unblinded consensus review.
Clinical data, brain, and vascular imaging findings were compared between patients with and without PFO and between patients with small to moderate and those with large PFO. Comparisons between groups were performed by a χ2 test, Fisher’s exact test, or t test for unpaired data whenever applicable. Factors independently associated with PFO were identified by logistic regression analysis. ORs with 95% CIs were calculated.
The association between migraine and PFO was examined through logistic regression analysis with migraine as a dependent variable and age, sex, PFO, and ASA as independent variables. To assess the role of PFO and ASA in isolation from or in association with each other, we used an independent variable divided into 4 categories: no PFO or ASA, isolated PFO, isolated ASA, and PFO and ASA.
This study included 581 consecutive patients. Of these, 267 patients (45.9%) had PFO and 61 (10.5%) had ASA. Of the 267 patients with PFO, 48% had small to moderate shunts and 52% had large shunts. PFO was strongly associated with ASA: 51 of the 267 patients with PFO (19.1%) also had ASA compared with 10 of 314 of those without PFO (3.2%; P<0.0001). The prevalence of ASA increased with the degree of shunt: 4.4%, 12.5%, and 25% in patients with small, moderate, and large shunts, respectively (P<0.0001).
Patients Characteristics and Stroke Patterns
Baseline characteristics and risk factors of stroke, according to the presence of a PFO, are shown in Table 1. Patients with PFO were younger and less likely to have traditional risk factors of stroke than those with no PFO. In logistic regression analysis (with age, sex, stroke risk factors, and ASA as independent variables), age (OR, 0.95; 95% CI, 0.93 to 0.97), hypertension (OR, 0.49; 95% CI, 0.28 to 0.85), hypercholesterolemia (OR, 0.56; 95% CI, 0.34 to 0.93), and current smoking (OR, 0.67; 95% CI, 0.47 to 0.97) were inversely associated with PFO, whereas ASA (OR, 7.4; 95% CI, 3.6 to 15.2) was positively associated with PFO.
Table 2 shows the characteristics of stroke. Patients with PFO had on average a less severe stroke as assessed by the Rankin scale at the time of inclusion, and this remained significant after adjustment for age, sex, vascular risk factors, and presence of ASA (P=0.01). Overall, no difference was found regarding arterial territories. Neuroimaging features of previous stroke were less frequent in patients with PFO, but this association was not significant after adjustment for age and risk factors for stroke. No significant difference was found between small to moderate and large shunts in terms of patient and stroke characteristics.
Mechanism of PFO-Associated Stroke
Clinical and imaging features suggestive of cardiogenic embolism did not differ significantly between groups, except for a higher frequency of cortical signs and of infarcts in the superior cerebellar artery territory in the PFO group. Features consistent with the diagnosis of paradoxical embolism were not found more frequently in patients with PFO than in patients without PFO, except for a higher frequency of circumstances predisposing to deep venous thrombosis (Table 3). Features suggesting cardioembolism or paradoxical embolism were not associated with the degree of shunt.
Of the 267 patients with PFO, 122 (45.7%) had a search for latent deep venous thrombosis or pulmonary embolism within 4 weeks after stroke onset (43% within 8 days). Investigations consisted of Doppler ultrasonography (n=112), phlebography (n=3), and pulmonary scintigraphy (n=43). A latent deep venous thrombosis or pulmonary embolism was found in 5 patients (4.1%) 4 to 12 days after stroke onset.
Palpitations preceding or accompanying stroke onset were rare and even less common in patients with PFO (0.7% versus 3.5% in those without PFO; P=0.06). Twenty-four-hour ECG recording did not reveal emboligenic arrhythmias in the 84 consecutive patients with PFO in whom it was performed.
Migraine was more common in patients with PFO (27.3%) than in patients without PFO (14.0%; P<0.0001). No significant relation was found between migraine and the degree of shunt: 22.2%, 29.2%, and 28.1% in patients with small, moderate, and large shunts, respectively (P=0.7). Migraine was also more frequent in patients with ASA (34.4%) than in those without ASA (18.5%; P=0.003). Logistic regression analysis showed that younger age, female sex, and PFO were significantly associated with migraine (Table 4). The association of PFO with migraine was stronger in patients with both PFO and ASA.
This is the first prospective, multicentric study comparing stroke risk factors and stroke patterns in a large population of young cryptogenic stroke patients with or without PFO. We selected patients ≤55 years of age because the higher prevalence of large-vessel atherosclerosis or small-artery disease in the elderly makes the diagnosis of cryptogenic stroke less frequent than in the young. In addition, the association of PFO and cryptogenic stroke has been consistently reported in this age group, whereas this association in those >55 years of age remains unconfirmed.6 Independent review of transesophageal echocardiography examinations by 2 experienced sonographers provided prevalences of PFO and ASA that were in the range of reported rates.6 To minimize information biases, clinical and imaging data were reviewed by 2 neurologists and 2 neuroradiologists who were blinded to the presence of PFO.
The first finding of our study is that cryptogenic stroke patients with PFO were younger and less likely to have traditional risk factors for stroke than patients without PFO. They also had on average a less severe stroke. These differences in stroke risk factors and stroke severity have not been stressed previously and suggest different stroke mechanisms in patients with and without PFO.
Some features suggestive of cardiogenic embolism were more frequent in the PFO group, which also suggests different stroke mechanisms in patients with and without PFO. These features, however, have a limited positive or negative predictive value for the diagnosis of cardioembolism.30 In contrast to the study by Steiner et al,15 we did not find embolic features to be more common in patients with large shunts than in patients with small to moderate shunts. On the whole, features suggesting paradoxical embolism were not more frequent in patients with PFO than in those without PFO, suggesting that paradoxical embolism might not be the prevalent mechanism of PFO-associated stroke, a finding consistent with our previous experience.8 These features, however, may be insufficient or inaccurate and therefore not useful in clinical practice. In the present study, investigators were not blinded to the presence of PFO when recording features suggesting paradoxical embolism. However, if this unblinded recording had introduced a bias, a higher prevalence of these features in patients with PFO would have been expected. The frequency of latent deep venous thrombosis in stroke patients with PFO8,31,32⇓⇓ was not an objective of our study, and the search for deep venous thrombosis was left to the investigator in charge of the patient. It was performed in fewer than half of the patients with PFO by use of various diagnostic techniques. This low rate of search for deep venous thrombosis suggests that many investigators either are not convinced that paradoxical embolism is a prevalent mechanism of PFO-associated stroke or are concerned about the low yield and pitfalls of a systematic search for deep venous thrombosis.7 Indeed, the source of emboli may remain undetected because of its location or the size of the thrombus. Venous thrombi may disappear either spontaneously or after anticoagulation before investigations are performed. Finally, venous thrombosis may be a mere consequence of immobilization resulting from stroke rather than a cause of stroke. The low rate of latent deep venous thrombosis found in the present study is consistent with the result of our previous study in young stroke patients.8 The role of a hypercoagulable state in the pathophysiology of PFO-associated stroke cannot be evaluated from this study because patients with a definite coagulopathy were not included in the study.
Our study does not provide argument for transient arrhythmia as a mechanism of PFO-associated stroke. Palpitations preceding or accompanying stroke onset were rare, and 24-hour ECG recording did not reveal emboligenic arrhythmias. It should be stressed, however, that arrhythmias are often clinically silent12 and that a single 24-hour Holter monitoring is not the optimal method to detect potentially more spaced-out episodes of transient arrhythmias.
Another interesting finding of our study is that migraine was more common in cryptogenic stroke patients with PFO (27.3%) than in those without PFO (14.0%). This result is consistent with recent studies in stroke patients and nonstroke individuals that showed an association between migraine with aura and PFO as detected by transcranial Doppler33,34⇓ or transthoracic echocardiography.35 At the time of the present study, the relation between migraine with aura and PFO had not emerged in the literature. Therefore, investigators were not asked to differentiate migraine with and without aura and were not blinded to the presence of PFO. However, because investigators were not aware of a potential relation between migraine and PFO, an information bias seems unlikely. The present study is the first to confirm the association between PFO and migraine with contrast transesophageal echocardiography as the diagnostic procedure with independent review of transesophageal echocardiographies by 2 sonographers who were blinded to the presence of migraine. In addition, our study, based on a large population of young stroke patients, shows that PFO is independently associated with migraine and that this association is stronger when PFO is associated with ASA. We did not find an increasing prevalence of migraine with the degree of shunt. No statistically significant association was found between migraine and isolated ASA, but the number of patients with isolated ASA was small. The link between migraine, PFO, and ASA is unclear. A particular genetic substrate might determine both atrial septal abnormalities36 and migraine.33 Another hypothesis, recently suggested by Wilmshurst et al,35 is that PFO might allow trigger substances of migraine (such as vasoactive chemicals) in the venous blood to bypass the pulmonary filter and to reach the systemic circulation in amounts large enough to induce a migraine attack. Further studies are needed to elucidate the relationships between migraine and atrial septal abnormalities.
J.L. Mas, C. Arquizan, C. Lamy, M. Zuber, C. Gianesini, D. Trystram, and J.F. Méder, Sainte-Anne Hospital, Paris, France.
J.L. Mas (chairman), Y. Bernard, B. Bertrand, J. Bogousslavsky, F. Chollet, L. Cabanes, A. Cohen, J.M. Ferro, H. Kwiecinski, J.P. Lesbre, D. Leys, T. Moulin, J.F. Pinel, R. Roudaut, D. Saudeau, and F. Woimant.
L. Cabanes, G. Derumeaux, X. Jeanrenaud, and A. Cohen.
P. Césaro, M. Giroud, F. Nicoli, and S. Weber.
Participating Institutions and Investigators
Centre Hospitalier Universitaire (CHU), Besançon, France. Neurology: T. Moulin and L. Tatu. Cardiology: A. Vuillemenot, D. Magnin, M.F. Seronde, F. Apffel, N. Meneveau, and Y. Bernard.
CHU Lariboisière and Saint-Antoine, Paris, France. Neurology: F. Woimant, P. Amarenco, N. Kubis, K. Vahedi, I. Crassard, G. Ast, H. Chabriat, M. Sarazin, and M. Haguenau. Cardiology: J.M. de Kermadec, A. N’Guyen Van Cao, S. Mazouz, B. Benhalima, C. Albo, A. Cohen, M. Khireddine, and N. Lamisse.
Sainte-Anne Hospital and CHU Cochin Port-Royal, Paris, France. Neurology: Coordinating Center. Cardiology: L. Cabanes, I. Cornuejols, and E. Lombard.
CHU, Nancy, France. Neurology: X. Ducrocq and J.C. Lacour. Cardiology: J.F. Bruntz and I. Magnin-Poull.
CHU, Tours, France. Neurology: D. Saudeau. Cardiology: A. Sirinelli.
CHU, Rennes, France. Neurology: J.F. Pinel. Cardiology: C. de Place, M. Laurent, and C. Bossee-Pilon.
Medical University of Warsaw, Warsaw, Poland. Neurology: H. Kwiecinski, B. Szyluk, A. Opuchlik, and J. Mieszkowski. Cardiology: A. Torbicki, P. Pruszczyk, and A. Kuch-Wocial.
CHU, Poitiers, France. N: J.P. Neau and C. Couderq. Cardiology: D. Coisne, G. Bacque, L. Christiaens, C. Couderq, and P. Raud-Raynier.
CHU Vaudois, Lausanne, Switzerland. Neurology: P. Arnold, M. Altieri, and J. Bogousslavsky. Cardiology: M. Nasratullah, N. Aebischer, and X. Jeanrenaud.
CHU, Rouen, France. Neurology: E. Guégan-Massardier and B. Mihout. Cardiology: D. Thomas, G. Derumeaux, and J.L. Gauthier.
CHU Pitié-Salpêtrière, Paris, France. Neurology: R. Manai, Y. Samson, and G. Rancurel. Cardiology: E. Coignard and R. Isnard.
CHU La Timone, Marseille, France. Neurology: L. Milandre. Cardiology: G. Habib.
Centre Hospitalier Général, Meaux, France. Neurology: F. Chedru and A. Ameri. Cardiology: J.F. Lefort.
CHU, Lille, France. Neurology: D. Leys and C. Lucas. Cardiology: C. Savoye, L. Goullard, and E. Chammas.
CHU, Saint-Etienne, France. Neurology: P. Garnier and D. Michel. Cardiology: C. Comtet and I. Cusey.
Santa Maria Hospital, University of Lisbon, Portugal. Neurology: T. Pinho e Melo, J.M. Ferro, P. Canhao, and F. Falcao. Cardiology: F. Pais, I. Dionisio, and M. Fiuza.
CHU Purpan, Toulouse, France. Neurology: J.F. Albucher. Cardiology: M. Andrieu, N. Blot-Souletie, M. Elbaz, and E. Maupas.
CHU, Bordeaux, France. Neurology: P. Gaïda and F. Rouanet. Cardiology: M. Marazanof, R. Roudaut, and P. Laffort.
CHU, Grenoble, France. Neurology: G. Besson, A. Jaillard, and M. Hommel. Cardiology: B. Bertrand and J.P. Baguet.
Klinikum Grossharden, Ludwig Maximiliams University, München, Germany. Neurology: T. Pfefferkorn and G. Hamann. Cardiology: W. von Scheidt.
Dr Schaffner Hospital, Lens, France. Neurology: F. Mounier-Véhier. Cardiology: R. Deturck.
CHU Rangueil, Toulouse, France. Neurology: V. Larrue. Cardiology: P. Massabuau.
CHU, Brest, France. Neurology: Y. Mocquard, F. Rouhart, and P. Diraison. Cardiology: Y. Jobic.
CHU, Nice, France. Neurology: M.H. Mahagne. Cardiology: P. Gibelin.
CHU, Angers, France. Neurology: H. Brugeilles, C. Moreau. Cardiology: J. Laporte.
Jolimont Hospital, Haine-Saint-Paul, Belgium. Neurology: G. Devuyst, P. Bara, M. Vandooren. Cardiology: R. Luwaert.
CHU Saint-Antonie. Tenon Hospital, Paris, France. Neurology: S. Alamowitch, C. Roos. Cardiology: E. Carbarz, B. Cormier.
E. Muller Hospital, Mulhouse, France. Neurology: G. Rodier. Cardiology: S. Hass.
Université libre de Bruxelles, Hôpital Erasme, Bruxelles, Belgium. Neurology: S. Blecic. Cardiology: P. Unger.
CHU, Clermont-Ferrant, France. Neurology: A. Coustes-Durieux. Cardiology: P. Marcollet.
This work was supported by grants from the Program Hospitalier de Recherche Clinique of the French Ministry of Health (AOM95059). The Assistance Publique-Hôpitaux de Paris had legal responsibility for the study.
Study members are listed in the Appendix.
- Received October 2, 2001.
- Revision received November 15, 2001.
- Accepted November 21, 2001.
- ↵Di Tullio M, Sacco RL, Gopal A, Mohr JP, Homma S. Patent foramen ovale as a risk factor for cryptogenic stroke. Ann Intern Med. 1992; 117: 461–465.
- ↵Cabanes L, Mas JL, Cohen A, Amarenco P, Cabanes PA, Oubary P, Chedru F, Guérin F, Bousser MG, de Recondo J. 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.
- ↵Overell JR, Bone I, Lees KR. Interatrial septal abnormalities and stroke: a meta-analysis of case-control studies. Neurology. 2000; 55: 1172–1179.
- ↵Mas JL. Patent foramen ovale, atrial septal aneurysm and ischaemic stoke in young adults. Eur Heart J. 1994; 15: 446–449.
- ↵Ranoux D, Cohen A, Cabanes L, Amarenco P, Bousser MG, Mas JL. Patent foramen ovale: is stroke due to paradoxical embolism? Stroke. 1993; 24: 31–34.
- ↵Mügge A, Daniel WG, Angermann C, Spes C, Khandheria BK, Kronzon I, Freeberg RS, Keren A, Dennig K, Engberding R, Sutherland GR, Vered Z, Erbel R, Visser CA, Lindert O, Hausmann D, Wenzlaff P. Atrial septal aneurysm in adults patients: a multicenter study using transthoracic and transesophageal echocardiography. Circulation. 1995; 91: 2785–2792.
- ↵Berthet K, Lavergne T, Cohen A, Guize L, Bousser MG, Le Heuzey JY, Amarenco P. Significant association of atrial vulnerability with atrial septal abnormalities in young patients with ischemic stroke of unknown cause. Stroke. 2000; 31: 398–403.
- ↵Bogousslavsky J, Garazi S, Jeanrenaud X, Aebischer N, Van Melle G, for the Lausanne Stroke With Paradoxical Embolism Study Group. Stroke recurrence in patients with patent foramen ovale. Neurology. 1996; 46: 1301–1305.
- ↵Steiner MM, Di Tullio MR, Rundek T, Gan R, Chen X, Liguori C, Brainin M, Homma S, Sacco RL. Patent foramen ovale size and embolic brain imaging findings among patients with ischemic stroke. Stroke. 1998; 29: 944–948.
- ↵Adams HP, Bendixen BH, Kappelle LJ, Biller J, Lowe BB, Gordon DL, Marsh EE, for the TOAST Investigators. Classification of subtype of acute ischemic stroke: definitions for use in a multicenter clinical trial. Stroke. 1993; 24: 35–41.
- ↵Headache Classification Committee of the International Headache Society. Classification and diagnostic criteria for headache disorders, cranial neuralgias and facial pain. Cephalalgia. 1988; 8: 19–28.
- ↵Van Swieten JC, Koudstaal PJ, Visser MC, Schouten HJA, Van Gijn J. Interobserver agreement for the assessment of handicap in stroke patients. Stroke. 1988; 19: 604–607.
- ↵Kittner SJ, Sharkness CM, Price TR, Plotnick GD, Dambrosia JM, Wolf PA, Mohr JP, Hier DB, Kase CS, Tuhrim S. Infarcts with a cardiac source of embolism in the NINCDS Stroke Data Bank: historical features. Neurology. 1990; 40: 281–284.
- ↵Bogousslavsky J, Cachin C, Regli F, Despland PA, Van Melle G, Kappenberger L. Cardiac sources of embolism and cerebral infarction: clinical consequences and vascular concomitants: the Lausanne Stroke Registry. Neurology. 1991; 41: 855–859.
- ↵Tatu L, Moulin T, Bogousslavsky J, Duvernoy H. Arterial territories of the human brain: brainstem and cerebellum. Neurology. 1996; 47: 1125–1135.
- ↵Tatu L, Moulin T, Bogousslavsky J, Duvernoy H. Arterial territories of the human brain: cerebral hemispheres. Neurology. 1998; 50: 1699–1708.
- ↵Caplan LR, Hier DB, D’Cruz I. Cerebral embolism in the Michael Reese Stroke Registry. Stroke. 1983; 14: 530–536.
- ↵Kittner SJ, Sharkness CM, Sloan MA, Price TR, Dambrosia JM, Tuhrim S, Wolf PA, Mohr JP, Hier DB. Features on initial computed tomography scan of infarcts with a cardiac source of embolism in the NINDS Stroke Data Bank. Stroke. 1992; 23: 1748–1751.
- ↵Amarenco P, Hauw JJ. Cerebellar infarction in the territory of the superior cerebellar artery: a clinicopathological study of 33 cases. Neurology. 1990; 40: 1383–1390.
- ↵Motto C, Aritzu E, Boccardi E, De Grandi C, Piana A, Candelise L. Reliability of hemorrhagic transformation diagnosis in acute ischemic stroke. Stroke. 1997; 28: 302–306.
- ↵Anzola GP, Magoni M, Guindani M, Rozzini L, Dalla Volta G. Potential source of cerebral embolism in migraine with aura: a transcranial Doppler study. Neurology. 1999; 52: 1622–1625.
- ↵Arquizan C, Coste J, Touboul PJ, Mas JL. Is patent foramen ovale a family trait? A transcranial Doppler sonographic study. Stroke. 2001; 32: 1563–1566.