Pathogenesis of anterior circulation stroke in patients with nonvalvular atrial fibrillation: The Lausanne Stroke Registry Julien Bogousslavsky, MD; Guy Van Melle, MD; Franco Regli, MD; and Lukas Kappenberger, MD
Article abstract-We studied coexisting potential arterial and cardiac causes of stroke in 159 patients with nonvalvular atrial fibrillation (AF), who were admitted to a population-based primary care center for an anterior circulation infarct. Systematic investigations included brain CT, carotid-Doppler ultrasounds with frequency analysis and echotomography, and mono- and bidimensional echocardiography. Lacunar infarction due to small-artery disease was at least as likely as an AF-related stroke in 13% of the patients who had hypertension and a small deep infarct. In 67% of the patients, internal carotid artery disease ipsilateral to infarct was present, but it was severe ( 1 5 0 %stenosis or occlusion) in only 11%.There was a potential cardiac source of embolism other than AF in 14%. Overall, although only 18% of the patients had AF as the only potential cause of stroke, embolism from the heart remained the most likely etiology of infarct in 76%. Our findings emphasize the role of AF-related hemodynamic disturbances, which were often associated with embolic phenomena, and a rather low early risk of recurring embolism (4%)within the 1st month after stroke. NEUROLOGY 1990;401046-1050
Nonvalvular atrial fibrillation (NVAF) is associated with an increased risk of stroke,’-I but it is not known how many of the strokes in patients with NVAF are actually due to atrial fibrillation (AF). Hart et a18 suggested that less than 213 of strokes in NVAF patients were probably embolic from the heart. Britton and Gustafssong emphasized both the cardiac source and associated large-artery atherosclerosis. Preliminary studies of carotid Doppler ultrasounds in asymptomaticlo and “cerebrovascular”l1 patients with NVAF have shown that internal carotid artery (ICA) disease is not uncommon. Using systematic brain CT, Doppler ultrasounds, and echocardiography, we have studied the coexisting potential causes of stroke in a series of patients with NVAF and acute infarct in the carotid territory. Methods. All patients with NVAF who were admitted to our center for an acute infarct in the territory of the ICA were studied over a 4-year period. These patients were part of 1,698 patients included consecutively in the Lausanne Stroke Registry,’2 a computed prospective registry of patients with firstever stroke admitted to a population-based primary care center. NVAF was diagnosed on the basis of ECG-documented AF prior to stroke or on admission, with no mitral stenosis on echocardiography. ICA territory infarcts were CT-proven and included the middle cerebral artery (MCA) territory, anterior cerebral artery territory, and the territory of deep perforators of the carotid system as defined in the templates of DamasioI3
and from our group.14 Systematic investigations included brain CT (up to 4 examinations, the 1st within 7 days of the stroke), carotid Doppler ultrasounds with frequency spectral analysis and B-mode echotomography, 12-lead ECG, and mono- and bidimensional echocardiography. Intracranial studies (angiography or transcranial Doppler) were done only in selected patients. Results of these investigations and risk factors (hypertension, diabetes, smoking, hypercholesterolemia, elevated hematocrit, as defined following previously reported criteria in the Lausanne Stroke Registry=), previous TIA(s), type of stroke onset, clinical findings, evolution in hospital, and functional disability were coded into the Registry and analyzed. Doppler findings were grouped into 4 categories: normal, plaques or (50% stenosis, 2 5 0 % stenosis, occlusion. Potential cardiac sources of embolism, other than AF, included akinetic left ventricular segment with or without thrombus due to myocardial infarct, cardiac tumor, global cardiac hypo- or dyskinesia (dilated cardiomyopathy), and other rarer diseases.12
Results. There were 89 (56%) men and 70 (44%) women, with a mean age of 68.6 k 7 years (range, 33 to 87 years). Seventy-six (48%) patients had hypertension, 22 (14%) had diabetes mellitus, 41 (26%) smoked cigarettes regularly, and 15 (9%) had hypercholesterolemia. Venous hematocrit on admission was higher than 0.45 in 44 (27%) patients. Ten (6%) patients had a history of vascular claudication. Heart disease. History of heart disease and ECG and
From the Departmentof Neurology (Drs. Bogousslavskyand Regli) and Division of Cardiology (Dr. Kappenberger), Centre Hospitalier Universitaire Vaudois, and the University Institute of Social and Preventive Medicine (Dr. Van Melle), Lauaanne, Switzerland. Received October 31,1989. Accepted for publication in final form December 26,1989. Address correspondence and reprint requests to Dr. Julien Bogousslavsky, Department of Neurology, CHUV, 1011 Lausanne, Switzerland. 1046 NEUROLOGY 40 July 1990
Table 1. History of heart disease and ECG History of heart disease Angina pectoris Myocardial infarct (MI) Atrial fibrillation None Electrocardiogram Anterior MI Inferior MI ST-Tchanges (with no MI) 1st-degree AV block Right bundle branch block Left bundle branch block Left axial deviation
I 71 (45%) 17*(11%) 125 (79%) 18 (11%) 13t(9%) 6 (4%) 80 (50%) 6 (4%) 8 (5%) 5 (3%) 22 (14%)
* Acute in 2 patients. t Asymptomatic in 2 patients.
Table 2. Echocardiography N Suspected left atrial thrombus Suspected left ventricular thrombus Left ventricular akinetic segment without thrombus Left atrial enlargement Other abnormality Mitral annulus calcification with no stenosis Left atrial myxoma Dilated cardiomyopathy Aortic sclerosis Normal
Table 3. Internal carotid artery (ICA) Doppler ultrasounds
139 (87%) 10 (7%) 3 (2%) 14 (10%) 81 (58%) 25 (18%) 1 1 3 20 14 (10%)
echocardiogram findings are summarized in tables 1 and 2. Thirty-four (21%) patients had no history of AF, which was discovered a t the time of admission for stroke. AF had been known for 4 f 3 years in the 125 patients with a history of AF. At the time of stroke, 9 (7%)of these 125 patients were on anticoagulant therapy (International Normalized Ratio 2.5 to 4), and 6 (5%)were taking antiplatelet agents. A positive history of coronary heart disease was present in 88 (55%)patients and occurred more often (77 of 125 [62%])in the patients with known AF than in those with presumed recent onset (11 of 34 [32%],p < 0.05 [Fisher test]). Echocardiography was not performed in 8 patients who died before it could be done and could not be assessed in 12 patients due to technical problems (mainly difficulty with positioning the patient and associated pulmonary disorders). Among the remaining 139 patients, a direct source of embolism was suspected in 14 (10%) instances (left atrial thrombus in 10, left ventricular thrombus in 3, left atrial myxoma in 1). Suspected left atrial thrombi were seen only in 10 (12%)of the 81 (58%)patients with enlarged left atrium. Indirect sources of embolism were found in 18 (13%) patients (left ventricular akinetic segment without thrombus in 14, dilated cardiomyopathy in 3, mitral annulus calcification in 1). Overall, 22 (14%) patients had a possible cardiac source of embolism not associated with AF. Thirty-four (24%) pa-
Normal Plaque/24 (36 hours in 7 patients), and fluctuated (>24 (30 hours) in 11 (7%)patients. Eight (5%) patients had a syncope (sudden transient loss of consciousness) as 1st cerebral symptom, including 3 with watershed infarction. Ten (6%)patients had inaugural seizures (hemiconvulsions in 8, generalized seizure in 2). Bradycardia below 50 beats per minute at stroke onset or on admission was observed in 17 (11%) patients, and 17 (11%) patients (including 6 with bradycardia) had mean blood pressure measured a t stroke onset or on admission below their lowest mean blood pressure during the preceding 2 years (mean of lowering, 20 & 12 mm Hg). Cerebral infarct. The topography of cerebral infarcts is summarized in table 4. Superficial MCA territory July 1990 NEUROLOGY 40 1047
Table 4. Topography of infarct
I
Left Right Both sides Territory Inferior division of MCA Deep ICA Superior division of MCA Superficial deep MCA ACA Watershed Complete ACA MCA
+
+
83 (52%) 62 (39%) 14 (9%) 51 (32%) 41 (26%) 31 (19%) 24 (15%) 6 (4%) 4 (3%) 2 (1%)
MCA Middle cerebral artery. ICA Internal carotid artery. ACA Anterior cerebral artery.
infarcts were the most common type of infarcts (51%), but infarcts limited to the deep territory of the MCA were not uncommon (26%). Infarcts in the territory of the inferior division of the MCA (32%) were more common than infarcts in the territory of the superior division of the MCA (19%).A large infarct involving the deep and superficial MCA territory or the whole ICA territory was observed in 26 (16%) patients. An infarct at the borderzone between the MCA and posterior cerebral artery territory (in 3) or between the MCA and anterior cerebral artery territory (in 1)was present in 4 (3%) patients. Fourteen (9%) patients had associated nonsymptomatic infarcts, which involved the superficial MCA territory in 12 patients. A hemorrhagic component of infarct was present in 10 (6%) patients (on CT 4 f 4 days after stroke). Coexisting arterial and cardiac diseuse. In 18 (11%) patients, AF coexisted with a 250% stenosis or an occlusion of ICA ipsilateral to stroke. In another 89 (56%) patients, AF coexisted with a 4 0 % stenosis or nonstenosing plaques of the ICA. In 21 (13%) patients with long-standing hypertension, CT showed a small deep infarct in the territory of MCA perforators, which was compatible with a lacunar infarct due to smallartery disease. We did not find any significant difference in the prevalence of ECG (ischemia, conduction defects) or echocardiography (atrial enlargement, ventricular akinetic segment, other abnormalities) abnormalities between the patients with or without largeartery disease or presumed small-artery disease, but this analysis was limited by the small number of patients in the subgroups. Short-term evolution. By preestablished policy, no patient received intravenous or oral anticoagulant therapy within 14 days of stroke (including those previously taking oral anticoagulants, but subcutaneous heparin (3 X 5,000 IU/d) and acetylsalicylic acid (1,000 mg/d) were given systematically. After that period, 47 patients were given acenocoumarol, and the remaining patients continued to take acetylsalicylic acid. Fifteen (9%)patients died within the 1st month after stroke (6 k 4 days), from cerebral mass effect (7), pneumonia (3), myocardial infarct (1), cardiac failure (2), and pulmonary embolism (2). The 30-day functional disability in the 144 survivors was as follows: no disability (4 pa1048 NEUROLOGY 40 July 1990
tients; 3%), return to previous activities with some difficulty (49 patients; 31%),could return to only some previous activities (65 patients; 41%), and could not return to previous activities (26 patients; 16%). Five (3%) patients sustained a recurrence of ischemic stroke within the 1st month after stroke, all during the 1st 2 weeks. Two of these strokes involved the same territory as the 1st stroke, while 3 involved another territory (contralateral ICA territory in 2, vertebrobasilar territory in 1).Recurrence of infarct in the carotid territory was not associated with >50% stenosis or occlusion of the ipsilateral ICA (0/4 versus 18/155). Stroke recurrence contributed to disability but not to 30-day mortality. Another patient had a systemic embolism (in 1 leg) 7 days after stroke. Thus, the event recurrence rate among survivors was 4% a t 2 weeks and at 1month. No complication of anticoagulant therapy was observed.
Discussion. Our study is the 1st in which systematic CT, Doppler ultrasounds, and echocardiography have been performed, thus providing an accurate evaluation of the coexistence of potential causes of stroke other than AF, such as arterial disease and concomitant heart disease. The key findings include relatively high frequency of lacunar infarcts, relatively low prevalence of severe carotid occlusive disease, high prevalence of accompanying cardiac lesions that are potentially also sources of clot, presence of circulatory changes that support a hemodynamic factor in causing stroke, and low recurrence rate of stroke in first 2 weeks while the patients were treated with acetylsalicylic acid and lowdose subcutaneous heparin. Lacunar infarcts. Thirteen percent of the patients had long-standing hypertension and a small infarct limited to the territory of deep perforators of the carotid system; in these, a lacunar infarct was at least as likely as an AFrelated infarct. Flegel et a17reported an increased risk of stroke in NVAF patients with hypertension, compared with NVAF patients without hypertension, which may be due to coexisting small-artery disease. Large-artery disease. Another 11% of the patients had 2 5 0 % stenosis or occlusion of ICA; in these, artery-to-artery embolism was as likely as cardioembolism. However, one cannot exclude that some of the 11 ICA occlusions ipsilateral to infarct were cardioembolic in that angiography and ultrasounds may not be reliable in differentiating cardioembolic from atherosclerotic occlusion. In an autopsy series, 20% of ICA occlusions were cardi~embolic.'~ However, even without considering the cases with ICA occlusion, we found that the frequency of 2 5 0 % ICA stenosis was higher on the side ipsilateral to stroke than on the contralateral side. In another 56% of our patients, AF coexisted with carotid plaques or (50% stenosis. Nevertheless, it is unlikely that this nonsevere large-artery disease played a role in the pathogenesis of most of the strokes, because it had the same frequency on the contralateral "asymptomatic" ICA. Previous studies have cast doubt upon nonsevere carotid disease as a significant etiology for stroke in general.'6J7 Also, carotid artery disease may be detected in up to 17% of unselected asymptomatic patients with AF.I0
Table 6 . Presumed pathogenesis Clinical group
Presumed pathogenesis
N
%
Small deep infarct plus hypertension
Penetrating artery disease
21
13%
Severe carotid artery occlusive disease
Atherosclerosis
18
11%
No or minimal carotid disease
Cardioembolic
120
76%'
Including cardiac sources of embolism unrelated to AF in 22 (14%) patients.
Accompanying cardiac sources of embolism. There was a potential cardiac source of embolism other than AF in 22 (14%)patients (left ventricular akinetic segment with or without thrombus, globally dilated cardiomyopathy, left atrial myxoma, mitral annulus calcification), including 3 with >50% ICA disease and 16 with (50% ICA disease. The likelihood of AF being the cause of stroke remains uncertain in these patients. Causes of stroke. Thus, only 29 (18%)of our patients had AF as the only potential cause of stroke (absence of evidence for coexisting large- or small-artery disease and another cardiac source of embolism). However, if we add to these 29 patients those patients with nonsevere ICA disease (plaques or (50% stenosis), we can assume that NVAF was the most likely cause of infarct in 101 (63.5%) patients. In the remaining 58 (36.5%) patients, another etiology for stroke was as likely or more likely than AF. Overall, cardioembolism was the most likely cause of stroke in 76% of the patients, including cardiac sources of embolism unrelated to AF in 14% (table 5). Hemodynamic factors.The mechanism of AF-related stroke is usually assumed to be embolic.2J8-21 We found that more than '/4 of the patients who had intracranial vascular assessment (angiography or transcranial Doppler) showed evidence of embolism. Embolic occlusions usually clear up shortly,22and it is likely that this figure is too low. However, we emphasize hemodynamic factors as an important contributor to stroke pathogenesis, since these factors largely have been overlooked in the literature. Britton and Gustafssong found that chronic heart failure increased the risk of stroke in NVAF patients and mentioned that hemodynamic disturbances were responsible. Acute loss of consciousness is not a feature of carotid territory ischemia, but 5% of our patients had a syncope a t stroke onset, which suggested global cerebral hypoperfusion due to hemodynamic failure. Moreover, 18% of our patients had bradycardia below 50 beats per minute or decreased blood pressure (compared with their usual blood pressure) at the time of admission, contrary to what would be expected in patients with acute stroke and sympathetic discharge. Cerebral blood flow may be decreased in association with AF, as compared with sinus rhythm, but the reason is unclear.23Our findings suggest that
hemodynamic dysfunction may be prominent at the time of stroke onset in many NVAF patients, even though the primary cause of stroke appears to be embolic. The prevalence of watershed infarct was not lower in our NVAF patients than in the Lausanne Stroke Registry in generaL2 In a previous study, we reported that 12% of unilateral watershed infarcts occurred in patients with NVAF, of whom had coexisting severe carotid disease.24 Although some authors have suggested that watershed infarct associated with AF may be emboli^,^^ the mechanism of watershed infarction is generally considered to be h e m o d y n a r n i ~ . * ~ That . ~ ~ 3- ~ ~ of our NVAF patients with watershed infarct had an inaugural syncope supports this hypothesis. On the other hand, the most commonly involved territory was the inferior division of the MCA, a finding consistent with embolism from the heart.28,29 Early recurrence. We observed a low rate of event recurrence without intravenous anticoagulant therapy. With a combination of antiaggregant therapy and lowdose subcutaneous heparin during the first 2 weeks, our patients had a recurrence rate of 4%.Though we confirmed that most of the recurrences occur during the first 2 weeks after stroke, our frequency of recurrence was lower than other^.^^^^ This issue is c o n t r ~ v e r s i a l , ~ ~ because the assessment of stroke recurrence may be difficult in the patients who die shortly after stroke and in the patients with a nonsudden stroke onset. Twenty percent of our patients had a progressive or fluctuating stroke onset, often over longer than 24 hours, which is not incompatible with early multiple embolization. However, after the neurologic deficit had stabilized, new events were uncommon during the followingweeks. The exact role for anticoagulant and antiaggregant therapy in the secondary prevention of stroke in NVAF patients cannot be resolved by our study, but is currently being evaluated in the European Atrial Fibrillation Trial and in North American studies.
Acknowledgment We thank Robert G. Hart, MD, San Antonio, TX, for his helpful comments.
References 1. Fairfax AJ, Lambert CD, Leatham A. Systemic embolism in chronic sinoatrial disorder. N Engl J Med 1976;295:190-192.
2.Hinton RC, Kistler JP, Fallon JT, Friedlich AC, Fisher CM. Influence of etiology of atrial fibrillation on incidence of systemic embolism. Am J Cardiol 1977;40509-513. 3. Wolf PA, Dawler TR, Thomas HE, Kannel WB. Epidemiologic assessment of chronic atrial fibrillation and risk of stroke: The Framingham Study. Neurology 1978;28973-977. 4. Wolf PA, Kannel WB, McGee DL, Meeks SL, Bharucha NE, McNamara PM. Duration of atrial fibrillation and imminence of stroke: The Framingham Study. Stroke 1983;14664-667. 5. Treseder AS, Sastry BSD, Thomas TPL, Yates MA, Pathy MSJ. Atrial fibrillation and stroke in elderly hospitalized patients. Age Ageing 1986;15:89-92. 6. Onundarson PT, Thorgeirsson G, Jonmundsen E, Sigfusson N, Handarson T. Chronic atrial fibrillation: epidemiologic features and 14 year follow-up. A case-control study. Eur Heart J July 1990 NEUROLOGY 40 1049
1987;8:521-527. 7. Flegel KM, Shipley MJ, Rose G. Risk of stroke in non-rheumatic atrial fibrillation. Lancet 1987;1:526-529. 8. Hart RG, C o d BM, Hart PD. Early recurrent embolism associated with nonvalvular atrial fibrillation: a retrospective study. Stroke 1983;14688-693. 9. Britton M, Gustafsson C. Non-rheumatic atrial fibrillation as a risk factor for stroke. Stroke 1985;16182-186. 10. Tegeler CH, Stroke Prevention in Atrial Fibrillation Study: Carotid Stenosis Study Group. Carotid stenosis in atrial fibrillation. Abstract. Neurology 1989;39(suppl 1):159. 11. Weinberger J , Rotlauf E, Materese E, Halperin J. Noninvasive evaluation of the extracranial carotid arteries in patients with cerebrovascular events and atrial fibrillations. Arch Intern Med 1988;148:1785-1788. 12. BogousslavskyJ , Van Melle G, Regli F, Lausanne Stroke Registry Group. The Lausanne Stroke Registry: analysis of 1000 consecutive patients with first stroke. Stroke 1988;19:1083-1092. 13. Damasio H. A CT guide to the identification of cerebral vascular territories. Arch Neurol1983;40138-142. 14. Ghika J , Bogousslavsky J, Regli F. Infardsin the territory of the deep perforators h m the carotid system. Neurology 1989,39507-512. 15. Castaigne P, Lhermitte F, Gautier JC, Escourolle R, Derouesnk C. Internal carotid artery occlusion: a study of 61 instances in 50 patients with post-mortem data. Brain 1970;93:231-258. 16. Harrison MJG, Marshall J. Angiographic appearance of carotid bifurcation in patients with completed stroke, transient ischemic attacks and cerebral tumor. Br Med J 1976;1:205-207. 17. Bogousslavsky J , Hachinski VC, Boughner DR, Fox AJ, Vinuela F, Bamett HJM. Cardiac and arterial lesions in carotid transient ischemic attacks. Arch Neurol1986;43:223-228. 18. Mohr J P , Caplan LR, Melski JW, et al. The Harvard Cooperative S t r o k e Registry: a p r o s p e c t i v e r e g i s t r y . Neurology 1978;28:754-762.
1060 NEUROLOGY 40 July 1990
19. Caplan LR, Hier DB, D’Cruz I. Cerebral embolism in the Michael Reese stroke registry. Stroke 1983;14:530-536. 20. Sherman DG, Goldman L, Whiting RB, Jurgensen K, Kaste M, Easton JD. Thromboembolism in patients with atrial fibrillation. Arch Neurol 1984;41:708-710. 21. Halperin JL, Hart RG. Atrial fibrillation and stroke: new ideas, persisting dilemmas. Stroke 1988;19:937-941. 22. Fieschi C, Bozzao L. Transient embolic occlusion of middle cerebral and internal carotid arteries in cerebral apoplexy. J Neurol Neurosurg Psychiatry 1969;32:236-240. 23. Boysen G, Petersen P, Kastrup J , Videback R, Schutten HJ. CBF during chronic atrial fibrillation and after conversion to sinus rhythm. J Cereb Blood Flow Metab 1989;9(suppl 1):365. 24. Bogousslavsky J, Regli F. Unilateral watershed cerebral infarcts. Neurology 1986;36:373-377. 25. Graeber MC, Jordan E, Mishra SK, Nadeau SE. Watershed infarction on C T an unreliable sign of hemodynamic stroke. Abstract. Neurology 1989;39(suppl1):183-184. 26. Zulch KJ. Uber die Entstehung und Lokalisation der Hirninfarkte. Neurochirurgia (Stuttg) 1961;21:158-178. 27. Adams JH, Brierley JB, Connor RCR, Treip C. The effects of systemic hypotension upon human brain: clinical and neuropathological observations in 11 cases. Brain 1966;89:235-268. 28. BogousslavskyJ , Van Melle G, Regli F. Middle cerebral artery pial temtory infarcts: a study of the Lausanne Stroke Registry. Ann Neurol 1989;25:555-560. 29.Caplan LR, Kelly M, Kase CS, et al. Infarcts of the inferior division of the right middle cerebral artery: mirror image of Wernicke’s aphasia. Neurology 1986;361015-1020. 30. Kelley RE, Berger JR, Alter M, Kovacs AG. Cerebral ischemia and atrial fibrillation:prospective study. Neurology 1984;34:1285-1291. 31. Sage JI. Prospective study of stroke associated with non-valvular atrial fibrillation: low frequency of early recurrence. Abstract. Ann Neurol 1986;20155.
Pathogenesis of anterior circulation stroke in patients with nonvalvular atrial fibrillation: The Lausanne Stroke Registry Julien Bogousslavsky, Guy Van Melle, Franco Regli, et al. Neurology 1990;40;1046 DOI 10.1212/WNL.40.7.1046 This information is current as of July 1, 1990 Updated Information & Services
including high resolution figures, can be found at: http://www.neurology.org/content/40/7/1046.full.html
Citations
This article has been cited by 13 HighWire-hosted articles: http://www.neurology.org/content/40/7/1046.full.html##otherarticles
Permissions & Licensing
Information about reproducing this article in parts (figures,tables) or in its entirety can be found online at: http://www.neurology.org/misc/about.xhtml#permissions
Reprints
Information about ordering reprints can be found online: http://www.neurology.org/misc/addir.xhtml#reprintsus
Neurology ® is the official journal of the American Academy of Neurology. Published continuously since 1951, it is now a weekly with 48 issues per year. Copyright © 1990 by the American Academy of Neurology. All rights reserved. Print ISSN: 0028-3878. Online ISSN: 1526-632X.
Article abstract-We studied coexisting potential arterial and cardiac causes of stroke in 159 patients with nonvalvular atrial fibrillation (AF), who were admitted to a population-based primary care center for an anterior circulation infarct. Systematic investigations included brain CT, carotid-Doppler ultrasounds with frequency analysis and echotomography, and mono- and bidimensional echocardiography. Lacunar infarction due to small-artery disease was at least as likely as an AF-related stroke in 13% of the patients who had hypertension and a small deep infarct. In 67% of the patients, internal carotid artery disease ipsilateral to infarct was present, but it was severe ( 1 5 0 %stenosis or occlusion) in only 11%.There was a potential cardiac source of embolism other than AF in 14%. Overall, although only 18% of the patients had AF as the only potential cause of stroke, embolism from the heart remained the most likely etiology of infarct in 76%. Our findings emphasize the role of AF-related hemodynamic disturbances, which were often associated with embolic phenomena, and a rather low early risk of recurring embolism (4%)within the 1st month after stroke. NEUROLOGY 1990;401046-1050
Nonvalvular atrial fibrillation (NVAF) is associated with an increased risk of stroke,’-I but it is not known how many of the strokes in patients with NVAF are actually due to atrial fibrillation (AF). Hart et a18 suggested that less than 213 of strokes in NVAF patients were probably embolic from the heart. Britton and Gustafssong emphasized both the cardiac source and associated large-artery atherosclerosis. Preliminary studies of carotid Doppler ultrasounds in asymptomaticlo and “cerebrovascular”l1 patients with NVAF have shown that internal carotid artery (ICA) disease is not uncommon. Using systematic brain CT, Doppler ultrasounds, and echocardiography, we have studied the coexisting potential causes of stroke in a series of patients with NVAF and acute infarct in the carotid territory. Methods. All patients with NVAF who were admitted to our center for an acute infarct in the territory of the ICA were studied over a 4-year period. These patients were part of 1,698 patients included consecutively in the Lausanne Stroke Registry,’2 a computed prospective registry of patients with firstever stroke admitted to a population-based primary care center. NVAF was diagnosed on the basis of ECG-documented AF prior to stroke or on admission, with no mitral stenosis on echocardiography. ICA territory infarcts were CT-proven and included the middle cerebral artery (MCA) territory, anterior cerebral artery territory, and the territory of deep perforators of the carotid system as defined in the templates of DamasioI3
and from our group.14 Systematic investigations included brain CT (up to 4 examinations, the 1st within 7 days of the stroke), carotid Doppler ultrasounds with frequency spectral analysis and B-mode echotomography, 12-lead ECG, and mono- and bidimensional echocardiography. Intracranial studies (angiography or transcranial Doppler) were done only in selected patients. Results of these investigations and risk factors (hypertension, diabetes, smoking, hypercholesterolemia, elevated hematocrit, as defined following previously reported criteria in the Lausanne Stroke Registry=), previous TIA(s), type of stroke onset, clinical findings, evolution in hospital, and functional disability were coded into the Registry and analyzed. Doppler findings were grouped into 4 categories: normal, plaques or (50% stenosis, 2 5 0 % stenosis, occlusion. Potential cardiac sources of embolism, other than AF, included akinetic left ventricular segment with or without thrombus due to myocardial infarct, cardiac tumor, global cardiac hypo- or dyskinesia (dilated cardiomyopathy), and other rarer diseases.12
Results. There were 89 (56%) men and 70 (44%) women, with a mean age of 68.6 k 7 years (range, 33 to 87 years). Seventy-six (48%) patients had hypertension, 22 (14%) had diabetes mellitus, 41 (26%) smoked cigarettes regularly, and 15 (9%) had hypercholesterolemia. Venous hematocrit on admission was higher than 0.45 in 44 (27%) patients. Ten (6%) patients had a history of vascular claudication. Heart disease. History of heart disease and ECG and
From the Departmentof Neurology (Drs. Bogousslavskyand Regli) and Division of Cardiology (Dr. Kappenberger), Centre Hospitalier Universitaire Vaudois, and the University Institute of Social and Preventive Medicine (Dr. Van Melle), Lauaanne, Switzerland. Received October 31,1989. Accepted for publication in final form December 26,1989. Address correspondence and reprint requests to Dr. Julien Bogousslavsky, Department of Neurology, CHUV, 1011 Lausanne, Switzerland. 1046 NEUROLOGY 40 July 1990
Table 1. History of heart disease and ECG History of heart disease Angina pectoris Myocardial infarct (MI) Atrial fibrillation None Electrocardiogram Anterior MI Inferior MI ST-Tchanges (with no MI) 1st-degree AV block Right bundle branch block Left bundle branch block Left axial deviation
I 71 (45%) 17*(11%) 125 (79%) 18 (11%) 13t(9%) 6 (4%) 80 (50%) 6 (4%) 8 (5%) 5 (3%) 22 (14%)
* Acute in 2 patients. t Asymptomatic in 2 patients.
Table 2. Echocardiography N Suspected left atrial thrombus Suspected left ventricular thrombus Left ventricular akinetic segment without thrombus Left atrial enlargement Other abnormality Mitral annulus calcification with no stenosis Left atrial myxoma Dilated cardiomyopathy Aortic sclerosis Normal
Table 3. Internal carotid artery (ICA) Doppler ultrasounds
139 (87%) 10 (7%) 3 (2%) 14 (10%) 81 (58%) 25 (18%) 1 1 3 20 14 (10%)
echocardiogram findings are summarized in tables 1 and 2. Thirty-four (21%) patients had no history of AF, which was discovered a t the time of admission for stroke. AF had been known for 4 f 3 years in the 125 patients with a history of AF. At the time of stroke, 9 (7%)of these 125 patients were on anticoagulant therapy (International Normalized Ratio 2.5 to 4), and 6 (5%)were taking antiplatelet agents. A positive history of coronary heart disease was present in 88 (55%)patients and occurred more often (77 of 125 [62%])in the patients with known AF than in those with presumed recent onset (11 of 34 [32%],p < 0.05 [Fisher test]). Echocardiography was not performed in 8 patients who died before it could be done and could not be assessed in 12 patients due to technical problems (mainly difficulty with positioning the patient and associated pulmonary disorders). Among the remaining 139 patients, a direct source of embolism was suspected in 14 (10%) instances (left atrial thrombus in 10, left ventricular thrombus in 3, left atrial myxoma in 1). Suspected left atrial thrombi were seen only in 10 (12%)of the 81 (58%)patients with enlarged left atrium. Indirect sources of embolism were found in 18 (13%) patients (left ventricular akinetic segment without thrombus in 14, dilated cardiomyopathy in 3, mitral annulus calcification in 1). Overall, 22 (14%) patients had a possible cardiac source of embolism not associated with AF. Thirty-four (24%) pa-
Normal Plaque/24 (36 hours in 7 patients), and fluctuated (>24 (30 hours) in 11 (7%)patients. Eight (5%) patients had a syncope (sudden transient loss of consciousness) as 1st cerebral symptom, including 3 with watershed infarction. Ten (6%)patients had inaugural seizures (hemiconvulsions in 8, generalized seizure in 2). Bradycardia below 50 beats per minute at stroke onset or on admission was observed in 17 (11%) patients, and 17 (11%) patients (including 6 with bradycardia) had mean blood pressure measured a t stroke onset or on admission below their lowest mean blood pressure during the preceding 2 years (mean of lowering, 20 & 12 mm Hg). Cerebral infarct. The topography of cerebral infarcts is summarized in table 4. Superficial MCA territory July 1990 NEUROLOGY 40 1047
Table 4. Topography of infarct
I
Left Right Both sides Territory Inferior division of MCA Deep ICA Superior division of MCA Superficial deep MCA ACA Watershed Complete ACA MCA
+
+
83 (52%) 62 (39%) 14 (9%) 51 (32%) 41 (26%) 31 (19%) 24 (15%) 6 (4%) 4 (3%) 2 (1%)
MCA Middle cerebral artery. ICA Internal carotid artery. ACA Anterior cerebral artery.
infarcts were the most common type of infarcts (51%), but infarcts limited to the deep territory of the MCA were not uncommon (26%). Infarcts in the territory of the inferior division of the MCA (32%) were more common than infarcts in the territory of the superior division of the MCA (19%).A large infarct involving the deep and superficial MCA territory or the whole ICA territory was observed in 26 (16%) patients. An infarct at the borderzone between the MCA and posterior cerebral artery territory (in 3) or between the MCA and anterior cerebral artery territory (in 1)was present in 4 (3%) patients. Fourteen (9%) patients had associated nonsymptomatic infarcts, which involved the superficial MCA territory in 12 patients. A hemorrhagic component of infarct was present in 10 (6%) patients (on CT 4 f 4 days after stroke). Coexisting arterial and cardiac diseuse. In 18 (11%) patients, AF coexisted with a 250% stenosis or an occlusion of ICA ipsilateral to stroke. In another 89 (56%) patients, AF coexisted with a 4 0 % stenosis or nonstenosing plaques of the ICA. In 21 (13%) patients with long-standing hypertension, CT showed a small deep infarct in the territory of MCA perforators, which was compatible with a lacunar infarct due to smallartery disease. We did not find any significant difference in the prevalence of ECG (ischemia, conduction defects) or echocardiography (atrial enlargement, ventricular akinetic segment, other abnormalities) abnormalities between the patients with or without largeartery disease or presumed small-artery disease, but this analysis was limited by the small number of patients in the subgroups. Short-term evolution. By preestablished policy, no patient received intravenous or oral anticoagulant therapy within 14 days of stroke (including those previously taking oral anticoagulants, but subcutaneous heparin (3 X 5,000 IU/d) and acetylsalicylic acid (1,000 mg/d) were given systematically. After that period, 47 patients were given acenocoumarol, and the remaining patients continued to take acetylsalicylic acid. Fifteen (9%)patients died within the 1st month after stroke (6 k 4 days), from cerebral mass effect (7), pneumonia (3), myocardial infarct (1), cardiac failure (2), and pulmonary embolism (2). The 30-day functional disability in the 144 survivors was as follows: no disability (4 pa1048 NEUROLOGY 40 July 1990
tients; 3%), return to previous activities with some difficulty (49 patients; 31%),could return to only some previous activities (65 patients; 41%), and could not return to previous activities (26 patients; 16%). Five (3%) patients sustained a recurrence of ischemic stroke within the 1st month after stroke, all during the 1st 2 weeks. Two of these strokes involved the same territory as the 1st stroke, while 3 involved another territory (contralateral ICA territory in 2, vertebrobasilar territory in 1).Recurrence of infarct in the carotid territory was not associated with >50% stenosis or occlusion of the ipsilateral ICA (0/4 versus 18/155). Stroke recurrence contributed to disability but not to 30-day mortality. Another patient had a systemic embolism (in 1 leg) 7 days after stroke. Thus, the event recurrence rate among survivors was 4% a t 2 weeks and at 1month. No complication of anticoagulant therapy was observed.
Discussion. Our study is the 1st in which systematic CT, Doppler ultrasounds, and echocardiography have been performed, thus providing an accurate evaluation of the coexistence of potential causes of stroke other than AF, such as arterial disease and concomitant heart disease. The key findings include relatively high frequency of lacunar infarcts, relatively low prevalence of severe carotid occlusive disease, high prevalence of accompanying cardiac lesions that are potentially also sources of clot, presence of circulatory changes that support a hemodynamic factor in causing stroke, and low recurrence rate of stroke in first 2 weeks while the patients were treated with acetylsalicylic acid and lowdose subcutaneous heparin. Lacunar infarcts. Thirteen percent of the patients had long-standing hypertension and a small infarct limited to the territory of deep perforators of the carotid system; in these, a lacunar infarct was at least as likely as an AFrelated infarct. Flegel et a17reported an increased risk of stroke in NVAF patients with hypertension, compared with NVAF patients without hypertension, which may be due to coexisting small-artery disease. Large-artery disease. Another 11% of the patients had 2 5 0 % stenosis or occlusion of ICA; in these, artery-to-artery embolism was as likely as cardioembolism. However, one cannot exclude that some of the 11 ICA occlusions ipsilateral to infarct were cardioembolic in that angiography and ultrasounds may not be reliable in differentiating cardioembolic from atherosclerotic occlusion. In an autopsy series, 20% of ICA occlusions were cardi~embolic.'~ However, even without considering the cases with ICA occlusion, we found that the frequency of 2 5 0 % ICA stenosis was higher on the side ipsilateral to stroke than on the contralateral side. In another 56% of our patients, AF coexisted with carotid plaques or (50% stenosis. Nevertheless, it is unlikely that this nonsevere large-artery disease played a role in the pathogenesis of most of the strokes, because it had the same frequency on the contralateral "asymptomatic" ICA. Previous studies have cast doubt upon nonsevere carotid disease as a significant etiology for stroke in general.'6J7 Also, carotid artery disease may be detected in up to 17% of unselected asymptomatic patients with AF.I0
Table 6 . Presumed pathogenesis Clinical group
Presumed pathogenesis
N
%
Small deep infarct plus hypertension
Penetrating artery disease
21
13%
Severe carotid artery occlusive disease
Atherosclerosis
18
11%
No or minimal carotid disease
Cardioembolic
120
76%'
Including cardiac sources of embolism unrelated to AF in 22 (14%) patients.
Accompanying cardiac sources of embolism. There was a potential cardiac source of embolism other than AF in 22 (14%)patients (left ventricular akinetic segment with or without thrombus, globally dilated cardiomyopathy, left atrial myxoma, mitral annulus calcification), including 3 with >50% ICA disease and 16 with (50% ICA disease. The likelihood of AF being the cause of stroke remains uncertain in these patients. Causes of stroke. Thus, only 29 (18%)of our patients had AF as the only potential cause of stroke (absence of evidence for coexisting large- or small-artery disease and another cardiac source of embolism). However, if we add to these 29 patients those patients with nonsevere ICA disease (plaques or (50% stenosis), we can assume that NVAF was the most likely cause of infarct in 101 (63.5%) patients. In the remaining 58 (36.5%) patients, another etiology for stroke was as likely or more likely than AF. Overall, cardioembolism was the most likely cause of stroke in 76% of the patients, including cardiac sources of embolism unrelated to AF in 14% (table 5). Hemodynamic factors.The mechanism of AF-related stroke is usually assumed to be embolic.2J8-21 We found that more than '/4 of the patients who had intracranial vascular assessment (angiography or transcranial Doppler) showed evidence of embolism. Embolic occlusions usually clear up shortly,22and it is likely that this figure is too low. However, we emphasize hemodynamic factors as an important contributor to stroke pathogenesis, since these factors largely have been overlooked in the literature. Britton and Gustafssong found that chronic heart failure increased the risk of stroke in NVAF patients and mentioned that hemodynamic disturbances were responsible. Acute loss of consciousness is not a feature of carotid territory ischemia, but 5% of our patients had a syncope a t stroke onset, which suggested global cerebral hypoperfusion due to hemodynamic failure. Moreover, 18% of our patients had bradycardia below 50 beats per minute or decreased blood pressure (compared with their usual blood pressure) at the time of admission, contrary to what would be expected in patients with acute stroke and sympathetic discharge. Cerebral blood flow may be decreased in association with AF, as compared with sinus rhythm, but the reason is unclear.23Our findings suggest that
hemodynamic dysfunction may be prominent at the time of stroke onset in many NVAF patients, even though the primary cause of stroke appears to be embolic. The prevalence of watershed infarct was not lower in our NVAF patients than in the Lausanne Stroke Registry in generaL2 In a previous study, we reported that 12% of unilateral watershed infarcts occurred in patients with NVAF, of whom had coexisting severe carotid disease.24 Although some authors have suggested that watershed infarct associated with AF may be emboli^,^^ the mechanism of watershed infarction is generally considered to be h e m o d y n a r n i ~ . * ~ That . ~ ~ 3- ~ ~ of our NVAF patients with watershed infarct had an inaugural syncope supports this hypothesis. On the other hand, the most commonly involved territory was the inferior division of the MCA, a finding consistent with embolism from the heart.28,29 Early recurrence. We observed a low rate of event recurrence without intravenous anticoagulant therapy. With a combination of antiaggregant therapy and lowdose subcutaneous heparin during the first 2 weeks, our patients had a recurrence rate of 4%.Though we confirmed that most of the recurrences occur during the first 2 weeks after stroke, our frequency of recurrence was lower than other^.^^^^ This issue is c o n t r ~ v e r s i a l , ~ ~ because the assessment of stroke recurrence may be difficult in the patients who die shortly after stroke and in the patients with a nonsudden stroke onset. Twenty percent of our patients had a progressive or fluctuating stroke onset, often over longer than 24 hours, which is not incompatible with early multiple embolization. However, after the neurologic deficit had stabilized, new events were uncommon during the followingweeks. The exact role for anticoagulant and antiaggregant therapy in the secondary prevention of stroke in NVAF patients cannot be resolved by our study, but is currently being evaluated in the European Atrial Fibrillation Trial and in North American studies.
Acknowledgment We thank Robert G. Hart, MD, San Antonio, TX, for his helpful comments.
References 1. Fairfax AJ, Lambert CD, Leatham A. Systemic embolism in chronic sinoatrial disorder. N Engl J Med 1976;295:190-192.
2.Hinton RC, Kistler JP, Fallon JT, Friedlich AC, Fisher CM. Influence of etiology of atrial fibrillation on incidence of systemic embolism. Am J Cardiol 1977;40509-513. 3. Wolf PA, Dawler TR, Thomas HE, Kannel WB. Epidemiologic assessment of chronic atrial fibrillation and risk of stroke: The Framingham Study. Neurology 1978;28973-977. 4. Wolf PA, Kannel WB, McGee DL, Meeks SL, Bharucha NE, McNamara PM. Duration of atrial fibrillation and imminence of stroke: The Framingham Study. Stroke 1983;14664-667. 5. Treseder AS, Sastry BSD, Thomas TPL, Yates MA, Pathy MSJ. Atrial fibrillation and stroke in elderly hospitalized patients. Age Ageing 1986;15:89-92. 6. Onundarson PT, Thorgeirsson G, Jonmundsen E, Sigfusson N, Handarson T. Chronic atrial fibrillation: epidemiologic features and 14 year follow-up. A case-control study. Eur Heart J July 1990 NEUROLOGY 40 1049
1987;8:521-527. 7. Flegel KM, Shipley MJ, Rose G. Risk of stroke in non-rheumatic atrial fibrillation. Lancet 1987;1:526-529. 8. Hart RG, C o d BM, Hart PD. Early recurrent embolism associated with nonvalvular atrial fibrillation: a retrospective study. Stroke 1983;14688-693. 9. Britton M, Gustafsson C. Non-rheumatic atrial fibrillation as a risk factor for stroke. Stroke 1985;16182-186. 10. Tegeler CH, Stroke Prevention in Atrial Fibrillation Study: Carotid Stenosis Study Group. Carotid stenosis in atrial fibrillation. Abstract. Neurology 1989;39(suppl 1):159. 11. Weinberger J , Rotlauf E, Materese E, Halperin J. Noninvasive evaluation of the extracranial carotid arteries in patients with cerebrovascular events and atrial fibrillations. Arch Intern Med 1988;148:1785-1788. 12. BogousslavskyJ , Van Melle G, Regli F, Lausanne Stroke Registry Group. The Lausanne Stroke Registry: analysis of 1000 consecutive patients with first stroke. Stroke 1988;19:1083-1092. 13. Damasio H. A CT guide to the identification of cerebral vascular territories. Arch Neurol1983;40138-142. 14. Ghika J , Bogousslavsky J, Regli F. Infardsin the territory of the deep perforators h m the carotid system. Neurology 1989,39507-512. 15. Castaigne P, Lhermitte F, Gautier JC, Escourolle R, Derouesnk C. Internal carotid artery occlusion: a study of 61 instances in 50 patients with post-mortem data. Brain 1970;93:231-258. 16. Harrison MJG, Marshall J. Angiographic appearance of carotid bifurcation in patients with completed stroke, transient ischemic attacks and cerebral tumor. Br Med J 1976;1:205-207. 17. Bogousslavsky J , Hachinski VC, Boughner DR, Fox AJ, Vinuela F, Bamett HJM. Cardiac and arterial lesions in carotid transient ischemic attacks. Arch Neurol1986;43:223-228. 18. Mohr J P , Caplan LR, Melski JW, et al. The Harvard Cooperative S t r o k e Registry: a p r o s p e c t i v e r e g i s t r y . Neurology 1978;28:754-762.
1060 NEUROLOGY 40 July 1990
19. Caplan LR, Hier DB, D’Cruz I. Cerebral embolism in the Michael Reese stroke registry. Stroke 1983;14:530-536. 20. Sherman DG, Goldman L, Whiting RB, Jurgensen K, Kaste M, Easton JD. Thromboembolism in patients with atrial fibrillation. Arch Neurol 1984;41:708-710. 21. Halperin JL, Hart RG. Atrial fibrillation and stroke: new ideas, persisting dilemmas. Stroke 1988;19:937-941. 22. Fieschi C, Bozzao L. Transient embolic occlusion of middle cerebral and internal carotid arteries in cerebral apoplexy. J Neurol Neurosurg Psychiatry 1969;32:236-240. 23. Boysen G, Petersen P, Kastrup J , Videback R, Schutten HJ. CBF during chronic atrial fibrillation and after conversion to sinus rhythm. J Cereb Blood Flow Metab 1989;9(suppl 1):365. 24. Bogousslavsky J, Regli F. Unilateral watershed cerebral infarcts. Neurology 1986;36:373-377. 25. Graeber MC, Jordan E, Mishra SK, Nadeau SE. Watershed infarction on C T an unreliable sign of hemodynamic stroke. Abstract. Neurology 1989;39(suppl1):183-184. 26. Zulch KJ. Uber die Entstehung und Lokalisation der Hirninfarkte. Neurochirurgia (Stuttg) 1961;21:158-178. 27. Adams JH, Brierley JB, Connor RCR, Treip C. The effects of systemic hypotension upon human brain: clinical and neuropathological observations in 11 cases. Brain 1966;89:235-268. 28. BogousslavskyJ , Van Melle G, Regli F. Middle cerebral artery pial temtory infarcts: a study of the Lausanne Stroke Registry. Ann Neurol 1989;25:555-560. 29.Caplan LR, Kelly M, Kase CS, et al. Infarcts of the inferior division of the right middle cerebral artery: mirror image of Wernicke’s aphasia. Neurology 1986;361015-1020. 30. Kelley RE, Berger JR, Alter M, Kovacs AG. Cerebral ischemia and atrial fibrillation:prospective study. Neurology 1984;34:1285-1291. 31. Sage JI. Prospective study of stroke associated with non-valvular atrial fibrillation: low frequency of early recurrence. Abstract. Ann Neurol 1986;20155.
Pathogenesis of anterior circulation stroke in patients with nonvalvular atrial fibrillation: The Lausanne Stroke Registry Julien Bogousslavsky, Guy Van Melle, Franco Regli, et al. Neurology 1990;40;1046 DOI 10.1212/WNL.40.7.1046 This information is current as of July 1, 1990 Updated Information & Services
including high resolution figures, can be found at: http://www.neurology.org/content/40/7/1046.full.html
Citations
This article has been cited by 13 HighWire-hosted articles: http://www.neurology.org/content/40/7/1046.full.html##otherarticles
Permissions & Licensing
Information about reproducing this article in parts (figures,tables) or in its entirety can be found online at: http://www.neurology.org/misc/about.xhtml#permissions
Reprints
Information about ordering reprints can be found online: http://www.neurology.org/misc/addir.xhtml#reprintsus
Neurology ® is the official journal of the American Academy of Neurology. Published continuously since 1951, it is now a weekly with 48 issues per year. Copyright © 1990 by the American Academy of Neurology. All rights reserved. Print ISSN: 0028-3878. Online ISSN: 1526-632X.
