Platelet Activation in the Cerebral Circulation in Different Subtypes of Ischemic Stroke and Binswanger’s Disease
Background and Purpose The sites of platelet activation in ischemic stroke are still unclear because previous reports have not identified them and various factors accompanied by stroke can activate platelets. We therefore examined the sites of platelet activation in patients with various types of ischemic stroke.
Methods The ratio of the plasma concentration of β-thromboglobulin (BTG) in the internal jugular vein to that in the antecubital vein (BTG-B) was calculated as an indicator of platelet activation in the cerebral circulation. Plasma BTG concentration was determined in 75 control subjects and in 186 patients with various subtypes of ischemic stroke including lacunar, atherothrombotic, and cardioembolic strokes, transient ischemic attacks, and Binswanger’s disease. The BTG ratio was evaluated with regard to subtype of stroke, time of blood sampling, size of infarct, presence of vascular lesions, and the effect of ticlopidine administration.
Results The mean BTG ratio was increased even in the chronic phase of most subtypes of stroke with the exception of cardioembolic stroke, which exhibited a persistent elevation of BTG-B concentrations. Patients with Binswanger’s disease showed a significant (P<.01) and frequent elevation of BTG ratio. High BTG ratios occurred in cases with vascular lesions observed on cerebral angiography. There was no correlation between the BTG ratio and infarct size. Use of ticlopidine was partially associated with a lower BTG ratio.
Conclusions Platelets were activated in the cerebral circulation of patients with stroke even in the chronic phase, which suggests the development of underlying vascular lesions and of thrombogenesis with or without infarction. Platelets were activated mainly within the heart in cases of cardioembolic stroke. An enhanced release reaction secondary to platelet activation was often seen in patients with Binswanger’s disease, which indicates that its pathophysiology differs from that of other subtypes of stroke.
β-Thromboglobulin (BTG) is a platelet-specific protein that is released into the circulation from activated platelets.1 Several quantitative studies of patients with cerebrovascular disease have used plasma BTG concentrations as a broad indicator for the activation of platelets. These studies found enhanced platelet activation, which suggests that the platelets participate in the development of most ischemic strokes.2 3 4 5 BTG concentrations are also influenced by the presence of atherothrombotic lesions, arrhythmias,6 renal functional impairment,7 and aging.1 8 Such factors as well as other forms of cerebral disease are so common in the elderly that the site of the release reaction in these studies is not necessarily the brain, since the blood samples obtained for BTG determination were usually obtained from the antecubital vein. As blood is returned from the brain via the internal jugular veins, usually to the heart, the ratio of plasma BTG concentration between the internal jugular vein and the carotid artery is considered to reflect the activation of platelets in the cerebral circulation, even in the presence of confounding factors.9 We therefore studied the BTG ratio in patients with ischemic stroke including Binswanger’s disease (ischemic leukoencephalopathy) to identify the sites of platelet activation and the pathophysiological role of the platelets in these ischemic disorders.
Subjects and Methods
We studied 186 Japanese patients with ischemic cerebrovascular disease (n=156) or Binswanger’s disease (n=30) (100 men, 86 women; mean age, 75.0 years).
We also studied 75 other subjects who were divided into three control groups: (1) 25 healthy, elderly subjects without significant vascular disease, diabetes, or hypertension; (2) 25 patients with atrial fibrillation alone, who were age-matched with the stroke patients; and (3) 25 volunteer patients with diseases other than stroke (nonstroke control subjects). Their diagnoses included Parkinson’s disease, myocardial infarction, atrial fibrillation, and glaucoma.
Stroke was classified according to the Classification of Cerebrovascular Diseases III of the National Institute of Neurological Disorders and Stroke10 and consisted of the three subtypes of lacunar, atherothrombotic, and cardioembolic stroke, as based on results of computed tomography (CT) and clinical findings.
Lacunar stroke was defined as a small infarction less than 1.5 cm in diameter in a deep location consistent with the clinical neurological deficit and with the absence of significant heart disease. Atherothrombotic stroke was defined as the presence of infarcts presumably caused by lesions of the extracranial or major intracranial arteries with characteristic clinical manifestations. Cardioembolic stroke was defined as the presence of both atrial fibrillation and infarcts involving the cerebral cortex. Seven patients experienced ischemic neurological symptoms typical of carotid system transient ischemic attacks (TIAs). Thirty patients with various degrees of dementia and CT evidence of diffuse changes in white matter were diagnosed as having Binswanger’s disease, based on the criteria proposed by Bennett et al.11 Hypertension and diabetes mellitus were diagnosed according to the criteria of the World Health Organization.12 13 Patients with any degree of renal failure (serum creatinine >1.5 mg/dL) were excluded from the study.
Blood specimens (2.5 mL) were withdrawn simultaneously from the internal jugular vein and the antecubital vein with a 21-gauge needle, taking care to avoid stasis or frothing. The internal jugular vein was usually punctured in the right carotid triangle along the pulsating carotid artery after the dominant internal jugular vein had been detected by 7.5-MHz B-mode sonography, since intravenous digital subtraction angiography has revealed that approximately two thirds of cases have a dominant right internal jugular vein.9 Paired blood samples were immediately placed into chilled tubes containing an anticoagulant (citric acid and sodium citrate) and a platelet-inhibitor solution (theophylline, adenosine, and dipyridamole). The tubes were inverted and chilled on ice for 30 minutes, then centrifuged at 2000g for 30 minutes at 4°C. Next, 0.3-mL aliquots of plasma were used to determine BTG concentrations by means of an enzyme-linked immunosorbent assay (Asserachrom β-TG, Diagnostica Stago). Concentrations of platelet factor 4 were determined at the same time in each sample (Asserachrom PF4, Diagnostica Stago). The ratio of BTG to platelet factor 4 was calculated as a means of checking sampling errors. Fourteen BTG concentrations were used in this study when the ratios exceeded 2.0. Twenty-one samples were excluded, leaving 230 samples to be evaluated for BTG ratios. Thus, BTG concentrations obtained for the internal jugular vein (A) and the antecubital vein (B) were used to calculate the ratio A/B as an indicator of platelet activation in the cerebral circulation. This method is based on two assumptions: (1) the ratio of BTG concentration between B and A reflects the activation of platelets in the cerebral circulation, with the latter representing the venous return from the brain, and (2) the BTG-B concentration in the antecubital vein is equivalent to that in the carotid artery if the platelets are activated only in the cerebral circulation. To document the latter point, we examined BTG ratios at six different times without splitting the samples in a patient with acute myocardial infarction, although high BTG-B concentrations were characteristic during a disease phase. The mean value in this case was 0.96±0.17 (range, 0.78 to 1.23). We used 25 nonstroke patients as control subjects for evaluating the normal BTG ratios. The mean range of normal for the BTG ratio was 0.96±0.42, with an upper limit of 1.8 (mean+2 SD), since the ratio is not affected in conditions with intact cerebral vessels.
The BTG-B concentration was used as an indicator of platelet activation in the systemic circulation.
We obtained informed consent from each subject. BTG ratios were measured in 205 paired blood samples from stroke patients, with one sample obtained in the chronic phase with or without antiplatelet treatment and additionally in the acute phase. Values were analyzed according to the type of infarction, the timing of sampling relative to the ischemic event, angiographic findings, sizes of infarcted areas, and the effect of administering ticlopidine, an antiplatelet drug.
The acute phase was defined as 7 days or less from the onset of the ischemic event, and the chronic phase was defined as more than 28 days from its onset. In TIAs, values were obtained in the acute phase. Values obtained in Binswanger’s disease were all considered to be in the chronic phase because the time of onset was unclear.
The size of infarct was estimated on CT in patients with lacunar, atherothrombotic, and cardioembolic strokes. The size was classified as small (<1.5 cm in diameter), large (more than half of the hemispheric area), or intermediate.
Some of the 27 selected patients who underwent intravenous subtraction angiography because they were suspected of having vascular lesions in major cerebral arteries subsequently underwent conventional angiography to evaluate lesions indicated by subtraction angiography. Findings were classified as normal, stenotic changes (<90% occlusion of the lumen, including any irregularities of the arterial vessel wall), and occlusive changes.
Patients in the chronic phase of lacunar (n=22) and atherothrombotic (n=27) stroke and those with Binswanger’s disease (n=8) were treated with ticlopidine 200 mg/d for more than 1 month. None of the subjects in this study received antiplatelet drugs such as aspirin or anticoagulants such as warfarin or heparin since elderly Japanese are often intolerant of those agents. To adjust for any effect of ticlopidine on BTG values, statistical analyses were performed on data from untreated subjects by unpaired t tests and χ2 analyses. The comparisons of the BTG-B concentrations and the BTG ratios between the normal control groups and all types of stroke were performed by a Dunnett’s post hoc procedure. A one-way ANOVA and Fisher’s protected least significant difference was used to assess differences among the stroke groups. The influence of ticlopidine administration on the BTG-B concentrations and the BTG ratios was also tested by means of unpaired t tests. These calculations were performed with the StatView 4.01 and SuperANOVA statistical programs on a Macintosh personal computer. A value of P<.05 was considered statistically significant.
Background of Stroke Subtypes
CT images and clinical findings verified the following diagnoses: 62 cases of lacunar, 56 of atherothrombotic, and 31 of cardioembolic strokes; 7 cases of TIA; and 30 cases of Binswanger’s disease (Table 1⇓). There were no significant differences in age between patients with different subtypes of stroke. Overall, more than half of the patients (n=96) had hypertension, except for those with cardioembolism (n=4). Of the 30 patients with Binswanger’s disease, 26 suffered from hemiparesis, 24 had moderate dementia, and 6 had mild dementia. Serum concentrations of BTG-B and BTG ratios were determined in 63 samples obtained from the 62 patients with lacunar stroke, 62 samples from the 56 patients with atherothrombotic stroke, 47 samples from the 31 patients with cardioembolic stroke, 8 samples from the 7 patients with TIA, and 33 samples from the 30 patients with Binswanger’s disease. Ticlopidine was administered during the chronic phase to 22 of the 62 patients with lacunar stroke, 27 of the 56 patients with atherothrombotic stroke, 3 of the 7 patients with TIA, and 8 of the 30 patients with Binswanger’s disease.
BTG in Differing Subtypes of Stroke
BTG-B concentrations and BTG ratios were evaluated in patients during each phase of stroke and in the control subjects (Table 2⇓). Serum BTG-B concentrations were significantly elevated during the acute phases of lacunar and cardioembolic strokes and in TIA, as well as in control patients with atrial fibrillation, compared with findings in healthy elderly control subjects. No significant difference in BTG-B concentration was seen in the acute versus the chronic phase of each type of stroke. However, the BTG-B concentration tended to rise during the chronic phase of atherothrombotic and cardioembolic strokes. The serum concentration of BTG-B in cardioembolic stroke exceeded that in lacunar stroke during the chronic phase.
The mean BTG ratio in patients with most subtypes of ischemic stroke significantly exceeded that of the control subjects, even during the chronic phase. The mean BTG ratio and the incidence of a high BTG ratio were increased in the chronic phase of lacunar stroke, although no distinct differences were observed between the acute and the chronic phases. The BTG ratio and the incidence of an elevated BTG ratio were significantly increased in both the acute and the chronic phases of atherothrombotic stroke. The group with cardioembolic stroke showed an elevation of BTG ratios during the acute phase. However, their median BTG ratio was 1.00 in the acute as well as the chronic phase. While 5 of the 7 patients with TIA showed high BTG ratios, the number of cases was too small for statistical evaluation.
Considering the chronic phase of each subtype of stroke, the mean BTG ratio was significantly higher (P<.05) in patients with Binswanger’s disease (4.08±6.88), although the ratio fluctuated widely. High BTG ratios (>1.8) were seen more often in patients with Binswanger’s disease as well as in those with lacunar or atherothrombotic strokes.
BTG Ratio and Infarct Size
CT imaging of atherothrombotic stroke in the chronic phase showed large infarcts in 5 patients and medium-sized infarcts in 20 patients. Of 25 patients with cardioembolic stroke, large infarcts were seen in 16 and medium-sized infarcts in 9. Small infarcts were seen in the 25 patients with lacunar stroke. The 5 patients with TIA showed high BTG ratios despite the absence of infarction. There was no correlation between BTG ratio and the size of the infarct.
BTG Ratio and Angiographic Findings
According to angiography, stenotic changes were seen in 5 of the 12 atherothrombotic cases (stenotic changes at both the extracranial carotid bifurcation and the siphon in 4 and at the extracranial carotid bifurcation in 1), in 3 of the 5 TIA cases (stenotic changes at the extracranial carotid bifurcation, the siphon, and the middle cerebral artery in each case), and in 4 of the 6 cases of Binswanger’s disease (stenotic changes at both the extracranial carotid bifurcation and the siphon in all 4 cases). Occlusion was demonstrated in 7 of the atherothrombotic cases (occlusion at the middle cerebral artery in 4, at the extracranial carotid bifurcation in 2, and at the siphon in 1), in 1 of the 5 TIA cases at the extracranial carotid bifurcation, and in 2 of the 6 cases of Binswanger’s disease.
The BTG ratio was significantly elevated in patients with a vascular lesion found on arteriography compared with those with normal angiographic findings (Table 3⇓). A high BTG ratio tended to be more frequent in patients with stenotic changes than in patients with normal angiographic findings.
Effect of Ticlopidine on BTG Ratio
Ticlopidine administration was associated with lower BTG ratios in patients with each stroke subtype, although the effect was not statistically significant (Table 4⇓). The incidence of an elevated BTG ratio tended to be decreased in patients with atherothrombotic stroke who received this agent.
Platelet activation is usually associated with thrombosis. Clinical evidence for the efficacy of antiplatelet agents in preventing a recurrence of ischemic stroke shows that the platelets are major participants in the etiology of thromboembolic disease.15 16 Ischemia itself can also lead to platelet activation.17 18 Thus, the platelets can be activated variously with alterations in platelet function,19 platelet morphology,20 21 and releasing factors2 3 4 5 during the acute phase of ischemic stroke.
Studies in humans show that activated platelets release BTG, which is then rapidly cleared from the plasma by renal excretion.22 Because BTG has a half-life of approximately 100 minutes, its plasma concentration reflects real-time platelet activation in vivo.22 Shah et al3 have shown that plasma concentrations of BTG-B are significantly elevated in the acute phase of atherothrombotic and cardioembolic strokes (50.2±4.1 ng/mL and 56.3±5.2 ng/mL, respectively) compared with findings in elderly control subjects (31.1±2.5 ng/mL). We found no differences between the acute and the chronic phases of stroke and found higher BTG-B concentrations for each subtype of stroke than previously reported.2 3 4 One possible reason is that our patients were older than those previously described. Plasma BTG concentrations increase with age, being approximately 40 ng/mL in healthy people between the ages of 71 and 90 years.1 Zahavi et al8 reported mean BTG values of 42.4±4.0 ng/mL in healthy elderly men, which is supported by our data. Zahavi et al suggested the presence of abnormal hemostatic mechanisms with an enhanced platelet activation in these individuals.
The timing at which blood is sampled after a stroke influences the BTG concentration. A marked elevation of plasma BTG has been observed within 3 days of the beginning of the acute phase in patients with atherothrombotic or cardioembolic stroke,5 and a persistent elevation of BTG concentration has been reported 6 weeks after onset.4 We observed an enhancement of platelet activation throughout the course of most subtypes of stroke. These data help to define the relation between platelet activation and stroke onset as well as the site of platelet activation.
It is obviously difficult to determine the extent of platelet activation before stroke onset in that BTG concentrations cannot be measured before the ischemic event. The sites of activation are also uncertain since strokes are frequently accompanied by such cardiovascular disorders as coronary heart disease or arteriosclerosis, especially in the elderly, which also can affect their plasma BTG concentrations.8 Thus, BTG may not always be derived from the cerebral circulation, even in a patient with acute stroke. Clearly, a specific method is needed to detect platelet activation in the cerebral circulation.
The ratio of the plasma BTG concentration between the efflux from A and the influx to B in the brain (A/B) is considered to be useful in assessing platelet activation in the cerebral circulation.9 In the present study BTG ratios were elevated in patients with all subtypes of stroke with the exception of those with TIA versus the nonstroke control subjects. The number of patients with TIA was too small for statistical evaluation. These findings indicate that platelets are often activated in the cerebral circulation, especially in patients with lacunar and atherothrombotic strokes or with Binswanger’s disease. In contrast, a high BTG ratio was infrequent in either phase of cardioembolic stroke despite large infarcts. The finding of elevated BTG-B concentrations in patients with atrial fibrillation without stroke indicates that their platelet activation was mainly intracardiac. The BTG ratio was thus more sensitive than the BTG-B concentration in detecting platelet activation in the cerebral circulation.
Such elevations were seen in patients with lacunar or atherothrombotic stroke even in the chronic phase, suggesting that vascular lesions are widely present in the cerebral circulation. Vascular lesions promote platelet activation (Table 3⇑), presumably because of the injured endothelium associated with atheromatous plaque and the change in hemodynamics due to shear stress by the stenosed lumen, which activates the circulating platelets.23 In fact, we observed high BTG ratios in patients with vascular lesions documented on cerebral angiography. The presence of an elevated BTG ratio suggests the presence of arteriosclerosis and atherosclerosis of the cerebral vessels associated with a tendency toward thrombosis.
Lacunar infarction, which is commonly caused by small-vessel disease, may be incidentally accompanied by atherothrombotic changes in the main arterial trunks.24 These data are compatible with electron microscopic findings showing an increase in platelets with pseudopod formation and a reduced number of dense bodies in the internal jugular vein of patients with cerebral arteriosclerosis.21 However, we observed no significant difference in the BTG ratio throughout the course of patients with lacunar strokes, either because platelet activation was only temporary in the acute phase or because platelet activation cannot be detected in small vessels. In atherothrombotic strokes, the relatively high BTG ratio observed in the acute phase indicates a high level of platelet activation as the causative vascular lesion responsible for the major ischemic area, although the small increase in BTG ratio in the chronic phase of that disorder resembled that observed in lacunar stroke. These changes were prominent in Binswanger’s disease, and high BTG ratios were frequently observed.
Elevation of the BTG ratios was more closely correlated with the severity of vascular lesions in the cerebral circulation than with infarct size. The wide variation in BTG ratio showed that platelets in the cerebral circulation were activated only variably, unrelated to the recurrence of ischemic stroke in most types of stroke, although the BTG ratio could be lowered to some extent by ticlopidine administration.
This activation suggests that underlying vascular lesions promote platelet release reactions, inducing vessel wall damage and causing neuronal impairment downstream.25 26 27 In the present study the enhanced platelet release reaction observed in patients with Binswanger’s disease might imply a different pathophysiology, although chronic ischemia in the hemispheric white matter has been considered as a cause.28 29 In this disorder, large amounts of platelet-derived substances are often released into the bloodstream, resulting in tissue injury in the downstream area. Further studies are needed to evaluate the brain damage induced by long-term activation of platelets in the cerebral circulation.
All funds for this study were obtained from within the authors’ department.
- Received August 15, 1994.
- Revision received October 10, 1994.
- Accepted October 19, 1994.
- Copyright © 1995 by American Heart Association
Taomoto K, Asada M, Kanazawa Y, Matsumoto S. Usefulness of the measurement of plasma β-thromboglobulin (β-TG) in cerebrovascular disease. Stroke. 1983;14:518-524.
Shah AB, Beamer N, Coull BM. Enhanced in vivo platelet activation in subtypes of ischemic stroke. Stroke. 1985;16:643-647.
Woo E, Huang CY, Chan V, Chan YW, Yu YL, Chan TK. Beta-thromboglobulin in cerebral infarction. J Neurol Neurosurg Psychiatry. 1988;51:557-562.
Waki R, Okada Y, Tashiro M, Miyashita T, Yamaguchi T. The plasma level of β-thromboglobulin in the acute phase of cerebral infarction. Jpn J Stroke. 1987;9:433-439.
Gustafsson C, Blomback M, Britton M, Hamsten A, Svensson J. Coagulation factors and the increased risk of stroke in nonvalvular atrial fibrillation. Stroke. 1990;21:47-51.
Iwamoto T, Uno M, Kiuchi A, Fukaya S, Yanagawa K. Plasma β-thromboglobulin gradient between the internal jugular and antecubital veins in cerebrovascular disease. J Jpn Coll of Angiol. 1991;31:1269-1273.
National Institute of Neurological Disorders and Stroke. Classification of cerebrovascular diseases III. Stroke. 1990;21:637-676.
Bennet DA, Wilson RS, Gilley DW, Fox JH. Clinical diagnosis of Binswanger’s disease. J Neurol Neurosurg Psychiatry. 1990;53:961-965.
World Health Organization Expert Committee on Arterial Hypertension. Report. Geneva, Switzerland: World Health Organization; 1978. WHO Technical Report Series 231.
World Health Organization. WHO Study Group on Diabetes Mellitus. Geneva, Switzerland: World Health Organization; 1985. Technical Report Series 727.
Kaplan KL, Owen J. Plasma levels of β-thromboglobulin and platelet factor 4 as indices of platelet activation in vivo. Blood. 1981;57:199-202.
Antiplatelet Trialists’ Collaboration. Secondary prevention of vascular disease by prolonged antiplatelet treatment. Br Med J. 1988;296:320-331.
Albers GW. Role of ticlopidine for prevention of stroke. Stroke. 1992;23:912-916.
Dougherty JH Jr, Levy DE, Weksler BB. Experimental cerebral ischemia produces platelet aggregates. Neurology. 1979;29:1460-1465.
Jafar JJ, Menoni R, Feinberg H, LeBreton G, Crowell RM. Selective platelet deposition during focal cerebral ischemia in cats. Stroke. 1989;20:664-667.
Joseph R, D’Andrea G, Oster SB, Welch KMA. Whale blood platelet function in acute ischemic stroke. Stroke. 1989;20:38-44.
Joseph R, Riddle JM, Welch KMA, D’Andrea G. Platelet ultrastructure and secretion in acute ischemic stroke. Stroke. 1989;20:1316-1319.
Imaizumi M, Ashida K, Isaka Y, Nakayama H, Matsushita K. Electron microscopic study of blood platelets in ischemic cerebrovascular disease. Jpn J Stroke. 1990;12:279-287.
Kappelle LJ, Koudstaal PJ, van Gijn J, Ramos LMP, Keunen JEE. Carotid angiography in patients with lacunar infarction. Stroke. 1988;19:1093-1096.
Fujimoto T, Suzuki H, Tanoue K, Fukushima Y, Yamazaki H. Cerebrovascular injuries induced by activation of platelets in vivo. Stroke. 1985;16:245-250.
Joseph R, Tsering C, Welch KMA. Study of platelet-mediated neurotoxicity in rat brain. Stroke. 1992;23:394-398.
Wester P, Dietrich WD, Prado R, Watson BD, Globus MYT. Serotonin release into plasma during common carotid artery thrombosis in rats. Stroke. 1992;23:870-875.
Babikian V, Ropper AH. Binswanger’s disease: a review. Stroke. 1987;18:2-12.