Risk factors and antiplatelet therapy in TIA and stroke patients

Journal of Neurological Sciences 154 (1998) 200–204

Risk factors and antiplatelet therapy in TIA and stroke patients a, b a a Juha Puranen *, Markku Laakso , Paavo Riekkinen Sr. , Juhani Sivenius a

Department of Neurology, Kuopio University, Kuopio, Finland Department of Medicine, Kuopio University, Kuopio, Finland

b

Received 6 February 1997; received in revised form 29 July 1997; accepted 31 July 1997

Abstract Antiplatelet treatment is well established in the secondary prevention of ischemic cerebrovascular events. The reasons for the occurrence of stroke, even though the patient receives drug treatment, have remained unclear. We performed a subgroup analysis of the European Stroke Prevention Study (ESPS 1) to investigate the efficacy of treatment in patients with different cardiovascular risk factors. The ESPS 1 recruited 1306 patients with TIA, RIND or stroke in one single center of Kuopio in Finland. The patients were treated with aspirin 990 mg / day1dipyridamole 225 mg / day or placebo for 2 years or until an endpoint. The endpoints were stroke or death from any cause. The risk factors for endpoints were analyzed by Cox univariate regression analyses. The effect of a single risk factor on the efficacy of antiplatelet therapy was analyzed by Cox proportional hazards model in subgroups according to the significant risk factors found in the univariate analysis. The efficacy of treatment was statistically significant in all subgroups except diabetics and current smokers for the stroke endpoint. When the combined endpoint of stroke or death was used, the treatment failed also to show statistical significance in patients with coronary heart disease. This lack of efficacy might be due to the small sample size. The efficacy of treatment was significantly better in TIA patients than in stroke patients when analyzed with stroke as an end point. This study provides strong evidence that the effectiveness of antiplatelet treatment in ischemic stroke and TIA patients is independent of most cardiovascular risk factors. Furthermore, all patients with TIA and ischemic stroke should receive secondary prevention regardless of whether or not there are risk factors.  1998 Elsevier Science B.V. Keywords: Antiplatelet treatment; Risk factors; Stroke; Transient ischemic attack

1. Introduction Antiplatelet therapy is well established in the secondary prevention of cerebrovascular events including both transient ischemic attacks (TIAs) and cerebral infarction (Antiplatelet Trialists’ Collaboration, 1994). However, antiplatelet treatment fails to prevent ischemic strokes in patients at risk despite the fact that most strokes are thromboembolic or atherothrombotic in origin. Age, male sex, arterial hypertension, diabetes mellitus, ischemic heart disease, and cigarette smoking have been shown to be risk factors for atherothrombotic cerebrovas*Corresponding author. Department of Neurology, South Karelia ¨ ¨ katu 14 A, 53130 Lappeenranta, Finland. Central Hospital, Valto Kakelan Tel.: 1358 56 1111; fax: 1358 56 115240. 0022-510X / 98 / $19.00  1998 Elsevier Science B.V. All rights reserved. PII S0022-510X( 97 )00229-3

cular events (Haberman et al., 1981; Wolf et al., 1983; Davis et al., 1987; Donnan et al., 1989; Sacco et al., 1989; Tuszynski et al., 1989; MacMahon et al., 1990; Shinton and Beevers, 1989). In the recent report of the Antiplatelet Trialists’ Collaboration, reductions in vascular events with antiplatelet therapy as secondary prevention were statistically significant in patients of middle and old age, in both sexes, in normotensive and hypertensive patients, and in both diabetic and nondiabetic patients (Antiplatelet Trialists’ Collaboration, 1994). However, the patients in the Antiplatelet Trialists’ Collaboration study represent a rather heterogeneous group of patients with various antiplatelet therapy regimens. There are very few single studies evaluating whether or not the treatment is equally efficient in patients with different cardiovascular risk factors indicating a good therapeutic response.

J. Puranen et al. / Journal of Neurological Sciences 154 (1998) 200 – 204

The European Stroke Prevention Study (ESPS 1) recruited 1306 patients with stroke or transient ischemic attack (TIA) in a single center, Kuopio, Finland. We performed a sub-group analysis based on these Finnish patients. The purpose of this study was to evaluate the efficacy of antiplatelet treatment in patients with and without known risk factors for ischemic stroke.

2. Subjects and methods The ESPS 1 was a randomized, double-blind, placebocontrolled European multicenter study with the participation of 16 centers in six countries. The aim of the ESPS 1 was to investigate whether the combination of dipyridamole (DP, Persantin), 75 mg three times a day, and acetylsalicylic acid (ASA), 330 mg three times a day, is more effective than placebo for the secondary prevention of stroke or death in patients with previous ischemic cerebral lesions. The study design and main results of this study have been previously published (ESPS Group, 1990). Of the 2500 patients included, 1306 patients (52%) were recruited by the University Hospital of Kuopio, 652 patients in the treatment group and 654 patients in the placebo group. The patients included had suffered a recent (,90 days) TIA, reversible ischemic neurological deficit (RIND), or stroke. TIA was defined as a neurological ischemic disorder with clinical deficits resolving fully within 24 h, and in RIND recovery occurred within 2–7 days. A vascular event was classified as stroke if the neurological deficit persisted for .7 days. In the present study, the patients with RIND were included in the stroke group. The treatment lasted 24 months, with follow-up visits every 3 months. The results were analyzed by the intention-to-treat analysis (Table 1). The end-points were stroke or stroke and death from any cause occurring during the 2-year follow-up. No differences were found between the treatment group and the placebo group with respect to age, sex, qualifying event, arterial territory, blood pressure, ischemic heart disease, previous myocardial infarction, diabetes, alcohol consumption or smoking habits (data not shown). A detailed medical history was obtained from the patients and their hospital records. Cox univariate regression analysis was applied to investigate the association of different risk factors and end-points. Age, male sex, Table 1 Distribution of qualifying events by treatment groups: intention-to-treat analysis (1306 patients)

TIA Stroke Total

DP-ASA (n)

Placebo (n)

262 390 652

254 400 654

DP-ASA, dipyridamole1aspirin; TIA, transient ischemic attack.

201

history of clinical diagnosis of hypertension, history of clinical diagnosis of diabetes, current cigarette smoking (vs. not smoking), alcohol consumption (vs. total abstinence), stroke at entry (vs. TIA), hemispheric event (vs. brainstem event), hypercholesterolemia (total cholesterol .6.5 mmol / l at entry), hypertriglyceridemia (triglycerides .2.5 mmol / l at entry), and a history of coronary heart disease (CHD) were included as independent variables. Definition of hypertension, diabetes, and coronary heart disease was based on clinical diagnosis. Cox proportional hazards model was performed to evaluate the treatment effect in all patients, and separately in subgroups including patients with and without a significant risk factor, the treatment effect being adjusted for all other confounding risk factors. Data analyses were conducted with the SPSS (Statistical Package for Social Sciences) program (SPSS for Unix: Advanced Statistics, Release 5.0). Student’s t-test for independent samples and x 2 -test were used to compare baseline data.

3. Results The mean (SD) age among the 1306 patients was 62.5 (10.2) years, 717 (54.9%) of the patients being men. During the 2-year follow-up, 148 patients suffered from a fatal or nonfatal stroke, 78 patients died from some other cause. Baseline characteristics at entry are shown in Table 2. Age, male sex, hypertension, diabetes, current smoking, stroke at entry, hemispheric event, and hypertriglyceridemia were associated with stroke only or stroke or death as end points by logistic regression analysis (Table 3). Coronary heart disease was a significant risk factor for the combined endpoint of stroke or death. Hypercholes-

Table 2 Baseline characteristics of study subjects included in intention-to-treat analysis Risk factor Qualifying event Stroke TIA Area of qualifying event Carotid Vertebrobasilar Male sex Hypertension Coronary heart disease Diabetes Current cigarette smoking Alcohol use Cholesterol level at entry .6.5 mmol / l Triglyceride level at entry .2.5 mmol / l

% 60.5 39.5 62.6 37.4 54.9 39.6 39.4 14.3 24.0 43.1 49.8 23.1

The risk factor for age was 62.5610.2 years (6standard deviation).

202

J. Puranen et al. / Journal of Neurological Sciences 154 (1998) 200 – 204

Table 3 Association of baseline risk factors with stroke or stroke and death in 1306 patients (Cox univariate regression analysis) Variable

Stroke

Age Stroke as qualifying event Carotid event a Male sex Hypertension Coronary heart disease Diabetes Current cigarette smoking Alcohol consumption Cholesterol .6.5 mmol / l Triglyceride .2.5 mmol / l

Stroke or death

HR (95% CI)

P value

HR (95% CI)

P value

1.02 (1.02, 1.03) 2.31 (1.60, 3.57) 1.90 (1.29, 2.83) 1.52 (1.08, 2.20) 1.47 (1.08, 2.14) 1.34 (0.95, 1.89) 2.07 (1.35, 3.11) 1.51 (1.06, 2.15) 1.36 (0.97, 1.92) 1.07 (0.76, 1.51) 2.01 (1.36, 2.83)

0.026 ,0.001 ,0.001 0.014 0.018 0.078 ,0.001 0.020 0.060 0.660 ,0.001

1.03 (1.01, 1.05) 2.10 (1.52, 2.90) 1.76 (1.28, 2.42) 1.71 (1.27, 2.32) 1.47 (1.05, 1.87) 1.38 (1.03, 1.84) 2.25 (1.58, 3.20) 1.38 (1.03, 1.85) 1.17 (0.88, 1.56) 1.01 (0.86, 1.35) 2.00 (1.46, 2.72)

,0.001 ,0.001 ,0.001 ,0.001 0.012 0.016 ,0.001 0.026 0.237 0.924 ,0.001

HR, hazard ratio; CI, confidence interval. a Stroke or TIA of carotid arterial distribution as qualifying event.

terolemia and alcohol consumption were not found to be risk factors for either endpoints. The treatment effect of DP-ASA for the reduction of stroke or a combined endpoint of stroke or death was evaluated by Cox proportional hazards regression analysis (Table 4). Unadjusted hazard ratio for treatment vs. placebo was 0.50 (CI (95% confidence interval) 0.36, 0.71, P,0.001)) when analyzed with stroke as an endpoint, and 0.61 (CI 0.46, 0.79, P,0.001) with a combined endpoint of stroke or death, indicating a risk reduction in the DP-ASA group. The treatment factor was also adjusted for all significant risk factors found in univariate evaluation.

Hazard ratios were 0.52 (CI 0.37, 0.73, P,0.001) and 0.63 (CI 0.48, 0.83, P,0.001), respectively. Likewise, the effect of a single risk factor on the efficacy of antiplatelet therapy was analyzed in subgroups according to the significant risk factors found in univariate analysis (Table 4). The efficacy of treatment was statistically significant in all subgroups except diabetics and current smokers. There were relatively few subjects in the diabetics and smoking subgroups, 187 and 313, respectively. When the combined endpoint of stroke or death was used, the treatment failed also to show statistical significance in patients with coronary heart disease.

Table 4 The effect of antiplatelet treatment on risk of stroke or stroke and death (Cox proportional hazards regression analysis) n

Unadjusted All patients Adjusted for confounding variables All patients Age $63 years Age ,63 years Stroke at entry TIA at entry Hemispheric event Brainstem event Male sex Female sex Hypertension No hypertension CHD No CHD Diabetes No diabetes Smoking No smoking Triglyceride .2.5 mmol / l Triglyceride #2.5 mmol / l

Stroke

Stroke or death

HR (95% CI)

P

HR (95% CI)

P

1306

0.50 (0.36, 0.71)

,0.001

0.61 (0.46, 0.79)

,0.001

1306 663 643 790 516 818 488 717 589 517 789 515 791 187 1119 313 993 302 1004

0.52 (0.37, 0.73) 0.56 (0.36, 0.88) 0.42 (0.25, 0.72) 0.61 (0.41, 0.89) 0.25 (0.11, 0.58) 0.58 (0.39, 0.85) 0.34 (0.16, 0.70) 0.55 (0.36, 0.85) 0.40 (0.22, 0.73) 0.37 (0.22, 0.63) 0.61 (0.38, 0.97) 0.60 (0.37, 0.99) 0.43 (0.26, 0.69) 0.63 (0.31, 1.26) 0.46 (0.31, 0.69) 0.67 (0.36, 1.23) 0.47 (0.31, 0.72) 0.43 (0.24, 0.77) 0.56 (0.37, 0.86)

,0.001 0.012 0.002 0.011 0.001* 0.006 0.004 0.007 0.003 ,0.001 0.035 0.046 ,0.001 NS ,0.001 NS ,0.001 0.005 0.008

0.63 (0.48, 0.83) 0.70 (0.50, 0.99) 0.51 (0.32, 0.79) 0.68 (0.50, 0.93) 0.46 (0.26, 0.81) 0.65 (0.48, 0.89) 0.53 (0.31, 0.91) 0.67 (0.48, 0.93) 0.51 (0.32, 0.83) 0.55 (0.37, 0.83) 0.64 (0.44, 0.93) 0.74 (0.50, 1.10) 0.51 (0.35, 0.74) 0.69 (0.40, 1.18) 0.59 (0.43, 0.81) 0.87 (0.53, 1.42) 0.56 (0.41, 0.78) 0.55 (0.35, 0.87) 0.67 (0.48, 0.94)

,0.001 0.042 0.003 0.017 0.007 0.008 0.020 0.017 0.007 0.004 0.018 NS ,0.001 NS 0.001 NS ,0.001 0.011 0.019

The treatment effect was adjusted for all other confounding risk factors listed in the table. HR, hazard ratio; CI, confidence interval; CHD, coronary heart disease; NS, not significant. *P,0.05 (the efficacy of treatment, TIA patients vs. stroke patients).

J. Puranen et al. / Journal of Neurological Sciences 154 (1998) 200 – 204

The efficacy of treatment was significantly better in TIA patients than stroke patients, when analyzed with stroke as the endpoint (P50.038). There was also a trend towards a better response in patients with hypertension vs. no hypertension, and in patients with a brainstem rather than a hemispheric event. This difference was not statistically significant.

4. Discussion The main result of our study based on secondary analysis of the ESPS 1 was that the effectiveness of antiplatelet treatment in ischemic stroke and TIA patients is quite independent of known cardiovascular risk factors for stroke. Most studies indicate that antiplatelet treatment provides protection resulting in a 15–30% reduction in the incidence of stroke after TIA or cerebral infarction (Antiplatelet Trialists’ Collaboration, 1994). It has been suggested that there are probably subgroups of patients in whom this treatment is particularly efficient and other subgroups in which prophylactic treatment might be expected to give only partial benefit (Chyatte and Chen, 1990). The rationale of our study was to evaluate the efficacy of antiplatelet treatment in patients with different cardiovascular risk factors in one single Finnish center of the ESPS 1. Drop-out rate among Finnish patients was much lower (19%) than in the other centers (46%), and the 1306 patients from the region’s 250 000 inhabitants are most probably more homogeneous than in the other 15 study centers in five countries that recruited another half of the total patient population. Arterial hypertension is the most important risk factor for ischemic stroke (Kannel et al., 1976, 1981; Hypertension Detection and Follow-up Program Cooperative Group, 1982; Davis et al., 1987; Sacco et al., 1989; Tuszynski et al., 1989; Collins et al., 1990; MacMahon et al., 1990). In addition, age (Sivenius et al., 1985; Davis et al., 1987; Passero et al., 1987; Dennis et al., 1989a), male sex (Haberman et al., 1981; Caplan et al., 1986; Davis et al., 1987), diabetes (Kannel and McGee, 1979; Fuller et al., 1983; Kannel and Wolf, 1983; Abbott et al., 1987; Davis et al., 1987; Sacco et al., 1989; Tuszynski et al., 1989; Kuusisto et al., 1994), and cigarette smoking (Abbott et al., 1986; Colditz et al., 1988; Wolf et al., 1988; Shinton and Beevers, 1989) have been found to be predictors of ischemic cerebral events. All of these variables were risk factors for stroke or stroke and death endpoint in the univariate analysis also in our study. Stroke event and hemispheric localization were strongly associated with both stroke and the combined endpoint. This finding is in accordance with previous reports (Turney et al., 1984; Dennis et al., 1989b; Hankey et al., 1991; Sivenius et al., 1991). Hypercholesterolemia showed no

203

significant relation to the risk of stroke or stroke and death. However, hypertriglyceridemia appeared to be a strong predictor for both endpoints. Hypertriglyceridemia often is associated with increased risk for thrombus formation and impaired fibrinolysis. A history of coronary heart disease showed a nonsignificant association with stroke endpoint. However, there was a significant association between coronary artery disease and the combined endpoint of stroke or death. We used the Cox proportional hazards model to evaluate the treatment effect as an independent variable for the endpoints. The treatment factor was not statistically significant in diabetics and in current smokers, when analyzed with stroke or stroke and death as the endpoint. The subgroups of patients with diabetes or current smoking were probably too small to permit statistical comparisons. However, there was a trend towards weaker response to treatment in both of these subgroups. Previous reports evaluating the efficacy of antiplatelet therapy in the secondary prevention of stroke in diabetics are contradictory (Baudoin et al., 1985; Sivenius et al., 1992; Antiplatelet Trialists’ Collaboration, 1994), and data concerning smokers are lacking. In a recent report (Bornstein et al., 1994), hyperlipidemia and ischemic heart disease predicted risk for stroke even though patients received aspirin treatment. In our study, the treatment effect was statistically nonsignificant in patients with CHD when analyzed with combined endpoints, and it was weaker also with stroke only as the endpoint in comparison with patients without CHD. The efficacy of treatment was significantly better in TIA patients versus stroke patients, when analyzed with stroke as the endpoint. However, the effect of therapy was significant in both patient groups. Patients with hypertriglyceridemia and also hypertensives actually responded better to DP-ASA than subjects without these risk factors. In addition, patients with a brainstem arterial event showed a better response than those with a hemispheric event. However, these differences between subgroups in efficacy were nonsignificant. The results according to the vascular territory and the presence of diabetes or the type of qualifying event agree with earlier findings in the ESPS (ESPS Group, 1990; Sivenius et al., 1991, 1992). We conclude that antiplatelet therapy seems to play a significant role in the secondary prevention of stroke independently of the existence of most cardiovascular risk factors. According to the results of this study, antiplatelet therapy is significantly better than placebo in patients older and younger than 63 years, in men and women, in TIA and stroke patients, in patients with hemispheric or brainstem lesion, and in patients with normal or elevated triglyceride level. Therapy was not effective in diabetic patients and in smokers. Patients with CHD seemed also not to respond to the therapy when stroke and death was the endpoint. The reason for this may be the fact that the effect of antiplatelet drugs is weak in the prevention of death. More data about

204

J. Puranen et al. / Journal of Neurological Sciences 154 (1998) 200 – 204

diabetics and smokers may be available from other large multicenter trials in the future. Meanwhile, all patients with TIA and ischemic stroke should receive secondary prevention with antiplatelet drugs.

Acknowledgements We would like to thank Pirjo Halonen, M.Sc., for her assistance in statistical analysis.

References Abbott, R.D., Yin, Y., Reed, D.M., Yano, K., 1986. Risk of stroke in male cigarette smokers. New Engl. J. Med. 315, 717–720. Abbott, R.D., Donahue, R.P., MacMahon, S.W., Reed, D.M., 1987. Diabetes and the risk of stroke. The Honolulu heart program. J. Am. Med. Assoc. 251, 949–952. Antiplatelet Trialists’ Collaboration 1994. Collaborative overview of randomised trials of antiplatelet therapy I: Prevention of death, myocardial infarction, and stroke by prolonged antiplatelet therapy in various categories of patients. Br. Med. J. 308, 81–106. Baudoin, C., Bousser, M.-G., Haguenau, M., Lefauconnier, J.-M., Eschwege, E., 1985. Secondary prevention of strokes: Role of platelet antiaggregant drugs in diabetic and non-diabetic patients. Diabetic Med. 2, 145–146. Bornstein, N.M., Karepov, V.G., Aronovich, B.D., Gorbulev, A.Y., Treves, T.A., Korczyn, A.D., 1994. Failure of aspirin treatment after stroke. Stroke 25, 275–277. Caplan, L.R., Gorelick, P.B., Hier, D.B., 1986. Race, sex and occlusive cerebrovascular disease: a review. Stroke 17, 648–655. Chyatte, D., Chen, T.L., 1990. Patterns of failure of aspirin treatment in symptomatic atherosclerotic carotid artery disease. Neurosurgery 26, 565–569. Colditz, G.A., Bonita, R., Stampfer, M.J., Willett, W.C., Rosner, B., Speizer, F.E., Hennekens, C.H., 1988. Cigarette smoking and risk of stroke in middle-aged women. New Engl. J. Med. 318, 937–941. Collins, R., Peto, R., MacMahon, S., Herbert, P., Fiebach, N.H., Eberlein, K.A., Godwin, J., Qizilbash, N., Taylor, J.O., Hennekens, C.H., 1990. Blood pressure, stroke, and coronary heart disease: Part 2, short-term reductions in blood pressure: overview of randomised drug trials in their epidemiological context. Lancet 335, 827–838. Davis, P.H., Dambrosia, J.M., Schoenberg, B.S., Schoenberg, D.G., Pritchard, D.A., Lilienfeld, A.M., Whisnant, J.P., 1987. Risk factors for ischemic stroke: a prospective study in Rochester, Minnesota. Ann. Neurol. 22, 319–327. Dennis, M.S., Bamford, J.M., Sandercock, P.A., Warlow, C.P., 1989. Incidence of transient ischemic attacks in Oxfordshire, England. Stroke 20, 333–339. Dennis, M.S., Bamford, J.M., Sandercock, P.A., Warlow, C.P., 1989. A comparison of risk factors and prognosis for transient ischemic attacks and minor ischemic strokes. The Oxfordshire Community Stroke Project. Stroke 20, 1494–1499. Donnan, G.A., McNeil, J.J., Adena, M.A., Doyle, A.E., O’Malley, H.M., Neill, G.C., 1989. Smoking as a risk factor for cerebral ischaemia. Lancet ii, 643–647. ESPS Group 1990. European stroke prevention study. Stroke 21, 1122– 1130. Fuller, J.H., Shipley, M.J., Rose, G., Jarrett, R.J., Keen, H., 1983. Mortality from coronary heart disease and stroke in relation to degree of glycaemia: the Whitehall study. Br. Med. J. 287, 867–870.

Haberman, S., Capildeo, R., Rose, F.C., 1981. Sex differences in the incidence of cerebrovascular disease. J. Epidemiol. Community Health 35, 45–50. Hankey, G.J., Slattery, J.M., Warlow, C.P., 1991. The prognosis of hospital-referred transient ischaemic attacks. J. Neurol. Neurosurg. Psychiatry 54, 793–802. Hypertension Detection and Follow-up Program Cooperative Group 1982. Five-year findings of the Hypertension Detection and Follow-up Program: III. Reduction in stroke incidence among persons with high blood pressure. J. Am. Med. Assoc. 247, 633–638. Kannel, W.B., McGee, D.L., 1979. Diabetes and cardiovascular disease: The Framingham Study. J. Am. Med. Assoc. 241, 2035–2038. Kannel, W.B., Wolf, P.A., 1983. Epidemiology of cerebrovascular disease. In: Ross-Russell, R.W. (Ed.), Vascular Disease of the Central Nervous System, 2nd ed. Churchill Livingstone, Edinburgh, pp. 1–24. Kannel, W.B., Dawber, T.R., Sorlie, P., Wolf, P.A., 1976. Components of blood pressure and risk of atherothrombotic brain infarction: The Framingham Study. Stroke 7, 327–331. Kannel, W.B., Wolf, P.A., McGee, D.L., Dawber, T.R., McNamara, P., Castelli, W.P., 1981. Systolic blood pressure, arterial rigidity, and risk of stroke: The Framingham Study. J. Am. Med. Assoc. 245, 1225– 1229. ¨ ¨ ¨ ¨ K., Laakso, M., 1994. Non-insulinKuusisto, J., Mykkanen, L., Pyorala, dependent diabetes and its metabolic control are important predictors of stroke in elderly subjects. Stroke 25, 1157–1164. MacMahon, S., Peto, R., Cutler, J., Collins, R., Sorlie, P., Neaton, J., Abbott, R., Godwin, J., Dyer, A., Stamler, J., 1990. Blood pressure, stroke, and coronary heart disease: part I, prolonged differences in blood pressure: prospective observational studies corrected for the regression dilution bias. Lancet 335, 765–774. Passero, S., Rossi, G., Nardini, M., Bonelli, G., D’Ettorre, M., Martini, A., Battistini, N., Albanese, V., Bono, G., Brambilla, G.L., Candelise, L., De Zanche, L., Inzitari, D., Rasura, M., Fieschi, C., 1987. Italian multicenter study of reversible cerebral ischemic attacks. Part 5. Risk factors and cerebral atherosclerosis. Atherosclerosis 63, 211–224. Sacco, R.L., Foulkes, M.A., Mohr, J.P., Wolf, P.A., Hier, D.P., Price, T.R., 1989. Determinants of early recurrence of cerebral infarction. The Stroke Data Bank. Stroke 20, 983–989. Shinton, R., Beevers, G., 1989. Meta-analysis of relation between cigarette smoking and stroke. Br. Med. J. 298, 789–794. ¨ ¨ ¨ K., Salonen, J., Riekkinen, P., 1985. Sivenius, J., Heinonen, O.P., Pyorala, The incidence of stroke in the Kuopio area of East Finland. Stroke 16, 188–192. Sivenius, J., Riekkinen, P.J.S., Smets, P., Laakso, M., Lowenthal, A., 1991. The European Stroke Prevention Study (ESPS): results by arterial distribution. Ann. Neurol. 29, 596–600. Sivenius, J., Laakso, M., Riekkinen, P., Smets, P., Lowenthal, A., 1992. European Stroke Prevention Study: Effectiveness of antiplatelet therapy in diabetic patients in secondary prevention of stroke. Stroke 23, 851–854. Turney, T., Garraway, W.M., Whisnant, J.P., 1984. The natural history of hemispheric and brainstem infarction in Rochester, Minnesota. Stroke 15, 790–794. Tuszynski, M.H., Petito, C.K., Levy, D.E., 1989. Risk factors and clinical manifestations of pathologically verified lacunar infarctions. Stroke 20, 990–999. Wolf, P.A., Kannel, W.B., Verter, J., 1983. Current status of risk factors for stroke. In: Barnett, H.J.M. (Ed.), Neurologic Clinics. Cerebrovascular Disease, vol. 1, no. 1. W.B. Saunders Co., Philadelphia, pp. 1317– 1343. Wolf, P.A., D’Agostino, R.B., Kannel, W.B., Bonita, R., Belanger, A.J., 1988. Cigarette smoking as a risk factor for stroke. J. Am. Med. Assoc. 259, 1025–1029.