Lipoprotein(a) in Cerebrovascular and Coronary Atherosclerosis

Transcription

1 Hiroshima J. Med. Sci. Vol.44, No.4, , December, 1995 HIJM Lipoprotein(a) in Cerebrovasclar and Coronary Atherosclerosis Shichi NOMURA First Department of Internal Medicine, Hiroshima University School of Medicine, Kasmi, Minami-k, Hiroshima 734, Japan ABSTRACT Serm lipoprotein(a) [Lp(a)] was determined in 85 healthy controls (control grop), in 49 patients with cerebrovasclar disease (CVD grop), and in 87 patients with acte myocardial infarction (AMI grop). Lp(a) concentration was measred sing a single radial immnodiffsion method. Becase Lp(a) showed a skewed distribtion, Lp(a) was plotted in each grop on a box plot and analyzed sing non-parametric methods. The following reslts were obtained: 1) Lp(a) levels were significantly higher in both the AMI and CVD grops than in the control grop. 2) The CVD grop was divided into 4 grops: [1] cerebral hemorrhage (HEM); [2] cardioembolic infarction (EMB); [3] lacnar infarction (LAC); and [ 4] atherothrombotic infarction (THR). In the THR grop, Lp(a) concentration was higher than those in the control and LAC grops. 3) In the CVD grop, patients with an elevated Lp(a) vale showed severe lesions in the major cerebral arteries evalated by cerebral arteriograms. 4) In the AMI patients who nderwent coronary angiography, the Lp(a) level showed a marked increase with an increase in the nmber of affected vessels. The correlation between coronary index (CI) and Lp(a) was also investigated. The lesion severity of coronary arteries was correlated with serm Lp(a) concentration. These reslts sggest that a high Lp(a) vale is linked to atherosclerosis of the cerebral and coronary arteries, and inflences its severity. Key words: Lipoprotein( a), Atherosclerosis, Angiography, Box plot Lipoprotein(a) [Lp(a)] is a serm lipoprotein first reported by Berg in ). It is similar to low density lipoprotein (LDL) in lipid composition and the presence of apolipoprotein B-1 (apo B-1). However, in addition to apo B-1, Lp(a) contains a highly glycosylated protein called apolipoprotein (a) [apo(a)], which is associated with apo B-1 via a dislfide link 1,l 2, 29 ). Since the stdy of Dahlen et al in 1972, Lp(a) has attracted the attention of many researchers becase a high Lp(a) level is associated with coronary atherosclerosis6,l5,23). In addition to coronary disease, evidence is accmlating that Lp(a) may be of importance in cerebrovasclar disease 2 o, 3 o, 3 2). Frthermore, the cdna seqence of apo(a) has been determined and shows a remarkable homology with hman plasminogen 9,l9). This finding may provide a direct link between thrombogenesis and progressive atherosclerosis 5, 25 ). In ischemic heart disease, some reports sggest that high levels of serm Lp(a) are related to the significantly greater severity of lesions qantified by coronary angiography 7,ll)_ However, there are few reports on the association between Lp(a) con- centration and lesion severity, evalated by angiography, in cerebrovasclar disease. In or department, Ohtski stdied the correlation between serm Lp(a) level and the stenosis of coronary lesions by angiography, and Konemori examined Lp(a) level in cerebral infarction, diagnosed by brain magnetic resonance imaging (MRI) 14, 22 ). Their analyses differ from those of others in sing box-and-whisker plots (box plots) 28 >. This method is sefl in analyzing data which is not of normal distribtion, sch as Lp(a) concentration. In the present stdy, the athor investigated the association between the Lp(a) level and angiographically assessed atherosclerosis in cerebrovasclar disease and coronary artery disease, sing box plots. MATERIALS AND METHODS Sbjects. All the sbjects were males nder 8 years of age. Serm lipid and Lp(a) concentrations were determined in 49 consective patients with cerebrovasclar disease (CVD grop) admitted to the Stroke Care Unit (SCU) of the National Cardio-

2 134 S. Nomra vasclar Center (NCVC), and in 87 consective patients with acte myocardial infarction (AMI grop) admitted to the Coronary Care Unit (CCU) of NCVC. The CVD patients were diagnosed by angiography as well as by brain compted tomography (CT) and magnetic resonance imaging (MRI). They were divided into 4 sbgrops: 1) cerebral hemorrhage (HEM); 2) cardioembolic infarction (EMB); 3) lacnar infarction (LAC); and 4) atherothrombotic infarction (THR), according to the 199 Classification of Cerebrovasclar Diseases III developed by the Ad Hoc Committee of the National Institte of Nerological Disorders and Stroke (NINDS) 21 ). The AMI sbjects were diagnosed by electrocardiography, serm CPK level, echocardfography, and angiography. A total of 85 apparently healthy individals examined for a medical check p were sed as controls (control grop). No patients received any drgs affecting the serm Lp(a) vale, sch as nicotinic acid. Lipid and lipoprotein determinations. Blood samples were drawn from the fasting patients early in the morning following hospital admission. The serm was isolated by low-speed centrifgation. Very low density lipoprotein (VLDL) was separated after ltracentrifgation for 18 hors at 4, rpm at a solvent density of 1.6 g/ml, as described previosly 31 ). The levels of cholesterol (Ch) and triglycerides (TG) in the serm and the VLDL fraction were determined by the enzymatic method. High density lipoprotein (HDL) cholesterol was measred by a precipitation method. The low density lipoprotein (LDL) cholesterol level was calclated as the serm cholesterol concentration mins the sm of VLDL-Ch and HDL-Ch levels. Qantitation of Lp(a) concentration. Serm Lp(a) levels were measred sing a single radial immnodiffsion (SRID) techniqe with a mono-specific antibody against hman Lp(a). This antibody was kindly provided by Dr. S. N. Pokrovsky (Institte of Experimental Car-. diology, Cardiology Research Center, Moscow, Rssia). The gel, at a volme of 6 ml per plate, consisted of.5m barbital bffer containing 1 % (W N) agarose and 12 µl of antibody against Lp(a) (2 mg/dl). Wells of 2.5 mm diameter were pnched ot, and 4 µl of standard or sample was added to each well. After 72 hors, the ringshaped immno-precipitates were measred in a tenth of a millimeter. The relationship between the Lp(a) level and the sqared diameter of the precipitate ring was linear in the range of mg/dl. Samples exceeding this range were dilted with.15 M NaCl. Angiography. A total of 42 CVD patients nderwent for-vessel cerebral angiography. The cerebral arteries were divided into extracranial and intracranial arteries as described by Mathew et al 18 ). In this stdy, the evalation of atherosclerotic lesions was restricted to the extracranial arteries and major intracranial arteries. In 76 AMI patients, coronary angiography was performed sing the Jdkins or Sones techniqe13,26). The angiographic findings were evalated by two different methods: 1) Patients were sbdivided into grops with one-, two-, and threevessel disease based on 75% or greater narrowing of the translminal diameter of the left anterior descending artery, the left circmflex artery, and the right coronary artery. 2) Coronary arteries were divided into segments according to the American Heart Association Committee Report 2 ). In each segment, the extent of stenosis was evalated and the coronary index (Cl) was calclated by modified Balcon's method 3 ). In CI, the lower the score, the greater the severity of the lesions. Data analysis. Lp(a) was plotted in each grop on a box plot, as reported by Ohtski 22 ) and Konemori 14 ). With the box plot method, a box is formed by the 25th percentile and 75th percentile vales that enclose the median. The distribtion is expressed by widening the box to 1.5 times the 25th percentile and 75th percentile vales, respectively. Comparison of the distribtion of Lp(a) and other lipids between the two grops was performed by the Mann-Whitney test, and among the three grops the Krskal-Wallis test was applied. The relation between Lp(a) level and the coronary index Age (yr) Cholesterol Triglyceride HDL-Ch Table 1. Characteristics of the sbjects control (n=85) AMI (n=87) 48 61* (42-53) (52-68) ( ) ( ) (74-163) (74-162) 45 34* (38-53) (31-43) CVD (n=49) 59* (55-63) 187 (154-21) 124 (96-188) 35* (3-42) AMI, acte myocardial infarction grop; CVD, cerebrovasclar disease grop; Ch, cholesterol; n=nmber of sbjects. Lipid vales are expressed in mg/dl. Vales are median (25th percentile - 75th percentile). *p<.1 (vs. contol, Mann-Whitney test)

3 Lipoprotein(a) in Cerebrovasclar and Coronary Atherosclerosis p<.1 p<.5 I ~ 8 s c 6 ~ E Q) 4 c c 2 a:...j 8,..., =a... C> 5 4.p rn 6 p< I- +-' Q) 2,..., ~ 1 a....j control AMI CVD Fig. 1. Distribtions of serm Lp(a) concetrations in the control, AMI, and CVD grops. Control, control grop; AMI, acte myocardial infarction grop; CVD, cerebrovasclar disease grop. control HEM EMB LAC THR Fig. 2. Distribtions of serm Lp(a) levels in the CVD sbgrops. HEM, cerebral hemorrhage; EMB, cardioembolic infarction; LAC, lacnar infarction; THR, atherothrombotic infarction. was investigated by Spearman's rank correlation coefficient. The level of significance was set at. 5 for all analyses. RESULTS The serm lipoprotein profile in each grop is shown in Table 1. AMI and CVD grops were older than the control grop. HDL-Ch was significantly lower in the AMI and CVD grops compared to that in the control grop, whereas there were no differences in the serm cholesterol and triglyceride levels. The distribtions of serm Lp(a) concentration in the three grops are shown by parallel box plot in Fig. 1. In the control grop, the 25th percentile vale, 5th percentile vale and 75th percentile vale of serm Lp(a) were 2. mg/dl, 6. mg/dl, and 11.5 mg/dl, respectively. Lp(a) did not show a normal distribtion. The median bf Lp(a) in the AMI and CVD grops was 9.3 mg/dl and 9. mg/dl, respectively. The serm Lp(a) concentrations were significantly higher in both the AMI and CVD grops than in the control grop. Since the age distribtion of the AMI and CVD grops was different from that of the control grop, the sbjects in each grop were groped by decade. There was no difference in the Lp(a) level among these 1-year age grops, and the Lp(a) concentration was higher in all the AMI and CVD sbgrops compared to the control grop (data not shown). The CVD grop was frther divided into the following 4 sbgrops: 1) 1 patients with cerebral hemorrhage (HEM grop); 2) 12 patients with cardioembolic infarction (EMB grop); 3) 18 patients with lacnar infarction (LAC grop); and 4) 9 patients with atherothrombotic infarction (THR grop). The lipid levels in each CVD sbgrop were not so different from those of the CVD patients (Table 2). The median of Lp(a) concentration in the HEM, EMB, LAC, and THR grops was 13.5 mg/dl, 8.7 mg/dl, 7.2 mg/dl, and 17.2 mg/dl, respectively. In the THR grop, the serm Lp(a) level was significantly higher than those in the control and LAC grops, bt Lp(a) levels in both the HEM and EMB grops did not significantly differ from that in the control grop (Fig. 2). Table 2. Serm lipid levels in the CVD sbgrops Age (yr) Cholesterol Triglyceride VLDL-Ch LDL-Ch HDL-Ch HEM (n=lo) 59* (52-6) 194 (178-2) 17 (7-128) 13 (6-17) 135 ( ) 43 (35-56) Abbreviations are as in Table 1. *p<.1, **p<.5 (vs. contol, Mann-Whitney test) EMB (n=12) 62* (53-64) 159** (15-191) 122 (19-2) 22 (16-25) 98 (9-12) 33* (25-39) LAC (n=18) 62* (57-63) 184 (142-2) 118 (96-14) 18 (13-24) 123 (79-145) 37* (28-45) THR (n=9) 57* (52-58) 197 ( ) 184 (11-26) 22 (14-32) 136 ( ) 32* (31-34)

4 136 S. Nomra Table 3. Relationship between lipid level and severity of cerebral artery lesions by angiography Stenosis ~ 5% (n=31) Stenosis>5% (n=ll) Age (yr) 59 (56-63) 59 (56-63) Cholesterol 187 (157-22) 181 (15-199) Triglyceride 118 (93-151) 177 (16-189) VLDL-Ch 18 (11-23) 22 (11-3) LDL-Ch 122 (9-145) 128 (1-147) HDL-Ch 38 (31-47) 32 (25-35) Abbreviations are as in Table 1..--,, 6 5 =a... Cl 5 4 ~ (I).--,, Ill : ' p<.5 In the 42 CVD sbjects, for-vessel cerebral angiography was performed and the relationship between serm lipid level and the extent of lesions in the major extracranial and intracranial arteries was investigated. These patients were divided into two grops according to the extent of lesions. Eleven patients had more than 5% stenosis of translminal diameter (severe stenosis) and 31 patients had mild stenosis ( <5%). There was no difference among these sbgrops in the lipid and lipoprotein parameters (Table 3). The median of Lp(a) level in patients with mild stenosis and severe stenosis was 9.5 mg/dl, and 16. mg/dl, respectively. The Lp(a) vale in patients with severe stenosis was significantly higher compared to the control grop (Fig. 3). In the AMI patients who nderwent coronary angiography, the association between the nmber of affected vessels and serm lipid was examined (Table 4 and Fig. 4). Thirty two patients had onevessel disease, 32 had two-vessel disease, and 12 had three-vessel disease. The median vale of Lp(a) was 6. 7 mg/dl in the one-vessel disease grop, 11. mg/dl in the two-vessel disease grop, and 22.1 mg/dl in the three-vessel disease grop, indicating that the serm Lp(a) concentations increased with an increase in the nmber of affected vessels (Fig. 4). On the other hand, no relationship was observed between the nmber of control Stenosis:;;so% Stenosis>SO% Fig. 3. Distribtions of Lp(a) levels in the CVD sbjects with mild and severe stenosis, stdied by cerebral angiography. 1 8,... =a... C) 5 6.;:::; e +' (I),... Ill : ' VD 2VD 3VD Fig. 4. Distribtions of Lp(a) levels in the AMI patients with one-, two-, and three-vessel disease. lvd, one-vessel disease; 2VD, two-vessel disease; 3VD, three-vessel disease. Table 4. Relationship between nmber of affected vessels and serm lipid level nmber of affected vessels 1 (n=32) 2 (n=32) 3 (n=12) Age (yr) 62 (51-68) 59 (52-65) 61 (52-65) Cholesterol 186 (158-29) 23 ( ) 193 ( ) Triglyceride 123 (72-168) 15 (75-172) 14 (88-119) VLDL-Ch 16 (7-27) 15 (8-24) 13 (11-18) LDL-Ch 128 (17-152) 147 ( ) 138 (126-17) HDL-Ch 36 (31-44) 33 (31-39) 4 (31-48) Abbreviations are as in Table 1.

5 Lipoprotein(a) in Cerebrovasclar and Coronary Atherosclerosis ~~m> ~~ >< Q) "'C 6 i_> oo<po..6 om> o c9 ~ 5 8 «I Oo e 4 «l>~oog Lp(a) concentration (mg/di) Fig. 5. Correlation between Lp(a) concentration and coronary index in the AMI grop. vessels with stenosis and variables other than Lp(a) (Table 4). Correlation between the coronary index and Lp(a) level was also evalated in the AMI grop (Fig. 5). Lp(a) vale was inversely correlated with CI; CI tended to decrease with an increase in the Lp(a) vale (r 8 =-.41, p<.1). DISCUSSION The distribtion of serm Lp(a) concentrations in the control grop showed no linearity on the Q-Q plot, and some markedly high vales were observed (data not shown). Ths, Lp(a) did not show a normal distribtion. This reslt is consistent with Ohtski's report 22 ). Therefore, it was not appropriate in this analysis to se the moment indicators, sch as mean or standard deviation. Other lipids also did not necessarily show normal distribtions. The distribtions of lipids were therefore assessed sing qantile indicators instead of moment indicators. Lp(a) was plotted in three grops on parallel box plots (Fig. 1). The median of Lp(a) in the control, AMI, and CVD grops was 6. mg/dl, 9.3 mg/dl, and 9. mg/dl, respectively. Lp(a) level was higher in the AMI and CVD grops than in the control grop. This is consistent with other reportsls, 23, 32 ). Althogh the serm Lp(a) concentration has been shown to increase gradally after the acte episode of myocardial infarction, reaching a maximm at 11 days, the Lp(a) level within 24 hors after the onset of the event changes little from the initial vale 17 ). Therefore, sbjects in this stdy were limited to patients admitted ::; 24 hors after onset, and samples were collected from the fasting patients in the morning following admission. Moreover, or reslts seem to be naffected by age differences among the control, AMI, and CVD grops, becase no relationship cold be demonstrated between Lp(a) level and age8,16,24). Since cerebrovasclar disease is cased by several different etiologies, the CVD individals were divided into 4 sbgrops: HEM grop; EMB grop; LAC grop; and THR grop. The categorization was based on the Classification of Cerebrovasclar Diseases III developed by the Nationaf Institte of Nerological Disorders and Stroke (NINDS-III) of the National Instittes of Health (NIH) 21 ). In this classification, lacnar infarction is treated as an independent category of cerebral infarction for the first time. Atherothrombotic infarction is mainly cased by atherosclerotic lesions, which are fond in extracranial and major intracranial arteries, whereas lacnar infarction is small infarct mainly cased by lipohyalinosis and microatheroma of penetrating arteries 1, 27 ). In or stdy, the Lp(a) concentration was higher in the THR grop than in the control grop and LAC grop. These data are consistent with the reports of Mrai et al and Woo et a1 2 o, 3 o). However, Konemori reported that, by mltiple logistic regression analysis, Lp(a) did not show a significant odds ratio in cerebral infaction 14 ). Konemori diagnosed cerebral infarction by MRI, whereas in the present stdy, cerebral angiography as well as MRI was applied for the diagnosis of the CVD patients, and the atherosclerosis of the large cerebral arteries was investigated in the atherothrombotic infarction. Therefore, the discrepancy seems to be not only de to the difference in the method of analysis bt also de to the size and the portion of analyzed vessels. Zenker et al performed B-mode and Doppler ltrasonography to examine the vasclar stenosis of extracranial cerebral arteries, and showed that Lp(a) correlated with the score of lesion severity in cerebrovasclar disease 32 ). However, there are few reports on the association between the Lp(a) level and the lesion severity qantified by cerebral arteriography. In this stdy, the stenosis of extracranial arteries and major intracranial arteries was evalated by angiography in 42 CVD patients, and the relationship between Lp(a) vale and the extent of lesions was investigated. The Lp(a) level in sbjects with severe stenosis was significantly higher than that of the control grop. Ths, patients with a high Lp(a) concentration had severe stenosis of the large cerebral arteries, and this reslt is similar to Zenker's report where the evalation was made by ltrasonography. In the AMI grop, the association between Lp(a) concentration and angiographically assessed coronary artery stenosis was also stdied. Two different approaches were sed to investigate the coronary artery stenosis. In the first approach, patients were divided into grops based on the nmber of arteries with 75% or greater stenosis. The serm Lp(a) concentration

6 138 S. Nomra increased with an increase in the nmber of vessels with stenosis. On the other hand, there was no relationship between the severity of lesions and other lipid levels. These are the same as Ohtski's reslts in or department 22 ). This approach is sally sed in the evalation of lesion severity of coronary arteries. However, it is not possible to assess the atherosclerosis in mltiple small lesions by this method. Therefore, in the second approach, the stenosis of each segment was determined, and each score was smmed as the coronary index. There are some papers on Lp(a) in which the severity of lesions was scored 7 ' ll). However, the method sed in the present stdy was different from theirs. The size of the vessel in the analysis of lesions was also considered, becase the weighting of atherosclerosis in large vessels is not similar to that in small vessels. The severity of coronary artery disease appeared to correlate well with the Lp(a) level. In this stdy, the athor determined Lp(a) concentrations in the CVD and AMI grops and evalated the relationship between atherosclerosis and Lp(a) levels by cerebral and coronary angiography. The observations sggest that Lp(a) is not only a risk factor for coronary and cerebral atherosclerosis bt may also be related to the progression of atherosclerosis. ACKNOWLEDGMENTS The athor wold like to thank Dr. Goro Kajiyama, Professor of the First Department of Internal Medicine, Hiroshima University School of Medicine, and Dr. Tak Yamamra and Dr. Akira Yamamoto, Department of Etiology and Pathophysiology, National Cardiovasclar Center Research Institte, for their excellent advice and gidance. (Received September 1, 1995) (Accepted November 3, 1995) REFERENCES 1. Armstrong, V.W., Walli, A.K. and Seidel, D Isolation, characterization, and ptake in hman fibroblasts of an apo(a)-free lipoprotein obtained on redction of lipoprotein(a). J. Lipid Res. 26: Asten, W.G., Edwards, J.E., Frye, R.L., Gensini, G.G., Gott, V.L., Griffith, L.S.C., McGoon, D.C., Mrphy, M.L. and Roe, B.B A reporting system on patients evalated for coronary artery disease. Circlation 51(Sppl.): Balcon, R., Catell, M.R., Stone, D.L. and Ferlicht, G A compter generated index for the assessment of coronary angiography. Acta Med. Scand. 615(Sppl.): Berg, K A new serm type system in man - the Lp system. Acta Pathol. Microbial. Scand. 59: Brown, M.S. and Goldstein, J.L Teaching old dogmas new tricks. Natre 33: Dahlen, G., Ericson, C., Frberg, C., Lndkvist, L. and Svardsdd, K Stdies on an extra pre-beta lipoprotein fraction. Acta Med. Scand. 53l(Sppl.): Dahlen, G.H., Gyton, J.R., Attar, M., Farmer, J.A., Katz, J.A. and Gotto, A.M. Jr Association of levels of lipoprotein Lp(a), plasma lipids, and other lipoproteins with coronary artery disease docmented by angiography. Circlation 74: Dahlen, G.H Incidence of Lp(a) lipoprotein among poplations, p In A. M. Scan (ed.), Lipoprotein(a). Academic Press, San Diego. 9. Eaton, D.L., Fless, G.M., Kohr. W.J., McLean, J.W., X, Q-T., Miller, C.G., Lawn, R.M. and Scan, A.M Partial amino acid seqence of apolipoprotein(a) shows that it is homologos to plasminogen. Proc. Natl. Acad. Sci. USA. 84: Fisher, C.M Lacnaes: small, deep cerebral infarcts. Nerology 15: Frick, M.H., Dahlen, G., Berg, K., Valle, M. and Hekali, P Serm lipids in angiographically assessed coronary atherosclerosis. Chest 73: Gabatz, J.W., Heideman, C., Gotto, A.M. Jr., Morrisett, J.D. and Dahlen, G.M Hman plasma lipoprotein(a): strctral properties. J. Biol. Chem. 258: Jdkins, M.P Selective coronary arteriography. Part 1. A perctaneos transfemoral technic. Radiology 89: Konemori, G Lipoprotein(a) and other risk factors for cerebral infarction. Hiroshima J. Med. Sci. 44: Kostner, G.M., Avogaro, P., Cazzolato, G., Marth, E., Bittolo-Bon, G. and Qinici, G.B Lipoprotein Lp(a) and the risk for myocardial infaction. Atherosclerosis 38: Laber, C., De Bacqer, D., De Backer, G., Vincke, J., Myldermans, L., Vandekerckhove, Y., Van der Stichele, E. and Rossene, M Plasma lipoprotein(a) vales and severity of coronary artery disease in a large poplation of patients ndergoing coronary angiography. Clin. Chem. 38: Maeda, S., Abe, A., Seishima, M., Makino, K., Noma, A. and Kawade, M Transient changes of serm lipoprotein(a) as an acte phase protein. Atherosclerosis 78: Mathew, N.T., Davis, D., Meyer, J.S. and Chandar, K Hyperlipoproteinemia in occlsive cerebrovasclar disease. JAMA 232: McLean, J.W., Tomlinson, J.E., Kang, W-J., Eaton, D.L., Chen, E.Y., Fless, G.M., Scan, A.M. and Lawn, R.M cdna seqence of hman apolipoprotein(a) is homologos to plasminogen. Natre 33: Mrai, A., Miyahara, T., Fjimoto, N., Matsda, M. and Kameyama, M Lp(a) lipoprotein as a risk factor for coronary heart disease and cerebral infarction. Atherosclerosis 59: National Institte of Nerological Disorders

7 Lipoprotein(a) in Cerebrovasclar and Coronary Atherosclerosis 139 and Stroke Ad Hoc Committee Classification of cerebrovasclar diseases III. Stroke 21: Ohtski, T Distribtion of serm lipoprotein(a) levels - a non-parametric analysis-. Hiroshima J. Med. Sci. 42: Rhoads, G.G., Dahlen, G., Berg, K., Morton, N. E. and Dannenberg, A.L Lp(a) lipoprotein as a risk factor for myocardial infarction. JAMA 256: Schriewer, H., Assmann, G., Sandkamp, M. and Schlte, H The relationship of lipoprotein(a) (Lp(a)) to risk factors of coronary heart disease: initial reslts of the prospective epidemiological stdy on company employees in Westfalia. J. Clin. Chem. Clin. Biochem. 22: Scott, J Thrombogenesis linked to atherogenesis at last? Natre 341: Sones, F.M. Jr. and Shirey, E.K Cine coronary arteriography. Mod. Concepts. Cardiovasc. Dis. 31: Toole, J.F Hypertension, lacnar infarction, and hypertensive encephalopathy, p In Cerebrovasclar disorders, 3rd ed. Raven Press, New York. 28. Tkey, J.W Exploratory data analysis, 1st ed. p.39-43, Addison Wesley. 29. Utermann, G. and Weber, W Protein composition of Lp(a) lipoprotein from hman plasma. FFBS. Lett. 154: Woo, J., La, E., Lam, C.W.K., Kay, R., Teoh, R., Wong, H.Y., Prall, W.Y., Kreel, L. and Nicholls, M.G Hypertension, lipoprotein(a), and apolipoprotein A-I as risk factors for stroke in the Chinese. Stroke 22: Yamamra, T., Sdo, H., Ishikawa, K. and Yamamoto, A Familial type I hyperlipoproteinemia cased by apolipoprotein C-II deficiency. Atherosclerosis 34: Zenker, G., Koltringer, P., Bone, G., Niederkorn, K., Pfeiffer, K. and Jrgens, G Lipoprotein(a) as a strong indicator for cerebrovasclar disease. Stroke 17: