Reduced serum lipoprotein(a) levels in patients with primary biliary cirrhosis

ATHEROSCLEROSIS ELSEVIER SCIENCE ,RFLAUD

Atherosclerosis IO5 ( 1994) 43-50

Reduced serum lipoprotein(a) levels in patients with primary biliary cirrhosis Wendy L. Gregory*a, Frances L. Gameb, Martin Farrerc, Michael F. Lakerb, Oliver F.W. Jamesd

Jeffrey R. Idle”,

Departments of aPharmacological Sciences. bClinical Biochemistry, ‘Cardiology, dA4edicine. University of Newcastle upon Qne, The Medical School, Newcastle upon Tyne. NE2 4HH. UK

(Received 16 July 1993; accepted 16 September 1993)

Abstract Lipoprotein(a) (Lp(a)) is a unique lipoprotein, elevated serum levels of which are independently associated with an increased risk of coronary heart disease (CHD). Primary billary cirrhosis (PBC) is often associated with high serum

cholesterol, itself a risk factor for CHD. Despite this, patients with PBC are thought to have a lower than expected incidence of CHD. We hypothesised that this may be related to low serum levels of Lp(a) in PBC patients. This was investigated by collecting fasting blood samples from 42 patients with PBC, 39 age- and sex-matched subjects with nonPBC liver disease and 432 community control subjects. Serum was analysed for total cholesterol, triglycerides, high density lipoprotein (HDL) cholesterol and apolipoproteins Al and B (apo Al and apo B). Lp(a) was measured by an enzyme-liked immunosorbent assay (ELISA) technique. There was a significant reduction of Lp(a) concentrations in the PBC group compared with the healthy controls (median value 28.5 mg/l vs. 75.0 mg/l, P < 0.005) and between the non-PBC liver disease group (median value 52.0 mg/l) and control group (P = 0.001). Within both the liver disease and PBC patient groups there were significant negative correlations between Lp(a) levels and bilirubin (R = -0.564, P < 0.001 and R = -0.395, P = 0.010 respectively). This preliminary study has demonstrated reduced Lp(a) levels in PBC patients which may be a contributory factor to explain a possible cardioprotective effect in such patients, despite elevated LDL cholesterol levels. Kqv words:

Liver disease; Lipoproteins;

Cholestasis; Coronary heart disease

1. Introduction Primary biliary cirrhosis is a chronic liver disease of unknown aetiology characterised by intrahepatic cholestasis [I]. It is thought to be * Corresponding author. Tel.: 091 222 6000 ext. 6770; Fax: 091 222 7230. 0021-91SO/94/$07.00 0 1994 Elsevier Science Ireland SSDI 002 l-9 I 50( 93)05 152-U

autoimmune in nature and affects predominantly women in their middle years. Lipoprotein patterns have been extensively studied in PBC; however, a great variety of findings have been published [2] and lipoprotein abnormalities appear to vary with stage of disease [3]. As with any chronic cholestatic liver disease, gross hyperlipidaemia may occur and can be as severe as that in familial

Ltd. All rights reserved.

44

homozygous hypercholesterolaemia [2]. Indeed, cutaneous xanthomata are commonly described in PBC and xanthomatous neuropathy can develop [4]. The possible mechanisms for this are not understood but several have been proposed [2]. Paradoxically, patients are thought to be relatively protected from atherosclerosis [5-91. However, much of the earlier evidence was based on case histories and small post mortem studies. Recently, Crippen et al. studied 312 PBC patients over 7 years and found the incidence of atherosclerotic death was not significantly different from that predicted by USA mortality data in subjects matched for sex and age [8]. The authors concluded that the hyperlipidaemia associated with PBC does not place them at increased risk of atheroma. However, the evidence is conflicting [2,10]. Since Berg first described Lp(a) in 1963 [l 11, several studies have found elevation of its serum concentration or its specific marker ape(a) to be an independent risk factor for the development of atherosclerosis [ 12,131. Serum Lp(a) concentration varies widely from undetectable levels to over 1.0 g/l and levels above 300 mg/l are associated with an increased risk of CHD [14]. The levels vary little with sex, age and body mass index (BMI) in adults, but black people have higher levels than white people [ 141. The physiological function of Lp(a) is not established, although it has thrombotic and atherogenic properties [14]. Lp(a) is a low density lipoprotein (LDL) particle in which apo B-100, the protein moiety of LDLs, is linked to one or two molecules of the unique glycoprotein, ape(a), by disulphide bridges [ 151. The structure of ape(a) shows considerable homology with plasminogen but it lacks fibrinolytic activity [16]. This has led to the suggestion that Lp(a) may compete with plasminogen for tissue plasminogen activator, and therefore exert a prothrombotic effect [17]. Alternatively, Lp(a) may deliver cholesterol to proliferating cells at sites of vascular injury by binding to fibrin and subsequently being taken up by macrophages, thereby contributing to foam cell formation [ 18,191. The liver plays a crucial role in lipoprotein metabolism and studies in patients undergoing liver transplantation suggest that this organ is the site of Lp(a) synthesis [20]. There have been very

W.L. Gregory et al. /Atherosclerosis

105 (1994) 43-50

few studies so far to determine the influence of liver disease on serum Lp(a) [21,22] and only one small study in PBC patients to determine the effect of ursodeoxycholic acid (UDCA) on Lp(a) levels [23]. The aim of the present study was to test the hypothesis that patients with PBC may be relatively protected from atherosclerosis despite elevated serum cholesterol levels because of lowered serum Lp(a) levels. 2. Patients and methods 2.1. Patients and controls The diagnosis of PBC in 42 patients was based on the triad of cholestatic liver blood tests, serum anti-mitochondrial antibody positivity and liver histology diagnostic of or consistent with the disease. These were age- and sex-matched to 39 patients with other forms of chronic liver disease (alcohol-induced, 18; autoimmune chronic active hepatitis, 10; cryptogenic, 7; other, 5). From a community cardiovascular risk-screening study based on one local general practice, 432 control volunteers were recruited. All subjects and patients were Caucasian and were excluded if they had a history of diabetes mellitus or uncontrolled thyroid disease, or were taking lipid-lowering drugs, including cholestyramine. Patients in the two disease groups who admitted to or were suspected of current excess alcohol consumption were excluded. Full clinical details of patients and volunteers are given in Tables 1 and 2. Smoking history was recorded, resting blood pressure measured and hypertension defined using WHO criteria (diastolic > 90 and systolic > 160 mmHg). Symptoms of CHD were recorded using a standard questionnaire [24] and a resting electrocardiogram (ECG) was analysed for ischaemia using the Minnesota coding [25]. Height and weight were recorded and BMI calculated. 2.2. Lipoprotein measurements

Blood was collected after a 14 h overnight fast with the subject in a semi-recumbent position. Blood, taken into plain tubes, was centrifuged within 1 h. Aliquots of serum were stored at -20°C prior to Lp(a) analysis. All other tests were done on fresh samples. All patients had cholesterol, tri-

W.L. Gregory et al. /Atherosclerosis

105 (1994) 43-50

Table I Clinical details of control

and patients

Number (male/female) Age (years)a BMI (kg/m*)” Smokers (%) CHD (%) Hypertensives (%) Alcohol consumption (units/week)b

volunteers

45

with PBC and other liver disease

Controls (n = 432)

Other liver disease (n = 39)

PBC (n = 42)

2261206 46.2 (kO.50) 24.9 (kO.17) 69 (16.0) 20 (4.6) 84 (19.4) 6 (O-80)

4135 58.6 (+2.18) 25.3 ( +0.88) 7 (17.9) l(17.9) 6 (15.4) 0 (O-28)

4138 62.2 (*ISI) 24.0 (*0.80) 13 (31.0) 6 (14.3) 14 (33.3) 0 (O-24)

“Mean value (*S.E.M.). bMedian value (range).

glycerides, HDL cholesterol, apo Al, apo B, glucose and TSH measured. Patients in the two disease groups also had liver function tests, full blood count and prothrombin time determined. Lp(a) was measured by an ELISA technique (Biopool, Umea, Sweden) with a lower limit of sensitiviy of 4.0 mg/l. Ape(a) phenotypes were classified using sodium dodecyl sulphate polyacrylamide gel electrophoresis [26,27]. Total serum cholesterol was measured using a cholesterol oxidase method with BCL kits (Lewes, UK). The serum triglyceride concentrations were determined using a lipase-glycerol kinase end point

Table 2 Clinical details of patients groups

Histological

stage

Prothrombin time (s)” Bilirubin (pmolil)” Albumin (g/l)” Alanine transaminase (IUII)” Alkaline phosphatase (HI/l)” “Median

value (range).

in the PBC and other liver disease

Other liver disease (n = 39)

PBC (n = 42)

Precirrhotic = 10 Cirrhotic = 27 (No biopsy = 2) 15 (12-25) 14 (4-385) 37 (20-48) 24 (6-233)

I=7 II-III = 19 IV= 16 13 (12-19) 12 (3-413) 41 (23-45) 61 (12-250)

120 (62-384)

351 (95-999)

reaction method with kits supplied by Roche (Welwyn Garden City, UK). The HDL cholesterol fraction was assayed following heparinmanganese precipitation. Apo Al and apo B were determined by rate nephelometry (Technicon DPA, Tarrytown, NY). LDL cholesterol was calculated using Friedewald’s formula [28]. 2.3. Statistical methods Distribution of Lp(a) was positively skewed and could not be logarithmically transformed to a normal distribution due to the high number of undetectable serum levels found in the two patient groups (Fig. 1). Therefore, the non-parametric Mann Whitney U-test was used to compare Lp(a) distributions. The proportions of groups with undetectable levels of Lp(a) were compared by using the x2 test with Yates’ correction. Correlation between parameters was by Spearman’s rank. For the normally distributed data one-way analysis of variance was used. 3. Results Full details of lipid and lipoprotein results are given in Table 3. Median Lp(a) concentrations were lowest in the PBC group (28.5 mg/l) and highest in the community control group (75.0 mg/l); there was a significant reduction in serum levels in the PBC and other liver disease groups compared with controls (Fig. la). However, the median Lp(a) concentration in the non-PBC liver

46

W.L. Gregory et al. /Atherosclerosis

105 (1994) 43-50

Table 3 Serum concentrations of lipids and lipoproteins in the community control, PBC and other liver disease groups Controls (n = 432)

Lr4a) btsW Log W-Gb Number (%) with undetectable levels of Lp(a) Total cholesterol (mmol~)~ Triglyceride (mmol/Qa Log triglycerideb HDL cholesterol (mmolily VLDL cholesterol (mmol/l)c LDL cholesterol (mmolil)c APO Al (g/B’ APO B WC

Other liver disease (OLD) (n = 39)

PBC (n = 42)

75.0 (<4-1033) 74.0 13 (3.0)

52.0 (<4-597) 35.1 11 (28.2)

5.9 (ztO.06)

Signiticanced PBC vs. controls

PBC vs. OLD

OLD vs. controls

28.5 (<4-756) 31.2 12 (28.6)

P < 0.005 P < 0.005 P < 0.001

NS NS NS

P = 0.01 P = 0.01 P < 0.001

5.1 (hO.27)

7.0 (*0.30)

P < 0.001

P < 0.001

P = 0.001

1.17 (0.5-12.4) 1.25 1.44 ( * 0.02)

1.10 (0.4-4.7) 1.21 1.30 (*0.09)

1.10 (0.5-3.2) 1.16 1.95 (*0.12)

NS NS P < 0.001

NS P < 0.001

NS NS NS

0.65 (iO.02)

0.64 (*0.07)

0.58 (*O&l)

NS

NS

NS

3.78 (*0.06)

3.04 (kO.18)

4.43 ( f 0.02)

P = 0.001

P < 0.001

P < 0.001

1.57 (hO.01) 1.25 (iO.02)

0.84 ( zt 0.06) 0.81 (kO.06)

1.01 (*0.05) 0.89 ( * 0.05)

P < 0.001 P < 0.001

P = 0.032

NS

P < 0.001 P < 0.001

NS

aMedian value (range). bGeometric mean value. CMean value ( ?? S.E.M.). dNS, not significant (for details see Patients and methods section).

disease group (52.0 mg/l) was not significantly different to the PBC group. There was a larger proportion of patients with undetectable serum Lp(a) levels in the two disease groups (Fig. 1b) compared with the control group (P < 0.001 for both PBC vs. controls and other liver disease group vs. controls). There was no significant difference between numbers with undetectable levels in the two disease group. Paradoxically, out of a total of six patients with evidence or symptoms of CHD in the PBC group, three (50%) had undetectable levels of Lp(a). Similarly, out of a total of seven patients with evidence or symptoms of CHD in the other liver disease group three (43%) had undetectable levels of Lp(a). Of the 20 individuals in the community control group with evidence or symptoms of CHD only one had an undetectable Lp(a) level. Ape(a) phenotypes did not appear to differ between the three groups. Within the other liver disease group there was a negative correlation between serum Lp(a) levels

and markers of disease severity: for bilirubin R = -0.611 (P < 0.001) and for prothrombin time R = -0.505(P = 0.001). Within the PBC group there was a negative correlation between Lp(a) and bilirubin (R = -0.409, P = 0.007) but a positive correlation between Lp(a) and prothrombin There was no correlatime (R = 0.311, P = 0.048). tion between Lp(a) levels and histological stage of PBC or other markers of disease severity such as serum transaminsases, alkaline phosphatase or albumin. In view of differing lipoprotein patterns with stage of PBC [3], Lp(a) levels were compared between histological stages I, II-III and IV and no differences were found (data not shown). 4. Discussion Serum Lp(a) levels were lower in the PBC and other liver disease groups compared with healthy controls. Furthermore, there was a significantly higher proportion of patients with undetectable

W.L. Gregor_vet al. /Atherosclerosis IOS (1994) 43-50

??Controls 50

(n=432)

PBC (n=42)

lzl

? ?Other liver disease (n=39) 40

Ii

iI a0

u)

20

10

0 50

250

450

6Oa

800

1050

SERUM Lp(a) (mgfl)

la

lb Fig. 1. (a) Percentage frequency distribution of serum Lp(a) levels in the three study groups. (b) Frequency distribution of log transformed serum Lp(a) concentrations for the community control, PBC and other liver disease groups demonstrating the apparent bimodal distribution consisting of subjects with detectable serum levels (log Lp(a) > 0.6) and those with undetectable levels (log Lp(a) < 0.6).

Lp(a) levels in the two patient groups when compared with controls. Although the reduction was not specific to PBC, there was a trend for the PBC patients to have lower serum levels than the non-PBC liver disease group. There were no apparent differences between those with and those without detectable levels within the two patient

groups to explain the apparent bimodal distribution of the logarithmically transformed data (Fig. 1b). The disease severity of PBC ranged from life-threatening hepatic failure to asymptomatic disease with consequent effects on duration of illness and its concomitant metabolic changes [ 11. This heterogeneity makes PBC a difficult disease

48

to study. Certainly patients with hepatic cirrhosis seem less liable to develop CHD than the rest of the population [29]. One post mortem study suggested a 25% reduction in the incidence of myocardial infarction in patients with cirrhosis compared with those without [29]. Feely et al. suggest that Lp(a) levels are reduced in patients with cirrhosis of various aetiologies when compared with healthy controls and patients with CHD and this may exert a cardioprotective effect [22]. It would be a gross over-simplification to suggest that serum Lp(a) concentration alone could be responsible for a possible reduction in the risk of CHD in PBC. Other factors are important. The disease predominantly affects women, who themselves have a lower risk of atherosclerosis. The hypercholesterolaemia associated with PBC may be severe [2], but the duration of such metabolic changes will be considerably shorter than those with an inherited condition which is expressed in childhood. Furthermore, there are qualitative differences in the cholesterol associated with cholestasis which will influence its atherogenic properties [2]. As expected, serum HDL concentrations were higher in our PBC patients and this is thought to have a protective effect [30]. However, the HDL precipitation assay which was used may be less accurate in the dyslipoproteinaemia of PBC, possibly because of apo Al -containing particles occurring in the LDL fraction [3]. It is thought that patients with relatively early PBC tend to have increased levels, especially in the HDL2 subfraction compared with reduced levels in advanced disease [3] and this subclass is thought to offer greater cardioprotection [30]. Although lower HDL cholesterol concentrations were found in our patients with stage IV disease compared with those with stage I-III, this was not significant (data not shown). Furthermore, this HDL subfraction is apolipoprotein E (apo E) rich [3] and apo E may also be protective because of its antithrombotic effect [31]. In addition, raised levels of apo B have been correlated with the risk of myocardial infarction [ 121. Apo B levels were reduced in the PBC and non-PBC liver disease groups compared with the healthy controls and therefore may exert an additional cardioprotective effect. However, elevated apo Al levels are thought to be protective

W.L. Gregory et al. /Atherosclerosis

IO5 (1994) 43-50

against myocardial infarction [ 121 and these were significantly lower in the two patient groups, but less so in the PBC group. The liver is thought to be the site of Lp(a) production [20] and one would expect serum concentrations to decrease as severity of liver disease increases. Much emphasis has been placed on tinding more sensitive markers of disease severity [32,33]. The search has already included components of lipid metabolic pathways such as serum lecithin cholesterol acyltransferase (LCAT) activity [32,34]. No differences in Lp(a) concentrations were found when comparing between different histological stages (I, II-III and IV) in the PBC group. However, a negative correlation between Lp(a) levels and serum bilirubin in both the PBC and liver disease groups was found. Bilirubin is accepted as having a prognostic role in PBC patients [35]. A negative correlation was found with regard to prothrombin time, another measure of synthetic capability, but in the liver disease group only. This presumably reflects the heterogeneity of the PBC patients; in some, prothrombin time will be prolonged because of impaired synthesis of clotting factors while in others it is a result of vitamin K malabsorption. These data suggest that Lp(a) is a measure of liver function. Further work is required to assess a possible use of Lp(a) as a prognostic liver function test, although the strong genetic control of serum concentrations [36] will adversely influence this, There was no difference in the number of PBC patients with CHD ‘(14.3%) compared with the other liver disease group (17.9%). However, this study was not designed to assess the incidence of ischaemic heart disease in such patients. In addition to increased total cholesterol and LDLcholesterol in our PBC group, these patients smoked more and there was a higher incidence of hypertension, two further risk factors for CHD. These factors and the small size of the study may contribute to the paradoxical over-representation of subjects with undetectable levels of Lp(a) having evidence of CHD in both patient groups. Review of the literature [5-lo] would suggest that further studies are required to assess the incidence of atherosclerosis in patients with PBC. If such patients are relatively protected then this preliminary

W. L. Gregory

et al. /Atherosclerosis

IOS (1994)

49

43-50

study would suggest that low Lp(a) levels may play a role in a very complex interaction. 5. References Sherlock, S. and Dooley, J., Diseases of the Liver and Biliary system, Blackwell Scientific Publications, Oxford, 9th Edn., 1993. 2 Miller. J.P., Dyslipoproteinaemia of liver disease, Bailliere’s Clin. Endocrinol. Metab., 4 (1990) 807. 3 Jahn. C.E., Schaefer, E.J., Taam. L.A. et al., Lipoprotein biliary cirrhosis, abnormalities in primary Gastroenterology. 89 (1985) 1266. 4 Turnberg, L.A.. Mahoney, M.P., Gleeson, M.H.. Freeman, C.B. and Gowenlock, A.H., Plasmaphoresis and plasma exchange in the treatment of hypcrlipaemia and xanthomatous neuropathy in patients with primary biliary cirrhosis, Gut, 13 (1972) 976. be5 Ahrens, E.H. and Kunkel. H.G., The relationship tween serum lipids and skin xanthomata in eighteen patients with primary blliary cirrhosis, J. Clin. Invest.. 28 (1949) 1565. 6 Ahrens, E.H.. The lipid disturbance in biliary obstruction and its relationship to the genesis of arteriosclerosis, Bull. NY Acad. Sci., 26 (1950) 151. S., Primary biliary cirrhosis (chronic in7 Sherlock, trahepatic obstructive jaundice), Gastroenterology, 37 (1959) 574. 8 Crippen, J.S., Lindor, K.D., Jorgensen, R. et al., Hypercholesterolaemia and atherosclerosis in primary biliary cirrhosis: what is the risk? Hepatology, 15 (1992) 858. 9 Propst. A., Propst, T., Lechleitner, M. et al., Hypercholesterolaemia in primary biliary cirrhosis is no risk factor for atherosclerosis, Dig. Dis. Sci., 38 (1993) 379. 10 Schaffner, F., Serum cholesterol in primary biliary cirrhosis, Gastroenterology, 56 (1969) 1111. II Berg, K., A new serum type system in man: the Lp system, Acta Pathol. Microbial. Stand., 59 (1963) 369. 12 Durrington. P.H., Ishola, M., Hunt, L.. Arrol, S. and Bhatnagar. D., Apolipoproteins (a). Al, and B and parental history in men with early onset ischaemic heart disease, Lancet, 2 (1988) 1070. 13 Murai, A.. Miyhara. T.. Fujimoto, M., Matsuda. M. and Kameyama, M.. Lp(a) lipoprotein as a risk factor for coronary heart disease and cerebral infarction, Atherosclerosis, 59 (1986) 199. Br. Med. J., 303 (1991) 663. 14 Scott, J., Lipoprotein(a), 15 Scanu. A.M., Lipoprotein (a), Bailliere’s Clin. Endocrinol. Metab.. 4 (1990) 939. J.E., Kuang, W.-J. et al.. 16 McLean, J.W.. Tomlinson. cDNA sequence of human apolipoprotein (a) is homologous to plasminogen, Nature, 300 (1987) 132. 17 Miles, L.A.. Fless, G.M., Levin, E.G., Scanu, A.M. and Plow, E.F., A potential basis for the thrombotic risks associated with lipoprotein (a), Nature, 339 (1989) 301. 18 Rath. M.. Niendorf. A.. Reblin, T., Dietel, M., Krebber.

19

1

20

21

22

23

24

25

26

27

28

29

30

31

32

H.J. and Beisiegel, U., Detection and quantification of lipoprotein(a) in the arterial wall of 107 coronary bypass patients, Arteriosclerosis, 9 (I 989) 579. Zioncheck, T.F., Powell, F.M., Rice, G.C., Eaton, D.L. and Lawn, R.M., Interaction of recombinant apolipoprotein(a) and lipoprotein(a) with macrophages, J. Clin. Invest., 87 (1991) 767. Kraft, H.G., Menzel, H.J.. Hoppichler. F., Vogel, W. and Utermann, G., Changes in genetic apolipoprotein phenotypes caused by liver transplantation, J. Clin. Invest.. 83 (1989) 137. Marth, E., Cazzolato, G., Bon, G.B.. Avogaro, P. and Kostner, G.M., Serum concentrations of Lp(a) and other lipoprotein parameters in heavy alcohol consumers, Ann. Nutr. Metab., 26 (1982) 56. Feely, J., Barry, M., Keeling, P.W.N., Weir, D.G. and Cooke, T., Lipoprotein(a) in cirrhosis, Br. Med. J.. 304 (1992) 545. Beuers, U.. Ritter, M.M., Richter, W.O. and Paumgartner. G., Lipoprotein(a) serum levels in chronic cholestatic liver disease during treatment with ursodeoxycholic acid, Arch. Intern. Med., 150 (1990) 1542. Rose. G.A.. The diagnosis of ischaemic

heart pain and in-

termittent claudication in field studies, Bull, WHO, 27 (1962) 645. Prineas, R.J.. Crow, R.S. and Blackburn, H.. The Minnesota Code Manual of Electrocardiographic Findings, John Wright PSG Inc., Boston, MA, 1982. Gaubatz. J. W., Ghanem, K.I., Guevara, J.J., Nava, M.L., Patsch, W. and Morrisett, J.D., Polymorphic forms of human apolipoprotein(a), J. Lipid Res., 31 (1990) 603. Farrer. M., Game, F.L., Adams, P.C., Laker, M.F. and Alberti, K.G.M.M., A simple sensitive technique for the classification of apolipoprotein(a) isoforms by sodium dodecyl sulphate polyacrylamide gel electrophoresis (SDS PAGE), Clin. Chim. Acta. 207 (1992) 215. Friedewald, W.T.. Levy, R. and Fredrickson, D.S., Estimations of serum low density lipoprotein cholesterol without use of preparative ultracentrifuge, Clin. Chem., 18 (1972) 499. Howel. W.L. and Manion, W.C., The low incidence of myocardial infarction in patients with portal cirrhosis of the liver: a review of 639 cases of cirrhosis of the liver from 17,731 autopsies, Am. Heart J., 60 (1960) 341. Miller, NE., Association of high density lipoprotein subclasses and apolipoproteins with ischaemic heart disease and coronary atherosclerosis. Am. Heart J.. 113 (1990) 589. Desai, K., Mistry, P., Bagget. C., Burroughs. A.K., Bellamy, M.F. and Owen, J.. Inhibition of platelet aggregation by abnormal high density lipoprotein particles in plasma from patients with hepatic cirrhosis, Lancet. 1 (1989) 693. Simko. V., Kellet, R.E. and Dincsoy, H.P., Predicting severity of liver disease: twelve laboratory tests evaluated by multiple regression, J. Int. Med. Res., 13 (1985) 249

50

33

34

W.L. Gregory et al. /Atherosclerosis

Keiding, S., Ericzon, B.G., Eriksson, S. et al., Survival after liver transplantation of patients with primary biliary Stand. J. cirrhosis in the Nordic countries, Gastroenterol., 25 (1990) Il. Higashi, H., Yanaga, K., Shimada, M., Makowka, L., van Thiel, D.H. and Starzl, T.E., Plasma LCAT activity in multiple organ donors: a predictor of allograft viability in clinical liver transplantation, Trans. Proc., 25 (1990) 433.

35

36

105 (1994) 43-50

Rydning, A., Schrumdf, E., Abdelnoor, M., Elgjo, K. and Jenssen, E., Factors of prognostic importance in primary biliary cirrhosis, Stand. J. Gastroenterol., 25 (1990) 119. Boerwinkle, E., Leffert, C.C., Lin, J.P., Lackner, C., Chiesa, G. and Hobbs, H.H., Apolipoprotein(a) gene accounts for greater than 90-percent of the variation in plasma lipoprotein(a) concentrations, J. Clin. Invest.. 90 (1992) 52.