Plasma triglycerides related decrease in high-density lipoprotein cholesterol and its association with myocardial infarction in heterozygous familial hypercholesterolemia

Plasma Triglycerides Related Decrease in High-Density Lipoprotein and Its Association With Myocardial Infarction in Heterozygous Hypercholesterolemia Sital Moorjani,

Cholesterol Familial

Claude Gagnb, Paul J. Lupien, and Daniel Brun

Recent studies suggest that decreased levels of high-density liopoprotein (HDLj may contribute to the risk of premature occlusive atherosclerosis in familial hypercholesterolemia (FH). To investigate further, we have analyzed the concentration as well as distribution of HDL cholesterol in relation to plasma triglycerides and their influence on ischaemic heart disease in FH subjects. The study was carried out in 71 men with heterozygous FH and 46 matched controls. FH subjects were relatively young with a mean age of 36 r 11 years. Tendon xanthomatas were observed in 57% of the subjects, whereas ischemic heart disease was identified in 33%. Compared to normals, the mean value of HDL cholesterol is significantly reduced by 21% in FH heterozygotes (42 2 12 v 33 + 9 mg/dL, P < 0.001). The decrease in HDL cholesterol is highly correlated to the levels of plasma triglycerides (r = -0.50, P < 0.001) and VLDL cholesterol (r = -0.53, P < 0.001). Moreover, HDL cholesterol decrease is not associated with elevated levels of LDL cholesterol (r = -0.20, NS), which is the primary characteristic feature of FH subjects. However, HDL cholesterol decrease is weakly related to total plasma cholesterol concentration (r = -0.24, P i 0.05). The body weight is also contributory to the reduction of HDL cholesterol (r = --0.42, P < 0.01). probably due to its strong positive correlation to plasma triglycerides (r = t0.54, P < 0.001). Grouping of subjects on the basis of triglyceride levels of less than 200 mg/dL (Ila phenotype) and more than 200 mg/dL (Ilb phenotype) shows that the concentration of HDL cholesterol undergoes a further significant decrease in the latter group (36 t 9 v 30 + 11 mg/dL, P < 0.001). Since the level of LDL cholesterol is similar in both groups (314 ? 68 v 316 + 76 mg/dL, NS), a further reduction in HDL cholesterol concentration results in an increased LDL/HDL ratio in Ilb phenotypes. Although HDL cholesterol is normally distributed in controls and type Ila phenotypes, its distribution is skewed to lower values in type Ilb. In addition to similar levels of LDL cholesterol, the presence of tendon xanthomatas is equally observed in both type Ila and type Ilb subjects (51 .I % and 62.5%. respectively). Similarly, the incidence of angina pectoris (19.2% and 12.5% in type Ila and type Ilb, respectively) is also approximately the same in both groups. However, the differences are striking in the incidence of myocardial infarction (MI), which is increased three-fold (25% v 8.5%) in type Ilb subjects, as compared to type Ila. These findings indicate that in addition to LDL excess, HDL deficiency associated with elevated plasma triglycerides contributes to the severity of ischemic heart disease, as revealed from the manifestation of Ml in some FH heterozygotes. D 1986 by Grune & Stratton, Inc.

F

AMILIAL hypercholesterolemia (FH) is a common genetic disorder with complete expression of elevated concentrations of both plasma and low-density lipoprotein (LDL) cholesterol in early childhood.’ The most devastating clinical manifestation of FH is premature occurrence of ischemic heart disease, which is attributed to increased levels of cholesterol in the LDL fraction.‘.* However, recent findings in FH’” have also shown a reduction in the concentration of high-density lipoprotein (HDL) cholesterol, thus raising the possibility that low HDL levels may also contribute to premature ischemic heart disease, due to the inverse relationship usually noted in population studies between HDL and atherosclerosis.’ In fact, a considerable diminution in the concentration of HDL cholesterol has been observed in FH homozygotes with coronary artery disease.‘-” However, the mean value of HDL cholesterol is less depressed in FH heterozygotes. The concentration of LDL cholesterol is increased to the same extent in FH heterozygotes of both sexes; however, the manifestation of ischemic heart disease is precocious and more severe in men than in women.4-6V’oZ’1 On the other hand, differences have been reported in the concentration of plasma triglycerides, which is higher in heterozygous men than in women and HDL cholesterol, which is lower in heterozygous men than in women.5,6 These findings have prompted us to analyze the concentration as well as the distribution of HDL cholesterol in relation to plasma triglycerides and their possible influence on ischemic heart disease in male heterozygotes with FH. Metabolism,

Vol 35, No 4 (April), 1986: pp 3 1 l-3 16

MATERIALS

AND

METHODS

Subjects

Seventy-one men with heterozygous FH and 46 age-matched controls of the same sex were studied. The subjects, mostly French Canadian (98%) were above 19 years of age (range 20 to 71 years) at the time of the study. The diagnosis of FH was made according to criteria described elsewhere’,’ after excluding all secondary causes of hypercholesterolemia, as well as other known forms of familial hyperlipidemia. Briefly, FH was diagnosed in the presence of tendon xanthomatas in patients with hypercholesterolemia; in subjects in whom tendon xanthomatas were absent or not yet developed due to young age,5 the diagnosis was ascertained if LDL cholesterol was above 190 mg/dL and tendon xanthomatas were observed in at least one first degree relative. The subjects were classified as having a Ha phenotype if plasma triglyceride concentrations were less than 200 mg/dL. A IIb phenotype was ascribed if triglyceride levels were equal to or above 200 mg/dL and a prominent prebetalipoprotein band was seen on agarose gel electrophoresis. In addition, 25 unaffected male family members and 21 unrelated normolipidemic individuals free from cardiovascular disease were also analyzed for control lipid values. The criteria for ischemic heart disease were angina pectoris and myocardial infarction (MI); angina was recog-

From the Departments of Biochemistry and Medicine, Faculty of Medicine. Lava1 University, Qutbec. Canada. Address reprint requests to Lipid Research Unit. Le Centre Hospitalier de 1’Vniversitb Laval, 2705 Boulevard Laurier, Qukbec. Canada. Gl V 4G2. Q 1986 by Grune & Stratton, Inc. 0026-0495/s6/3504~005$03.00/0

311

312

MOORJANI ET AL

nized by clinical assessment of patient’s history of typical chest pain, whereas MI was diagnosed on the basis of the history, EKG, and/or documented medical record including serum enzymes.

Table 1. Prevalence of Clinical Manifestations in Men With Heterozygous Familial Hypercholesterolemia FH Heterozygotyes 118Phenotype

Blood Samples and Laboratory Methods All subjects were sampled in the reclining position after an overnight 12- to 14-hour fast. The venous blood was withdrawn from an anticubital vein under ethylenediaminetetracetate (1 mg/mL, final concentration) and was immediately centrifuged at 1,000 x g for 10 minutes. The plasma was decanted and stored at 4 OC in the presence of thimerosal (0.15 mg/mL, final concentration) before lipoprotein analyses were performed. Plasma VLDL fraction was isolated by ultracentrifugation using a 50.3 rotor (Beckman Instruments Inc, Palo Alto, Calif) at a density of 1.006 g/mL as described by Have1 et al.‘” The infranatant of density more than 1.006 g/mL was used for the determination of HDL cholesterol after precipitation of LDL with heparin (412 USP units per mL, Sigma Chemical Co, St Louis) and manganese chloride (0.092 molar, final concentration), as previously described.13 The concentration of LDL cholesterol was obtained by difference from the values of cholesterol in the infranatant, measured before and after the precipitation was performed. The cholesterol and triglyceride in plasma and lipoprotein fractions were quantified on an Auto Analyser II (Technicon Instruments Corp, Tarrytown, NY) using isopropanol extracts treated with zeolite, as described by Rush et al.l4 Agarose gel electrophoresis of plasma and lipoprotein fractions was carried out by a modified procedure of Noble” as described previously.5 Recovery of cholesterol in the lipoprotein fractions isolated by ultracentrifugation averaged 95% (range, 90% to 102%) of the total plasma cholesterol value. The coefficient of variation (CV) for the measurement of control sera was 2.1% for high cholesterol value, 2.7% for low cholesterol value, and 3.0% for triglycerides. Duplicate analysis of 55 sera for HDL cholesterol determinations on different days yielded analytical variation of 3.3%. Data Analysis Only the results obtained at first visit of patient to clinical center (while on an ad libitum diet) were included in the analyses. All lipoprotein data was excluded from analysis in the cases of (1) MI during the last three months of blood sampling and (2) use of drugs known to have an effect on plasma lipoprotein metabolism. Means and standard deviations were calculated and the significance of differences between means in various groups was tested by analysis of the variance. ‘6 The hypothesis of equal proportions in different groups was analyzed by the normal approximation of binomial distribution. The relationship between two variables was estimated using Spearman’s nonparametric rank correlation procedure.16 The coefficient of variation for duplicate measurements of HDL cholesterol samples was computed from the formula [2: (di%)‘/2n)}“* where “di%” is the difference between a pair of determinations in percent of its mean value and “n” is the number of duplicate samples. RESULTS

Heterozygous FH patients in the present study are relatively young, having a mean age of 38 + 11 years with a range of 20 to 7 1 years. This is probably because fewer male individuals with FH often survive past 60 years of age’ due to early mortality from premature ischemic heart disease. The presence of tendon xanthomatas observed in 57% of subjects (Table 1) is comparable to our previously reported finding of 63% in men aged 30 to 39 years’ and 58% for both sexes with

Number (n)

47

Ilb Phenotype

P-VdlJe

24

Age (vr)

36 + 11

40+

10

NS

Tendon Xanthomatas

24 (51.1)

15 (62.5)

NS

lschaemic Heart Disease

NS

13 (28.01

9 (37.5)

Angina Pectoris

9 (19.2)

3 (12.5)

NS

Myocardial Infarction

4 (8.5)

6 (25.0)

< 0.05

Values in parenthesis are percent of n.

a mean age of 35 yearsI Similarly, 33% occurrence of ischemic heart disease in the present study compares well with 39% incidence observed in 140 men of 20 to 49 years reported previously.’ As shown in Table 1, the frequency of occurrence of tendon xanthomatas, as well as ischemic heart disease, is approximately similar in both groups of FH subjects, irrespective of the IIa or IIb phenotype. Moreover, the incidence of angina pectoris is also approximately equal in both groups although its occurrence is slightly lower in IIb phenotypes (the data is not statistically significant). On the contrary, important differences are noted in the prevalence of MI, which is approximately three times more frequent in type IIb as compared to IIa phenotype. The data on plasma lipid and lipoprotein levels in controls and FH subjects is shown in Table 2. As expected, in comparison to normals the mean concentration of LDL cholesterol is increased more than twofold in both the IIa and IIb phenotypes. Consequently, plasma cholesterol levels are also highly increased in both groups. However, there is a further significant rise in cholesterol concentration of 9% in type IIb as compared to IIa, and this increase is accounted for by the higher levels of VLDL cholesterol due to elevated plasma triglycerides in IIb phenotypes. Most striking changes are noticed in the concentration of HDL (Table 2), which is decreased by 14% and 29% in IIa and IIb phenotypes, respectively, as compared to controls. There is a significant difference of 15% in HDL cholesterol levels between IIa and IIb subjects. This further decrease in HDL cholesterol concentration in IIb is also reflected in a significantly higher LDL/HDL ratio, in spite of similar levels of LDL cholesterol in both groups. The frequency distribution of HDL cholesterol in normals and FH subjects is shown in Fig 1. The controls have an approximately normal distribution (skewness, 0.63; kurtosis, 0.44) with a median level of 40 mg/dL and a mean value of 42 mg/dL. The distribution is also unimodal and normal for FH subjects with IIa phenotype (skewness, 0.60; kurtosis, 0.43), but it is shifted to lower values and has a median of 35 mg/dL and a mean of 36 mg/dL. For FH subjects with IIb phenotype the HDL cholesterol distribution is slightly skewed (skewness, 0.94; kurtosis, 1.18) towards lower values with a median and mean levels of 27 mg/dL and 30 mg/dL, respectively. As shown in Fig 2, HDL cholesterol has a marked negative correlation to plasma triglycerides. Similarly, a highly signif-

HETEROZYGOUS FAMILIAL HYPERCHOLESTEROLEMIA

Table 2.

Plasma Lipid and Lipoprotein

Concentrations

313

and Correlation

and Those With Heterozygous

Coefficient

Values for HDL Cholesterol

Subjects With Heteroqgous Ila Phenotype

lib Phenotype

46

47

24

40*

Age (vr)

FH

Normal’

Number

ll$

in Normal Men

Familial Hypercholesterolemia

36 i

40r

11

rt If-value1

10

0.23 INS)

102 + 10

109 ?r 1 ill

-0.42

(< 0.01)

Triglycerides (mg/dL)

123 i- 35

113 i 31

271

-0.50

(< 0.001)

Cholesterol (mg/dL)

198+30

363 + 6411

397 + 681117

Body Weight (% ideal)5

15 + 6

VLDL Cholesterol (mg/dL) HDL Cholesterol (md/dL) LDL/HDL Cholesterol Ratio

316 k 7611

314 r 6811

42 t 12

36 k 911

3.6 ? 1.1

9.4 + 3.611

- 0.24 (< 0.05)

50 t 10011ll

14 t 9

139 k 27

LDL Cholesterol bng/dL)

+ 801/ll

-0.53

k

-0.20

(NS)

0.001)

-0.84

(0.001)

30 ? 11 Iill 11.8 t 4.31(ll

‘The lipid and lipoprotein values in normals are comparable to data from Lipid Research Clinics, Toronto/Hamilton study for 189 men aged 35 to 39 years (triglycerides, 148 + 86; cholesterol, 203 + 36; VLDL-C, 24 ? 18; LDL-C, 135 + 33; HDL-C, 45 t 1O).38 tCorrelations between HDL cholesterol and other variables were calculated from the combined data for Ila and Ilb phenotypes. *Values are mean + SD. §From Metropolitan Life Insurance tables (the ideal body weight for height for the midpoint of the range for medium body frame). //Significantly different from normals, P < 0.01. nsignificantly different from Ila phenotype, P <: 0.01

lil

3s

60

85

FH. ub

icant negative relationship is also noticed between VLDL and HDL cholesterol (Table 2). On the other hand, decrease in HDL cholesterol concentration is unrelated to LDL cholesterol, whereas a weak but significant association is noticed between HDL and total plasma cholesterol. This later relationship can be explained by a significant increase in VLDL cholesterol, which in fact is responsible for the higher levels

400

l*

l

l

12

l l

l

W a

24

4

16

300

l *

.

2 s II LL

l

6

200

e

l

1

I

0 CONTROL 0

l

le l

e

l

O

och% 00

OO@O 0 00 8 OO ooOoo% 0 00 0

HDL- Cholesterol, mg per dl Fig 1. The frequency distributions of HDL cholesterol in controls and FH heterozygotes with Ila and Ilb phenotype. For the number of subjects, see table 1. Arrows indicate median values: controls, 46 mg/dL: FH Ila phenotype, 35 mg/dL, and FH Ilb phenotype, 27 mg/dL.

0

10

20

FH,IIa FH, IIb

s l

0

C

24

0 0

l

PH. IIa

30

O0

40

HDL-Cholesterol,

0

.?T

50

O

60

70

mg per dl

Fig 2. Scattergram showing the relationship between plasma triglycerides and HDL cholesterol in FH heterozygotes (r = 0.50, P < 0.001).

314

MOORJANI ET AL

of total cholesterol in IIb phenotypes, in spite of a corresponding decrease in HDL cholesterol. In addition, there is a highly significant negative correlation between LDL/HDL cholesterol ratio and HDL cholesterol levels. Finally, a significant negative relationship is also noticed between HDL cholesterol and body weight.

DISCUSSION

FH is a relatively common autosomal dominant disorder characterized by increased LDL and total plasma cholesterol. This disorder results from an abnormally poor removal of LDL due to a primary defect in the high affinity LDL cell receptor sites.’ As a result, LDL accumulates in the plasma up to four to six times normal in homozygotes and two to three times normal in heterozygotes. In the rare homozygous state, severe atherosclerosis develops in early childhood, and death from coronary heart disease usually occurs before the third decade of life. In the more frequent heterozygous state, 50% of the men and women develop clinical ischemic heart disease by the age of 40 to 50 years, respectively.‘*” By age 60, 50% of males and 30% of females with heterozygous FH sustain MIL8 as compared to a frequency of 14% and 9% for normals. The incidence of MI in this study is 14% in a group of young male heterozygotes with a mean age of 38 years. It is now well established that as major risk factors, both cholesterol and LDL levels are strongly and directly related to ischemic heart disease.“**’ In addition, the early development of coronary heart disease in FH is also attributed to elevated concentration of cholesterol in the LDL fraction. On the other hand, the concentration of HDL cholesterol has been recognized as an important, independent, and preventive factor in coronary heart disease.‘~*‘~** The HDL particles contribute to the removal of esterified cholesterol from the arterial smooth muscle cells23 and transport back to the liver for catabolism. Thus, HDLs appear to play a role in reducing the risk of coronary heart disease. The findings from our present study clearly affirms that the characteristic elevation of LDL cholesterol is associated with premature ischemic heart disease in heterozygous FH men. However, the data also shows that these individuals have abnormally low concentrations of HDL cholesterol, and corroborates our previously reported results in 209 families with FH5s6and those of Streja et al4 obtained from a study of a large kindred. Thus, the opposite changes in both LDL and HDL may be responsible for the premature ischemic heart disease seen in FH patients. In this respect, it has recently been demonstrated24 that the normal direction of cholesterol net transport is reversed in cultured cells incubated in the presence of sera from patients with heterozygous FH. So instead of a net cholesterol transport from the cells to the medium, there is a net uptake of cholesterol from the plasma by the cells. This is an abnormality consistent with the combined effects of opposite changes in LDL and HDL in the pathogenesis of atherosclerosis. Based on these findings and others,G there are now good indications that the major clinical characteristics of FH cannot be entirely due to raised LDL cholesterol concentration. Furthermore, Heiberg and Slack*’ (see also reference 2)

have previously drawn our attention to the existence of familial aggregation of a positive family history of coronary heart disease irrespective of the similarities in plasma cholesterol concentration between families. Our present data on FH heterozygotes clearly indicates that the levels of LDL cholesterol are similar in both IIa and IIb phenotypes, and also, the characteristic appearance of tendon xanthomatas is equally observed in both groups. The prevalence of ischemic heart disease is also similar in both groups. However, differences are noticed in the mean concentration of HDL cholesterol, which is associated with a three-fold increase in the incidence of MI in subjects with IIb phenotype as compared to IIa. Taken together, these findings indicate that in addition to LDL excess, HDL deficiency contributes to the severity of ischemic heart disease in some patients. The mean decrease of 2 1% in HDL cholesterol observed in this study is comparable to the results reported previously.ti The quantitative change in HDL cholesterol is small as compared to LDL, and furthermore, there is no correlation between the two parameters. A 14% HDL cholesterol decrease in IIa phenotypes as compared to controls with similar levels of plasma triglycerides cannot be explained from our present data. However, the lack of correlation between opposite changes in LDL and HDL should not be taken as evidence that the observed low levels of HDL cholesterol are not due to phenotype expression of a primary accumulation of LDL. A further decrease of 17% in IIb phenotypes (29% in comparison to controls) is partially explained by significant negative relationships found between HDL cholesterol and plasma triglycerides, and between VLDL cholesterol and HDL cholesterol. The inverse relation among triglycerides, VLDL cholesterol, and HDL cholesterol has been documented for normolipidemic and various hyperlipidemic statesz6’*’ These reciprocal relationships result from the transfer of surface components from VLDL to HDL during triglyceride lipolysis.23 It is also likely that increased ponderousness in type IIb phenotypes (109%) as compared to type IIa (102%) may also be a contributing factor to a further decrease in HDL cholesterol in this group. Several environmental, clinical, and pharmacologic factors, as well as lifestyles, can influence HDL cholesterol levels. The changes in HDL resulting from secondary clinical disorders and those resulting from drugs affecting lipid metabolism were properly controlled in this study. Since the main object of the study was to establish relationships between HDL, other lipids, lipoprotein profiles, and their influence on ischemic heart disease, the effects of lifestyles (smoking, alcohol intake, physical activity) were not taken into account due to the fewer number of subjects in the study. The influence of body weight was particularly studied since it is closely associated with triglyceride metabolism and is also known to affect concentration of plasma HDL. In our FH heterozygotes, a significant negative correlation (r = -0.42, P < 0.01) is observed between HDL cholesterol and body weight, as seen in a number of crossectional population studies.28~29 On the other hand, a positive significant relationship is noted between body weight and plasma triglycerides (r = +0.54, P < 0.001). Although it has been suggested3’

HETEROZYGOUS FAMILIAL

315

HYPERCHOLESTEROLEMIA

that abnormal plasma triglyceride metabolism is not a feature of heterozygous FH, this study was based on data obtained in siblings from two small kindreds and thus may not represent all families with FH. In fact, our data demonstrates that hypertriglyceridemia is a common feature in FH, and along with excess body weight, it contributes to a decrease in HDL cholesterol in some FH patients. Similar findings have been reported for adult men in randompopulation studies.3’ Since the liver32 plays an important role in the catabolism of LDL, hypertriglyceridemia in some FH patients may develop from delayed clearance of VLDL or its remnant at the B and E receptor sites, due to an excess of LDL. Although hypertriglyceridemia is very frequently observed in patients with premature coronary heart disease,33s34 several epidemiologic studies have failed to show an independent association between plasma triglyceride levels and the incidence of coronary heart disease.3s However, it has been pointed out that differences in the choice of coronary endpoints36 in various epidemiologic studies may be responsible for some of the discrepancy. The issue is further complicated as hypertriglyceridemia itself is heterogenous in origin. It is therefore of considerable interest that even in FH, heterozy-

gotes with accompanying hypertriglyceridemia (type IIb) are at a further risk of premature MI, although the incidence of angina pectoris is not different than in type IIa. These observations are consistent with the recent findings from the Uppsala Primary Preventive Study,37 in which risk factor patterns are similar in both angina pectoris and MI group, except for plasma triglycerides, which remain as a discriminator between the two groups. The concentration of HDL cholesterol was not reported in this study. Since individuals with FH are already at a higher risk of premature atherosclerotic disease due to primary elevation of LDL, it is therefore important to identify the factors influencing both plasma triglyceride and HDL levels, so as to include necessary therapeutic measures in the treatment. Further studies are also required to clarify the role of environmental and genetic determinants of HDL in FH subjects.

ACKNOWLEDGMENT

We thank the nursing staff and laboratory personnel of the Lipid Research Unit for their valuable contribution. We would also like to acknowledge the skillful secretarial assistance of Doris Moreau in the preparation of this manuscript.

REFERENCES

1. Fredrickson DS. Goldstein JL, Brown MS: The familial hypercholesterolemia, in Standubry JB, Wyngaarden JB, Fredriskson DS (eds): The Metabolic Basis of Inherited Diseases. New York, McGraw-Hill, 1978, pp 604-655 2. Slack J: Inheritance of familial hypercholesterolemia, in Paolletti R, Gotto AM (eds): Atherosclerosis Reviews. New York, Raven Press, 1979, pp 35-66 3. Kwiterovich Jr PO, Fredrickson DS, Levy RI: Familial hypercholesterolemia (one form of familial type II hyperlipoproteinemia). A study of its biochemical, genetic and clinical representation in childhood. J Clin Invest 53:1237-1249, 1974 4. Streja D. Steiner G, Kwiterovich PO: Plasma high density lipoprotein and ischaemic heart disease. Ann Intern Med 89:871880, 1978 5. Gagnt C, Moorjani S, Brun D, et al: Heterozygous familial hypercholesterolemia: Relationship between plasma lipids, lipoproteins, clinical manifestations and ischaemic heart disease in men and women. Atherosclerosis 34: 13-24, 1979 6. Moorjani S. Gagni C, Brun D, et al: Les cu-lipoproteines (cholesterol-HDL) dans la forme hetirozygote de I’hypercholesttrolemie familiale. Union Med Can 109:.530-534, 1980 7. Miller GJ, Miller NE: Plasma high density lipoprotein concentration and development of ischaemic heart disease. Lancet 1:16-19, 1975 8. Moorjani S, Lupien PJ, Lepage M, et al: Decreased serum levels of high density lipoproteins in familial hyperbetalipoproteinemia associated with early atherosclerosis. XI International Biochemical Meeting, Salamanca, April 2427, 1973 (abstr j 194) 9. Goldberg RB, Fless GM, Baker SG, et al: Abnormalities of high density lipoproteins in homozygous familial hypercholesterolemia. Arteriosclerosis 4~4722478, 1984 10 Hirobe K, Matsuzawa Y, Ishikawa K, et al: Coronary artery disease in heterozygous familial hypercholesterolemia. Atherosclerosis 44:201-210, 1982 1I. Beaumont V, Jacotot B, Beaumont JL: Ischaemic heart disease in men and women with familial hypercholesterolemia and xanthomatosis: A comparative study of genetic and environmental

factors in 274 heterozygous cases. Atherosclerosis 24:441-450, 1976 12. Have1 RJ, Eder HA, Bragdon JH: The distribution and chemical composition of ultracentrifugally separated lipoproteins in human serum. J Clin Invest 34:1345-1353, 1955 13. Burstein M, Samaille J: Sur un dosage rapide du cholesterol lie aux @-lipoproteines du strum. Clin Chim Acta 5:609-610. 1960 14. Rush RL, Leon L, Turrel J: Automated simultaneous cholesterol and triglyceride determination on the Auto Analyser instrument, in Advances in Automated Analysis. Miami, Fla, Thurman, 1970, pp 503-507 15. Noble RP: Electrophoretic separation of plasma lipoproteins in agarose gels. J Lipid Res 9:693-700, 1968 16. Snedecor GW, Cochran WG: Statistical Methods. Ames, Iowa, Iowa State University Press, 1980, pp 107-193 17. Mathon G, Gagne C, Brun D, et al: Articular manifestations of familial hypercholesterolemia. Ann Rheum Dis 44:599-602, 1985 18. Stone NJ, Levy RI, Fredrickson DS, et al: Coronary artery disease in I I6 kindred with familial type II hyperlipoproteinemia. Circulation 49:476-488, 1974 19. Kannel WB, Dawber TR, Friedman GD, et al: Risk factors in coronary heart disease: An evaluation of several serum lipids as predictors of coronary heart disease. Ann Intern Med 61:888-899, 1964 20. The Lipid Research Clinics Coronary Primary Prevention Trial Results II. The relationship of reduction in incidence of coronary heart disease to cholesterol lowering. JAMA 25 1:365-374, 1984 21. Gordon T, Castelli WP, Hjortland MC, et al: High density lipoprotein as a protective factor against coronary heart disease: The Framingham study. Am J Med 62:707-714. 1977 22. Miller NE, Thelle DS, Forde CH, et al: The Tromso study. High density lipoprotein and coronary heart disease: A prospective case control study. Lancet 1:965-967, 1977 23. Eisenberg S: High density lipolprotein metabolism. J Lipid Res 25:1017-1058,1984

316

24. Fielding PE, Fielding CJ, Have1 RJ, et al: Cholesterol net transport, esterification and transfer in human hyperlipidemic plasma. J Clin Invest 71:449-460, 1983 25. Heiberg A, Slack J: Family similarities in the age at coronary death in familial hypercholesterolemia. Br Med J 2:493-495, 1977 26. Meyers LH, Phillips NR, Have1 RJ: Mathematical evaluation of methods for estimation of the concentration of the major lipid components of human serum lipoproteins. J Lab Clin Med 88:491505,1976 27. Schaefer EJ, Anderson D, Brewer HB, et al: Plasma triglycerides in regulation of HDL cholesterol levels. Lancet 2:391-393, 1978 28. Garrison RJ, Wilson PW, Castelli WP, et al: Obesity and lipoprotein cholesterol in the Framingham offspring study. Metabolism 29:1053-1060, 1980 29. Glueck CJ, Taylor HL, Jacobs D, et al: Plasma high density lipoprotein cholesterol: Association with measurements of body mass. The Lipid Research Clinics Program Prevalence Study. Circulation 62:62-69, 1980 (suppl4) 30. Angelin B: Metabolism of endogenous plasma triglyceride in familial hypercholesterolemia: Studies of affected and unaffected siblings of two kindreds. Eur J Clin Invest 10:23-26, 1980 31. Connor SL, Conner WE, Sexton G, et al: The effect of age, body weight and family relationships on plasma lipoproteins and

MOORJANI ET AL

lipids in men, women and children of randomly selected families. Circulation 65:129&1298, 1982 32. Brown MS, Goldstein JL: Lipoprotein receptors in the liver: Control signals from plasma cholesterol traffic. J Clin Invest 72:743-747,1983 33. Carlson LA, Bottiger LE: Ischaemic heart disease in relation to fasting values of plasma triglycerides and cholesterol. Lancet 1:865-868,1972 34. Sniderman A, Shapiro S, Marpole D, et al: Association of coronary atherosclerosis with hyperapobetalipoproteinemia (increased protein but normal cholesterol levels) in human plasma low density lipoproteins. Proc Natl Acad Sci USA 77:604-608, 1980 35. Lippel K, Tyroler H, Eder H, et al: Relationship of hypertriglyceridemia to atherosclerosis. Arteriosclerosis 1:406-417, I98 I 36. Carlson LA, Bottiger LE: Serum triglycerides, to be or not to be a risk factor for ischaemic heart disease. Atherosclerosis 39:287291, 1981 37. Aberg H, Lithell H, Selinus I, et al: Serum triglycerides are a risk factor for myocardial infarction but not for angina pectoris: Results from a IO-year follow-up of Uppsala primary preventive study. Atherosclerosis 54:89-97, 1985 38. Jones GJL, Hewitt D, Godin GJ, et al: Plasma lipoprotein and the prevalence of hyperlipoproteinemia in a Canadian working population. Can Med Assoc J 122:3746, 1980