- Email: [email protected]
Elevated plasma lipoprotein(a) associated with abnormal stress thallium scans in children with familial hypercholesterolemia
PREv5NnvE cARDlol.oGY
Elevated Plasma Lipoprotein(a) Associated with Abnormal Stress Thallium Scans in Children with Familial Hypercholesterolemia RobertA. Hegele,MD, PhilipW.Connelly,PhD, GeraldineCulletiean, MSc, andVeraRose,MD ipoprotein(a) (Lp(a)) is a cholesteryl ester-rich particle consisting of 2 attached components: a low-density lipoprotein (LDL) particle whose apolipoprotein B-100 moiety is linked by disulphide bonds to a single large polymorphic glycoprotein, apolipoprotein(a).’ Plasma Lp(a) concentrations are strongly correlated with an increased prevalence of coronary artery disease, especially in patients younger than age 60 years and in those with strong family histories (reviewed in reference 1). The mechanisms underlying the associations with atherosclerosis are still unknown, although in vitro studies suggest that Lp(a) can be involved both in the inhibition of fibrinolysis acutely and in the formation of atheromatous plaques chr~nically.~ Plasma concentrations of Lp(a) vary over a wide range among persons, but remain stable in any particular subject.* Genetic and environmental factors affect the plasma concentration of Lp(a). For example, size variation at the apolipoprotein(a) gene is a major determinant of Lp(a) concentrations.* Also, diabetes mellitus,3 renal disease,4 postoperative state~,~medical treatment with plasma exchange,6 niacin7 stanozolol,8 danaz01,~ and tissue-type plasminogen activatorlO can also affect plasma concentrations of Lp(a). Studies in tissue culture and in vivo are inconclusive regarding role of the LDL receptor in the catabolism of Lp(a).l Medications that up-regulate the LDL receptor, such as cholestyramine and lovastann, do not lower Lp(a). 11-13Although the relation between LDL receptor function and plasma Lp(a) is not clear, elevated Lp(a) can accelerate the expression of coronary artery disease in adults with familial hypercholesterolemia (FH).14 We hypothesized that increased plasma Lp(a) was associated with more severe coronary atherosclerosis in children with FH. We studied asymptomatic heterozygotes with FH and homozygotes with stress thallium scintigraphy and plasma Lp(a) determinations.
L
Plasma lipoprotein(a) (@(a)) conce&Wi BCB associatedwithaniBrQskofcoronary* twy disease in adults with fsmilial hypmehoksterolemia (FH). We hypothesized that Lp(a) f%nmdMions in children with FH were higher among those with myacardial ischemia on stress thallium scans and among those with a family bib comnary artery dii. mem toryofprematu~ ty-nhm asymptometie heterozygotes with FH (mnge9to23years)and7homozygotes(range4 to 13 years) were evaluated with clinical asses!+ merit, llpoprobin measurement and stress mallC umsums.Comparedwithsubjectswithnormal stress thallhun scans, mean @(a) was sigMicantly hm in homozygotes with stress thallium ab normaltties (79 + l6 vs 15 f 5.5 mg/dl, p q 0.03), andtemkdtobehiiinhebrozygot~with stress thallium alMomIalitii(39+12vs20 -r4.2 mg/dl, p q 0.10). @(a)tendedto be hii in heterozygotes with a family history of premature ~arteydisease(30+6.4vsl4&4.1 mg/dl, p q 0.12). It is concluded that Q(a) is high er in hypeMolesterolemic children who have ab normal stress thallium e Lp(a) may be useful inassessingcoronary artery disease risk in chic dren with M. (Am J Cardiil1993;72:402-403)
MEWlOOS From the Departments of Medicine and Clinical Biochemistry, St. Michael’s Hospital, the Division of Cardiology, Department of Paediiatrics, The Hospital for Sick Children, University of Toronto, Toronto, Ontario, Canada. This study was supported by grant Al663 from the Heart and Stroke Foundations of Ontario and Canada. Dr. Hegele is a McDonald Scholar of the Heart and Stroke Foundation of Canada. Manuscript received November 30, 1992; revised manuscript received December 18, 1992, and accepted March 10,1993. Address for reprints: Robert A. Hegele, MD, DNA Research Laboratory, St. Michael’s Hospital, 38 Shuter Street, Toronto, Ontario, Canada M5B lA6. This work was presented in part at the 65th Scientific Sessions of the American Heart Association (abstract #0047). 402
THE AMERICAN JOURNAL OF CARDIOLOGY VOLUME 72
Patients A total of 36 asymptomatic children, and adolescent and young adult subjects with FH were evaluated with single-photon emission computed tomography 201~thallium scanning after exercise and plasma determinations of lipoproteins. Twenty-two male and 14 female subjects (mean age 16 years, range 4 to 27) were assessed. Seven patients (6 males, 1 female) were homozygotes with FH and 29 patients (16 males, 13 females) were heterozygotes with FH. All subjects were nonsmokers. All FH homozygotes were ascertained through referral to the lipid clinic based on dermatologic tindings
AUGUST 15.1993
TABLE
I Baseline
Biochemical
Variables
Heterozygotes Females/males Age (years) Total cholesterol (mmol/L) Triglycerides (mmol/L) LDL cholesterol (mmol/L) HDL cholesterol (mmol/L) ApoB (g/L) ApoA-I (g/L) Lp(a) (mg/dl)
Homozygotes
p Value
l/6
16/13 16.8 f 0.70 7.23 -t 0.28 (279 2 11 mg/dl)
13.6 + 3.0 14.9 k 1.31 (577
k 51 mg/dl)
NS 0.001
1.42 + 0.16 (126
1.75 f 0.33 (155 2 29 mg/dl)
NS
f 14 mg/dl)
5.49 f 0.27 (212 2 10 mg/dl)
13.4 -t 1.40 (519
f 54 mg/dl)
0.001
1.09 f 0.05 (42 f 1.9 mg/dl)
0.72 2 0.04 (28 -+ 1.5 mg/dl)
0.0001
1.85 f 0.07 1.25 + 0.04 24.3 +- 4.46
3.79 f 0.30 0.90 t 0.08 51.8 + 16.2
0.0005 0.0007 0.08
Apo = apolipoprotein; HDL = high-density lipoprotein: LDL = low-densky lipopmtein; Lp = plasma lipoprotem.
TABLE
II
Thallium
Scans
in Heterozygotes
with
Familial
Hypercholesterolemia
Normal Females/males Age (years) Cholesterol (mmol/L) Triglycerides (mmol/L) LDL cholesterol (mmol/L) HDL cholesterol (mmol/L) ApoB (g/L) ApoA-I (g/L) Lp(a) (mg/dl)
8714 17.1 7.05 1.49 5.30 1.06 1.81 1.21 19.5
k f f + f + 2 +
0.76 0.31 0.21 0.30 0.05 0.08 0.05 4.2
Abnormal
(273 2 14.3 mgldl) (129 f 18.7 mg/dl) (205 f 11.6 mgldl) (41 + 1.9 mg/dl)
512 15.7 7.81 1.17 6.11 1.17 1.98 1.38 39.1
2 $ k + + $ ” -c
1.72 0.63 0.17 0.59 0.11 0.18 0.08 11.9*
(302 (104 (237 (45
-c 24.4 mg/dl) k 15.1 mg/dl) +- 22.8 mg/dl) -t 4.2 mg/dl)
*p = 0.10. All differences are nonsignificant (p >O.lO). Abbrewations as in Table I.
such as planar or tendinous xanthomata, or both. All heterozygotes with PI-I were ascertained through a family history of hyperlipidemia. The diagnosis was established by family history and lipid prolile, and, additionally in homozygotes, typical clinical features such as planar and tendinous xanthomas. Parental and family histories were obtained with regard to myocardial infarction, coronary bypass surgery, cardiac death, or a combination of all 3. With use of these criteria, 18 heterozygotes with FH had a positive parental history of premature vascular disease. All homozygotes with FH had parents who were alive and well with no history of coronary artery disease. Medical treatment included a fat-modified diet with cholestyramine; compliance with treatment was variable. ThaNurn scarmin%: Informed consent was obtained from all subjects. Bach patient was asked to fast on the study day. Exercise was performed as described,t5 with termination either because of maximum work load, electrocardiographic abnormalities or fatigue. The maximal 201&a&m dose of 74 MBq, based on weight, was administered intravenously 1 minute before the exercise was stopped. Poststress images were obtained 4 minutes after and redistribution images 3 hours after the exercise ended. Patients were allowed a light meal between poststress and redistribution images to minimize splanchnic blood flow. Studies were performed using either a General Electric Starcam M or Starcam gamma camera (Milwaukee, Wisconsin) with technical spectications as de~cribed.~~An abnormality was diagnosed based on photon-deficient areas in the myocardium in poststress
images.15 In homozygotes with PI-I, there was complete concordance between the presence of a perfusion defect observed with stress thallium scanning and the presence of anatomically appropriate coronary arterial lesions observed with angiography.15 Heterozygotes with PI-I did not undergo angiography. Normolipidemic children without heart disease were not studied with stress thallium scanning. Biochemkal anatysis: Aliquots of fasting plasma frozen at -70°C were used for determinations of total cholesterol, triglycerides, LDL cholesterol and high-density lipoprotein (IIDL) cholesterol, apolipoproteins A-I and B and Lp(a) using described methodsto StatIstical malysk SAS was used for all statistical analyses.l6 Data were summarized descriptively with mean + SE. Lp(a) was transformed for ana.lysis.‘ONormal probability plots were used to assess the normality of continuous variables. In the case of equal variances, means of continuous variables were compared with a 2tailed Student’s t test. When there was evidence of unequal variance, the p value of an approximate t test statistic was used. Multivariate analysis to discriminate between subjects based on stress thallium imaging was done using biochemical values as independent variables in the SAS logistic regression procedure. RESULTS m-h --lheteroqrm with famiiid m Baseline biochemical data are summarized in Table I for homozygotes and heterozygotes with PH. Mean plasma concenLIPOPROTEIN(A)IN FAMILIAL HYPERCHOLESTEROLEMIA 403
TABLE III Family Hypercholesterolemia
Females/males Age (years) Cholesterol (mmol/L) Triglycerides (mmol/L) LDLcholesterol (mmol/L) HDLcholesterol (mmol/L) ApoB (g/L) ApoA-I (g/L) Lp(al (mg/dl)
History
of Coronary
Heart
Disease
in Heterozygotes
with
Familial
Negative
Positive
p Value
5/6 18.0 f 1.17 7.95 2 0.42 (308 2 16.3 mg/dl)
B/l0 16.1 -t 3.65 6.80 + 0.35 (263 * 13.5 mg/dl)
NS 0.05
1.30 2 0.27 (115 + 23.9 mg/dl)
1.49 k 0.21 (132
f 18.6 mg/dl)
NS
6.24 k 0.35 (241 f 13.5 mg/dU
5.03 * 0.35 (195 k 13.5 mg/dl)
0.02
1.10 f 0.09 (43 f 3.5 mg/dU
1.08 f 0.06 (42 2 2.3 mg/dl)
2.06 f 0.11 1.30 f 0.06 14.4 2 4.1
1.73 f 0.08 1.21 + 0.06 30.3 ” 6.4
NS 0.03 NS NS (0.12)
I
Abbreviations as in Table 1.
TABLE
IV Thallium
Scans
in Homozygotes
with
Familial
Hypercholesterolemia
Normal Females/males Age (years) Cholesterol (mmol/L) Triglycerides (mmol/L) LDL cholesterol (mmol/L) HDLcholesterol (mmol/L) ApoB (g/L) ApA- (g/L) Lp(a) (mg/dL)
1
l/3 20.0 2 4.0 11.6 + 1.50 (449
Abnormal
+ 58.1 mg/dl)
O/3 8.75 2 2.46 17.4 + 0.31 (673 k 12 mg/dl)
p Value 0.08 0.05
1.99 k 0.65 (176 k 57.6 mg/dl)
1.58 2 0.39 (140
2 15.1 mg/dl)
NS
9.87 f 1.70 (382
16.0 of: 0.39 (619
f 15.1 mg/dl)
0.03
f 65.8 mg/dl)
0.81 f 0.04 (31 f 1.5 mg/dl)
0.65 2 0.05 (25 + 1.9 mg/dU
3.04 2 0.35 1.07 f 0.02 15.4 f 5.48
4.35 2 0.09 0.77 2 0.08 79.0 2 18.1
1
NS 0.03 0.03 0.03
Abbreviations as in Table I.
nations of total cholesterol, LDL cholesterol and apo--in--w lipoprotein B were signilicantly higher (approximately hlsW&s d heat draarrr Biochemical data are listed double) in homozygotes with FH than in heterozygotes in Table III for FH heterozygotes with and without fam(for each variable, p ~0.001). In addition, mean plasma ily histories of coronary artery disease. Mean plasma concentrations of HDL cholesterol and apolipoprotein concentrations of total cholesterol, LDL cholesterol and A-I were significantly lower (by approximately one apolipoprotein B were signilicantly lower (by 15%) in third) in homozygotes with FH than in heterozygotes subjects with than in those without a positive family his(p = 0.0001 and 0.0007, respectively). Mean plasma con- tory (p = 0.05, 0.02 and 0.03, respectively). Mean plascentrations of Lp(a) tended to be higher in homozygotes ma concentrations of Lp(a) tended to be higher in subwith FH than in heterozygotes (p = 0.08). Mean plasma jects with than without a positive family history (30 + concentrations of triglycerides were not signiticantly dif- 6.4 vs 14 + 4.1, p = 0.12). There were no signiticant different between the 2 groups. ferences in any other variable. --h--m -Upopotehrhho-mw=-m &WSS thallium scm~ Biochemical data am summarized stressthaHlumrcas:Biochemicaldata~summarized in Table II for l?H heterozygotes with and without evi- in Table IV for FH homozygotes with and without evidence of myocardial ischemia on stress thallium scans. dence of myocardial ischemia on stress thallium scans. There were no significant differences in any of the vari- Mean plasma concentrations of total cholesterol were ables, although there were trends to higher total choles- signiticantly higher (by 50%) in the FH homozygotes terol, LDL cholesterol and apolipoprotein B in subjects with positive stress thallium scans (p = 0.05), and there with positive stress thallium scans. There was a non- were nonsignikant trends toward higher mean plasma signiftcant trend toward a higher mean Lp(a) in het- concentrations of LDL cholesterol and apolipoprotein B. erozygotes who had a positive stress thallium scan than Mean plasma concentrations of apolipoprotein A-I were in those who had a negative stress thallium scan (39 + signiticantly lower (by 25%) in the FH homozygotes 12 vs 20 f 4.2 mg/dl, p = 0.10). Multivariate analysis with positive stress thallium scans @ = 0.03). Mean using Lp(a), LDL cholesterol and HDL cholesterol as plasma concentrations of Lp(a) were significantly highindependent variables failed to show sign&ant dis- er in FII homozygotes with than in those without stress crimination between subjects with normal and abnormal thallium abnormalities (79 f 18 vs 15 f 5.5 mg/dl, stress thallium scans (data not shown). p = 0.03). When data from homozygotes and heterozy404
THE AMERICAN JOURNAL OF CARDIOLOGY VOLUME 72
AU(;UST 15.1993
gotes with FH were combined, subjects with stress thallium abnormalities had signilicantly higher plasma concentrations of Lp(a) than those without (54 f 11 vs 20 + 3.8 mg/dl, p = 0.01). DISCUSSION The principal novel tinding of this study is that stress thallium abnormalities in asymptomatic children with FH, particularly homozygotes, were associated with significant elevations in plasma Lp(a) concentrations. In 2 previous studies of adult subjects with heterozygous FI-I, Lp(a) was also found to be a signiftcant risk factor for coronary artery disease.i4J7 In a third study, Lp(a) did not distinguish FH heterozygotes who had coronary artery disease from those who did not.18 The discrepancies between these studies may have resulted from differences in defining both FH and coronary artery disease. Our studies suggest that asymptomatic children with FH are more likely to have myocardial ischemia on stress thallium scans when their Lp(a) is high. However, larger prospective studies using established criteria to detine both FH genotypicahy and the cardiac abnormality angiogmphically may be the best approach to resolve these apparent discrepancies. The expression of stress thallium abnormalities in our study occurred at a younger age in homozygotes with FH than in heterozygotes (8.75 f 2.46 vs 15.7 f 1.72 years, p = 0.003). It appears that the development of stress thallium abnormalities may depend both on the patient’s age and on the severity of the lipoprotein abnormalities. Concentrations of total cholesterol, LDL cholesterol and apolipoprotein B were lower in subjects with parental histories of coronary artery disease. One interpretation of this observation is that other genetic factors, such as Lp(a), may be strong enough to confer an increased coronary artery disease risk even when the plasma concentrations of other apolipoprotein B-containing lipoproteins are relatively more favorable. Because Lp(a) is under strong genetic control, it is not surprising that FH heterozygotes with a parental history of coronary artery disease tended to have higher plasma Lp(a) than heterozygotes with no family history. Mbewu et all8 reported that Lp(a) in heterozygotes with FH (i.e., subjects with a single defective LDL receptor allele) is higher than Lp(a) in unaffected relatives. Our study appears to extend this conclusion to homozygotes with FH (i.e., subjects with 2 defective LDL receptor alleles), whose Lp(a) we found tended to be higher than in heterozygotes with FH. This observation lends some support to the concept that at least some functional LDL receptor activity is important in regulating plasma Lp(a). The relation between elevated plasma total cholesterol in FH and elevations in plasma Lp(a) is not firmly established. In some instances, mutations of the LDL receptor gene that cause F+I have been reported to be associated with elevated plasma Lp(a) concentrations.i7J9 Other reports suggest no effect of dysfunctional LDL receptors on plasma Lp(a).20*21 The discrepancies between the associations of various defects in the LDL receptor and plasma Lp(a) may be due to the fact that different subjects with FH have different gene mutations that in turn affect different functional domains.
Our homozygotes with FH had signilicantly lower plasma concentrations of HDL cholesterol and apolipoprotein A-I than the heterozygotes. The very lowest levels of HDL cholesterol and apolipoprotein A-I were seen in FH homozygotes with positive stress thallium scans. The observation of an inverse relation between I-IDL cholesterol and LDL cholesterol in FH has been noted previously,22 and others have reported that homozygotes have lower HDL cholesterol levels than heterozygotes.23 Recent studies of the in vivo catabolism of apolipoprotein A-I in a homozygote with FH indicate that low levels of HDL cholesterol and apolipoprotein A-I resulted both from defective synthesis and increased catabolism of apolipoprotein A-I.24 The unfavorable HDL cholesterol and apolipoprotein A-I could have been another independent factor associated with the stress thallium abnormalities in our homozygotes. There were no significant differences in other components of the lipoprotein profile between heterozygotes with and without stress thallium abnormalities. Apolipoprotein(a) is a polymorphic glycoprotein whose multiple alleles are encoded by a single 10~~1s.~ An added layer of complexity to the challenge of the use of Lp(a) in clinical decision-making stems from the fact that variation in the size of apolipoprotein(a) isoforms is associated with variation in plasma Lp(a) levels.’ We did not assessapolipoprotein(a) isoforms either by genetic analysis or by protein gel electrophoresis in our subjects with FH. Thus, it is possible that the differences in Lp(a) seen between the groups may have been due to an unequal distribution of apolipoprotein(a) alleles associated with extremes of Lp(a) concentrations in addition to the potential impact of defective LDL receptors on Lp(a). It could thus be argued that apolipoprotein(a) isoform determination should be included with Lp(a) phenotype determinations. However, there was no selection or ascertainment bias for those in our study sample; all subjects were asymptomatic. This means that if apolipoprotein(a) isoform distribution differed between subjects with normal and abnormal stress thallium scans, such a difference would have been random. In addition, analysis of apolipoprotein(a) isoforms is currently technically diflicult and expensive; furthermore no consensus exists for the role of apolipoprotein(a) isoform determination in the clinical assessment of coronary artery disease risk. Given these limitations, it seems reasonable to evaluate Lp(a) concentrations alone as an independent risk factor for atherosclerosis, and to accept the possibility that the assay of apolipoprotein(a) isoforms may be shown in the future to yield additional information. Acknowledgmentz We acknowledge the expert assistance of Susan McIntyre, Camilla Vezina, Teresa Lippingwell and Cheri Tully. 1. Scam AM. Lipoprotein(a). JAhfA 1992;267:332&3329. 2. Lackner C, Boawinkle E, L&fats C, Ranhig T, Hobbs H. Molecular basis of apoli~pmtein(a) size isofmm heterogeneity as revealed by pulsed field gel elechophoresis. J Clin Invesf 1991;87:215~2161. a. Ram&z LC, Amuz-Pacheco C, Lackner C, Albright G, Adams BV, R&in P. Lipoprotein(a) levels in diabetes mellihs: relationship to metabolic control. Ann Infern Med 19921 lWt2-47.
LIPOPROTEIN(A)IN FAMILIAL HYPERCHOLESTEROLEMIA 405
4. Black IW, Wilcken DEL. Decreases in apolipopmtein(a) after renal tmnsplantation: implications for lipoprotein(a) metabolism. C/in Chem 1992;38:35>357. 5. Maeda S, Abe A, Seishima M, Makino K, Noma A, Kawade M. Transient changes of serum lipoprotein(a) as an acute phase protein. Atherosclerosis 1989;78: 145-150. 6. Schenk I, Keller C, Hailer S, Wolfram G, zolner N. Reduction of Lp(a) by diferent methods of plasma exchange. K&I Wochenschr 1988;66:1191-1201. 7. Carlson LA, Hamsten A, Asplund A. Pronounced lowering of semm levels of lipoprotein Lp(a) in hyperlipidemic subjects treated with nicotinic acid. J Intern Med 1989;226:271-276. 8. Albers JJ, Taggart HMc, Applebaum-Bowden D, Haffner S, Chestnut CH III, Hazard WR. Reduction of lecithin-cholesterol acyltransferase, apolipoprotein D and Lp(a) lipoprotein with the anabolic steroid stanozoIo1. Biochem Siophys Acta 1984; 795~293-296. 9. Crook D, Sidhu M, Seed M, O’Donnell M, Stevenson JC. Lipoprotein Lp(a) levels are reduced by danazol, an anabolic steroid. Atherosclerosis 1992;92:4147. 10. Hegele RA, Freedman MR, Langer A, Connelly PW, Armstrong PW. Acute reduction of lipoprotein(a) by tissue-type plasminogen activator. Circukztion 1992; 85:20342038. 11. Albers JJ, Cabana VG, Wamick GR, Hazard WR. Lp(a) lipoprotein: relationship to sinking pre-beta lipoprotein, hyperlipoproteinemia and apolipqrotein B. Metabolism 1975;24:1047-1054. l2. Vessby B, Kosmer G, Lithell H, ‘Ihomis J. Diverging effects of cholestyramine on apolipopmtein B and Lp(a). Atherosclerosis 1982;44:61-71. 13. Kosmer GM, Gavish D, L.&pold B, Bolanm K, Weintraub MS, Breslow JL. HMG CoA reductase inhibitors lower LDL cholesterol without reducing Lp(a) Ievels. Circukztion 1989;80:1313-1319. 14. Seed D, Hopplicher F, McCarthy S, Thompson GR, Boerwinkle E, Utermann G. Relation of serum lipoprotein(a) conccntmtion and apolipqrotein(a) phenotype in coronary heart disease in patients with familial hypercholesterolemia. N Engr J Med 1990;322:1494-1499. 15. Mouratidis B, Vaughan-Neil EF, Gilday DL, Ash JM, Cullen-Dean G, McIntyre S, MacMillan JH, Rose V. Detection of silent myocardial ischemia in adoles-
406
THE AMERICAN JOURNAL OF CARDIOLOGY VOLUME 72
cents and young adults with familial hypercholestemlemia with single-photonemission computed tomography 201-thallium scanning. Am J Cardiol 1992;70: 1109-1112. 16. SAS Institute Incorporated SAS Language Guide for Personal Computers. Version 6.03. Cxy NC. pp 1409. 17. Wiiund 0, Angelin B, Oloffson S, Eriksson M, Fager G, Berglund L, Bondjers G. Apolipopmtein(a) and ischemic heart disease in familial hypercholesterolemia. Lancet 1990,2:136&1363. 18. Mbewu AD, Bshamagar PN, Durington PN, Hunt L, Ishola M, Arm1 S, Mackness M, Lockley P, Miller JP. Serum lipoprotein(a) in patients heterozygous for familial hypercholesterolemia, their relatives and unrelated control populations. Arterioscler Thromb 1991;11:94&946. 19. Utermann G, Hopplicher F, Dieplinger H, Seed H, Thompson G, B&e E. Defects in the low density lipoprotein receptor gene affect lp(a) levels: multiplicative interaction of two gene loci associated with premature atherosclerosis. froc Nat1 Acad Sn’ USA 1989;86:4171-4174. 20. Hegele RA, Sutherland S, Robertson M, Wu LL, Emi M, Hopkins PN, Williams RR, Lalouel JM. The effect of genetic determinants of low density lipoprotein levels on Lipoprotein(a). Chin Invest Med 1991;14:146-152. 21. Soutar AK, McCarthy SN, Seed M, Knight BL. Relationship between apolipqm t&n(a) phenotype, lipoprotein(a) concentration in plasma, and low density lipopmtein receptor function in a large kindred with familial hypercholeaterolemia due to the pro664--1eu mutation in the LDL receptor gene. J Chin Invest 1991;86:487492. 22. Streja D, Steiner G, Kwitemvich PO. Plasma high density lipoproteins and heart disease: studies in a large kindred with familial hypercholestemkxnia. Ann Intern Med 1978;89:871-880. 23. Kwiterovich PO, Fredrickson DS, Levy RI. Familial hypercholesterolemia: a study of its biochemical, genetic and clinical presentation in childhood. J Clin Invest 3974;53:1237-1246. 24. Schaefer JR, Rader DJ, Ikewaki K, Fairwell T, Zech LA, Kmdt MR, Davignon J, Gregg RE, Brewer HB Jr. In viva metabolism of apolipopmtein AI in a patient with homozygous familial hypexcholesterolemia. ArterioscLr Thromb 1992; 12&l?-848.
AUGUST 15,1993
Elevated Plasma Lipoprotein(a) Associated with Abnormal Stress Thallium Scans in Children with Familial Hypercholesterolemia RobertA. Hegele,MD, PhilipW.Connelly,PhD, GeraldineCulletiean, MSc, andVeraRose,MD ipoprotein(a) (Lp(a)) is a cholesteryl ester-rich particle consisting of 2 attached components: a low-density lipoprotein (LDL) particle whose apolipoprotein B-100 moiety is linked by disulphide bonds to a single large polymorphic glycoprotein, apolipoprotein(a).’ Plasma Lp(a) concentrations are strongly correlated with an increased prevalence of coronary artery disease, especially in patients younger than age 60 years and in those with strong family histories (reviewed in reference 1). The mechanisms underlying the associations with atherosclerosis are still unknown, although in vitro studies suggest that Lp(a) can be involved both in the inhibition of fibrinolysis acutely and in the formation of atheromatous plaques chr~nically.~ Plasma concentrations of Lp(a) vary over a wide range among persons, but remain stable in any particular subject.* Genetic and environmental factors affect the plasma concentration of Lp(a). For example, size variation at the apolipoprotein(a) gene is a major determinant of Lp(a) concentrations.* Also, diabetes mellitus,3 renal disease,4 postoperative state~,~medical treatment with plasma exchange,6 niacin7 stanozolol,8 danaz01,~ and tissue-type plasminogen activatorlO can also affect plasma concentrations of Lp(a). Studies in tissue culture and in vivo are inconclusive regarding role of the LDL receptor in the catabolism of Lp(a).l Medications that up-regulate the LDL receptor, such as cholestyramine and lovastann, do not lower Lp(a). 11-13Although the relation between LDL receptor function and plasma Lp(a) is not clear, elevated Lp(a) can accelerate the expression of coronary artery disease in adults with familial hypercholesterolemia (FH).14 We hypothesized that increased plasma Lp(a) was associated with more severe coronary atherosclerosis in children with FH. We studied asymptomatic heterozygotes with FH and homozygotes with stress thallium scintigraphy and plasma Lp(a) determinations.
L
Plasma lipoprotein(a) (@(a)) conce&Wi BCB associatedwithaniBrQskofcoronary* twy disease in adults with fsmilial hypmehoksterolemia (FH). We hypothesized that Lp(a) f%nmdMions in children with FH were higher among those with myacardial ischemia on stress thallium scans and among those with a family bib comnary artery dii. mem toryofprematu~ ty-nhm asymptometie heterozygotes with FH (mnge9to23years)and7homozygotes(range4 to 13 years) were evaluated with clinical asses!+ merit, llpoprobin measurement and stress mallC umsums.Comparedwithsubjectswithnormal stress thallhun scans, mean @(a) was sigMicantly hm in homozygotes with stress thallium ab normaltties (79 + l6 vs 15 f 5.5 mg/dl, p q 0.03), andtemkdtobehiiinhebrozygot~with stress thallium alMomIalitii(39+12vs20 -r4.2 mg/dl, p q 0.10). @(a)tendedto be hii in heterozygotes with a family history of premature ~arteydisease(30+6.4vsl4&4.1 mg/dl, p q 0.12). It is concluded that Q(a) is high er in hypeMolesterolemic children who have ab normal stress thallium e Lp(a) may be useful inassessingcoronary artery disease risk in chic dren with M. (Am J Cardiil1993;72:402-403)
MEWlOOS From the Departments of Medicine and Clinical Biochemistry, St. Michael’s Hospital, the Division of Cardiology, Department of Paediiatrics, The Hospital for Sick Children, University of Toronto, Toronto, Ontario, Canada. This study was supported by grant Al663 from the Heart and Stroke Foundations of Ontario and Canada. Dr. Hegele is a McDonald Scholar of the Heart and Stroke Foundation of Canada. Manuscript received November 30, 1992; revised manuscript received December 18, 1992, and accepted March 10,1993. Address for reprints: Robert A. Hegele, MD, DNA Research Laboratory, St. Michael’s Hospital, 38 Shuter Street, Toronto, Ontario, Canada M5B lA6. This work was presented in part at the 65th Scientific Sessions of the American Heart Association (abstract #0047). 402
THE AMERICAN JOURNAL OF CARDIOLOGY VOLUME 72
Patients A total of 36 asymptomatic children, and adolescent and young adult subjects with FH were evaluated with single-photon emission computed tomography 201~thallium scanning after exercise and plasma determinations of lipoproteins. Twenty-two male and 14 female subjects (mean age 16 years, range 4 to 27) were assessed. Seven patients (6 males, 1 female) were homozygotes with FH and 29 patients (16 males, 13 females) were heterozygotes with FH. All subjects were nonsmokers. All FH homozygotes were ascertained through referral to the lipid clinic based on dermatologic tindings
AUGUST 15.1993
TABLE
I Baseline
Biochemical
Variables
Heterozygotes Females/males Age (years) Total cholesterol (mmol/L) Triglycerides (mmol/L) LDL cholesterol (mmol/L) HDL cholesterol (mmol/L) ApoB (g/L) ApoA-I (g/L) Lp(a) (mg/dl)
Homozygotes
p Value
l/6
16/13 16.8 f 0.70 7.23 -t 0.28 (279 2 11 mg/dl)
13.6 + 3.0 14.9 k 1.31 (577
k 51 mg/dl)
NS 0.001
1.42 + 0.16 (126
1.75 f 0.33 (155 2 29 mg/dl)
NS
f 14 mg/dl)
5.49 f 0.27 (212 2 10 mg/dl)
13.4 -t 1.40 (519
f 54 mg/dl)
0.001
1.09 f 0.05 (42 f 1.9 mg/dl)
0.72 2 0.04 (28 -+ 1.5 mg/dl)
0.0001
1.85 f 0.07 1.25 + 0.04 24.3 +- 4.46
3.79 f 0.30 0.90 t 0.08 51.8 + 16.2
0.0005 0.0007 0.08
Apo = apolipoprotein; HDL = high-density lipoprotein: LDL = low-densky lipopmtein; Lp = plasma lipoprotem.
TABLE
II
Thallium
Scans
in Heterozygotes
with
Familial
Hypercholesterolemia
Normal Females/males Age (years) Cholesterol (mmol/L) Triglycerides (mmol/L) LDL cholesterol (mmol/L) HDL cholesterol (mmol/L) ApoB (g/L) ApoA-I (g/L) Lp(a) (mg/dl)
8714 17.1 7.05 1.49 5.30 1.06 1.81 1.21 19.5
k f f + f + 2 +
0.76 0.31 0.21 0.30 0.05 0.08 0.05 4.2
Abnormal
(273 2 14.3 mgldl) (129 f 18.7 mg/dl) (205 f 11.6 mgldl) (41 + 1.9 mg/dl)
512 15.7 7.81 1.17 6.11 1.17 1.98 1.38 39.1
2 $ k + + $ ” -c
1.72 0.63 0.17 0.59 0.11 0.18 0.08 11.9*
(302 (104 (237 (45
-c 24.4 mg/dl) k 15.1 mg/dl) +- 22.8 mg/dl) -t 4.2 mg/dl)
*p = 0.10. All differences are nonsignificant (p >O.lO). Abbrewations as in Table I.
such as planar or tendinous xanthomata, or both. All heterozygotes with PI-I were ascertained through a family history of hyperlipidemia. The diagnosis was established by family history and lipid prolile, and, additionally in homozygotes, typical clinical features such as planar and tendinous xanthomas. Parental and family histories were obtained with regard to myocardial infarction, coronary bypass surgery, cardiac death, or a combination of all 3. With use of these criteria, 18 heterozygotes with FH had a positive parental history of premature vascular disease. All homozygotes with FH had parents who were alive and well with no history of coronary artery disease. Medical treatment included a fat-modified diet with cholestyramine; compliance with treatment was variable. ThaNurn scarmin%: Informed consent was obtained from all subjects. Bach patient was asked to fast on the study day. Exercise was performed as described,t5 with termination either because of maximum work load, electrocardiographic abnormalities or fatigue. The maximal 201&a&m dose of 74 MBq, based on weight, was administered intravenously 1 minute before the exercise was stopped. Poststress images were obtained 4 minutes after and redistribution images 3 hours after the exercise ended. Patients were allowed a light meal between poststress and redistribution images to minimize splanchnic blood flow. Studies were performed using either a General Electric Starcam M or Starcam gamma camera (Milwaukee, Wisconsin) with technical spectications as de~cribed.~~An abnormality was diagnosed based on photon-deficient areas in the myocardium in poststress
images.15 In homozygotes with PI-I, there was complete concordance between the presence of a perfusion defect observed with stress thallium scanning and the presence of anatomically appropriate coronary arterial lesions observed with angiography.15 Heterozygotes with PI-I did not undergo angiography. Normolipidemic children without heart disease were not studied with stress thallium scanning. Biochemkal anatysis: Aliquots of fasting plasma frozen at -70°C were used for determinations of total cholesterol, triglycerides, LDL cholesterol and high-density lipoprotein (IIDL) cholesterol, apolipoproteins A-I and B and Lp(a) using described methodsto StatIstical malysk SAS was used for all statistical analyses.l6 Data were summarized descriptively with mean + SE. Lp(a) was transformed for ana.lysis.‘ONormal probability plots were used to assess the normality of continuous variables. In the case of equal variances, means of continuous variables were compared with a 2tailed Student’s t test. When there was evidence of unequal variance, the p value of an approximate t test statistic was used. Multivariate analysis to discriminate between subjects based on stress thallium imaging was done using biochemical values as independent variables in the SAS logistic regression procedure. RESULTS m-h --lheteroqrm with famiiid m Baseline biochemical data are summarized in Table I for homozygotes and heterozygotes with PH. Mean plasma concenLIPOPROTEIN(A)IN FAMILIAL HYPERCHOLESTEROLEMIA 403
TABLE III Family Hypercholesterolemia
Females/males Age (years) Cholesterol (mmol/L) Triglycerides (mmol/L) LDLcholesterol (mmol/L) HDLcholesterol (mmol/L) ApoB (g/L) ApoA-I (g/L) Lp(al (mg/dl)
History
of Coronary
Heart
Disease
in Heterozygotes
with
Familial
Negative
Positive
p Value
5/6 18.0 f 1.17 7.95 2 0.42 (308 2 16.3 mg/dl)
B/l0 16.1 -t 3.65 6.80 + 0.35 (263 * 13.5 mg/dl)
NS 0.05
1.30 2 0.27 (115 + 23.9 mg/dl)
1.49 k 0.21 (132
f 18.6 mg/dl)
NS
6.24 k 0.35 (241 f 13.5 mg/dU
5.03 * 0.35 (195 k 13.5 mg/dl)
0.02
1.10 f 0.09 (43 f 3.5 mg/dU
1.08 f 0.06 (42 2 2.3 mg/dl)
2.06 f 0.11 1.30 f 0.06 14.4 2 4.1
1.73 f 0.08 1.21 + 0.06 30.3 ” 6.4
NS 0.03 NS NS (0.12)
I
Abbreviations as in Table 1.
TABLE
IV Thallium
Scans
in Homozygotes
with
Familial
Hypercholesterolemia
Normal Females/males Age (years) Cholesterol (mmol/L) Triglycerides (mmol/L) LDL cholesterol (mmol/L) HDLcholesterol (mmol/L) ApoB (g/L) ApA- (g/L) Lp(a) (mg/dL)
1
l/3 20.0 2 4.0 11.6 + 1.50 (449
Abnormal
+ 58.1 mg/dl)
O/3 8.75 2 2.46 17.4 + 0.31 (673 k 12 mg/dl)
p Value 0.08 0.05
1.99 k 0.65 (176 k 57.6 mg/dl)
1.58 2 0.39 (140
2 15.1 mg/dl)
NS
9.87 f 1.70 (382
16.0 of: 0.39 (619
f 15.1 mg/dl)
0.03
f 65.8 mg/dl)
0.81 f 0.04 (31 f 1.5 mg/dl)
0.65 2 0.05 (25 + 1.9 mg/dU
3.04 2 0.35 1.07 f 0.02 15.4 f 5.48
4.35 2 0.09 0.77 2 0.08 79.0 2 18.1
1
NS 0.03 0.03 0.03
Abbreviations as in Table I.
nations of total cholesterol, LDL cholesterol and apo--in--w lipoprotein B were signilicantly higher (approximately hlsW&s d heat draarrr Biochemical data are listed double) in homozygotes with FH than in heterozygotes in Table III for FH heterozygotes with and without fam(for each variable, p ~0.001). In addition, mean plasma ily histories of coronary artery disease. Mean plasma concentrations of HDL cholesterol and apolipoprotein concentrations of total cholesterol, LDL cholesterol and A-I were significantly lower (by approximately one apolipoprotein B were signilicantly lower (by 15%) in third) in homozygotes with FH than in heterozygotes subjects with than in those without a positive family his(p = 0.0001 and 0.0007, respectively). Mean plasma con- tory (p = 0.05, 0.02 and 0.03, respectively). Mean plascentrations of Lp(a) tended to be higher in homozygotes ma concentrations of Lp(a) tended to be higher in subwith FH than in heterozygotes (p = 0.08). Mean plasma jects with than without a positive family history (30 + concentrations of triglycerides were not signiticantly dif- 6.4 vs 14 + 4.1, p = 0.12). There were no signiticant different between the 2 groups. ferences in any other variable. --h--m -Upopotehrhho-mw=-m &WSS thallium scm~ Biochemical data am summarized stressthaHlumrcas:Biochemicaldata~summarized in Table II for l?H heterozygotes with and without evi- in Table IV for FH homozygotes with and without evidence of myocardial ischemia on stress thallium scans. dence of myocardial ischemia on stress thallium scans. There were no significant differences in any of the vari- Mean plasma concentrations of total cholesterol were ables, although there were trends to higher total choles- signiticantly higher (by 50%) in the FH homozygotes terol, LDL cholesterol and apolipoprotein B in subjects with positive stress thallium scans (p = 0.05), and there with positive stress thallium scans. There was a non- were nonsignikant trends toward higher mean plasma signiftcant trend toward a higher mean Lp(a) in het- concentrations of LDL cholesterol and apolipoprotein B. erozygotes who had a positive stress thallium scan than Mean plasma concentrations of apolipoprotein A-I were in those who had a negative stress thallium scan (39 + signiticantly lower (by 25%) in the FH homozygotes 12 vs 20 f 4.2 mg/dl, p = 0.10). Multivariate analysis with positive stress thallium scans @ = 0.03). Mean using Lp(a), LDL cholesterol and HDL cholesterol as plasma concentrations of Lp(a) were significantly highindependent variables failed to show sign&ant dis- er in FII homozygotes with than in those without stress crimination between subjects with normal and abnormal thallium abnormalities (79 f 18 vs 15 f 5.5 mg/dl, stress thallium scans (data not shown). p = 0.03). When data from homozygotes and heterozy404
THE AMERICAN JOURNAL OF CARDIOLOGY VOLUME 72
AU(;UST 15.1993
gotes with FH were combined, subjects with stress thallium abnormalities had signilicantly higher plasma concentrations of Lp(a) than those without (54 f 11 vs 20 + 3.8 mg/dl, p = 0.01). DISCUSSION The principal novel tinding of this study is that stress thallium abnormalities in asymptomatic children with FH, particularly homozygotes, were associated with significant elevations in plasma Lp(a) concentrations. In 2 previous studies of adult subjects with heterozygous FI-I, Lp(a) was also found to be a signiftcant risk factor for coronary artery disease.i4J7 In a third study, Lp(a) did not distinguish FH heterozygotes who had coronary artery disease from those who did not.18 The discrepancies between these studies may have resulted from differences in defining both FH and coronary artery disease. Our studies suggest that asymptomatic children with FH are more likely to have myocardial ischemia on stress thallium scans when their Lp(a) is high. However, larger prospective studies using established criteria to detine both FH genotypicahy and the cardiac abnormality angiogmphically may be the best approach to resolve these apparent discrepancies. The expression of stress thallium abnormalities in our study occurred at a younger age in homozygotes with FH than in heterozygotes (8.75 f 2.46 vs 15.7 f 1.72 years, p = 0.003). It appears that the development of stress thallium abnormalities may depend both on the patient’s age and on the severity of the lipoprotein abnormalities. Concentrations of total cholesterol, LDL cholesterol and apolipoprotein B were lower in subjects with parental histories of coronary artery disease. One interpretation of this observation is that other genetic factors, such as Lp(a), may be strong enough to confer an increased coronary artery disease risk even when the plasma concentrations of other apolipoprotein B-containing lipoproteins are relatively more favorable. Because Lp(a) is under strong genetic control, it is not surprising that FH heterozygotes with a parental history of coronary artery disease tended to have higher plasma Lp(a) than heterozygotes with no family history. Mbewu et all8 reported that Lp(a) in heterozygotes with FH (i.e., subjects with a single defective LDL receptor allele) is higher than Lp(a) in unaffected relatives. Our study appears to extend this conclusion to homozygotes with FH (i.e., subjects with 2 defective LDL receptor alleles), whose Lp(a) we found tended to be higher than in heterozygotes with FH. This observation lends some support to the concept that at least some functional LDL receptor activity is important in regulating plasma Lp(a). The relation between elevated plasma total cholesterol in FH and elevations in plasma Lp(a) is not firmly established. In some instances, mutations of the LDL receptor gene that cause F+I have been reported to be associated with elevated plasma Lp(a) concentrations.i7J9 Other reports suggest no effect of dysfunctional LDL receptors on plasma Lp(a).20*21 The discrepancies between the associations of various defects in the LDL receptor and plasma Lp(a) may be due to the fact that different subjects with FH have different gene mutations that in turn affect different functional domains.
Our homozygotes with FH had signilicantly lower plasma concentrations of HDL cholesterol and apolipoprotein A-I than the heterozygotes. The very lowest levels of HDL cholesterol and apolipoprotein A-I were seen in FH homozygotes with positive stress thallium scans. The observation of an inverse relation between I-IDL cholesterol and LDL cholesterol in FH has been noted previously,22 and others have reported that homozygotes have lower HDL cholesterol levels than heterozygotes.23 Recent studies of the in vivo catabolism of apolipoprotein A-I in a homozygote with FH indicate that low levels of HDL cholesterol and apolipoprotein A-I resulted both from defective synthesis and increased catabolism of apolipoprotein A-I.24 The unfavorable HDL cholesterol and apolipoprotein A-I could have been another independent factor associated with the stress thallium abnormalities in our homozygotes. There were no significant differences in other components of the lipoprotein profile between heterozygotes with and without stress thallium abnormalities. Apolipoprotein(a) is a polymorphic glycoprotein whose multiple alleles are encoded by a single 10~~1s.~ An added layer of complexity to the challenge of the use of Lp(a) in clinical decision-making stems from the fact that variation in the size of apolipoprotein(a) isoforms is associated with variation in plasma Lp(a) levels.’ We did not assessapolipoprotein(a) isoforms either by genetic analysis or by protein gel electrophoresis in our subjects with FH. Thus, it is possible that the differences in Lp(a) seen between the groups may have been due to an unequal distribution of apolipoprotein(a) alleles associated with extremes of Lp(a) concentrations in addition to the potential impact of defective LDL receptors on Lp(a). It could thus be argued that apolipoprotein(a) isoform determination should be included with Lp(a) phenotype determinations. However, there was no selection or ascertainment bias for those in our study sample; all subjects were asymptomatic. This means that if apolipoprotein(a) isoform distribution differed between subjects with normal and abnormal stress thallium scans, such a difference would have been random. In addition, analysis of apolipoprotein(a) isoforms is currently technically diflicult and expensive; furthermore no consensus exists for the role of apolipoprotein(a) isoform determination in the clinical assessment of coronary artery disease risk. Given these limitations, it seems reasonable to evaluate Lp(a) concentrations alone as an independent risk factor for atherosclerosis, and to accept the possibility that the assay of apolipoprotein(a) isoforms may be shown in the future to yield additional information. Acknowledgmentz We acknowledge the expert assistance of Susan McIntyre, Camilla Vezina, Teresa Lippingwell and Cheri Tully. 1. Scam AM. Lipoprotein(a). JAhfA 1992;267:332&3329. 2. Lackner C, Boawinkle E, L&fats C, Ranhig T, Hobbs H. Molecular basis of apoli~pmtein(a) size isofmm heterogeneity as revealed by pulsed field gel elechophoresis. J Clin Invesf 1991;87:215~2161. a. Ram&z LC, Amuz-Pacheco C, Lackner C, Albright G, Adams BV, R&in P. Lipoprotein(a) levels in diabetes mellihs: relationship to metabolic control. Ann Infern Med 19921 lWt2-47.
LIPOPROTEIN(A)IN FAMILIAL HYPERCHOLESTEROLEMIA 405
4. Black IW, Wilcken DEL. Decreases in apolipopmtein(a) after renal tmnsplantation: implications for lipoprotein(a) metabolism. C/in Chem 1992;38:35>357. 5. Maeda S, Abe A, Seishima M, Makino K, Noma A, Kawade M. Transient changes of serum lipoprotein(a) as an acute phase protein. Atherosclerosis 1989;78: 145-150. 6. Schenk I, Keller C, Hailer S, Wolfram G, zolner N. Reduction of Lp(a) by diferent methods of plasma exchange. K&I Wochenschr 1988;66:1191-1201. 7. Carlson LA, Hamsten A, Asplund A. Pronounced lowering of semm levels of lipoprotein Lp(a) in hyperlipidemic subjects treated with nicotinic acid. J Intern Med 1989;226:271-276. 8. Albers JJ, Taggart HMc, Applebaum-Bowden D, Haffner S, Chestnut CH III, Hazard WR. Reduction of lecithin-cholesterol acyltransferase, apolipoprotein D and Lp(a) lipoprotein with the anabolic steroid stanozoIo1. Biochem Siophys Acta 1984; 795~293-296. 9. Crook D, Sidhu M, Seed M, O’Donnell M, Stevenson JC. Lipoprotein Lp(a) levels are reduced by danazol, an anabolic steroid. Atherosclerosis 1992;92:4147. 10. Hegele RA, Freedman MR, Langer A, Connelly PW, Armstrong PW. Acute reduction of lipoprotein(a) by tissue-type plasminogen activator. Circukztion 1992; 85:20342038. 11. Albers JJ, Cabana VG, Wamick GR, Hazard WR. Lp(a) lipoprotein: relationship to sinking pre-beta lipoprotein, hyperlipoproteinemia and apolipqrotein B. Metabolism 1975;24:1047-1054. l2. Vessby B, Kosmer G, Lithell H, ‘Ihomis J. Diverging effects of cholestyramine on apolipopmtein B and Lp(a). Atherosclerosis 1982;44:61-71. 13. Kosmer GM, Gavish D, L.&pold B, Bolanm K, Weintraub MS, Breslow JL. HMG CoA reductase inhibitors lower LDL cholesterol without reducing Lp(a) Ievels. Circukztion 1989;80:1313-1319. 14. Seed D, Hopplicher F, McCarthy S, Thompson GR, Boerwinkle E, Utermann G. Relation of serum lipoprotein(a) conccntmtion and apolipqrotein(a) phenotype in coronary heart disease in patients with familial hypercholesterolemia. N Engr J Med 1990;322:1494-1499. 15. Mouratidis B, Vaughan-Neil EF, Gilday DL, Ash JM, Cullen-Dean G, McIntyre S, MacMillan JH, Rose V. Detection of silent myocardial ischemia in adoles-
406
THE AMERICAN JOURNAL OF CARDIOLOGY VOLUME 72
cents and young adults with familial hypercholestemlemia with single-photonemission computed tomography 201-thallium scanning. Am J Cardiol 1992;70: 1109-1112. 16. SAS Institute Incorporated SAS Language Guide for Personal Computers. Version 6.03. Cxy NC. pp 1409. 17. Wiiund 0, Angelin B, Oloffson S, Eriksson M, Fager G, Berglund L, Bondjers G. Apolipopmtein(a) and ischemic heart disease in familial hypercholesterolemia. Lancet 1990,2:136&1363. 18. Mbewu AD, Bshamagar PN, Durington PN, Hunt L, Ishola M, Arm1 S, Mackness M, Lockley P, Miller JP. Serum lipoprotein(a) in patients heterozygous for familial hypercholesterolemia, their relatives and unrelated control populations. Arterioscler Thromb 1991;11:94&946. 19. Utermann G, Hopplicher F, Dieplinger H, Seed H, Thompson G, B&e E. Defects in the low density lipoprotein receptor gene affect lp(a) levels: multiplicative interaction of two gene loci associated with premature atherosclerosis. froc Nat1 Acad Sn’ USA 1989;86:4171-4174. 20. Hegele RA, Sutherland S, Robertson M, Wu LL, Emi M, Hopkins PN, Williams RR, Lalouel JM. The effect of genetic determinants of low density lipoprotein levels on Lipoprotein(a). Chin Invest Med 1991;14:146-152. 21. Soutar AK, McCarthy SN, Seed M, Knight BL. Relationship between apolipqm t&n(a) phenotype, lipoprotein(a) concentration in plasma, and low density lipopmtein receptor function in a large kindred with familial hypercholeaterolemia due to the pro664--1eu mutation in the LDL receptor gene. J Chin Invest 1991;86:487492. 22. Streja D, Steiner G, Kwitemvich PO. Plasma high density lipoproteins and heart disease: studies in a large kindred with familial hypercholestemkxnia. Ann Intern Med 1978;89:871-880. 23. Kwiterovich PO, Fredrickson DS, Levy RI. Familial hypercholesterolemia: a study of its biochemical, genetic and clinical presentation in childhood. J Clin Invest 3974;53:1237-1246. 24. Schaefer JR, Rader DJ, Ikewaki K, Fairwell T, Zech LA, Kmdt MR, Davignon J, Gregg RE, Brewer HB Jr. In viva metabolism of apolipopmtein AI in a patient with homozygous familial hypexcholesterolemia. ArterioscLr Thromb 1992; 12&l?-848.
AUGUST 15,1993