Phenotypic expression of familial hypercholesterolaemia in Central and Southern Tunisia

Phenotypic expression of familial hypercholesterolaemia in Central and Southern Tunisia

ATHEROSCLEROSIS Atherosclerosis 104 (1993) 153-158 Phenotypic expression of familial hypercholesterolaemia Central and Southern Tunisia in M.N. S...

558KB Sizes 0 Downloads 22 Views



104 (1993) 153-158

Phenotypic expression of familial hypercholesterolaemia Central and Southern Tunisia


M.N. Slimane”, H. Pousseb, F. Maatougc, M. Hammami*a, M.H. Ben Farhatc ‘Laboraroire de Biochitnie, FacultP de Medecine 5019. Monastir, Tunisia Services de hPediarrie et de “Cardiologie, Hopiral de Monastir, Monastir. Tunisia


7 July 1993; accepted

27 August


Abstract We studied 14 families with familial hypercholesterolaemia (FH) from Central and Southern Tunisia. Twenty-six living homozygotes were identified in these areas of whom 24 homozygotes and 27 of their obligate heterozygote parents are the subject of this report. Ten of the 14 families are unrelated and in 9 of them there were consanguineous marriages. The mean age of homozygotes was 16 for females (range 2.5-40) and 12.5 for males (range 2-34). All the homozygotes had extensive xanthomatosis and showed variable clinical manifestations of coronary heart disease (CHD). Plasma total and LDL cholesterol levels averaged 18 and 16.9 mmol/l, respectively. Mean high density lipoprotein (HDL) cholesterol values were 0.48 mmol/l for males and 0.70 mmol/l for females. The mean age of the obligate heterozygotes was 44 (range 32-62 years) for mothers and 51 (range 35-80 years) for fathers. None of them had tendon xanthomas, not even the oldest, who was aged 80. Only 5 of the 27 obligate heterozygotes had developed CHD (aged 34-58). Plasma cholesterol levels varied more than twofold (4.1- 10 mmol/l) and averaged 6.79 and 7.51 mmolil for males and females, respectively. LDL cholesterol values were below the age- and sex-related 95th percentile from the Lipid Research Clinics Prevalence Study in 46% of male and 30% of female heterozygotes. The frequency of homozygotes was 1:125 000 and the minimum estimated frequency of heterozygotes was 1:165 in Central and Southern Tunisia. Only Afrikaners in South Africa and French Canadians have such high frequencies of FH. Key words: Serum lipids; Apolipoproteins;

Familial hypercholesterolaemia;

1. Introduction

Familial hypercholesterolaemia (FH) is an inherited disorder characterized by increased levels of low density lipoprotein (LDL). In most instances this is due to genetic deficiency or absence of cell surface receptors for LDL [l]. Subjects * Corresponding


0021-9150/93/%06.00 0 1993 Elsevier Scientific SSDl 0021-9150(93)05139-v



Xanthoma; Tunisia

homozygous for FH usually have LDL cholesterol levels above 15 mmol/l, develop tendon and tuberous xanthomas in childhood and die from severe atherosclerotic disease before aged 30. The diagnostic hallmark of heterozygotes is the presence of tendon xanthomas. Commonly these appear from the age of 20 onwards [2] and are related to elevated plasma cholesterol levels. The frequencies of FH homozygotes and Ltd. All rights reserved.


heterozygotes in European and North American white populations are about 1:106 and 1:500, respectively [3]. The prevalence of FH in Tunisia is unknown and there have been no reports which describe the clinical and biochemical features of FH in that country. In this paper we describe the clinical findings, together with lipid and lipoprotein analysis, of the affected members of 14 families in Central and Southern Tunisia diagnosed as having FH. 2. Patients and methods Blood samples were obtained from 51 members of the 14 families (27 parents - 14 mothers and 13 fathers, of 24 homozygotes - 17 females and 7 males). A history of symptoms of CHD, drug treatment and risk factors was obtained and the presence of xanthomas was noted. Diagnosis of the homozygous state was based on the presence of hypercholesterolaemia and the appearance of xanthomas in the first decade of life. Cardiac investigations were performed at the Monastir Hospital (Tunisia) including electrocardiography and echocardiography in all the patients, and cardiac catheterization and coronary angiography for homozygotes aged more than 8 years and for any symptomatic parents. Venous blood was collected into tubes containing EDTA (1 mg/ml) and plasma was separated by centrifugation at 2000 x g for 10 min. Plasma total cholesterol (TC) and triglyceride (TG) were determined by enzymatic methods [4,5] using commercial kits (Boehringer-Mannheim, Germany). HDL cholesterol was determined by assaying the concentration of cholesterol in the supernatant obtained after sodium phosphotungstate MgC12 precipitation of plasma [6]. LDL cholesterol was calculated using the formula of Friedewald as long as TG values were less than 4.5 mmol/l [7]. Plasma apolipoprotein A-I (apo A-I) and apolipoprotein B (apo B) were measured by electroimmunoassay

PI. 2.1. Statisrics The unpaired t-test was used to compare differences between means.

M.N. Slimane et al. /Atherosclerosis 104 (1993) 153-158

3. Results 3.1. Clinical data Homozygotes. The clinical characteristics of the 17 female and 7 male homozygotes are shown in Table 1. The ages of females ranged between 2.5 and 40 years with a mean of 16 years. The ages of males ranged between 2 and 34 years with a mean of 12.5 years. Ten male homozygotes aged lo-25 and 6 females aged 6-24 years were known to have died during the 15 years before the start of this study. Tuberose xanthomas appeared in all patients before the first decade of life, the sites of predilection being the elbows, knees and ankles. Tendon xanthomas were present in 65% and 85% of females and males, respectively. Cornea1 arcus was seen in 65% and 86% of females and males, respectively, mostly those aged over 10 years. Coronary heart disease (CHD), as shown by angina pectoris, myocardial infarction, electrocardiographic changes and/or abnormalities on coronary angiography, was noted in 53% and 30% of females and males, respectively. A loud ejection systolic murmur in the aortic area was heard in 94% of females and 57% of males, in whom the aortic root was shown to be narrowed and thickened with distortion of the aortic valve movement on echocardiography. Four females and 1 male had peripheral arterial disease. Family history showed that 59% of female and 85% of male homozygotes were born of a con-

Table 1 Clinical characteristics of 24 Tunisian homozygotes

Age (years)” Body mass index (kg/m’)” Tuberose xantbomas Tendon xanthomas lschaemic heart disease Aortic root disease Peripheral arterial disease “Mean * SD.

Female (n = 17)

Male (n = 7)

16.02 f 9.23 18.85 f 4.6

12.57 zk 10.7 18.66 f 3.53

17 11 9 16 4

7 6 2 4


M.N. Slimane et al. /Atherosclerosis

104 (1993)

Table 2 Clinical characteristics of obligate heterozygotes

(n = 14)

Fathers s (n = 13)

44.07 f 12.09 28.15 f 2.91

51.61 f 14.16 25.78 * 3.97

Mothers Age (years)” Body mass index (kg/m2)” Tendon xanthomas Xanthelasma Cornea1 arcus lschaemic heart disease Peripheral arterial disease

0 2 I 3 I

0 I I 2 2

“Mean f SD

sanguineous marriage as opposed to 30% of nonhomozygotes in Central and Southern Tunisia. In an additional 30% the parents were originally from the same area. Heterozygotes. The 14 mothers and 13 fathers of homozygous children were regarded as obligate heterozygotes. As shown in Table 2 the age range of mothers was 32-62 years with a mean of 44 years. Of these, 79% were less than 50 years old. The age range of fathers was 35-80 years with a mean of 5 1 years, 46% being less than 50 years old. None of the obligate heterozygote parents had tendon xanthomas. Cornea1 arcus was seen only in one mother and one father. Ischaemic heart

Table 3 Concentrations (mean (S.D.)) of serum lipids lipoproteins and apolipoproteins of 24 Tunisian homozygotes Female (n = 17) Age (years) Total cholesterol (mmohl) Triglycerides (mmohl) Anti log,, triglycerides (mmol/l) LDL cholesterol (mmol/l) HDL cholesterol (mmolil) Apo A-I (mgdl) Apo B (mg/dl)

disease and peripheral arterial disease were remarkably infrequent in the heterozygote parents in this study. Three mothers and two fathers had CHD and one mother and two fathers had peripheral arterial disease. 3.2. Laboratory data Values of serum total cholesterol in homozygotes varied from 12 to 26 mmol/l with mean values of 17.9 and 18.1 mmol/l in male and females, respectively, as shown in Table 3. The mean values of triglyceride were normal and similar but HDL cholesterol levels were distinctly low in both female and male homozygotes (0.70 f 0.33 mmol/l and 0.48 + 0.17 mmol/l, respectively). Serum LDL cholesterol concentrations varied from 11.6 to 22.9 mmol/l with a mean of 16.9 mmol/l in both sexes. Serum cholesterol levels of the obligate heterozygote parents varied from 4.1 to 10 mmol/l with a mean of 7.51 f 1.95 and 6.79 f 1.55 for females and males, respectively (Table 4). However, differences between males and females were not significant nor was there any correlation between the age of the patients and their plasma cholesterol. Heterozygote parents had higher levels of HDL cholesterol than their homozygous offspring (0.98 f 0.27 mmol/l and 0.91 f 0.14 mmol/l in

Table 4 Concentrations (mean (SD.)) of serum lipids, lipoproteins, and apolipoproteins of obligate heterozygotes

Male (n = 7)

16.02 * 9.2 18.09 f 2.97

12.57 f 10.7 17.92 f 5.24

0.96 f 0.38 I.13 f 0.19

I.17 f 0.63 1.27 f 0.22

16.87 f 2.86

16.90 f 5.15

?? 0.33

0.48 +Z0.17

87.58 f 18.2 235.17 f 45.6

75.42 f 8.36 238.0 f 68.7




Age (years) Total cholesterol (mmol/l) Triglyceride (mmolil) Anti log,,, triglycerides (mmol/l) LDL cholesterol (mmol/l) HDL cholesterol (mmolil) Apo A-I (mg/dl) Apo B (mg/dl) ‘P < 0.05

Mothers (n = 13)

Fathers (n = 14)

44.07 f 12.09 7.51 ?? 1.95

51.61 f 14.16 6.79 f 1.55

0.90 ?? 0.31 I.17 f 0.16

I.16 f 0.65 1.06 f 0.24

5.95 f 1.88

5.41 f I.18

0.98 f 0.27

0.91 f 0.14

128.5 f 15.1 136.9 f 23.4

I I I.3 f 19.7’ 124.3 f 31.3


M. N. Slimune

et al. /Atherosclerosis

104 (1993)


compared with normal values for North America, 46% and 30% of fathers and mothers, respectively, had an LDL cholesterol which was below the age and sex related 95th percentile [lo] (Fig. la and lb). Male and female homozygotes had higher apolipoprotein B (apo B) values but lower apo A-I value than their parents (Tables 3 and 4). Female heterozygotes had a higher mean apo A-I concentration than males (P < 0.05).


4. Discussion age (years) 40








age (years) 30






Patient's value 95th percede

of LRC values

Fig. 1. Serum LDL-cholesterol of obligate heterozygotes compared with age- and sex-related 95th percentile values derived from the Lipid Research Clinics (LRC) Prevalence Study [IO].

mothers and fathers, respectively) but these values are still below the normal range, as defined by the Lipid Research Clinics (LRC) Prevalence Study [9]. There were no significant differences in HDL cholesterol between the two sexes in homozygotes or heterozygotes. LDL cholesterol was markedly lower in parents and varied from 4.10 to 7.22 mmol/l with a mean of 5.41 f 1.18 mmol/l in fathers and from 3.04 to 8.99 mmol/l with a mean of 5.95 f 1.88 mmol/l in mothers (Table 4). As

Tunisia is in North Africa. It has known several different civilizations over the centuries: Roman, Carthaginian, Turkish,. Arab and French. Today the population consists mainly of people of Berber or Arab origin with some minorities such as Italians and Jews. The diet is cereal-based (wheat and barley) supplemented with vegetables, olive oil, fish and meat (in particular lamb). The population under 50 years old of Central and Southern Tunisia in 1992 was about 3 000 000 and in these areas 26 FH homozygotes aged 2.5-40 years were alive, of whom 60% were born of consanguineous marriages. The prevalence of homozygotes in the population at risk was therefore 0.0000087 or about 1 in 125 000. This prevalence is a conservative estimate since we did not ascertain all the live homozygotes in these areas. Furthermore, we are aware of other FH families in the north of Tunisia. Knowing the approximate prevalence of homozygotes, we estimated the prevalence of heterozygotes from the Hardy-Weinberg equation [l l] to be about 1 in 165. We are aware of two other large studies of homozygotes. Khachadurian and Uthman saw 52 such patients in Lebanon over 20 years [2], 58% of whom were born of consanguineous parents like many of our patients. Seftel et al. [ 121 reported 34 Afrikaner homozygotes in South Africa, the prevalence of homozygotes and heterozygotes in this population being 1 in 30 000 and 1 in 100. respectively. All our FH homozygotes exhibit tuberous xanthomas on the elbows, knees and buttocks, the spread and severity increasing with age. The presence of tendon xanthomas usually precedes CHD; it is common to have tendon xanthomas

M. N. Slimane

el al. /Atherosclerosis

IO4 (1993)



without CHD before the age of 10. The latter disease was present in 10% of homozygotes before 9 years old, in 71% between 10 and 19 years old and in 100% of those aged 20 years or above. Therefore, CHD in our homozygotes appears to have a later onset than in other homozygote populations (CHD occurs in 50% of Afrikaners before age 9 [12] and in 25% of homozygote Japanese before age 10 [ 131). Their mean life expectancy was 13 years compared with 17 years in Japan [13] and 21 years in Lebanon [2]. In our patients 94% of females and 57% of males had a loud ejection systolic murmur in the aortic area and 23% of females and 14% of males had peripheral vascular disease. By comparison, none of the Lebanese homozygotes and only 1 patient from an Afrikaner population had peripheral vascular disease. Mean cholesterol concentration for our homozygotes was 18 mmol/l, similar to those reported in the United States [ 141, Japan [13], China [15] and Lebanon [2]. HDL cholesterol levels were reduced in our FH homozygotes, particularly in males (0.48 mmol/l). Most previous studies in FH heterozygotes show that the prevalence of xanthomas increases with age until the end of the 5th decade although up to 20% of heterozygotes never develop tendon xanthomas [3,16]. All our FH obligate heterozygotes went through life without developing tendon xanthomas. The onset of CHD was correlated with total and LDL cholesterol in both male and female FH heterozygote parents. Irrespective of sex, the incidence of CHD was low and only occurred over 30 years of age (about 23% for women and 15% for men). The mean serum cholesterol levels in the parents of our homozygotes were 7.51 mmol/l in mothers and 6.79 mmol/l in fathers, similar to those reported in China by Cai et al. [ 151,but lower than the mean value of 8.32 mmol/l in the obligate heterozygotes studied by Mabuchi’s group in Japan [13]. As compared with the 95th percentile of age- and sex-related LRC values [lo], 7 fathers (46%) and 4 mothers (30%) had a normal LDL cholesterol. From these data we concluded that despite different environmental conditions and dietary habits, the severity of hypercholesterolaemia among homozygotes in Tunisia seems to be similar to homozygotes in different parts of the world. In

contrast, the situation is different for FH heterozygotes many of whom had total and LDL cholesterol levels much lower than in North America and Japan. Moreover, our heterozygotes had no tendon xanthomas and a low incidence of CHD. The mild clinical expression of FH in our heterozygotes is possibly due to the Mediterranean type of diet. In a recent study [17] we demonstrated that a Tunisian population eating a high fat diet had higher apo B and total cholesterol levels than one eating a fish and vegetarian diet. Although the incidence of CHD in FH heterozygotes and in the general population of Tunisia is much lower than that of Western countries it may start to rise if serum cholesterol levels increase to Western levels through changes in dietary habits. At present we are trying to identify which mutations of the LDL receptor gene are responsible for FH in Central and Southern Tunisia. The discovery of these mutations should allow us to identify heterozygotes better, as it is difficult to diagnose non-obligate heterozygotes solely on clinical and laboratory criteria. Assuming that the number of different mutations responsible for FH in Tunisia is relatively small, as is the case in other populations with a high gene frequency due to intermarriage, this should enable the majority of heterozygotes to be identified. This would be an essential precondition for any programme of genetic counselling aimed at reducing the frequency of homozygous FH. 5. Acknowledgements

This work was supported by the Secretariat d’fitat a la Recherche Scientifique et Technique Tunisien. We are grateful to Dr. G.R. Thompson (MRC Lipoprotein Team, Hammersmith Hospital) for his expert advice and encouragement. We thank Mrs Liz Manson for preparing the typescript and Mr. Ben Ammar Amor for technical assistance. 6. References I

Goldstein, J.L. and Brown, MS.. The LDL receptor defect in familial hypercholesterolemia: implications for pathogenesis and therapy, Med. Clin. N. Am.. 66 (1982) 335.

M.N. Slimane et al. /Atherosclerosis


Khachadurian, A.K. and Uthman, SM., Experiences with the homozygous cases of familial hypercholesterolemia, Nutr. Metab., 15 (1973) 132. Goldstein, J.L. and Brown, MS., Familial hypercholesterolemia. In: Stanbury, J.B., Wyngaarden, J.B., Frederickson, D.S., Goldstein, J.L. and Brown, M.S. (Eds.), The Metabolic Basis of Inherited Disease, 5th Edn., McGraw-Hill, New York, 1983, pp. 672-712. Fruchart, J.C., Marcovina, S. and Puchois, P., Laboratory measurement of plasma lipids and lipoproteins. In: Fruchart, J.C. and Shepherd, J. (Eds.), Walter de Gruyter, New York, 1989, pp. 79-l 12. Ziegenhorm, J., Bartl, K. and Deeg, R., Improved kinetic method for automated determination of serum triglycerides, Clin. Chem., 26 (1980) 973. Assmann, G., Schriever, H., Schmitz, G. and Hongele, E.D., Quantification of high density lipoprotein cholesterol by precipitation with phosphotungstic and MgCI,, Clin. Chem., 29 (1983) 2026. Friedewald, W.T., Levy, R.T. and Fredrickson, D.S., Estimation of the concentration of low density lipoprotein cholesterol in plasma without the use of preparative ultracentrifugation, Clin. Chem., I8 (1972) 499. Fruchart, J.C., Kora, I., Cachera, C, Clavey, V., Duthilleul, P. and Maschetto. Y., Simultaneous measurement of plasma apolipoproteins A-I and B by electroimmunoassay, Clin. Chem., 28 (1982) 59. Heiss, G., Johnson, N.J., Reiland, S., Davis, C.E. and Tyroler, J.A., The epidemiology of plasma high density lipoprotein cholesterol levels. The Lipid Research Clinics Program Prevalence Study Summary, Circulation, 62 (Suppl. IV) (1980) 116.

104 (1993) 153-158

10 Lipid Research Clinics Program Epidemiology Committee, Plasma lipid distributions in selected North American populations: the Lipid Research Clinics Program Prevalence Study, Circulation, 60 (1979) 427. II Emery, A.E.H., Methodology in Medical Genetics. Churchill Livingstone, Edinburgh, 1976. I2 Seftel, H.C., Baber, S.G., Sandler, M.P. et al., A host of hypercholesterolaemic homozygotes in South Africa, Br. Med. J., 281 (1980) 633. 13 Mabuchi, H., Tatami, R., Haba, T. et al., Homozygous familial hypercholesterolemia in Japan, Am. J. Med., 65 ( 1978) 290. 14 Sprecher, D.L., Boeg, J.M., Schaefer, E.J. et al., The association of LDL receptor activity, LDL cholesterol level and clinical course in homozygous familial hypercholesterolemia, Metabolism, 34 (1985) 294. 15 Cai, H.-J., Fan, L.-M., Huang, M.-G., Chen, X.Y., Liu, G.-Q. and Chen, Q., Homozygous familial hypercholesterolemic patients in China, Atherosclerosis, 57 (1985) 303. I6 Kwiterovich, P.O., Fredrickson, D.S. and Levy, R.I., Familial hypercholesterolemia (one form of familial type II hyperliproteinemia). A study of its biochemicals, genetic and clinical presentation in childhood, J. Clin. Invest., 53 (1974) 1237. I7 Slimane, M.N., Hammami, M., Abid, M. et al., Influence des modes alimentaires sur le profil lipidique et le risque atherogene chez les tunisiens, Rev. Fr. Corps Gras, 9/lO (1992) 265.