 |
|
 |
 |
 |
Primary Hyperlipoproteinaemias (Type IIa) involving
Hypercholesterolaemia
Incidence
High serum cholesterol levels are common in adults in much of Europe, the USA, Australia and New Zealand. In the UK, for example, 25-30% of middle-aged individuals have levels exceeding 6.5 mmol/L (250 mg/dL), while the equivalent proportion in the USA is at least 10%.
Causes
Most of this hypercholesterolaemia does not represent the effect of any
single cause, but is owing to some combination of dietary fat, obesity
and individual susceptibility to the development of hypercholesterolaemia.
This susceptibility is partly genetic, probably involving more than one
gene, and this common type of hypercholesterolaemia is usually referred
to as polygenic hypercholesterolaemia. At the very top end of the range
of cholesterol distribution are individuals who have the less common monogenic
condition, familial hypercholesterolaemia.
Familial hypercholesterolaemia
For a review of familial hypercholesterolaemia see references
[Durrington, 1995a; Goldstein,
1995].
Heterozygous familial hypercholesterolaemia
What is the incidence?
Familial hypercholesterolaemia is dominantly inherited. The heterozygous form of the condition affects about one in 500 people in Europe and the USA, making it the most common genetic disorder in these countries. In some populations (eg, Lebanese Christians, Afrikaners and Cape-coloured peoples of South Africa and French Canadians) it is considerably more common. This is because such people have descended from a relatively small number of early settlers, a few of whom, by chance, had familial hypercholesterolaemia. This is known as a founder effect. In other populations, such as Africans who have not intermingled with Europeans, familial hypercholesterolaemia is rare.
What is the lipoprotein phenotype?
Typically, the serum cholesterol in adult heterozygotes is 9-12 mmol/L (350-450 mg/dL). The condition is expressed regardless of diet or age, and elevated cholesterol levels are already present in cord blood and throughout childhood. The lipoprotein phenotype is usually IIa, but occasionally there is a moderate increase in fasting serum triglycerides to produce a IIb pattern. There is a tendency for HDL-C to be at the lower end of the range, particularly if triglycerides are elevated.
| Features of familial hypercholesterolaemia |
| Tendon xanthomata
The clinical hallmark of familial hypercholesterolaemia is the
presence of tendon xanthomata. These appear in heterozygotes from
the age of 20 years onwards. The most common sites for tendon
xanthomata are in the tendons overlying the knuckles and in the
Achilles tendons (see Figure 6). Less commonly, they may also
be found in other tendons. It is also common to find subperiosteal
xanthomata on the upper tibia where the patellar tendon inserts.
The skin overlying tendon xanthomata is of normal colour and they
do not appear yellow. The cholesteryl ester deposits are deep
within the tendons. Tendon xanthomata feel hard because they are
fibrotic. Indeed, it is not uncommon for those in the Achilles
tendons to become inflamed from time to time, sometimes presenting
as chronic Achilles tenosynovitis. More generalised tendinitis
may follow rapid therapeutic reductions in serum cholesterol levels.
Other occurrences of tendon xanthomata
Tendon xanthomata occur in only two disorders apart from familial
hypercholesterolaemia and these are so rare as not to pose any
diagnostic difficulty. The conditions are cerebrotendinous xanthomatosis,
in which plasma cholestanol is elevated and deposited in tendons,
and phytosterolaemia (beta-sitosterolaemia), in which there is
abnormal intestinal absorption of plant sterols, which are then
deposited in tendons.
Corneal arcus
Corneal arcus is also a frequent occurrence in familial hypercholesterolaemia.
When it occurs in adolescence or early adulthood, it is more likely
to be associated with familial hypercholesterolaemia than when
the condition occurs in middle age or later. It is, however, not
uncommon to encounter patients with familial hypercholesterolaemia
who have florid tendon xanthomata, but no arcus. It is, therefore,
not a very valuable physical sign.
Xanthelasmata palpebrarum
Xanthelasmata palpebrarum, although occurring with greater frequency
and at a younger age in familial hypercholesterolaemia affect
only a minority of heterozygous individuals and:
- are not specific for any particular type of hypercholesterolaemia;
- occur in polygenic hypercholesterolaemia, pregnancy, primary
biliary cirrhosis and hypothyroidism;
- are common in middle-aged women, often overweight, with no
very marked increase in serum cholesterol, if any; and
- may run in families, apparently independently of hypercholesterolaemia.
|
 |
| Figure 6. Types of xanthoma in familial
hypercholesterolaemia |
Identification of patients
Identifying familial hypercholesterolaemia heterozygotes as early as possible is important, because of their risk of CHD. Over 50% of affected men die before the age of 60 years if untreated. It is not uncommon for men to have their first myocardial infarction or develop angina in their thirties (occasionally even earlier). Some 15% of women with familial hypercholesterolaemia die of CHD before the age of 60 years, and the majority have symptomatic coronary disease by that age. Some evidence of cardiac ischaemia may be seen in perhaps as many as 10% of women before their menopause. However, whereas it is exceptional for a man with familial hypercholesterolaemia to live to 70 years of age without symptomatic CHD, almost 25% of women do so.
| Why are not all patients detected? |
| The above accounts largely for the reason why a family
history of premature CHD is absent in as many as 25% of patients
discovered to have familial hypercholesterolaemia on screening,
or in men who are discovered to have familial hypercholesterolaemia
when they present with a heart attack in early life. The latter
occurs because the condition has been inherited from their mother,
who has not yet developed coronary symptoms. Most people with familial
hypercholesterolaemia are not overweight and do not have risk factors
for CHD, such as obesity, hypertension, diabetes or a history of
smoking, other than hypercholesterolaemia and a family history of
premature coronary disease. Those without a family history of premature
CHD (approximately 25%) will be missed in screening programmes for
risk factors for CHD, in which cholesterol is only measured selectively.
|
CHD risk calculations
Methods of CHD risk calculation based on epidemiological studies (which
do not include significant numbers of people with familial hypercholesterolaemia)
will underestimate risk [Bhatnagar, 2000]
(see later). Thus, screening for high cholesterol in the relatives of
patients with known familial hypercholesterolaemia is important if new
cases are to be detected [Durrington,
2001].
Early development of CHD
Those patients with familial hypercholesterolaemia who develop CHD particularly
early, often come from families in which the affected members have also
developed the condition early. This may be because other genetic factors
in the family predispose to CHD. Thus, low serum HDL-C and increased
fasting triglycerides are associated with a worse prognosis. Serum Lp(a)
is increased in familial hypercholesterolaemia and any familial tendency
to run a high level of Lp(a) is exacerbated in those members who also
have familial hypercholesterolaemia. The apo E4 genotype is also associated
with more aggressive atheroma in familial hypercholesterolaemia. A knowledge
of the average age at which affected members of a family developed CHD
may be helpful in planning how actively to treat boys and young adult
women.
Occurrence of atheroma
There is an increased risk of atheroma in other parts of the arterial tree in heterozygous familial hypercholesterolaemia, but this is strikingly less so than in the coronary arteries. Some heterozygotes have aortic systolic cardiac murmurs caused by deposits of atheroma in the aortic root, sometimes involving the aortic cusps.
Homozygous familial hypercholesterolaemia
What is the incidence?
Most cases of homozygous familial hypercholesterolaemia occur in societies
in which consanguineous marriages and heterozygous familial hypercholesterolaemia
are frequent. The chance of marriage between unrelated heterozygotes,
meeting by chance in countries such as the UK or the USA is one in 5002,
and each of their children would stand a one in four chance of being
homozygotes. Assuming no adverse effect on the survival of the conceptus,
an incidence of homozygous familial hypercholesterolaemia of one in
106 births would be predicted. It is, thus, a rare condition
under these circumstances.
What is the lipoprotein phenotype?
Serum cholesterol is typically greater than 15 mmol/L (600 mg/dL).
| Features of homozygous familial hypercholesterolaemia
|
Cutaneous xanthomata
Clinically, homozygous familial hypercholesterolaemia is characterised
by the development of cutaneous xanthomata in childhood. Cutaneous
xanthomata may be present in the first year of life or may not develop
until later childhood. They are typically orange-yellow, subcutaneous,
planar xanthomata, occurring on the buttocks, antecubital fossae
and the hands, frequently in the webs between the fingers (see Figure
6c). Tuberose subcutaneous xanthomata on the knees, elbows and knuckles
are also a feature. |
Clinical course of the disease
Myocardial infarction and angina frequently occur in childhood, sometimes even in infancy. Atheromatous deposits at the aortic root, invariably present by puberty, are so marked as to produce significant aortic stenosis, which contributes to the risk of sudden death. Death before the age of 30 years, and often considerably younger, was the rule before the advent of plasmapheresis and similar techniques for the extracorporeal removal of LDL.
Polyarthritis
Polyarthritis, predominantly affecting the ankles, knees, wrists and proximal interphalangeal joints is common in homozygotes for familial hypercholesterolaemia.
The metabolic defect in familial hypercholesterolaemia
Decreased catabolism of LDL
In familial hypercholesterolaemia decreased catabolism of LDL means that it remains for longer in the circulation. Normally the plasma half-life of LDL is between two-and-a-half and three days, whereas in familial hypercholesterolaemia heterozygotes it is between four-and-a-half and five days (and even longer in homozygotes).
What causes the defect?
The molecular defect that causes this has been elucidated following the discovery of the LDL receptor (see above) by Goldstein and Brown in 1973, for which they received the Nobel Prize for Medicine in 1985. The gene encoding the LDL receptor is located on chromosome 19. Heterozygotes express only about half the quantity of LDL receptors that a person without familial hypercholesterolaemia expresses, while homozygotes have between none and 25% of normal receptor activity.
What are the effects of mutations?
Mutations in the LDL receptor gene have two general effects. They either:
- produce receptors with no binding activity (receptor negative), which causes problems because the receptor is not synthesised, is not transported to the cell surface, or, if it gets there, cannot be internalised after binding to LDL; or
- produce receptors with an abnormal binding site (receptor defective), which allows some LDL to be bound and to enter the cell, but this occurs only slowly because of the abnormal binding.
What is the value of DNA-based diagnostic services?
Some 200 mutations have been described and undoubtedly, more exist. In
societies such as the UK and the USA, therefore, the most frequent of
these mutations is likely to occur in no more than 3-4% of patients with
familial hypercholesterolaemia [Durrington,
2001]. This means that the prospect of developing a DNA test for this
condition in these countries is unrealistic. In Afrikaners, however, only
three mutations account for 90% of familial hypercholesterolaemia and
even in Holland, the range of mutations associated with familial hypercholesterolaemia
is sufficiently small that a DNA-based diagnostic service may be a realistic
possibility [Umans-Eckenhausen,
2001].
Significance of the type of mutation
Only in populations with a small number of LDL receptor mutations, or where intermarriage is common, are clinical homozygotes for familial hypercholesterolaemia truly homozygous in the sense that both their LDL gene mutations are identical. Most will be mixed heterozygotes. For clinical purposes it is reasonable to label those patients who have the clinical syndrome as homozygotes. It is, however, instructive to realise that some of the heterogeneity of the severity of the syndrome relates to the nature of the two LDL mutations present. Thus, the worst prognosis is associated with inheritance of two receptor-negative mutations, and the best with two receptor-defective mutations. The type of receptor mutation in heterozygotes is also probably of some importance, but here it is blurred against a background of other acquired or genetic factors, which can find expression over a much longer time than in homozygotes.
Familial defective apo B100
A small proportion (3%) of patients, who have the same clinical features
as heterozygotes for familial hypercholesterolaemia, do not have an LDL
receptor defect. Instead, they have a mutation of apo B in which glutamine
is substituted for arginine at amino acid residue 3500 (which is part
of the LDL-receptor-binding domain). This disorder has been termed familial
defective apo B100. It has a frequency of approximately one in 500-600
in the UK and the USA, but only a minority of affected individuals has
tendon xanthomata. The serum cholesterol level that tends to be associated
with this disorder is around 8 mmol/L (310 mg/dL), which is less than
in most heterozygotes for familial hypercholesterolaemia.
Common or polygenic hypercholesterolaemia
For a review of common or polygenic hypercholesterolaemia see reference
[Durrington, 1995a]. When a diagnosis
of familial hypercholesterolaemia can be made, either because hypercholesterolaemia
is present in childhood or an adult has the clinical features of the syndrome,
a reasonably accurate estimate of clinical risk can be made and appropriate
therapy given. In the UK, however, familial hypercholesterolaemia probably
accounts for no more than 3% of CHD mortality before the age of 60 years
in men. There is overlap between the range of LDL-C levels encountered
in familial hypercholesterolaemia and those owing to the more common,
polygenic hypercholesterolaemia. Epidemiological studies have not included
sufficient numbers of people with particularly high cholesterol levels
to be certain, but it is probable that the risk in familial hypercholesterolaemia
is greater than in polygenic hypercholesterolaemia with similar cholesterol
levels. This may be because, in the familial condition, the hypercholesterolaemia
has been present since birth, whereas polygenic hypercholesterolaemia
is frequently not fully developed until the third or fourth decade. Furthermore
familial hypercholesterolaemia, unlike many other types of hypercholesterolaemia,
is associated with increased serum concentrations of Lp(a).
CHD mortality
Table 2 gives estimates of the extent to which different levels of cholesterol
contribute to the overall cumulative male mortality from CHD by the
age of 60 years, and shows that the majority of such premature deaths
come from the mid range of cholesterol distribution. It has been argued
that if a significant reduction in the incidence of CHD is to be achieved
in countries such as the UK, efforts to lower cholesterol cannot simply
be confined to those individuals whose plasma cholesterol lies at the
upper end of the distribution.
 |
| Table 2. Estimates of UK mortality from
CHD in men |
How can the incidence be decreased?
Nevertheless, because the number of people with values in the mid range is so huge (the vast majority of whom are not at increased risk of premature CHD), a different strategy from that applied to those in the upper part of the cholesterol distribution must be applied to reducing their cholesterol. This is the 'low-risk' or 'population' strategy, which aims to lower serum cholesterol by public health measures directed at encouraging the adoption of a lower fat diet and the avoidance of obesity.
Who will develop CHD?
Some patients from the mid range of cholesterol are, however, at much
greater individual risk from their cholesterol level than the majority,
because they have other risk factors for CHD, which combine to increase
their susceptibility. Probably the most potent of these is that the
individual already has CHD. In middle-aged people with existing CHD,
serum cholesterol is an important indicator of cardiac prognosis (see
Figure 7) [Pekkanen, 1990], which
ranks after left ventricular function, but ahead of most of the other
risk factors for CHD. Lipoproteins are also the most important risk
factors for occlusion of coronary artery bypass grafts after the initial
postoperative period. In people, who have not yet developed CHD, the
effect of risk factors, such as cigarette smoking, hypertension and
diabetes, is also to increase the risk from any given level of cholesterol
(see Figure 4). A family history of CHD at an early age in a first-degree
relative also increases the likelihood of CHD, and part of this effect
is independent of other risk factors.
| When is intervention appropriate? |
| The combination of all of these factors, with a relatively
modestly increased serum cholesterol level, can increase individual
risk substantially to a level where clinical intervention is as
justified, as it is with more marked elevations in serum cholesterol.
This is the 'high-risk' or 'clinical' approach to prevention of
CHD. |
 |
| Figure 7. Age-adjusted deaths from CHD |
Measuring CHD risk
The equations derived from the results of the Framingham study permit
an assessment of CHD and stroke risk, and can assist clinical judgement
in assessing the likelihood that a patient can benefit from antihypertensive
or lipid-lowering therapy. These equations are reproduced in the Cardiac
Risk Assessor Programme of the Joint British Societies (available from
the British Heart Foundation and on the websites of the British
Hypertension Society and the British
National Formulary and in the recent American Treatment Panel third
report (ATP III). Several attempts have been made to portray these risk
equations as charts, tables or scoring systems [Expert
Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol
in Adults, 2001; British Cardiac Society,
2000; Wood, 1998]. In one study,
the Joint British Coronary Risk Charts (see Figure 8) [British
Cardiac Society, 1998] were the closest to the original equation
[Jones, 2001].
 |
| Figure 8a. CHD risk prediction charts: no
diabetes |
 |
| Figure 8b. CHD risk prediction charts: diabetes
|
Metabolic defect in polygenic hypercholesterolaemia
In polygenic hypercholesterolaemia, there is generally overproduction of VLDL by the liver. If this is converted rapidly to LDL, there is no increase in serum triglyceride levels. The LDL-receptor mechanism is probably overloaded in many individuals, and in any case, appears to catabolise only about one-third of LDL. The build-up of cholesterol in most patients is, therefore, not caused by any defect in the LDL receptor, but to the inability of non-receptor-mediated catabolism to cope without a rise in the serum cholesterol concentration.
What are the factors that affect polygenic hypercholesterolaemia?
Obesity and a high-fat diet (particularly one high in saturated fat) are probably the major reasons for the enormous differences in the prevalence of polygenic hypercholesterolaemia in different parts of the world. Undoubtedly, however, individual responses to diet vary tremendously, and there is probably a complex interplay between dietetic and genetic factors in the genesis of polygenic hypercholesterolaemia. The rise in cholesterol with age, which occurs in both men before and women after the climacteric, seems less evident in societies where the cholesterol level is, for dietetic reasons, lower.
Dietary modification
There is an impression that dietary modification, aimed at lowering cholesterol in middle age in societies where serum cholesterol is high, does not reduce it to the extent that might be anticipated from populations habitually consuming such a diet. Whether this is simply a matter of non-compliance with diet or represents some permanent change in metabolism caused by a high-fat diet in early life, is at present uncertain.
|
 |
 |