See article on page 569
In this issue of Hepatology, Failla et
al. show that patients with genetic hemochromatosis have significant
eccentric hypertrophy of the radial artery although none of them have
arterial hypertension or evidence of cardiovascular
disease.1 The investigators suggest that the structural
arterial changes may be accompanied by functional changes in arterial
distension (stiffening). The structural alterations were largely
reverted by iron depletion; similarly, artery wall stiffening improved,
although the initial abnormality was not significant. The investigators
conclude that in patients with hemochromatosis, arterial wall thickness
is increased before the onset of cardiovascular complications. They did
not study potential mechanisms by which iron overload could cause these
arterial changes. They did review recent literature, which explores the
association between iron and atherosclerosis.
The investigators chose to study arterial alterations in
hemochromatosis as a “human model” of the effect of marked iron
overload on cardiovascular disease and atherosclerosis. If iron excess
contributes to the atherosclerotic process, complications of
atherosclerosis should be an early and prominent clinical feature in
hemochromatosis. The finding that iron overload leads to marked
eccentric hypertrophy of middle-sized arteries in advance of other
apparent cardiovascular disease supports this hypothesis. For
hepatologists, and in particular for physicians and scientists who work
in the field of hemochromatosis and iron overload, the data reported by
Failla et al. are astonishing and beg the question whether (and if so
why) we have overlooked such an association for more than a century.
Indeed it has become trendy to speculate about the hazards of iron
excess, rather than the hazards of iron deficiency, which were a major
focus during centuries of limited nutritional iron supply. Today, in
significant parts of the industrialized world, particularly in men,
iron excess is found more often than iron deficiency.2 The
potential link between iron and atherosclerosis has attracted
increasing attention in recent years as shown by the number of
published studies (Fig. 
1). The
“Oxidative Stress Theory” suggests that the production of
tissue-damaging free radicals is an essential component in the
pathogenesis of numerous chronic diseases and that iron may help to
catalyze the reactions producing free radicals.
Fig. 1. Number of articles (in the English language) dealing with the link of
iron and atherosclerosis published during the last 20 years.
Publications were assessed as “PubMed” hits when the topics
“iron” and “atherosclerosis” were examined via the Internet on
June 22, 2000.
|
|
|
Click on Image to view full size
|
Excessive iron may promote cardiomyopathy, arthropathy, infection,
fibrosis, and diabetes mellitus, as well as some malignant, endocrine,
and neurodegenerative disorders (for review see
Weinberg3). Almost all of the disorders associated with
iron loading are manifest in genetic hemochromatosis in which iron
stores are markedly increased over a long period of time. Recent
discussion has focused on the question whether slight iron excess (in
subjects without genetic hemochromatosis) contributes to the two
diseases that cause most deaths, i.e., atherosclerosis and
neoplasia. Obviously hemochromatosis is an optimal situation in which
to study such a potential association. There are several reports that
construct links between iron and atherosclerosis or cancer including
the current report by Failla et al. All of these articles begin with
statements indicating that “genetic hemochromatosis (is) associated
with increased cardiovascular morbidity and mortality.” Although
literally this is true, it is misleading in the present context. There
is overwhelming evidence that cardiomyopathy occurs more frequently in
patients with hemochromatosis than in normal controls. However, there
is also overwhelming evidence that this cardiomyopathy is not caused by
coronary artery disease, but by iron-mediated damage to cardiomyocytes.
Coronary artery disease, other forms of ischemic heart disease, stroke,
and peripheral artery disease are not increased in hemochromatosis.
They may even be less frequent than in control populations as assessed
by large, partly prospective cohort studies, epidemiologic studies, and
autopsy studies. Although Failla et al. mention that “older pathology
reports on hereditary hemochromatosis” do not provide evidence of
changes in large- and middle-size arteries4,5 it should be
noted that more recent, large, prospective clinical and epidemiologic
studies also support the view that atherosclerosis is not prominent in
genetic hemochromatosis. Virtually all large cohort studies have shown
beyond any doubt that atherosclerosis, coronary artery disease, stroke,
and peripheral artery disease are rare clinical features and infrequent
causes of death in genetic hemochromatosis.6-10
A large autopsy study examined the prevalence of coronary artery
disease in patients with hemochromatosis and multiorgan hemosiderosis
from a registry of nearly 48,000 autopsies performed at The Johns
Hopkins Hospital (Baltimore, MD) between 1889 and 1992.11
In a 2:1 control-case ratio, 82 controls matched by age, race, and sex
were compared with 41 cases with iron overload. Pathologic description
of the coronary arteries was carefully recorded as advanced or severe
in 12% of iron-overload cases compared with 38% of controls
(P = .01). The prevalence of three-vessel disease
assessed by postmortem coronary arteriography was 11.1% in
iron-overload cases compared with 33.3% in controls (P
= .04). The odds ratio of coronary artery disease with iron overload
was 0.18 (95% confidence interval: 0.04-0.73). The investigators
cautiously concluded that iron overload resulting from hemochromatosis
or multiorgan hemosiderosis is not associated with an increased
prevalence of coronary artery disease.11 The data even
suggest that coronary artery disease may be less frequent and less
severe in iron overload syndromes when compared with control subjects.
In addition, there are several autopsy studies in patients with
hemochromatosis who died from cardiac causes. These deaths were almost
exclusively caused by cardiomyopathy with normal coronary arteries
(literature in von Herbay et al.12).
As correctly cited by Failla et al., peripheral artery disease is
uncommon in patients with genetic hemochromatosis and insulin-dependent
diabetes.13,14 These studies show that even in the
presence of strong risk factors for atherosclerosis, such as
insulin-dependent diabetes mellitus, iron overload does not promote
peripheral artery disease.
A recent study searched Multiple-Cause Mortality Files compiled by the
National Center for Health Statistics for the years 1979 to 1992 for
all records listing hemochromatosis.15 These data were
used to calculate age-adjusted and age-specific mortality rates,
identify medical conditions associated with a known diagnosis of
hemochromatosis at death, and calculate proportionate mortality ratios
for these medical conditions. As expected from previous studies, the
association of cardiomyopathy and hemochromatosis was increased about
4.8-fold over the expected ratio. However, deaths in which other
cardiac disorders were present, including acute myocardial infarction,
other ischemic heart disease, old myocardial infarction, angina
pectoris, and other chronic ischemic heart disease, were not higher
than expected in patients with hemochromatosis but tended to be lower
in patients who died with hemochromatosis.15
Failla et al. mention some experimental evidence that iron may
contribute to the atherosclerotic plaque formation and human studies
that have reported a correlation between iron stores and the risk of
coronary disease or atherosclerosis.16,17 Along these
lines, Sullivan suggested that iron depletion protects against ischemic
heart disease and argued that the difference in the incidence of heart
disease between men and women may be caused by differences in their
iron stores.18 The maximal gender difference in serum
ferritin level is reached at approximately 45 years of age and is about
300%. The maximal gender difference in heart disease is also reached
at approximately 45 years and is also about 300%.19 The
iron hypothesis could explain the gender difference in coronary disease
as well as the low prevalence of coronary disease in regions with a
high prevalence of iron deficiency.18 However, iron
deficiency may be only one consequence of a globally insufficient diet,
which limits atherogenesis for other reasons. The iron hypothesis could
also explain the protective effect of medication that causes
gastrointestinal blood loss, e.g., aspirin, or inhibits iron
absorption, e.g., cholestyramine, and the risk of an
increasing effect of oral contraceptives, which are known to decrease
menstrual blood loss (for review and literature see de Valk and
Marx20). If the iron hypothesis is true, easy and effective
ways to reduce the risk of cardiovascular events would be through blood
donation and abolition of iron fortification of food and multivitamin
preparations.20 If such an iron hypothesis is true,
coronary heart disease should be a prominent feature of hemochromatosis
and other forms of severe iron overload.
Some epidemiologic studies (mainly from Finland) support the hypothesis
that high iron levels or iron stores (usually serum iron or ferritin
were measured) and heterozygous presence of the hemochromatosis gene
Cys282Tyr mutation are associated with an increased risk of myocardial
infarction.16,17,21-26 As also mentioned by Failla et al.,
asymptomatic carotid atherosclerosis assessed by duplex sonography has
been linked to high iron stores.16 Lowering of iron stores
reduced, and iron accumulation increased, cardiovascular
risk.26 The latter results, although intriguing, could not
be confirmed by further studies measuring carotid intima-media
thickness as an indicator for early asymptomatic
atherosclerosis.27,28
The vast majority of recent epidemiologic data, including results from
prospective, cross-sectional, case-control and autopsy studies, have
failed to support the original hypothesis that high body iron stores or
heterozygosity for the Cys282Tyr mutation of HFE increase the risk of
coronary heart disease, myocardial infarction, or
atherosclerosis.29-48 In the First National Health and
Nutrition Examination Survey Epidemiologic Follow-up Study, serum iron
levels and transferrin saturation were not related to myocardial
infarction.37,38 Sempos et al. even found an inverse
association between transferrin saturation and coronary heart
disease.39 Regnstrom et al. studied serum ferritin levels
and serum iron levels in 94 young male survivors of myocardial
infarction 4 to 6 months after the event.40 They found no
differences in ferritin levels in myocardial infarction patients
compared with healthy controls and the patients had even lower serum
iron levels.
In sum, recent literature does not support the hypothesis that iron
contributes to atherosclerosis to a major degree although further
prospective studies are required to elucidate this association. There
is overwhelming evidence that atherosclerosis, coronary artery disease,
stroke, and peripheral artery disease are neither prominent clinical
features nor frequent causes of death in genetic hemochromatosis. In
view of this solid background, the new and provocative findings of
Failla et al. are difficult to reconcile. The investigators themselves
admit problems fitting their data to the well-known and well-studied
clinical features of hemochromatosis. Only further studies will be able
to resolve this apparent paradox. It must also be kept in mind that
many populations have a high prevalence of the hemochromatosis gene
Cys282Tyr mutation, which is a relatively young mutation (probably less
than 2,000 years old). If iron excess, even of a minor degree, is so
dangerous in contributing to atherosclerosis and cancer, it is
difficult to explain the extraordinary evolutionary advantage to HFE
heterozygotes.
)
