Neonatal hemochromatosis

The relationship between high dietary iron intake, mutations of the HFE gene, and iron status, and their effects on human health are reviewed. Prolonged high dietary intakes of iron are unlikely to result in iron overload in the general population. Homozygotes for the C282Y mutation of the HFE gene have elevated body iron levels. Heterozygotes have normal iron stores but some may be at increased risk for cardiovascular disease. There is no convincing evidence that elevated iron status increases the risk of coronary heart disease or type 2 diabetes, but high iron intakes may increase the risk of colorectal cancer. The dietary levels of iron associated with health risks in different HFE genotypes must be determined. Key Words: dietary iron, HFE mutations, cardiovascular disease, colorectal neoplasms, type 2 diabetes

(c) 2003 International Life Sciences Institute

doi: 10.131/nr.2003.febr.45-62

Introduction

For many years, public health concerns have focused on iron deficiency, but with the increasing interest in the role of pro-oxidants in chronic disease, and the discovery that the C282Y mutation of the HFE gene is involved in hereditary hemochromatosis, attention is now turning to the possibility that an excessive supply of iron may also pose a health risk (Figure 1).

What Is Iron Overload?

Genetically Determined Iron Overload Diseases

Most cases of iron overload in populations with Northern European ancestry are attributed to HFE-related hereditary hemochromatosis (HH). Most patients with HH are homozygous for the C282Y mutation of the HFE gene (a substitution of a cysteine with a tyrosine amino acid at position 282 in the HFE preprotein),1 but there is disagreement about the extent to which the mutation predicts disease (i.e., its clinical penetrance). In the United States, 0.3% of whites, 0.06% of blacks, and 0.03% of Mexican Americans are homozygous for the C282Y mutation; that is, 1 in 385 individuals, approximately 718,000 homozygous individuals, in the United States have this genotype.2 Heterozygosity for the C282Y mutation is present in 9.5% of whites, 2.3% of blacks, and 2.8% of Mexican Americans. A higher prevalence has been reported from nonrandom samples in the United Kingdom, with C282Y homozygosity rates of 1 in 138 in Eastern England,3 1 in 148 in Wales,4 and 1 in 102 in Northern Ireland.5 By contrast, the mutation is rare in non-Celtic European populations such as Greece in which the C282Y homozygosity rate is less than I in 100,000.(6)

The clinical symptoms of HH are usually seen in middle age, but patients may be as young as 20 years old.7 Whereas early symptoms are nonspecific (e.g., fatigue, depression, and arthralgias), later findings are more obviously related to the tissues that are accumulating iron; these include tender hepatomegaly, polydipsia with polyuria (owing to diabetes mellitus), cardiomyopathy, increased skin pigmentation, arthritis, and hypogonadism.8 Compared with the general population, cause of death for patients with hemochromatosis is considerably more likely to be liver cancer, cardiomyopathy, liver cirrhosis, or diabetes mellitus.9

Diagnosis of iron overload requires transferrin saturation greater than 55% in men and postmenopausal women, or greater than 50% in premenopausal women (repeated on a fasting sample), and serum ferritin concentration greater than 300 (mu)/L in men, or greater than 200 (mu)g/L in women. The diagnosis should be confirmed by liver biopsy or by removal of greater than 4 g iron by weekly venesection.7 Treatment of HH by regular phlebotomy to remove excess iron is highly effective and results in a normal lifespan and quality of life if instituted early.9 Although the efficacy and ease of phlebotomy is often stated as a reason for use as the sole treatment in individuals with HH, it is not without problems. For some individuals, repeated phlebotomies have a detrimental effect on quality of life;10 particularly in these individuals, dietary interventions to decrease both the intake and bioavailability11 of dietary iron may well play a role in reducing the number of phlebotomies required to manage the condition.

Although most patients with HH are homozygous for the C282Y mutation of the HFE gene (91% in the UK12), a number of other mutations have also been found in the HFE coding sequence. The most prevalent of these is the H63D mutation. Compound heterozygotes are heterozygous for both the C282Y and H63D mutations; they comprise approximately 2.6% of HH patients in the United Kingdom.12 Mutations other than C282Y, such as S65C, appear to have a lower clinical penetrance, and may result in milder forms of the condition.13 In European populations with a lower prevalence of the C282Y mutation, such as Italy, C282Y homozygosity may explain as few as 64% of HH cases.14

Messenger RNA for HFE is widely expressed in human tissues, the highest expression being found in the liver and small intestine.1 In the small intestine, expression of the gene is tightly localized in the epithelial cells of the intestinal crypt,15 which is consistent with HFE playing a role in iron absorption. Wild-type HFE has been shown to form a complex with beta^sub 2^-microglobulin (beta^sub 2^M)16 and transferrin receptor.17 Recent research suggests that HFE-beta^sub 2^M increases iron uptake by increasing the rate at which transferrin receptors are recycled back to the cell surface after delivering iron to the cell cytoplasm, thereby increasing the number of receptors at the cell surface at any one time.18

We know that in hereditary hemochromatosis, the C282Y mutation of the HFE gene prevents formation of a disulphide bridge in the protein product, preventing it from binding with beta^sub 2^M.16 When it is not stabilized by beta^sub 2^M, the HFE protein does not reach the cell surface.16 Investigators have hypothesized that the C282Y mutation would result in lower levels of the HFE-transferrin receptor complex at the cell surface, and that this would in turn lead to a decreased uptake of transferrin-bound iron and lower iron status in the cell.18,19 In the enterocyte, this low iron state would be indistinguishable from anemia and would stimulate iron absorption, eventually leading to systemic iron overload.19 Supporting this proposed mechanism, low iron status stimulates an increased expression of transporters on the apical membrane (to import iron from the gut lumen)20 and the basolateral membrane (to export iron out into the circulation),21 and hemochromatosis patients have been shown to over-express the iron importer DCT1(22) and the iron exporter IREG1.(21) The H63D mutation does not prevent HFE interaction with beta^sub 2^M, however, and must impair HFE function in some other way.16

Two additional types of hemochromatosis have been identified recently: type 2 hemochromatosis, or juvenile hemochromatosis, which is a severe rapidly developing iron-overload condition not related to mutations of the HFE gene, and type 3 hemochromatosis, which is associated with mutations of the transferrin receptor 2 gene, but is indistinguishable from HFE hemochromatosis in its presentation.23 African iron overload is now thought to be caused by an interaction between high dietary iron intake and an unidentified non-HFE iron-loading gene.24 Rare forms of genetic iron overload include neonatal hemochromatosis, aceruloplasminemia, and atransferrinemia.8

Secondary Iron Overload Conditions

The two main causes of secondary iron overload are repeated blood transfusion (iron from degraded erythrocytes accumulates in the body), and ineffective erythropoiesis (which leads to increased iron absorption in iron-loading anemias such as beta-thalassemia and sideroblastic anaemia).

Iron Overload Associated with Excessive Oral Intake. Acute iron overload can result in profound mental retardation or death. Unintentional overdose of iron tablets is one of the two most common causes of poisoning deaths in children under 6 years of age.25 The lethal oral dose of elemental iron is 200 to 250 mg/kg.26 The mechanism of iron toxicity is believed to be free radical production resulting in lipid peroxidation, particularly in the gastrointestinal epithelium, cardiovascular system, and liver.

Whether prolonged excessive oral intake also results in iron overload is less clear. It is generally agreed that in the absence of an absorption defect, such as that in HH, iron absorption falls with increasing iron status.28 However, because the body does not regulate iron excretion, and is not able to down-regulate iron absorption to zero even when body iron levels are very high, it is conceivable that excessively high iron intakes over a prolonged period of time may also result in iron overload. The relationship between dietary iron intake and iron status is discussed below.

One recent study in Oslo, Norway, found that of 120 patients with clinical, biochemical, and/or histologic evidence of hemochromatosis, 11 had neither the C282Y nor H63D mutations associated with hereditary hemochromatosis.29 Nine of these patients had a history of almost daily intake of oral iron supplementation lasting from 10 to 50 years, suggesting that prolonged iron supplementation may be associated with iron overload. Whether moderately elevated iron status, such as that which could be achieved by dietary intake alone, presents a health risk is uncertain.

Relationship between HFE Genotype and Iron Status

Before the C282Y mutation was identified in 1996, differences in iron status among homozygotes, heterozygotes, and controls were investigated using family studies. Individuals with a diagnosis of hemochromatosis were assumed to be homozygotes, and family members were designated heterozygotes on the basis of HLA typing. This method cannot ensure complete separation of C282Y genotypes, however, and it is likely that the inclusion of the occasional compound heterozygote or unidentified C282Y homozygote in the heterozygote population was responsible for the early reports of large differences in iron status between heterozygotes and controls. It is also possible that another, as yet unidentified, genetic or environmental iron loading factor was present in the heterozygote relatives of clinically diagnosed homozygotes.

C2a2Y Homozygotes

By definition, C282Y homozygotes who have been diagnosed with HH have considerably elevated iron status. To date there have only been three studies large enough to investigate iron status in C282Y homozygotes from the general population,4,30,31 and only one of these included a random sample.30 Transferrin saturation is considerably higher in both male and female C282Y homozygotes than in their wild-type counterparts, and serum ferritin concentration is increased approximately three- to fivefold.30,31 In a random population sample in New Zealand, the sex-adjusted mean serum ferritin concentration in homozygotes was 397 (mu)g/L compared with 87 (mu)/L in those without the mutation.30

One Australian study reported that a high proportion of C282Y homozygotes from the community have elevated liver iron concentrations.32 Of 16 homozygotes identified from a population of 3011 unrelated white Australian adults, all had elevated transferrin saturation and/or serum ferritin concentrations, and all 11 of those who agreed to undergo liver biopsy had a liver iron concentration above the normal range.

C282Y/H63D Compound Heterozygotes

Iron status indices for C282Y/H63D compound heterozygotes appear to he between those for C282Y homozygotes and heterozygotes. Transferrin saturation is significantly higher in both male and female compound heterozygotes than in their wild-type counterparts;31,33 sex-adjusted figures for a random population sample were 42% for compound heterozygotes compared with 31% for the wild-type genotype.30 Compound heterozygotes are also considerably more likely to have elevated transferrin saturation; more than 20% have transferrin saturation greater than 45%, compared with 3 to 7% of wild-types.30,33,34 Serum ferritin concentration appears to be higher in men, but not women.31,33

C282Y Heterozygotes

Most studies report that both male and female C282Y heterozygotes have slightly, but significantly, elevated transferrin saturation.3,31,35 Reports of combined male and female data show a threefold increase in the percentage of people with transferrin saturation >50%.30,34 For example, Distante et al.34 discovered that 11% of heterozygotes had elevated transferrin saturation compared with 3% of wild-type individuals. Serum ferritin concentration did not appear to differ.30,33,35

There are no data on the prevalence of elevated hepatic iron in C282Y heterozygotes taken from the general population. Bulaj et al.36 reported that of 39 liver biopsies in putative heterozygotes identified by family studies, three had increased stainable iron, and one showed hepatic damage, which could not be explained by excessive alcohol consumption, porphyria cutanea tarda, or hepatitis. A number of other studies have reported elevated mean hepatic iron concentration in putative heterozygotes, but they have not demonstrated liver damage in the absence of confounding factors such as excessive alcohol intake. It is impossible to determine whether heterozygotes are at any increased risk of liver damage without studies in which hepatic iron concentration and liver damage are investigated in heterozygotes who are defined by their HFE genotype and who are drawn from the general population.

H63D Mutations

An Australian study37 recently reported elevated transferrin saturation in individuals with H63D mutations in the absence of C282Y mutations. The difference in transferrin saturation was 7% (male H63D homozygotes 35% vs. wild-type 28%). There were no significant differences in mean serum ferritin concentration, but 9% of male H63D homozygotes and 3% of heterozygotes had elevated transferrin saturation and serum ferritin concentration, compared to 0.7% of the wild-type individuals.

Other Genetic Mutations

Whereas this discussion has concentrated on the effects of the C282Y and H63D mutations of the HFE gene, there is compelling evidence that polymorphisms in other, as yet unidentified, genes account for considerably more of the variation in iron status indices. A recently published twin study38 estimated that although the C282Y and H63D mutations account for less than 5% of the phenotypic variation in serum ferritin concentration in both sexes and transferrin saturation in men, other genetic factors are responsible for an additional 30% of the transferrin saturation and 45% of the serum ferritin variation.

Relationship between Dietary Iron Intake and Iron Status

Dietary Sources of Iron

The two forms of iron in the diet, heme and nonheme iron, are found in different foods, and enter the intestinal mucosal cell by two independent pathways. Heme iron is more efficiently absorbed from the diet (approximately 25%) than nonheme iron (5-15%), and is subject to the influence of fewer iron absorption modifiers.39

Heme iron comprises approximately 40% of the total iron in animal tissue,40 although it is likely that chicken and fish provide considerably less heme iron per portion than red meat.41,42 The main sources of native nonheme iron include cereals, pulses, vegetables, and meat. Fortification and contaminant iron that is soluble in the gut enters a common nonheme iron pool and is absorbed into the intestinal cells similarly to native nonheme iron. However, forms of inorganic iron used for fortification have widely varying absorption.43 Significant amounts of nonheme iron may also be added to food inadvertently, for example, from cast-iron or steel cookware,44,45 and a recent randomized controlled trial indicated that contaminant iron from cast-iron cookware may be sufficiently bioavailable to influence iron status.46

Iron Bioavailability

Bioavailability can be defined as "a measure of fractional utilization of orally ingested nutrient."47 Iron bioavailability is affected by numerous dietary and host-related factors that are discussed in detail in a recent review.28

Two factors have been identified as modifiers of heme iron absorption: iron status48 and calcium intake.49 Interestingly, the inhibitory effect of high iron status on heme iron absorption is less powerful than that on nonheme iron.48

By contrast, a large number of dietary components have been identified as potential modifiers of nonheme iron absorption.50 Enhancers include ascorbic acid, meat, fish, and poultry, organic acids (e.g., citric, lactic), fermented soy products (e.g., soy sauce), and cysteine-- containing peptides. Inhibitors include phytate, polyphenols (e.g., in tea, coffee, and red wine), calcium, avidin (in raw eggs), oxalic acid (e.g., in spinach), soy protein, and other inorganic elements (e.g., copper and manganese). Many of these factors have been identified as iron absorption modifiers in single-meal absorption studies, however, and single-meal studies tend to overestimate the effect of iron absorption modifiers on iron absorption.50 The extent to which these food components influence iron status may be more pertinent. Animal tissue,51,52 vitamin C,53,54 and tea55 are associated with iron status, particularly in populations at increased risk of low iron status.

The failure of most population studies to find an association between iron status and total dietary iron intake is not surprising in view of the range of dietary components known to modify nonheme iron absorption. The inhibitory effect of phytate has not been demonstrated in population studies, presumably because there are no comprehensive food composition data for phytate. Iron status does not appear to be affected by increased calcium intake,56 although, to date, studies have only been carried out in iron-replete individuals.

Iron Intake and Status in the General Population

In healthy people, iron homeostasis is controlled by changes in absorptive efficiency. This means that iron status cannot be predicted from iron intake in individuals with normal iron status. Although the mechanism for this control has not yet been determined, Hallberg et al.48 demonstrated a consistent inverse relationship between iron status and iron absorption to a level of iron storage indicated by a serum ferritin concentration of approximately 60 (mu)g/L. They reported that after this point, iron absorption decreases to a level just sufficient to cover basal iron losses. These findings explain the lack of association between iron intake and iron status, but more research is required to determine why the mean serum ferritin concentration in the general population is greater than 60 (mu)/L.39 A number of factors independent of iron may be responsible. Serum ferritin concentration is positively correlated with both BMI (kg/m^sup 2^) and alcohol intake in iron-replete men,57 and is elevated in inflammation and infection.58

The diet used in the study by Hallberg et al.48 was designed to maximize iron bioavailability, but it did not contain large amounts of fortification or contaminant iron.59 Both fortification60 and contaminant46 iron have been shown to improve iron status in iron-deficient populations in developing countries. However, the role of fortification iron in maintaining iron status is less clear in developed countries. For instance, stopping iron fortification of flour in Denmark does not appear to have lowered the iron status of Danish adults.57,61

One well-designed cross-sectional study showed a positive association between both heme and supplemental iron intake (but not dietary nonheme iron intake) and serum ferritin concentration in a population largely composed of iron-replete adults.54 Iron-replete men involved in a 2-year fortification trial, in which iron intakes were increased by 7.5 mg per day, showed no increase in iron status compared with controls,60 and one iron-replete man who consumed 10 mg of iron as ferrous sulfate with one main meal a day for 500 days showed no increase in serum ferritin concentration.62 There is also evidence that higher intakes of enhancers of nonheme iron absorption, such as vitamin C, do not improve iron status in iron-replete individuals.63 Moderately elevated intakes of nonheme iron therefore do not appear to influence iron status in healthy iron-replete individuals. Although heme iron absorption is under homeostatic control, the absorption response may be less sensitive to higher iron status than that for nonheme iron.

It is important to note that relationships between dietary iron intake and iron status in the general population are not necessarily the same in individuals with HFE mutations. It is likely that asymptomatic individuals with HFE mutations have been included in studies investigating the effects of dietary iron on iron status, but subgroup analyses have not been carried out to describe their particular response.

Iron Intake and Status in Individuals with HFE Mutations

Although many studies have investigated the relationship between HFE genotype and iron status, few have investigated the relationship between iron intake and iron status in people with HFE mutations.

Individuals with HH have decreased efficiency of iron absorption with increasing iron status, but absorption is higher in relation to their iron stores than in control individuals.64 However, this control may only apply to absorption of nonheme iron and not heme iron.65 Iron fortification does not appear to increase iron status in people with HH. Withdrawal of food iron fortification in Sweden has decreased the frequency with which patients need to be phlebotomized to maintain their iron status.66 There is some evidence that modifiers of nonheme iron absorption affect iron bioavailability in patients with hemochromatosis similarly to the normal population. For example, regular tea drinking with meals has been shown to reduce the frequency of phlebotomies needed to manage iron accumulation in patients with hemochromatosis.11

A single meal absorption study by Lynch et al.65 reported that heterozygous individuals (relatives of patients with hemochromatosis) exhibit normal control of iron absorption from a hamburger meal. When the hamburger meal was supplemented with 20 mg of iron (as ferrous sulfate) and 100 mg of vitamin C, however, nonheme iron absorption was significantly elevated such that the heterozygous individuals absorbed 1.35 mg more iron from the meal than controls. This suggests that heterozygotes may have poorer control of iron absorption when presented with large amounts of highly bioavailable iron. It is not known whether C282Y-defined heterozygotes fail to control iron absorption at levels of iron that are likely to be consumed in the diet of freeliving individuals.

Relationship between Elevated Iron Status and Health

Research suggests that elevated iron status may be associated with increased risk of a number of health conditions. Those likely to be of greatest public health concern are coronary heart disease (CHD), cancer, and type 2 diabetes.

Coronary Heart Disease

In 1981, Sullivan67 proposed the "iron hypothesis," which suggested that the reason premenopausal women had a lower prevalence of ischemic heart disease than either postmenopausal women or men was because of their lower iron stores. Differences in iron stores were also proposed as an explanation for the lower rates of heart disease in developing countries. The hypothesis that elevated iron status increases CHD is still being vigorously argued 20 years later.68

Basic scientific research suggests a number of mechanisms by which iron could increase CHD risk. Iron is a transition metal that can oscillate between ferrous (II) and ferric (III) states by either accepting or donating an electron. These reactions generate free radicals that are potentially highly damaging to both lipids and DNA. Iron-generated free radicals could increase CHD risk by oxidizing low-density lipoprotein (LDL) cholesterol, damaging the arterial endothelium directly, promoting thrombosis, or interfering with normal vasomotor regulation.69 Iron has been found in human atherosclerotic lesions,70 but whether it is there as cause or result of the lesion is uncertain. The degree of iron-catalyzed oxidation that occurs in vivo is unclear, particularly at physiologic levels of iron. Iron must be in its reduced (ferrous) state to catalyze free radical reactions, but the vast majority of iron in vivo is protein-bound in its oxidized (ferric) state. Oxidative stress may liberate small amounts of ferrous iron, but it is likely that this is a localized phenomenon in response to local oxidative stress, rather than a process that is dependent on total body iron levels.69 There was no association between serum ferritin and two markers of oxidative stress in 473 men and women who participated in the Atherosclerosis Risk in Communities study.71

In the absence of intervention studies that alter iron status and investigate the effects on CHD rates, a prospective study investigating associations between serum ferritin concentration and clinically apparent or fatal CHD (Table 1) provides the strongest data available to evaluate the "iron hypothesis." Cross-sectional, casecontrol and retrospective studies all have weaknesses that the prospective study avoids. In particular, because data on iron status and health outcomes are being collected at the same point in time in a nonprospective study, it is very difficult to determine whether a change in the level of the iron index causes, or is caused by, the health outcome being measured. Serum ferritin concentration is the iron index of choice because it is the only one that reflects the gradations of iron status within iron sufficiency and iron overload states.81

The first prospective study to suggest an association between elevated iron stores and CHD risk was by Salonen et al.80 This study is often cited in support of the "iron hypothesis." Salonen et al. reported a relative risk (RR) of acute myocardial infarction of 2.2 in Finnish men with a serum ferritin concentration of >=200 (mu)g/L compared with those with a serum ferritin of =5.0 mmol/L). Moreover, when serum ferritin was entered into the same model as a continuous variable instead of a dichotomous one, the RR was still statistically significant, but clinically meaningless (RR = 1.002, CI 1.001-1.003). No explanation is given for the choice of 200 (mu)g/L as the cut-off for dichotomizing serum ferritin concentration (the mean serum ferritin was 166 (mu)g/L). Another weakness of the study is that because only 3 years of follow-up were reported, the positive association may be due to the effects of subclinical heart disease at baseline on serum ferritin concentration.

Since publication of this study, there have been eight more reports of prospective studies investigating the relationship between serum ferritin concentration and CHD.72-79 Of these, only two support the hypothesis (excluding one paper74 reporting what is largely a reanalysis of the Salonen et al. data80). Both of these studies suggest that if there is an effect of elevated iron status on heart disease risk, it occurs in individuals already at increased risk because of the presence of other risk factors. Klipstein-Grobusch et al.73 found an increased risk of first myocardial infarction in men and women with serum ferritin in the highest tertile (>171 (mu)g/L) who were current or past smokers or had diabetes (there was a marginally significant increase in risk for those with a total cholesterol >6.5 mmol/L). Individuals with C-reactive protein (CRP) >6 mg/L were excluded from the analysis to remove possible confounding owing to infection and inflammation, but the full range of CRP concentrations were not controlled for, so the possible effects of mild elevations of CRP on CHD risk were not addressed. Although the other positive study76 reported an increased risk of carotid atherosclerosis incidence and progression with higher serum ferritin concentrations even when LDL cholesterol was at the 10th percentile, an association between serum ferritin and risk of cardiovascular disease or death was only reported in individuals with LDL cholesterol above the 90th percentile. CRP was not controlled for in this study.

Sempos et al. performed a prospective cohort study in a random population of 1604 men and women free of self-reported CHD at baseline (the NHANES II Mortality Study)72 Mean follow-up time was 13 years, analysis excluded the first 3 years of follow-up, and 220 deaths from cardiovascular disease were reported. There was no association between serum ferritin and death from myocardial infarction, CHD, or cardiovascular disease as a whole. A recent meta-analysis of 11 prospective studies did not find evidence of an association between measures of iron status and CHD.82 There was no increase in risk of CHD for individuals with serum ferritin >=200 (mu)g/L versus

Several other types of studies have failed to demonstrate a convincing association between iron status and CHD risk. Of the 13 nonprospective studies investigating serum ferritin concentration and atherosclerosis or myocardial infarction, only two report any association, and both of these studies found an association with atherosclerosis only in a subgroup.83,84

Although a number of studies have suggested that blood donors may have a lower risk of CHD than non-blood donors, it is unlikely that their lower iron status is responsible. Because of the stringent criteria required for blood donation, blood donors, as a group, are more likely to be healthy and less likely to be taking medication than the general population. A recent prospective study of 38,244 men found a strong association between lifetime blood donations and lower plasma ferritin levels, but found no association between blood donation and risk of myocardial infarction or fatal CHD.85

Finally, the extreme iron overload state hemochromatosis is not associated with excess deaths owing to atherosclerosis, coronary artery disease, stroke, or peripheral artery disease.9 Rather, it is cardiomyopathy that is seen in excess in people with hemochromatosis, and this is caused by iron-mediated damage to cardiomyocytes rather than by coronary artery disease.86 In a retrospective autopsy study of 41 cases with iron overload compared with 82 age-, sex-, and race-matched controls, the odds ratio (OR) of coronary artery disease with iron overload was 0.18 (CI 0.04-0.73).87

There are a number of potential confounders that may explain why some studies have reported an association between iron status and heart disease risk. For instance, infections have been shown to increase serum ferritin concentration-58 and the risk of atherosclerosis.88 Furthermore, physiologic stresses that cause an acutephase response elevate concentrations of both CRP and serum ferritin,89 and elevated CRP concentration has been shown to be associated with increased risk of CHD.90 A meta-analysis of 1053 CHD cases in longterm prospective studies reported a combined risk ratio of 1.7 (CI 1.4-2.1) for CRP in the top versus the bottom tertile.90 The difference between the tertiles was just 1.4 mg/L (2.4 mg/L vs. 1.0 mg/L). It remains to be seen whether such small differences in CRP concentration are associated with significant changes in serum ferritin, and whether the larger differences in CRP concentration that have been associated with falsely elevated serum ferritin are also associated with increased risk of CHD. Another possible explanation is that serum ferritin may be a surrogate indicator for a number of other cardiovascular disease risk factors, For example, BMI, fibrinogen, highdensity lipoprotein cholesterol, and diastolic blood pressure were all shown to be predictors of serum ferritin concentration in a cross-sectional study of 337 healthy Norwegian men.91

Cancer

Three mechanisms have been suggested for an increased risk of cancer in individuals with higher iron status: iron-mediated free-radical generation leading to DNA damage, the use of iron as a nutrient by microbial or neoplastic cells, and impaired immune competence.92 Although a number of studies have investigated iron status and cancer risk, results are conflicting. Unfortunately, there are no prospective studies that measure serum ferritin concentration at baseline, use cancer incidence or death as an outcome, adequately assess intake of foods such as meat, which may influence both iron status and cancer risk, and have a sufficiently long follow-up period to minimize the effects on the results of subclinical cancer at baseline. Without studies of this quality, it is not possible to draw any conclusions as to whether high iron status is associated with an increased risk of cancer.

Type 2 Diabetes

Three mechanisms have been proposed for an association between elevated iron status and increased risk of type 2 diabetes: an iron effect on insulin synthesis and excretion by pancreatic cells, oxidation of free fatty acids by free radicals leading to diminished glucose utilization by muscle tissue resulting in increased insulin resistance, and liver iron accumulation interfering with its capacity to remove insulin from the blood.93 Only one large well-controlled study has investigated whether nonhemochromatotic individuals with elevated iron status have a higher risk of type 2 diabetes.94 This crosssectional study of 9486 adults from NHANES III reported increased risk of newly diagnosed diabetes in both men (OR = 4.9, CI 3.1-8.0) and women (OR = 3.6, CI 2.0-6.5) with elevated serum ferritin concentrations (>=300 (mu)g/L in men, >=150 (mu)g/L in women). Every attempt was made to control for possible effects of inflammation on serum ferritin concentration. Unfortunately, with a cross-sectional design in which both risk factor and outcome are measured at the same point in time, it is impossible to be certain whether it is the elevated serum ferritin causing the diabetes, or diabetes induced inflammation causing elevated serum ferritin concentrations. CRP concentration was significantly higher amongst those with newly diagnosed diabetes (4.6 mg/L) than those with normal fasting blood glucose concentration (2.8 mg/L), and increased with increasing serum ferritin concentration. The increased risk of newly diagnosed diabetes was significant only in participants with elevated serum ferritin but normal transferrin saturation (

Health

C282Y Homozygotes

The relationship between HFE genotype and health in symptomatic C282Y homozygotes has been discussed in detail. However, the extent to which C282Y homozygosity predicts disease, that is, the clinical penetrance of the genotype, is unclear. The prevalence of diagnosed hemochromatosis is not as high as the prevalence of C282Y homozygosity. For instance, Eastern England has a C282Y allele frequency of 0.085, which should correspond with approximately 3500 homozygotes in the Norfolk and Norwich University Hospital area. However, only 18 patients were being treated for hemochromatosis in 1997.(3) Although a substantial proportion of these homozygotes will not yet be of an age to experience the clinical manifestations of HH, this cannot entirely explain the diagnosis of only 3% of C282Y homozygotes in the region; mass under-diagnosis is also not a likely explanation.

A recent U.S. study by Beutler et al.95 suggests that the clinical penetrance of the C282Y mutation may be as low as 1%. The study has a number of methodological limitations, however, which may have led to an underestimation of the true clinical penetrance of the mutation. Firstly, the C282Y homozygotes investigated were not representative of the population of homozygotes as a whole. The participants were members of the Kaiser-- Permanente medical care program so were either part of an employee group or had MediCare (government funded health care for >65-year olds). Those under the age of 65 were therefore a comparatively healthy group. Moreover, symptom data for 18% of the homozygotes identified were excluded from the analysis because they were already undergoing treatment for hemochromatosis when the questionnaire was administered. Both these factors are likely to have selectively excluded the less healthy homozygotes, biasing the sample so that the clinical penetrance of the C282Y mutation appeared lower. Secondly, the symptom definition was generally poor, making it difficult to determine the extent to which the symptoms reported reflected the true health status of C282Y homozygotes in general. In particular, a number of key "symptoms" were poorly defined. For instance, the incidence of diabetes mellitus was determined by self report or blood glucose concentration using an inappropriate CUtoff.96 Thirdly, the sample size may have been too small for some trends to reach statistical significance (e.g., general health and fatigue in homozygotes over age 55). Because the clinical manifestations of hemochromatosis are usually not seen until middle age, it is unfortunate that this study of 40,000 people included only 60 homozygotes over age 55, of whom only approximately 24 were men. Nevertheless, the study does suggest that the penetrance of the C282Y mutation is considerably lower than was once thought.

Further research is needed to determine the penetrance of C282Y homozygosity and the extent to which penetrance is modulated by physiologic factors, such as menstruation and pregnancy, and environmental factors, such as blood donation and diet (e.g., consumption of alcohol, meat, fortification iron, and iron absorption modifiers such as polyphenols in tea and coffee).

C282YIH63D Compound Heterozygotes

Compound C282Y/H63D heterozygotes appear to be at increased risk of clinical hemochromatosis. In addition, there is evidence of increased risk of liver disease in compound heterozygotes. Bacon et al.97 reported that of 132 patients with liver disease, 6% were compound heterozygotes (a higher prevalence than the 2% expected in the general Australian population33). Similarly, in a study of archived liver biopsies in Eastern England, the prevalence of the compound heterozygote genotype was significantly higher in people with cirrhosis (3%) than in the general population (0.9%).98 Using these figures, Willis et al.98 estimate that 1% of compound heterozygotes are diagnosed with liver disease.

C282Y Heterozygotes

A recent study in 1784 Danish 45- to 100-year olds suggests that C282Y heterozygosity may be associated with shorter life expectancy, particularly in women.99 This is especially surprising because massively elevated iron status is said to underlie the pathology of symptomatic C282Y homozygosity, yet heterozygosity is associated with only a small elevation in transferrin saturation, and no elevation in serum ferritin concentration. In the Danish study, there was a significant trend of lower carrier frequency with increasing age amongst women, and a significantly lower proportion of women aged 85 to 94 years were C282Y heterozygotes (9.4%) compared with the 45- to 54-year age group (19.5%). Interestingly, the 95% CIs for the carrier frequency in 65- to 74-year-- old men, and 85- to 94-year-old men and women, were lower than the frequency reported in Danish neonates (13.7%);100 these numbers suggest a lower frequency of C282Y heterozygotes in these age groups. These findings need to be confirmed, however, ideally in a large prospective study in which migration effects can be excluded and the causes of death determined, or in a larger cross-sectional study that could determine a firm baseline frequency of the mutation by emphasizing middle-aged and older individuals, but also including a younger age group.

With the identification of the C282Y mutation, it has become possible to investigate whether carriers are at increased risk of various diseases. To date, the majority of studies investigating this question have been casecontrol studies, and the majority of positive findings have not been repeated. This may be because the studies were biased as a result of failure to ensure that the controls were similar to the cases in all important respects except the disease. It may also be because a large study is required to determine the true prevalence of a mutation that may be present in less than 10% of the control population. The present discussion will concentrate on the two diseases that have been investigated in prospective cohort studies: cardiovascular disease and type 2 diabetes.

Thus far, three possible mechanisms have been proposed for an increased cardiovascular disease risk in C282Y heterozygotes. Firstly, C282Y heterozygosity could increase iron stores leading to free radical generation and, in particular, LDL oxidation. There is no evidence that heterozygotes have higher iron status, however, and no evidence that individuals with elevated iron status have an increased risk of CHD. Secondly, C282Y heterozygotes may have higher levels of non-transferrin bound iron (NTBI) than wild-type controls, and this iron may be more available to generate free radicals and free radical damage. In one recent study, NTBI concentration was significantly higher in 22 heterozygotes than in 17 controls, even though they had similar iron status.101 This finding is potentially important, but needs to be confirmed by other research groups. Thirdly, the region in which the HFE gene is located is very gene rich; it is possible that cardiovascular disease risk is being modified by a gene other than HFE that is co-inherited with the C282Y mutation.

The three prospective studies of C282Y heterozygosity and cardiovascular disease risk (Table 2) have found a positive association. Risk of CHD was increased twofold in a nested case-control study with 243 cases. 102 Another nested case-control study reported a population attributable risk for C282Y heterozygosity of 8.8% for cerebrovascular disease death and 4% for total cardiovascular disease death.103 In this study, risk of fatal myocardial infarction, cerebrovascular disease death, and total cardiovascular disease death were all markedly increased in C282Y heterozygotes compared with their wild-type counterparts, but only in those who were both smokers and hypertensive. Risk of myocardial infarction was increased twofold in Finnish men who were C282Y heterozygotes and had systolic blood pressure above the median.104 In this study, there were only eight myocardial infarction occurrences in C282Y heterozygotes, however, so the risk estimates are less certain. None of the numerous case-control studies published to date have found an association between C282Y heterozygosity and cardiovascular disease risk.

At present, the probability that C282Y heterozygosity increases risk of cardiovascular disease seems stronger than that for elevated iron status, but more prospective studies are needed to confirm the associations reported. Moreover, a coherent mechanism for a direct relationship must be defined. Even if these findings are confirmed, Roest et al. 103 demonstrated that any effect of C282Y heterozygosity on population cardiovascular disease risk (4% of total cardiovascular disease risk) is not likely to be as great as that of smoking (8%) or hypertension (30%), both of which are modifiable.

Patients with hereditary hemochromatosis are at increased risk of type 2 diabetes, and are seven times more likely to die from diabetes than the general population.9 This has lead to the suggestion that C282Y heterozygotes might also be at increased risk for diabetes. Salonen et al. 105 reported an increased risk of type 2 diabetes amongst men who had C282Y mutations than amongst those without the mutation in their prospective study of eastern Finnish men. However, this finding was only marginally significant (OR = 3.5, CI 1.02-12.1), had wide CIs because there were only four C282Y carriers who developed type 2 diabetes during the 4-year followup, and the 35 "carriers" with C282Y mutations included one homozygote. Numerous case-control studies have failed to report an increased prevalence of C282Y heterozygosity amongst patients with type 2 diabetes compared with nondiabetic controls. Current research suggests, therefore, that there is no increased risk of type 2 diabetes in C282Y heterozygotes.

Screening for HFE Mutations

Because HH is a serious disease, because it can be treated effectively, and because genetic testing for HFE mutations is affordable, it has been argued that genetic screening should be carried out for all ethnically relevant communities.106 In 1998, however, an expert panel convened by the Centers for Disease Control and Prevention and the National Human Genome Research Institute stated that they did not recommend population-based genetic screening for HH due to "uncertainties about prevalence and penetrance of HFE mutations and the optimal care of asymptomatic people carrying HFE mutations."107 In 2003 we have a clearer picture of the prevalence of the mutations, but their penetrance is still not known. It appears, however, that the penetrance of C282Y homozygosity may be substantially lower than was originally thought, and as stated by Worwood,108 the lower the clinical penetrance, the less justifiable population screening becomes. The situation is even less clear for C282Y heterozygotes. Genetic screening does, however, have a role to play in screening family members of clinically diagnosed hemochromatosis cases.

Relationship between High Iron Intake and Health

High dietary iron intake is proposed to increase the risk of two health conditions in particular: CHD and colorectal cancer.

Coronary Heart Disease

No convincing mechanism for an effect of high iron intake, independent of iron status, on CHD risk has been proposed. High heme iron intake is suggested to increase risk of CHD by promoting heme iron-induced oxidation of LDL cholesterol,109 although this ignores the observation that heme iron joins the common iron pool in the enterocyte and is not released into the blood stream intact.

Nevertheless, a number of epidemiologic studies have investigated the question of whether high iron intake is associated with higher CHD risk. Only three of these were prospective studies that attempted to assess habitual iron intake (i.e., used a food frequency questionnaire or diet history).109-111 Two of these studies 109,111 reported an association between heme, but not total, iron intake and myocardial infarction.

The study by Klipstein-Grobusch et al.109 in elderly Dutch men and women reported a multivariate adjusted RR of myocardial infarction in those with the highest intakes of heme iron of 1.86 (CI 1.14-3.09). Even though the authors reported that the increased risk of myocardial infarction was only present in those who had hypercholesterolemia, hypertension, or diabetes, however, they did not control for these conditions in their multivariate analysis. No conclusion can be made, therefore, that high heme iron intake was in itself responsible for the increased risk of myocardial infarction. Rather, it is possible that the association was confounded by the presence of these other cardiovascular risk factors in individuals who also had high heme iron intakes. Moreover, animal tissue is the sole source of heme iron, and it is likely that red meat was the main source of heme iron in this elderly population. The literature on vegetarianism suggests that people who consume less red meat tend to consume more fruits and vegetables and lead a more active lifestyle than individuals who consume larger amounts of red meat. Because these dietary and lifestyle practices may affect cardiovascular risk, they may also confound the putative association between heme iron intake and cardiovascular risk. Although Ascherio et al.111 reported a multivariate adjusted RR of myocardial infarction of 1.48 for individuals with the highest quintile of heme iron intake, it was only marginally significant (CI 1.01-2.16). Individuals in this study with the highest intake of heme iron were also twice as likely to be smokers, had the highest BMI, had the highest intake of saturated fat and calories from fat, and had the lowest intake of the antioxidant vitamins E and C and beta-carotene. Although all these variables were included in the multivariate analysis, it is questionable whether the effect of heme iron intake alone could be effectively teased out of such a mass of variables that are associated with both increased CHD risk and heme iron intake. This is of particular concern when the validity of the food frequency questionnaire estimates of heme iron and saturated fat intake are not reported. A recent meta-analysis carried out using data from three prospective studies investigating dietary iron intake and CHD risk published before 1998 reported no increased risk of CHD for individuals with iron intakes in the top third versus those in the bottom third at baseline (RR = 0.84, CI 0.66-- 1.06).82

Colorectal Cancer

In iron-replete individuals, iron absorption is downregulated such that almost 100% of nonheme iron consumed in the diet or as supplements will pass through the gastrointestinal tract unabsorbed. Raised iron levels within the lumen of the colon may increase free radical generation in conjunction with colonic microflora112 and, as a consequence, the risk of cancer may be increased by free radical damage to lipids, proteins, and DNA of colon cells, free radical-mediated activation of procarcinogens to carcinogens in the lumen, or stimulated crypt cell proliferation. 13 Lund et al.113 reported a significant increase in fecal iron content in healthy human volunteers consuming 19 mg elemental iron per day. In addition, the concentration of weakly bound iron increased fivefold, and the production of free radicals under aerobic in vitro conditions increased by 40%. In humans, there may be sufficient oxygen tension in a microclimate at the mucosal surface for free radical production.

The majority of epidemiologic studies that have attempted to determine whether there is any association between iron intake and colorectal cancer have been case-control studies. Data from these studies are considerably weaker than those from prospective cohort studies for a number of reasons. In particular, (a) cases are likely to believe that there is an association between diet and colorectal cancer and this awareness is likely to alter the way in which they recall their diet, (b) symptoms resulting from the cancer may have altered the patient's diet before diagnosis, and (c) cancer-induced bleeding from the gut may lead to poor iron status and iron supplementation. For these reasons, this discussion will concentrate on the prospective cohort data.

To date, only two prospective studies have investigated iron intake and risk of colorectal cancer: the New York University Women's Health Study114 and the NHANES I cohort study.115 Both studies reported an increased risk of colorectal cancer with increasing iron intake, particularly for the proximal colon. In one study, the increased risk occurred in individuals with higher fat intakes.114 The study by Wurzelmann et al.115 provided stronger evidence of an association between iron intake and colorectal cancer risk because the results were not materially altered by exclusion of cases arising in the first 3 years of follow-up (this is important because a gastrointestinal condition may cause dietary changes) and because meat intake was controlled for.116 The results must be treated with some caution, however, because iron intake from supplements was not assessed117 and the food frequency questionnaire does not appear to have been validated.

The proposed mechanisms for an increased risk of colorectal cancer with high iron intake are convincing; however, it is important to remember that iron in the gut is not necessarily able to act freely, The diet contains numerous potentially protective factors such as antioxidants, phytic acid, polyphenols, and fiber, and the mucosa has protective factors such as intracellular antioxidants and enzymatic protection against free radical damage. Although the results of the prospective studies suggest that there may be an effect of iron intake on colorectal cancer risk, neither study is definitive. What is needed is a large prospective study with a long follow-up period (to allow exclusion of the first few years of follow-up), using a validated dietary assessment method that is suitable for estimating habitual intake of dietary iron intake and meat intake, and that assesses iron supplement usage adequately. Future studies should also report the median iron intake in each quartile so that the public health implications of any positive finding can be determined. This is particularly important because the new U.S. DRIs for iron for premenopausal women are very high: 18 mg/day for nonpregnant premenopausal women and 27 mg/day during pregnancy,39 a figure that would require iron supplementation or widespread iron fortification.

Conclusion

The main causes of iron overload in developed countries are HH (largely owing to HFE mutations) in adults and unintentional overdose of iron tablets in children. Prolonged high dietary intakes of iron are unlikely to cause iron overload in the general population because iron absorption is so tightly controlled, although the control of heme iron absorption may not be as strict as that for nonheme iron. However, homozygotes for the C282Y mutation of the HFE gene have poorer control of iron absorption and hence elevated iron status. As a group, C282Y heterozygotes have marginally elevated transferrin saturation, but normal serum ferritin concentrations.

The evidence that elevated iron status increases risk of CHD or type 2 diabetes is not convincing. It has been suggested that C282Y heterozygotes may be at increased risk of cardiovascular disease, but this association needs to be confirmed, and is not likely to be as strong as the modifiable risk factors smoking and hypertension. There is some evidence that high iron intakes may increase the risk of colorectal cancer, but these findings must be confirmed in well designed prospective studies. Most importantly, the levels of intake associated with any risk must be determined to enable policy decisions to be made on dietary advice, food fortification and the use of supplements.

Acknowledgments

Dr. Heath was in receipt of an Overseas Postdoctoral Fellowship from the Health Research Council of New Zealand when she carried out research funded by the Food Standards Agency (UK) on the effect of HFE mutations on iron nutrition. She was also supported by the Biotechnology and Biological Sciences Research Council (UK). The authors prepared a review entitled "Iron overload and diet" for Kellogg Co. (Europe) Ltd. in April 2001.

1. Feder JN, Gnirke A, Thomas W, et al. A novel MHC class 1-like gene is mutated in patients with hereditary haemochromatosis. Nat Genet. 1996;13: 399-408.

2. Steinberg KK, Cogswell ME, Chang JC, et al. Prevalence of C282Y and H63D mutations in the hemochromatosis (HFE) gene in the United States. JAMA. 2001;285:2216-2222.

3. Willis G, Jennings BA, Goodman E, Fellows I, Wimperis JZ. A high prevalence of HLA-H 845A mutations in hemochromatosis patients and the normal population in Eastern England. Blood Cells Mol Dis. 1997;23:288-291.

4. Jackson HA, Carter K, Darke C, et al. HFE mutations, iron deficiency and overload in 10 500 blood donors. Br J HaematoL 2001;114:474-484.

5. Murphy S, Curran MD, McDougall N, Callender ME, O'Brien CJ, Middleton D. High incidence of the Cys 282 Tyr mutation in the HFE gene in the Irish population-implications for haemochromatosis. Tissue Antigens. 1998;52:484-488.

6. Papanikolaou G, Politou M, Terpos E, Fourlemadis S, Sakellaropoulos N, Loukopoulos D. Hereditary hemochromatosis: HFE mutation analysis in Greeks reveals genetic heterogeneity. Blood Cells Mol Dis. 2000;26:163-168.

7. Worwood M. Haemochromatosis. Clin Lab Haematol. 1998;20:65-75.

8. Andrews NC. Inherited iron overload disorders. Curr Opin Pediatr. 2000;12:596-602.

9. Niederau C, Fischer R, Sonnenberg A, Stremmel W, Trampisch HJ, Strohmeyer G. Survival and causes of death in cirrhotic and in non cirrhotic patients with primary hemochromatosis. N Engl J Med. 1985;313:1256-1262.

10. Seamark CJ, Hutchinson M. Controversy in primary care: should asymptomatic haemochromatosis be treated? Treatment can be onerous for patient and doctor. BMJ. 2000;320:1314-1316.

11. Kaltwasser JP, Werner E, Schalk K, Hansen C, Gottschalk R, Seidl C. Clinical trial on the effect of regular tea drinking on iron accumulation in genetic haemochromatosis. Gut. 1998;43:699-704.

12. The UK Haemochromatosis Consortium. A simple genetic test identifies 90% of UK patients with haemochromatosis. Gut. 1997;41:841-844.

13. Mura C, Raguenes 0, Ferec C. HFE mutations analysis in 711 hemochromatosis probands: evidence for S65C implication in mild form of hemochromatosis. Blood. 1999;93:2502-2505.

14. Carella M, D'Ambrosio L, Totaro A, et al. Mutation analysis of the HLA-H gene in Italian hemochromatosis patients. Am J Hum Genet. 1997;60:828-- 832.

15. Parkkila S, Waheed A, Britton RS, et al. Immunohistochemistry of HLA-H, the protein defective in patients with hereditary hemochromatosis, reveals unique pattern of expression in gastrointestinal tract. Proc Natl Acad Sci U S A. 1997;94:2534-- 2539.

16. Feder JN, Tsuchihashi Z, Irrinki A, et al. The hemochromatosis founder mutation in HLA-H disrupts b2-microglobulin interaction and cell surface expression. J Biol Chem. 1997;272:14025-14028.

17. Gross CN, Irrinki A, Feder JN, Enns CA. Co-trafficking of HFE, a nonclassical major histocompatibility complex class I protein, with the transferrin receptor implies a role in intracellular iron regulation. J Biol Chem. 1998;273:22068-22074.

18. Waheed A, Grubb JH, Zhou XY, et al. Regulation of transferrin-mediated iron uptake by HFE, the protein defective in hereditary hemochromatosis. Proc Natl Acad Sci U S A. 2002;99:3117-3122.

19. Wood RJ. The "anemic" enterocyte in hereditary hemochromatosis: molecular insights into the control of intestinal iron absorption. Nutr Rev. 2002; 60:144-148.

20. Gunshin H, Mackenzie B, Berger UV, et al. Cloning and characterization of a mammalian proton-coupled metal-ion transporter. Nature. 1997;388:482-- 488.

21. McKie AT, Marciani P, Rolfs A, et al. A novel duodenal iron-regulated transporter, IREG1, implicated in the basolateral transfer of iron to the circulation. Mol Cell. 2000;5:299-309.

22. Zoller H, Pietrangelo A, Vogel W,Weiss G. Duodenal metal-transporter (DMT-1, NRAMP-2) expression in patients with hereditary haemochromatosis. Lancet. 1999;353:2120-2123.

23. Roy CN, Andrews NC. Recent advances in disorders of iron metabolism: mutations, mechanisms and modifiers. Hum Mol Genet. 2001;10:2181-- 2186.

24. Gordeuk V, Mukiibi J, Hasstedt SJ, et al. Iron overload in Africa: interaction between a gene and dietary iron content. N Engl J Med. 1992;326:95-- 100.

25. McGuigan MA. Common culprits in childhood poisoning: epidemiology, treatment and parental advice for prevention. Paediatr Drugs. 1999;1:313-- 324.

26. Robotham JL, Lietman PS. Acute iron poisoning: a review. Am J Dis Child. 1980;134:875-879.

27. Tenenbein M. Toxicokinetics and toxicodynamics of iron poisoning. Toxicol Lett. 1998;102-103:653-- 656.

28. Fairweather-Tait SJ, Teucher B. Iron and calcium bioavailability of fortified foods and dietary supplements. Nutr Rev. 2002;360-367.

29. Bell H, Berg JP, Undlien DE, et al. The clinical expression of hemochromatosis in Oslo, Norway: excessive oral iron intake may lead to secondary hemochromatosis even in HFE C282Y mutation negative subjects. Scand J Gastroenterol. 2000; 35:1301-1307.

30. Burt MJ, George PM, Upton JD, et al. The significance of haemochromatosis gene mutations in the general population: implications for screening. Gut. 1998;43:830-836.

31. Beutler E, Felitti V, Gelbart T, Ho N. The effect of HFE genotypes on measurements of iron overload in patients attending a health appraisal clinic. Ann Intern Med. 2000;133:329-337.

32. Olynyk JK, Cullen DJ, Aquilia S, Rossi E, Summerville L, Powell LW. A population-based study of the clinical expression of the hemochromatosis gene. N Engl J Med. 1999;341:718-724.

33. Rossi E, Bulsara MK, Olynyk JK, Cullen DJ, Summerville L, Powell LW. Effect of hemochromatosis genotype and lifestyle factors on iron and red cell indices in a community population. Clin Chem. 2001;47:202-208.

34. Distante S, Berg JP, Lande K, Haug E, Bell H. High prevalence of the hemochromatosis-associated

Cys282Tyr HFE gene mutation in a healthy Norwegian population in the city of Oslo, and its phenotypic expression. Scand J Gastroenterol. 1999;34: 529-534.

35. Datz C, Haas T, Rinner H, Sandhofer F, Patsch W, Paulweber B. Heterozygosity for the C282Y mutation in the hemochromatosis gene is associated with increased serum iron, transferrin saturation, and hemoglobin in young women: a protective role against iron deficiency? Clin Chem. 1998;44: 2429-2432.

36. Bulaj ZJ, Griffen LM, Jorde LB, Edwards CQ, Kushner JP. Clinical and biochemical abnormalities in people heterozygous for hemochromatosis. N Engl J Med. 1996;335:1799-1805.

37. Gochee PA, Powell LW, Cullen DJ, Du Sart D, Rossi E, Olynyk JK. A population-based study of the biochemical and clinical expression of the H63D hemochromatosis mutation. Gastroenterology. 2002;122:646-651.

38. Whitfield JB, Cullen LM, Jazwinska EC, et al. Effects of HFE C282Y and H63D polymorphisms and polygenic background on iron stores in a large community sample of twins. Am J Hum Genet. 2000;66:1246-1258.

39. Panel on Micronutrients Food and Nutrition Board. Dietary reference intakes for vitamin A, vitamin K, arsenic, boron, chromium, copper, iodine, iron, manganese, molybdenum, nickel, silicon, vanadium, and zinc. Washington, DC: National Academy Press; 2001.

40. Monsen ER, Hallberg L, Layrisse M, et al. Estimation of available dietary iron. Am J Clin Nutr. 1978; 31:134-141.

41. Rangan A, Blight G, Binns C. Haem iron content in Australian meats and fish. Proceedings of the Nutrition Society of Australia. 1996;20:210.

42. Hallberg L, Hulten L. Prediction of dietary iron absorption: an algorithm for calculating absorption and bioavailability of dietary iron. Am J Clin Nutr. 2000;71:1147-1160.

43. Salgueiro MJ, Zubillaga M, Lysionek A, Caro R, Weill R, Boccio J. Fortification strategies to combat zinc and iron deficiency. Nutr Rev. 2002;60:52-- 58.

44. Brittin HC, Nossaman CE. Iron content of food cooked in iron utensils. JAm Diet Assoc. 1986;86: 897-901.

45. Fairweather-Tait SJ, Fox TE, Mallillin A. Balti curries and iron. BMJ. 1995;310:1368.

46. Adish AA, Esrey SA, Gyorkos TW, Jean-Baptiste J, Rojhani A. Effect of consumption of food cooked in iron pots on iron status and growth of young children: a randomised trial. Lancet. 1999;353:712-- 716.

47. Fairweather-Tait SJ. Iron. J Nutr. 2001;131:1383S-- 1386S.

48. Hallberg L, Hulten L, Gramatkovski E. Iron absorption from the whole diet in men: how effective is the regulation of iron absorption? Am J Clin Nutr. 1997;66:347-356.

49. Hallberg L, Rossander-Hulthen L, Brune M, Gleerup A. Inhibition of haem-iron absorption in man by calcium. Br J Nutr. 1992;69:533-540.

50. Heath A-LM, Fairweather-Tait SJ. Clinical implica

tions of changes in the modern diet: iron intake, absorption and status. In: Worwood M, ed. Bailliere's Best Practice & Research: Clinical Haematology; 2002;15:225-241.

51. Lyle RM, Weaver CM, Sedlock DA, Rajaram S, Martin B, Melby CL. Iron status in exercising women: the effect of oral iron therapy vs increased consumption of muscle foods. Am J Clin Nutr. 1992;56:1049-1055.

52. Heath A-LM, Skeaff CM, Williams SM,Gibson RS. The role of blood loss and diet in the aetiology of mild iron deficiency in premenopausal adult New Zealand women. Public Health Nutr. 2001;4:197-- 206.

53. Hunt JR, Gallagher SK, Johnson LK. Effect of ascorbic acid on apparent iron absorption by women with low iron stores. Am J Clin Nutr. 1994; 59:1381-1385.

54. Fleming DJ, Jacques PF, Dallal GE, Tucker KL, Wilson PW, Wood RJ. Dietary determinants of iron stores in a free-living elderly population: The Framingham Heart Study. Am J Clin Nutr. 1998;67: 722-733.

55. Temme EHM, Van Hoydonck PGA. Tea consumption and iron status. Eur J Clin Nutr. 2002;56:379-- 386.

56. Minihane AM, Fairweather-Tait SJ. Effect of calcium supplementation on daily nonheme-iron absorption and long-term iron status. Am J Clin Nutr. 1998;68:96-102.

57. Milman N, Ovesen L, Byg K-E,Graudal N. Iron status in Danes updated 1994. I: Prevalence of iron deficiency and iron overload in 1332 men aged 40-70 years. Influence of blood donation, alcohol intake, and iron supplementation. Ann Hematol. 1999;78:393-400.

58. Hulthen L, Lindstedt G, Lundberg PA,Hallberg L. Effect of mild infection on serum ferritin concentration-clinical and epidemiolgoical implications. Eur J Clin Nutr. 1998;52:376-379.

59. Hulten L, Gramatkovski E, Gleerup A,Hallberg L. Iron absorption from the whole diet. Relation to meal composition, iron requirements and iron stores. Eur J Clin Nutr. 1995;49:794-808.

60. Ballot DE, MacPhail AP, Bothwell TH, Gillooly M, Mayet FG. Fortification of curry powder with NaFe(III)EDTA in an iron-deficient population: report of a controlled iron-fortification trial. Am J Clin Nutr. 1989;49:162-169.

61. Osier M, Milman N, Heitmann BL. Consequences of removing iron fortification of flour on iron status among Danish adults: some longitudinal observations between 1987 and 1994. Prev Med. 1999;29: 32-36.

62. Sayers MH, English G, Finch C. Capacity of the store-regulator in maintaining iron balance. Am J Hematol. 1994;47:194-197.

63. Cook JD, Watson SS, Simpson KM, Lipschitz DA, Skikne BS. The effect of high ascorbic acid supplementation on body iron stores. Blood. 1984;64: 721-726.

64. Walters GO, Jacobs A, Worwood M, Trevett D, Thomson W. Iron absorption in normal subjects and patients with idiopathic haemochromatosis:

relationship with serum ferritin concentration. Gut. 1975;16:188-192.

65. Lynch SR, Skikne BS, Cook JD. Food iron absorption in idiopathic hemochromatosis. Blood. 1989; 74:2187-2193.

66. Olsson KS, Vaisanen M, Konar J,Bruce A. The effect of withdrawal of food iron fortification in Sweden as studied with phlebotomy in subjects with genetic hemochromatosis. Eur J Clin Nutr. 1997;51:782-786.

67. Sullivan JL. Iron and the sex difference in heart disease risk. Lancet. 1981;1:1293-1294.

68. Schumann K, Borch-Iohnsen B, Hentze MW,Marx JJM. Tolerable upper intakes for dietary iron set by the US Food and Nutrition Board. Am J Clin Nutr. 2002;76:499-500.

69. Corti M-C, Gaziano M, Hennekens CH. Iron status and risk of cardiovascular disease. Ann Epidemiol. 1997;7:62-68.

70. Lee F-Y, Lee T-S, Pan C-C, Huang A-L, Chau LY. Colocalization of iron and ceroid in human atherosclerotic lesions. Atherosclerosis. 1998;138:281288.

71. Iribarren C, Sempos CT, Eckfeldt JE, Folsom AR. Lack of association between ferritin level and measures of LDL oxidation: the ARIC study. Atherosclerosis. 1998;139:189-195.

72. Sempos CT, Looker AC, Gillum RF, McGee DL, Vuong CV, Johnson CL. Serum ferritin and death from all causes and cardiovascular disease: the NHANES II mortality study. Ann EpidemioL 2000; 10:441-448.

73. Klipstein-Grobusch K, Koster JF, Grobbee DE, et al. Serum ferritin and risk of myocardial infarction in the elderly: the Rotterdam Study. Am J Clin Nutr. 1999;69:1231-1236.

74. Tuomainen T-P, Punnonen K, Nyyssonen K, Salonen JT. Association between body iron stores and the risk of acute myocardial infarction in men. Circulation. 1998;97:1461-1466.

75. Marniemi J, Jarvisalo J, Toikka T, Raiha I, Ahotupa M, Sourander L. Blood vitamins, mineral elements and inflammation markers as risk factors of vascular and non-vascular disease mortality in an elderly population. Int J Epidemiol. 1998;27:799-807.

76. Kiechl S, Willeit J, Egger G, Poewe W, Oberhollenzer F. Body iron stores and the risk of carotid atherosclerosis: prospective results from the Bruneck study. Circulation. 1997;96:3300-3307.

77. Aronow WS, Ahn C. Three-year follow-up shows no association of serum ferritin levels with incidence of new coronary events in 577 persons aged >=62 years. Am J Cardiol. 1996;78:678-679.

78. Manttari M, Manninen V, Huttunen JK, et al. Serum ferritin and ceruloplasmin as coronary risk factors. Eur Heart J. 1994;15:1599-1603.

79. Magnusson MK, Sigfusson N, Sigvaldason H, Johannesson GM, Magnusson S, Thorgeirsson G. Low iron-binding capacity as a risk factor for myocardial infarction. Circulation. 1994;89:102-108.

80. Salonen JT, Nyyssonen K, Korpela H, Tuomilehto J, Seppanen R, Salonen R. High stored iron levels are associated with excess risk of myocardial infarction in Eastern Finish men. Circulation. 1992; 86:803-811.

81. Gibson RS. Principles of Nutritional Assessment. New York: Oxford University Press; 1990.

82. Danesh J, Appleby P. Coronary heart disease and iron status: meta-analyses of prospective studies. Circulation. 1999;99:852-854.

83. Rossi E, McQuillan BM, Hung J, Thompson PL, Kuek C, Beilby JP. Serum ferritin and C282Y mutation of the hemochromatosis gene as predictors of asymptomatic carotid atherosclerosis in a community population. Stroke. 2000;31:3015-3020.

84. Kiechl S, Aichner F, Gerstenbrand F, et al. Body iron stores and presence of carotid atherosclerosis: results from the Bruneck Study. Arterioscler Thromb. 1994;14:1625-1630.

85. Ascherio A, Rimm EB, Giovannucci E, Willett WC, Stampfer MJ. Blood donations and risk of coronary heart disease in men. Circulation. 2001;103: 52-57.

86. Niederau C. Iron overload and atherosclerosis. Hepatology. 2000;32:672-674.

87. Miller M, Hutchis GM. Hemochromatosis, multiorgan hemosiderosis, and coronary artery disease. JAMA. 1994;272:231-233.

88. Kiechl S, Egger G, Mayr M, et al. Chronic infections and the risk of carotid atherosclerosis: prospective results from a large population study. Circulation. 2001;103:1064-1070.

89. Rubin C, Wood PJ, Archer T, Rowe DJ. Changes in serum ferritin and other `acute phase' proteins following major surgery. Ann Clin Biochem. 1984; 21:290-294.

90. Danesh J, Collins R, Appleby P, Peto R. Association of fibrinogen, C-reactive protein, albumin, or leukocyte count with coronary heart disease: meta-analyses of prospective studies. JAMA, 1998;279:1477-1482.

91. Oshaug A, Bugge KH, Bjonnes CH, Borch-Iohnsen B, Neslein I-L. Associations between serum ferritin and cardiovascular risk factors in healthy young men. A cross sectional study. Eur J Clin Nutr, 1995;49:430-438.

92. Van Asperen IA, Feskens EJ, Bowles CH, Kromhout D. Body iron stores and mortality due to cancer and ischaemic heart disease: a 17-year follow-up study of elderly men and women. Int J Epidemiol. 1995;24:665-670.

93. Tuomainen T-P, Nyyssonen K, Salonen R, et al. Body iron stores are associated with serum insulin and blood glucose concentrations: population study in 1,013 eastern Finnish men. Diabetes Care. 1997;20:426-428.

94. Ford ES, Cogswell ME. Diabetes and serum ferritin concentration among U.S. adults. Diabetes Care. 1999;22:1978-1983.

95. Beutler E, Felitti VJ, Koziol JA, Ho NJ, Gelbart T. Penetrance of 845G/EA (C282Y) HFE hereditary haemochromatosis mutation in the USA. Lancet. 2002;359:211-218.

96. The Expert Committee on the Diagnosis and Classification of Diabetes Mellitus. Report of the Expert Committee on the Diagnosis and Classification of Diabetes Mellitus. Diabetes Care. 2002;25:S5S20.

97. Bacon BR, Olynyk JK, Brunt EM, Britton RS, Wolff RK. HFE genotype in patients with hemochroma

tosis and other liver diseases. Ann Intern Med. 1999;130:953-962.

98. Willis G, Wimperis JZ, Lonsdale R, et al. Incidence of liver disease in people with HFE mutations. Gut. 2000;46:401-404.

99. Bathum L, Christiansen L, Nybo H, et al. Association of mutations in the hemochromatosis gene with shorter life expectancy. Arch Intern Med. 2001;161:2441-2444.

100. Merryweather-Clarke AT, Simonsen H, Shearman JD, Pointon JJ, Norgaard-Pederson B, Robson KJH. A retrospective anonymous pilot study in screening newborns for HFE mutations in Scandinavian populations. Hum Mutat. 1999;13:154-- 159.

101. de Valk B, Addicks MA, Gosriwatana I, Lu S, Hider RC, Marx JJM. Non-transferrin-bound iron is present in serum of hereditary haemochromatosis heterozygotes. Eur J Clin Invest. 2000;30:248-- 251.

102. Rasmussen ML, Folsom AR, Catellier DJ, Tsai MY, Garg U, Eckfeldt JH. A prospective study of coronary heart disease and the hemochromatosis gene (HFE) C282Y mutation: the Atherosclerosis Risk in Communities (ARIC) Study. Atherosclerosis. 2001; 154:739-746.

103. Roest M, van der Schouw YT, de Valk B, et al. Heterozygosity for a hereditary hemochromatosis gene is associated with cardiovascular death in women. Circulation. 1999;100:1268-1273.

104. Tuomainen T-P, Kontula K, Nyyssonen K, Lakka TA, Hello T, Salonen JT. Increased risk of acute myocardial infarction in carriers of the hemochromatosis gene Cys282Tyr mutation: a prospective cohort study in men in Eastern Finland. Circulation. 1999;100:1274-1279.

105. Salonen JT, Tuomainen T-P, Kontula K. Role of C282Y mutation in haemochromatosis gene in development of type 2 diabetes in healthy men: prospective cohort study. BMJ. 2000;320:1706-- 1707.

106. Allen K, Williamson R. Should we genetically test everyone for haemochromatosis? J Med Ethics. 1999;25:209-214.

107. Burke W, Thomson E, Khoury MJ, et al. Hereditary hemochromatosis: gene discovery and its implications for population-based screening. JAMA. 1998;280:172-178.

108. Worwood M. What is the role of genetic testing in diagnosis of haemochromatosis? Ann Clin Biochem. 2001;38:3-19.

109. Klipstein-Grobusch K, Grobbee DE, den Breeijen JH, Boeing H, Hofman A, Witteman JCM. Dietary iron and risk of myocardial infarction in the Rotterdam Study. Am J Epidemiol. 1999; 149:421- 428.

110. Reunanen A, Takkunen H, Knekt P, Seppanen R, Aromaa A. Body iron stores, dietary iron intake and coronary heart disease mortality. J Intern Med. 1995;238:223-230.

111. Ascherio A, Walter CW, Rimm EB, Giovannucci EL, Stampfer MJ. Dietary iron intake and risk of coronary disease among men. Circulation. 1994;89: 969-974.

112. Babbs CF. Free radicals and the etiology of colon cancer. Free Radic Biol Med. 1990;8:191-200.

113. Lund EK, Wharf SG, Fairweather-Tait SJ, Johnson IT. Oral ferrous sulfate supplements increase the free radical-generating capacity of feces from healthy volunteers. Am J Clin Nutr. 1999;69:250-- 255.

114. Kato I, Dnistrian AM, Schwartz M, et al. Iron intake, body iron stores and colorectal cancer risk in women: a nested case-control study. Int J Cancer. 1999;80:693-698.

115. Wurzelmann JI, Silver A, Schreinemachers DM, Sandler RS, Everson RB. Iron intake and the risk of colorectal cancer. Cancer Epidemiol Biomarkers Prev. 1996;5:503-507.

116. Norat T, Riboli E. Meat consumption and colorectal cancer: a review of epidemiologic evidence. Nutr Rev. 2001;59:37-47.

117. Nelson RL. Iron and colorectal cancer risk: human studies. Nutr Rev. 2001;59:140-148.

Anne-Louise M. Heath, Ph.D., Susan J. Fairweather-Tait, D.Sc.

Dr. Heath is with the Department of Human Nutrition, University of Otago, Dunedin, New Zealand. Prof. Fairweather-Tait is with the Institute of Food Research, Norwich Research Park, Norwich, NR4 7UA, United Kingdom.

Copyright International Life Sciences Institute and Nutrition Foundation Feb 2003
Provided by ProQuest Information and Learning Company. All rights Reserved