Jump to ContentJump to Main Navigation
Coronary Heart Disease EpidemiologyFrom aetiology to public health$

Michael Marmot and Paul Elliott

Print publication date: 2005

Print ISBN-13: 9780198525738

Published to Oxford Scholarship Online: September 2009

DOI: 10.1093/acprof:oso/9780198525738.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2019. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use.  Subscriber: null; date: 24 October 2019

Serum homocysteine and coronary heart disease

Serum homocysteine and coronary heart disease

Chapter:
(p.239) Chapter 16 Serum homocysteine and coronary heart disease
Source:
Coronary Heart Disease Epidemiology
Author(s):

D. S. Wald

M. R. Law

N. J. Wald

J. K. Morris

Publisher:
Oxford University Press
DOI:10.1093/acprof:oso/9780198525738.003.0016

Abstract and Keywords

This chapter discusses the link between serum homocysteine and coronary heart disease (CHD). Genetic studies show a moderately higher risk of CHD for a moderately higher level of serum homocysteine. Prospective studies show a positive association between serum homocysteine and cardiovascular disease after allowance for confounding. Although these two types of study are susceptible to different sources of error, they show quantitatively similar associations, a result that is unlikely to have occurred through different potential sources of confounding acting independently. The homocystinurias cause high serum homocysteine levels and high risks of premature cardiovascular disease, and lowering serum homocysteine reduces this high risk. These observations provide a compelling case for a cause and effect relationship between homocysteine and CHD and, therefore, a protective role for folic acid on CHD prevention.

Keywords:   cardiovascular disease, CHD, folic acid, homocysteine levels

16.1 Introduction

In the 35 years since the link between serum homocysteine and cardiovascular disease was first made (McCully 1969) much evidence has accumulated on the subject. However, opinion on whether homocysteine causes cardiovascular disease remains divided. Resolving the uncertainty is important, as serum homocysteine levels can be lowered by taking additional folic acid, raising the prospect of a simple means of prevention (Homocysteine Lowering Trialists Collaboration 1998). This chapter examines the evidence for causality with respect to coronary heart disease (CHD).

16.2 Homocysteine and B vitamins

Homocysteine is an amino acid formed from the essential amino acid methionine. Methionine is the major methyl group donor in mammals and homocysteine is a by-product of this process. Homocysteine provides a reservoir for regenerating methionine (Fig. 16.1), thereby maintaining

                      Serum homocysteine and coronary heart disease

Fig. 16.1 Summary of the major pathways of homocysteine metabolism.

(p.240) the methylation process throughout the body (Hankey and Eikelboom 1999). Otherwise it serves no known useful function and is thought to be toxic to vascular endothelial cells, to increase blood coagulability, and to promote smooth muscle cell proliferation – processes central to atherosclerosis (Brown et al. 1998; Fryer et al. 1993; Bellamy et al. 1998).

Homocysteine is metabolized in two ways (Fig. 16.1). It is remethylated to methionine, with 5-methyl tetrahydrofolate acting as the methyl group donor, a process dependent on vitamin B12 and the enzyme methionine synthase. Tetrahydrofolate is then remethylated to replenish 5-methyl tetrahydrofolate, a process dependent on the enzyme methylenetetrahydrofolate reductase (MTHFR) and the amino acid serine. Natural folate and folic acid (its synthetic analogue helps to replenish tetrahydrofolate). Homocysteine is also metabolized by transulfuration to cystathionine, a process dependent on vitamin B6 and the enzyme cystathionine beta synthase. Impairment of either of these two processes can increase serum homocysteine concentrations.

The Homocysteine Trialists Collaboration (1998) showed that folic acid was the most effective of the B vitamins in lowering serum homocysteine. A dose of about 1 mg/day lowered serum homocysteine by 25% (or ~3 μmol/l from the population average level of 12 μmol/l); doses above 1 mg/day produced no additional benefit. Vitamin B12 (0.5 mg/day) produced only an additional 7% reduction and B6 had no further detectable effect. Subsequent trials have shown that the full homocysteine-lowering effect of folic acid is achieved with about 0.8 mg folic acid per day, and most of the effect with 0.4 mg (van Oort et al. 2003; Wald et al. 2001). Folic acid lowers serum homocysteine from all pre-treatment levels in Western populations, though the reduction is greater from higher levels (Wald et al. 2001). Folic acid supplementation is more effective than dietary change; an unrealistically large amount of folate-containing foods (such as ~4 kg of broccoli per day) would need to be eaten to reach the equivalent homocysteine-lowering effect of a daily 0.8 mg folic acid supplement. This is because the folate concentration of foods is relatively low and the bioavailability of natural folate is about half that of folic acid.

16.3 The link between homocysteine and CHD: homocystinuria

The association between homocysteine and CHD was identified in 1969 by McCully, who described premature atherosclerotic disease at autopsy in two children who died with the rare autosomal recessive condition, homocystinuria (McCully 1969). Homocystinuria is a deficiency of one of three enzymes involved in homocysteine metabolism (Fig. 16.1), leading to three distinct disorders: cystathionine beta synthase deficiency, methylenetetrahydrofolate reductase deficiency, and the B12 metabolic defects that result in impaired methionine synthase activity. Heterozygotes for these three disorders have about three times the population average serum homocysteine concentration and a high risk of cardiovascular disease. Homozygotes have serum homocysteine levels 10–50 times the population average and a very high risk of premature cardiovascular disease; ~50% of them experience an arterial or venous disease event by the age of 30 (Mudd et al. 1985). A high homocysteine level is the only biochemical change common to all three disorders; no other substance is consistently high or low. It follows, therefore, that the high homocysteine causes the increased risk of cardiovascular disease. Two studies among homozygotes with homocystinuria treated with vitamins B6, B12, and folic acid indicate that risk can be reduced (Table 16.1). Treatment with these vitamins led to only 2 vascular events when 30 would have been expected (from previous observation in untreated patients) in one study (Kluijtmans et al. 1999), and 0 events when 29 would have been expected in the other (Yap and Naughten 1998). While these were not randomized trials, selection bias could not reasonably explain so large a difference with 2 events observed versus 59 expected.

(p.241)

Table 16.1 Observed numbers of vascular events in two studies of patients (homozygotes) with homocystinuria treated with B vitamins, and the expected numbers of events calculated by the authors of the two studies from age-specific rates in 629 untreated patients with homocystinuria

Study (first author)

No. of patients

Median age at diagnosis of homocystinuria

Mean follow-up (years)

Vascular eventsa

p value

Observed

Expected

Kluijtmans 1999

29

23

13

2

30

< 0.001

Yap 1998

25

0 (newborn)

15

0

29

< 0.001

(a) Deep vein thrombosis, pulmonary embolism, myocardial infarction, stroke, or peripheral arterial disease).

16.4 Retrospective and prospective epidemiological studies

Retrospective and prospective studies provide evidence of the dose–response relationship across the range of serum homocysteine in the population. There is about a two-fold risk gradient from the highest to the lowest fifth of serum homocysteine values.

In retrospective studies, homocysteine is measured after the diagnosis of CHD in cases (generally after a myocardial infarction) and in unaffected controls. Over 30 such studies have been published and all show a positive association between CHD and serum homocysteine. Figure 16.2 shows the results of a meta-analysis of the 12 published retrospective studies (combining data from 1517 cases of myocardial infarction) that reported the proportional difference in risk for a specified serum homocysteine difference, adjusted for age and, in some studies, other cardiovascular risk factors (or reported data from which this could be calculated) (Chao et al. 1999; Genest et al. 1990; Hoogeveen et al. 1998; Hopkins et al. 1995; Israelsson et al. 1988; Joubran et al. 1998; Loehrer et al. 1996; Malinow et al. 1996; Pancharuniti et al. 1994; Schwartz et al. 1997; Thögersen et al. 2001; Verhoef et al. 1996). The risk of a CHD event (odds ratio) for 3 μmol/l lower serum homocysteine (achievable by taking 0.8 mg folic acid) is shown for each study together with the summary estimate for all studies combined. The summary odds ratio was 0.78 (95% confidence interval: 0.72–0.85), or 0.75 (0.68–0.82) adjusted for regression-dilution bias (i.e., the diminution of an association because of imprecise measurement) (Clarke et al. 2001). This result is likely to overestimate the true effect, as some of the studies did not adjust for confounding by cardiovascular risk factors such as smoking, serum cholesterol, and blood pressure and possibly because atherosclerotic disease may increase homocysteine, due to reduced renal function (so-called ‘reverse causality’) (Wald et al. 2003).

Prospective studies, by their design, guard against an effect of disease on homocysteine. In these studies blood is taken from healthy subjects who are then followed up for several years. An efficient design used in many of the prospective studies of homocysteine and vascular disease, which avoids testing many thousands of samples at the outset, is to store the blood and test stored samples from those who later develop CHD events and from matched controls, a so-called nested case control design. Figure 16.3 shows the results of a meta-analysis of 16 published prospective studies of serum homocysteine and CHD events (death or non-fatal myocardial infarction, n = 3144) (Wald et al. 2002). The odds ratios shown were adjusted for age, sex, smoking habits, blood pressure, and serum cholesterol in all the studies except one, which was adjusted for age and sex alone (Stampfer et al. 1993). The summary odds ratio was 0.89 (0.85–0.92) for a 3 μmol/l lower serum homocysteine, or 0.85 (0.80–0.90) adjusted (p.242)

                      Serum homocysteine and coronary heart disease

Fig. 16.2 Results of 12 retrospective studies of serum homocysteine and CHD events: values are odds ratios (95% confidence intervals) for a 3 μmol/l lower serum homocysteine. Results are adjusted for age, sex, and, in some studies, other cardiovascular risk factors, but not for regression-dilution bias.

                      Serum homocysteine and coronary heart disease

Fig. 16.3 Results of 16 prospective studies of serum homocysteine and CHD events: values are odds ratios (95% confidence intervals) for a 3 μmol/l lower serum homocysteine, adjusted for age, sex, smoking, serum cholesterol, and blood pressure (age and sex alone in one study (Stampfer et al. 1993)) but not for regression-dilution bias. Modified from Wald et al. 2002.

for regression-dilution bias. These results are similar to those published from another meta-analysis of 11 prospective studies (Homocysteine Studies Collaboration 2002).

The retrospective and prospective studies show a positive association between serum homocysteine and CHD. These studies on their own may be insufficient to determine whether (p.243) the association is one of cause and effect but with the additional evidence from genetic epidemiological studies of the thermolabile C677T MTHFR polymorphism, the uncertainty can be resolved.

16.5 Genetic epidemiology: the MTHFR studies

Moderately raised serum homocysteine levels (about 25% above average levels) occur as a result of a single mutation in the MTHFR gene (cytosine to thymidine (C→T) at base pair position 677) that renders the enzyme thermolabile with reduced activity (Frosst et al. 1995). The presence of this polymorphism in the population provides a natural experiment capable of testing whether moderately raised levels cause CHD. The C→T mutation is common (about 10% of individuals are homozygous (TT) and about 47% are heterozygous (CT)) such that it has been possible to conduct studies of the risk of CHD in persons with and without the mutation, and many are now available.

The estimated difference in homocysteine levels between persons homozygous for the abnormal allele (TT) and persons homozygous for the normal allele (CC), from a meta-analysis of 33 studies, is 2.7 μmol/l (Wald et al. 2002). The effect of the TT genotype on serum homocysteine levels varies between individuals and communities because it is subject to environmental influence, in particular serum folate (Kluijtmans et al. 1997). The variable differences in serum homocysteine mean that heterogeneity between studies is to be expected and has been observed. The heterogeneity and the relatively small difference in homocysteine between TT and CC mean that large numbers are needed to show a statistically significant association with CHD. It is only in 2002 that sufficient data have become available (through the publication of over 40 studies) to permit a meta-analysis with sufficient statistical power.

Figure 16.4 shows the odds ratios of CHD (95% confidence intervals) for CC homozygotes relative to TT homozygotes in order of increasing effect, from a meta-analysis of 46 studies combining data from 12 193 cases and 11 945 controls. The overall summary odds ratio is 0.83 (95% confidence interval: 0.72–0.94; p = 0.01), indicating that the risk of CHD is, on average, 17% lower in CC homozygotes than in TT homozygotes (Wald et al. 2002). Another meta-analysis yielded a similar result (0.86 (0.78–0.95)) (Klerk et al. 2002). The odds ratio of 0.83 for the average homocysteine difference of 2.7 μmol/l is equivalent to an odds ratio of 0.81 (0.69–0.88) for the 3 μmol/l decrease in homocysteine produced by folic acid (calculated by raising 0.83 to the power of 3/2.7). This is similar to the summary estimate from the prospective studies for the same difference in serum homocysteine (odds ratio 0.85 (0.80–0.90)).

16.6 Interpretation of the evidence on causality

The results from the prospective and the MTHFR studies can in principle be interpreted in one of two ways – a direct (or causal) explanation or an indirect (or non-causal) explanation. An indirect (non-causal) explanation would depend on the prospective and MTHFR studies both showing associations with homocysteine through confounding. In the MTHFR studies, the homocysteine difference arises from a single gene mutation effectively allocated at random throughout the population through the random segregation of alleles during gametogenesis and conception – known as Mendelian randomization. There is, therefore, no basis for expecting that persons with and without the mutant gene would systematically differ in other cardiovascular risk factors. The data from these studies confirm this; there were no statistically significant differences in serum cholesterol levels, blood pressure, or smoking habits between (p.244)

                      Serum homocysteine and coronary heart disease

Fig. 16.4 Results of published studies of association between methylenetetrahydrofolate reductase (MTHFR) mutation and CHD events: values are odds ratios (95% confidence intervals) for homozygotes for mutant allele (TT) vs. wild type (CC). Modified from Wald et al. 2002.

(p.245) persons with the TT and CC genotypes (Wald et al. 2002). The confounding would, therefore, have to involve some unknown cardiovascular risk factor, controlled by a gene linked to the MTHFR gene (i.e., at a neighbouring locus on the same chromosome). Importantly, the proposed genetic linkage could not account for confounding in the prospective studies because it would be too weak to do so, accounting for only one-quarter of the two-fold higher risk observed from the 10–90th centiles in the prospective studies. The indirect explanation relies on two separate explanations for the effects in the prospective and genetic epidemiological studies that produce nearly identical results for a given difference in serum homocysteine, even though any confounding would differ across the two types of study. This is so complex and improbable that it can be reasonably rejected, leaving the direct (causal) explanation as the simpler and more plausible interpretation of the results.

16.7 Dose-response relationship between homocysteine and CHD

Figure 16.5 shows dose-response plots of the incidence of CHD events against serum homocysteine. The figure shows plots of two prospective studies (Ridker et al. 1999; Wald et al. 1998), both of which reported summary estimates of risk close to the median for all prospective studies, a meta-analysis of retrospective studies (seven that published individual data on serum homocysteine in cases and controls – so permitting a plot of the combined results) (Genest et al. 1990; Hoogeveen et al. 1998; Joubran et al. 1998; Malinow et al. 1996; Pancharuniti et al. 1994; Thögersen et al. 2001; Verhoef et al. 1996) and a meta-analysis of the MTHFR studies (plots of relative risk over a narrower range of homocysteine in persons with the CC, CT, and TT genotypes) (Wald et al. 2002).

                      Serum homocysteine and coronary heart disease

Fig. 16.5 Dose-response plots of the relative odds of CHD events against serum homocysteine from two prospective studies, a meta-analysis of 46 methylenetetrahydrofolate reductase (MTHFR) studies, and a meta-analysis of seven retrospective studies.

(p.246) With relative odds of disease on the vertical axis using a proportional (or logarithmic) scale, the plots yield reasonably straight lines, indicating a constant proportional lower risk with lower serum homocysteine levels, from any starting point of serum homocysteine. It follows from the continuous dose-response plots in Fig. 16.5 that, as with the other cardiovascular risk factors, intervention to lower serum homocysteine should not be limited to people with a high serum homocysteine, but should be applicable to everyone at high risk whatever the reason for the high risk.

16.8 Implications for prevention

The summary results from the MTHFR and prospective studies are combined in Table 16.2. The effects on deep vein thrombosis and stroke are also shown (Wald et al. 2002). The overall (weighted average) odds ratio for CHD is 0.84 (0.80–0.89) for 3 μmol/l lower serum homocysteine – an expected reduction in risk of 16% (11–20%). This estimated risk reduction is relatively modest compared to the effect of treatments that lower serum cholesterol or blood pressure. Nonetheless, the public health impact would be large because CHD is so common (about 120 000 deaths in the UK and about 400 000 deaths in the USA each year).

16.9 The ongoing randomized trials of homocysteine reduction

It is widely felt that clinical practice should not change until there are randomized trials that show the effect of treatment on disease events. We believe that this is too simplistic. For example, there is no randomized trial evidence of the effect of giving up smoking on the risk of CHD, but causality is accepted. It is argued that we have previously been misled by associations in epidemiological studies (for example, the anti-oxidant, vitamin E, and CHD in the US Nurses Study and in the US male Health Professionals Study (Rimm et al. 1993; Stampfer et al. 1993)), but randomized trials showed otherwise (Heart Protection Study Collaborative Group 2002; The Heart Outcomes Evaluation Study Investigators 2000; see also the previous chapter). The inference from this that observational studies cannot be relied on is unjustified because in

Table 16.2 Summary results (95% confidence intervals) from the MTHFR studies and the prospective studies on CHD, deep vein thrombosis, and strokea

Study type

No. of studies

No. of cases

OR for 3 μmol/l lower

homocysteine

OR expressed as risk reduction

CHD

MTHFR

 Prospectiveb

46

16

12193

3144

0.81 (0.69–0.88)

0.85 (0.80–0.90)

0.84 (0.80–0.89)

16% (11–20%)

Deep vein thrombosis

MTHFR

26

3439

0.75 (0.62–0.92)

25% (8–38%)

Stroke

MTHFR

 Prospectiveb

7

8

1217

676

0.74 (0.43–1.28)

0.76 (0.67–0.86)

0.76 (0.67–0.85)

24% (15–33%)

(a) Results taken from Wald et al. 2002.

(b) Prospective studies adjusted for regression-dilution bias, and for age, sex, blood pressure, and serum cholesterol in all studies, except age and sex only in one.

(p.247)

Table 16.3 Randomized trials of homocysteine reduction with folic acid, Vitamin B6, and Vitamin B12 with vascular disease endpoints

Study

Population

Start

Primary outcome

Intervention

Size

Cambridge Heart Antioxidant Study, UK

MI/angina

1998

MI

Folic acid, 5 mg/dayvs. placebo

4000

Oxford Study of the Effectiveness Of Additional Reductions in Cholesterol and Homocysteine, UK

MI

1998

MI

Folic acid 2 mg/day and vitamin B12 1 mg/dayvs. placebo; [SEARCH] Study

12000

Norwegian Study of Homocysteine Lowering with B Vitamins in Myocardial Infarction [NORVIT] Study, Norway

MI

1998

MI

Folic acid 5 mg/day for 2 weeks, then 0.8 mg vs. placebo; vitamin B6, 40 mg/dayvs. placebo

3000

Bergan Vitamin study, Norway

Stroke/TIA

1997

Stroke

Folic acid 5 mg/day for 2 weeks, then 0.8 mg vs. placebo; vitamin B6 40 mg/day vs. placebo

2000

Prevention with a Combined Inhibitor and Folate in Coronary Heart Disease [PACIFIC] study, Australia

MI/angina and risk factors

1998

Peripheral vascular disease

Folic acid 0.2 or 2 mg/day, vs. placebo

10000

Vitamins To prevent Stroke [VITATOPS] study, Australia

Stroke/TIA

1999

Stroke

Folic acid 2 mg/day, B6 25 mg/day, and B12 0.4 mg/day, vs. placebo

5000

Vitamins in stroke prevention [VISP] study, USA

Stroke/TIA

1998

Stroke

Folic acid 2.5 mg/day, B6 25 mg/day, and B12 0.4 mg/day, vs. folic acid 0.2 mg/day and B12 0.06 mg/day

3600

Womens Antioxidant and Cardiovascular disease Study, USA

Vascular disease and risk factors

1998

Vascular disease

Folic acid 2.5 mg/day, B6 50 mg/day, and B12 1 mg/day, vs. placebo.

8000

Heart Outcomes Prevention Evaluation Study [HOPE-2], Canada

Vascular disease

1999

Peripheral vascular disease

Folic acid 2.5 mg/day, B6 50 mg/day, and B12 1 mg/day

5000

Vitamin and Thrombosis Trial, Netherlands

DVT/PE

2000

Venous thrombosis

Folic acid 5 mg/day, B12 0.4 mg/day, and B6 50 mg/day

600

DVT: deep vein thrombosis; PE: pulmonary embolism; TIA: transient ischaemic attack.

Adapted from Clarke and Collins 1998.

(p.248) these examples there was no basis to exclude confounding as the reason for the association in the observational studies, as acknowledged by the authors (Rimm et al. 1993; Stampfer et al. 1993). Also there was genuine uncertainty over whether hormone replacement therapy reduced cardiovascular risk, which is why randomized trials were needed (Grady et al. 2002). In the case of homocysteine and CHD the position is very different. As well as the evidence from prospective studies, there is genetic evidence from the MTHFR and the homocystinuria studies, the kind of corroboration which was lacking for vitamin E and hormone replacement therapy. In our view, randomized trials are not necessary to show that homocysteine levels are causally related to cardiovascular disease but, if sufficiently large, may provide an indication of the time required to realize the full potential 16% reduction in CHD risk, which is not known from current evidence.

Table 16.3 lists the ongoing large randomized trials of folic acid, vitamin B6, and vitamin B12 in relation to cardiovascular disease events (Clarke and Collins 1998). Because the expected effect of homocysteine reduction on CHD prevention is modest, the trials need to be extremely large to have sufficient statistical power to show it. A trial of recurrent CHD events (a secondary prevention trial) would require about 15 000 participants followed for 5 years to demonstrate the expected 16% reduction in CHD events. None of the ongoing trials are this large. The introduction of folic acid fortification of flour (to prevent neural tube defects) in the US and other countries has, furthermore, reduced the statistical power by reducing the already small homocysteine difference between treated and control groups (since the control as well as the treated individuals groups are now receiving some folic acid).

Three small randomized trials have been published. There is a tendency for these to be interpreted as either positive or negative when, in fact, they lack the statistical power to be informative (Baker et al. 2002; Liem et al. 2003; Schnyder et al. 2001); their confidence intervals are consistent with no effect and with the expected modest effect. Over the next few years, evidence from other trials will emerge and it will be important to avoid interpreting non-significant results as negative – that is, to avoid concluding that no evidence of an effect is evidence of no effect. The focus should be on whether the confidence intervals include the expected relative risk of 0.84 for a 3 μmol/l homocysteine reduction. Only if they include 1.0 but exclude 0.84 would there be reason to question our expectation.

16.10 Conclusions

Four observations arise from the evidence summarized in this chapter:

  1. 1 The genetic (MTHFR) studies show a moderately higher risk of CHD for a moderately higher level of serum homocysteine.

  2. 2 The prospective studies show a positive association between serum homocysteine and cardiovascular disease after allowance for confounding.

  3. 3 These two types of study are susceptible to different sources of error, but show quantitatively similar associations, a result that is unlikely to have occurred through different potential sources of confounding acting independently.

  4. 4 The homocystinurias cause high serum homocysteine levels and high risks of premature cardiovascular disease, and lowering serum homocysteine reduces this high risk.

Together, these observations provide a compelling case for a cause and effect relationship between homocysteine and CHD and, therefore, a protective role for folic acid on CHD prevention.

(p.249) References

Bibliography references:

Baker, F., Picton, D., Blackwood, S. et al. (2002). Blinded comparison of folic acid and placebo in patients with ischaemic heart disease: an outcome trial [Abstract]. Circulation, 106 (Suppl. II), 274.

Bellamy, M. F., McDowell, I. F., Ramsey, M. W. et al. (1998). Hyperhomocysteineia after an oral methionine load acutely impairs endothelial function in healthy adults. Circulation, 98, 1848–52.

Brown, J. C., Rosenquist, T. H., and Monaghan, D. T. (1998). ERK2 activation by homocysteine in vascular smooth muscle cells. Biochemical and Biophysical Research Communications, 251, 669–76.

Chao, C. L., Tsai, H. H., Lee, C. M. et al. (1999). The graded effect of hyperhomocysteinemia on the severity and extent of coronary atherosclerosis. Atherosclerosis, 147, 379–86.

Clarke, R. and Collins, R. (1998). Can dietary supplements with folic acid or vitamin B6 reduce cardiovascular risk? Design of clinical trials to test the homocysteine hypothesis of vascular disease. Journal of Cardiovascular Risk, 5, 249–55.

Clarke, R., Lewington, S., Donald, A. et al. (2001). Underestimation of the importance of homocysteine as a risk factor for cardiovascular disease in epidemiological studies. Journal of Cardiovascular Risk, 8, 363–9.

Frosst, P., Blom, H. J., and Milos, R. (1995). A candidate genetic risk factor for vascular disease: a common mutation in methylenetetrahydrofolate reductase. Nature Genetics, 10, 111–13.

Fryer, R. H., Wilson, B. D., Gubler, D. B. et al. (1993). Homocysteine, a risk factor for premature vascular disease and thrombosis, induces tissue factor activity in endothelial cells. Arteriosclerosis and Thrombosis, 13, 1327–33.

Genest, J. J., McNamara, J. R., Salem, D. N. et al. (1990). Plasma homocysteine levels in men with premature coronary artery disease. Journal of the American College of Cardiology, 16, 1114–19.

Grady, D., Herrington, D., Bittner, V. et al. (2002). Cardiovascular disease outcomes during 6.8 years of hormone therapy: Heart and Estrogen/progestin Replacement Study follow-up (HERS II). Journal of the American Medical Association, 288 (1), 49–57.

Hankey, G. J. and Eikelboom, J. W. (1999). Homocysteine and vascular disease. Lancet, 354, 407–13.

Heart Protection Study Collaborative Group (2002). MRC/BHF Heart Protection Study of antioxidant vitamin supplementation in 20 536 high-risk individuals: a randomised placebo-controlled study. Lancet, 360, 23–33.

Homocysteine Lowering Trialists Collaboration (1998). Lowering blood homocysteine with folic acid based supplements: meta-analysis of randomised trials. British Medical Journal, 316, 894–8.

Homocysteine Studies Collaboration (2002). Homocysteine and risk of ischaemic heart disease and stroke. Journal of the American Medical Association, 288, 2015–22.

Hoogeveen, E. K., Kostense, P. J., Beks, P. J. et al. (1998). Hyperhomocysteinemia is associated with an increased risk of cardiovascular disease, especially in non-insulin-dependent diabetes mellitus. Arteriosclerosis, Thrombosis, and Vascular Biology, 18, 133–8.

Hopkins, P. N., Wu, L. L., Hunt, S. C. et al. (1995). Higher plasma homocysteine and increased susceptibility to adverse effects of low folate in early familial coronary artery disease. Arteriosclerosis, Thrombosis, and Vascular Biology, 15, 1314–20.

Israelsson, B., Bratttstrom, L. E., and Hultberg, B. L. (1988). Homocysteine and myocardial infarction. Atherosclerosis, 71, 227–33.

Joubran, R., Asmi, M., Busjahn, A. et al. (1998). Homocysteine levels and coronary heart disease in Syria. Journal of Cardiovascular Risk, 5, 257–61.

Klerk, M., Verhoef, P., Clarke, R. et al. (2002). MTHFR 677C to T Polymorphism and risk of coronary heart disease. Journal of the American Medical Association, 288, 2023–31.

Kluijtmans, L. A. J., Kastelein, J. J. P., Lindemans, J. et al. (1997). Thermolabile methylenetetrahydrofolate reductase in coronary artery disease. Circulation, 96, 2573–7.

Kluijtmans, L. A. J., Boers, G. H. D., Kraus, J. P. et al. (1999). The molecular basis of cystathionine-synthase deficiency in Dutch patients with homocystinuria: effect of CBS genotype on biochemical and clinical phenotype and on response to treatment. American Journal of Human Genetics, 65, 59–67.

(p.250) Liem, A., Reynierse-Buitenwerf, G. H., Zwinderman, A. H. et al. (2003). Secondary prevention with folic acid: effects on clinical outcomes. Journal of the American College of Cardiology, 41, 2105–13.

Loehrer, F. M., Angst, C. P., Haefeli, W. E. et al. (1996). Low whole-blood s-adenosylmethionine and correlation between 5-methylenetetrahydrofolate and homocysteine in coronary artery disease. Arteriosclerosis, Thrombosis, and Vascular Biology, 18, 727–33.

Malinow, M. R., Ducimetiere, P., Luc, G. et al. (1996). Plasma homocysteine levels and graded risk for myocardial infarction: findings in two populations at contrasting risk for coronary disease. Atherosclerosis, 126, 27–34.

McCully, K. S. (1969). Vascular pathology of homocyteinemia: implications for the pathogenesis of arteriosclerosis. American Journal of Pathology, 56, 111–28.

Mudd, S. H., Skovby, F., Levy, H. L. et al. (1985). The natural history of homocystinuria due to cystathionine beta-synthase deficiency. American Journal of Human Genetics, 37, 1–31.

Pancharuniti, N., Lewis, C. A., Sauberlich, H. E. et al. (1994). Plasma homocysteine, folate, vitamin B12 concentrations and risk for early onset coronary artery disease. American Journal Clinical Nutrition, 59, 940–94.

Ridker, P. M., Manson, J. E., Buring, J. E. et al. (1999). Homocysteine and risk of cardiovascular disease among postmenopausal women. Journal of the American Medical Association, 281, 1817–21.

Rimm, E. B., Stampfer, M. J., Ascherio, A. et al. (1993). Vitamin E consumption and the risk of coronary heart disease in men. New England Journal of Medicine, 328, 1450–6.

Schnyder, G., Roffi, M., and Pin, R. (2001). Decreased rate of coronary restenosis after lowering of plasma homocysteine levels. New England Journal of Medicine, 345, 1593–600.

Schwartz, S. M., Siscovick, D. S., Malinow, M. R. et al. (1997). Myocardial infarction in young women in relation to plasma total homocysteine, folate, and a common variant in the methylenetetrahydrofolate reductase gene. Circulation, 96, 412–17.

Stampfer, M. J., Hennekens, C. H., Manson, J. E. et al. (1993). Vitamin E consumption and the risk of coronary disease in women. New England Journal of Medicine, 328, 1444–9.

The Heart Outcomes Evaluation Study Investigators (2000). Vitamin E supplementation and cardiovascular events in high-risk patients. New England Journal of Medicine, 342, 154–60.

Thögersen, A. M., Nilsson, T. K., Dahlen, G. et al. (2001). Homozygosity for the mutation C677T of 5, 10-methylenetetrahydrofolate reductase and total plasma homocysteine are not associated with greater than normal risk of a first myocardial infarction in northern Sweden. Coronary Artery Disease, 12, 85–90.

van Oort, F. V. A., Melse-Boonstra, A., Brouwer, I. A. et al. (2003). Folic acid and reduction of plasma homocysteine concentrations in older adults: a dose-response study. American Journal of Clinical Nutrition, 77, 1318–23.

Verhoef, P., Stampfer, M. J., Buring, J. E. et al. (1996). Homocysteine metabolism and risk of myocardial infarction: relation with vitamins B6, B12 and folate. American Journal of Epidemiology, 143, 845–59.

Wald, D. S., Bishop, L., Wald, N. J. et al. (2001). Randomised trial of folic acid supplementation on serum homocysteine levels. Archives of Internal Medicine, 161, 695–700.

Wald, D. S., Law, M., and Morris, J. (2002). Homocysteine and cardiovascular disease: evidence on causality from a meta-analysis. British Medical Journal, 325, 1202–6.

Wald, D. S., Law, M. L., and Morris, J. (2003). Is serum homocysteine measurement of value in predicting the severity of coronary artery disease? Thrombosis Research, 111, 55–7.

Wald, N. J., Watt, H. C., Law, M. R. et al. (1998). Homocysteine and ischaemic heart disease: results of a prospective study with implications regarding prevention. Archives of Internal Medicine, 158, 862–7.

Yap, S. and Naughten, E. (1998). Homocyseinemia due to cystatjionine beta synthase deficiency in Ireland: 25 years experience of a newborn screened and treated population with reference to a clinical outcome and biochemical control. Journal of Inherited Metabolic Disease, 21, 738–47.