|
| HOME | SUBSCRIPTIONS | CURRENT ISSUE | PAST ISSUES | CARDIOSOURCE | SEARCH | HELP | FEEDBACK |
|
|
|
|||||||||
|
|||||||||||
|
J Am Coll Cardiol, 1998; 32:353-359 © 1998 by the American College of Cardiology Foundation |
|| || || ||
||
* Department of Nutrition, Harvard School of Public Health, Boston, Massachusetts, USA
Department of Epidemiology, Harvard School of Public Health, Boston, Massachusetts, USA
Department of Cancer Biology, Harvard School of Public Health, Boston, Massachusetts, USA
Department of Environmental Health, Harvard School of Public Health, Boston, Massachusetts, USA
|| Channing Laboratory, Department of Medicine, Harvard Medical School and Brigham and Women’s Hospital, Boston, Massachusetts, USA
¶ Division of Human Nutrition and Epidemiology, Agricultural University, Wageningen, The Netherlands
Manuscript received August 13, 1997; revised manuscript received March 30, 1998, accepted April 16, 1998.
Address for correspondence: Dr. Eric B. Rimm, Department of Nutrition, Harvard School of Public Health, 665 Huntington Avenue, Boston, MA 02115
 |
Abstract |
|---|
 Top Abstract MethodsResultsDiscussionReferences |
|---|
Background. The mutation increases plasma homocysteine levels by impairing its remethylation. Increased plasma homocysteine is an independent risk factor for cardiovascular disease.
Methods. This was a case-control study nested within the Health Professionals Follow-up Study. In 1986, 44,940 U.S. male health professionals, aged 40 to 75 years and free from diagnosed cardiovascular disease, provided detailed information on usual dietary intake, including intake of folate, vitamins B2, B6 and B12, and methionine. Between 1993 and 1995, blood samples were provided by approximately 40% of the participants. We compared data from 500 men with nonfatal coronary heart disease, diagnosed between 1986 and 1992, with data from 500 age-matched control subjects who were free of diagnosed cardiovascular disease at the time of the matched case subject’s diagnosis.
Results. Frequencies of homozygosity (+/+) and heterozygosity (+/–) for the mutation were 12.2% and 41.8% in case subjects and 14.4% and 40.0% in control subjects. With subjects homozygous (–/–) or heterozygous (+/–) for the wildtype allele as a reference and matched by age, the odds ratio of coronary heart disease was 0.83 (95% confidence interval, 0.57 to 1.19) for +/+ subjects. The odds ratio was unchanged after adjustment for smoking and other risk factors for coronary heart disease. Odds ratios were also calculated within strata for intake of vitamins involved in homocysteine metabolism or methionine, the metabolic precursor of homocysteine. The +/+ genotype was not directly associated with risk of coronary heart disease among men with low (that is, within the lowest quartile) intake (<301 µg/d) of folate, the substrate for methylenetetrahydrofolate reductase; low intake (<1.8 mg/d) of vitamin B2, the cofactor for methylenetetrahydrofolate reductase; low intake (<8.0 µg/d) of vitamin B12, the cofactor for remethylation; low intake (<2.1 mg/d) of vitamin B6, the cofactor in the catabolic pathway of homocysteine; or high intake (>2.4 g/d) of methionine.
Conclusions. In this generally well-nourished population, men with the +/+ genotype for the C677T mutation in the methylenetetrahydrofolate reductase gene have no increase in risk of coronary heart disease, even when intake of folate or other B vitamins is low.
| ||||||||||||||
In the present investigation among 500 case subjects and 500 control subjects in the Health Professionals Follow-up Study, we studied risk of CHD associated with homozygosity for the mutation. We investigated whether the effect of the mutation could be moderated by dietary folate and various other factors that might affect homocysteine or folate metabolism, including age, and intake of vitamins B2, B6, B12, methionine (the metabolic precursor of homocysteine), and alcohol. We hypothesized that homozygosity for the mutation would be a stronger risk factor for CHD in subgroups with low vitamin intake levels (particularly low folate intake), high methionine intake, and high alcohol consumption (because of its effects on folate metabolism).
|
Methods |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
On the basis of a priori criteria we excluded 1,595 men whose reported daily energy intake was outside the range of 3,360 to 17,640 kJ or who left blank 70 or more items about food in the dietary questionnaire. Because men with CVD or related conditions may have recently changed their dietary pattern in response to diagnosis, we further excluded 4,994 men who reported CVD on the baseline 1986 questionnaire, leaving 44,940 men for follow-up.
Blood collection. In 1992, we sent a blood collection kit to each participant who was willing to provide a specimen. Participants were asked to return three 10-ml vials of whole blood. Between 1993 and 1995 we received blood samples from 18,025 participants. Upon arrival, the specimens were centrifuged at 4°C to obtain plasma, serum, and buffy coat. These samples were stored at –150°C in liquid nitrogen.
Case ascertainment and control identification. We selected subjects who had provided blood samples and in whom CHD had been diagnosed (either myocardial infarction, coronary artery bypass grafting [CABG], or percutaneous transluminal coronary angioplasty [PTCA]) after return of the baseline 1986 questionnaire but before January 31, 1992. The men reported the diagnosis on the follow-up questionnaires of 1988, 1990, or 1992. We requested medical records from a sample of participants who reported CABG or PTCA. In a sample of 102 self-reported events, 98 (96%) were confirmed by medical record, so we accepted self-report of all other CABGs or PTCAs as CHD end points. A myocardial infarction was confirmed by a study physician using the following World Health Organization criteria: compatible symptoms plus either typical electrocardiographic changes or elevation of cardiac enzymes (33). We identified 500 cases of nonfatal CHD, 220 with myocardial infarction and 280 with either CABG or PTCA. Because the blood samples were collected after diagnosis, we could not include fatal cases. One control subject was matched to each case subject by age (±1 year). Control subjects were randomly selected from participants who had also provided a blood sample, met the matching criteria and were free from clinical CHD at the time of diagnosis in the matched case subject.
Dietary assessment. The 1986 questionnaire inquired about average intake of 131 foods during the previous year. In addition to the structured questions on food intake, we asked participants to identify specific brands of cold breakfast cereals, multivitamins, and cooking oils. Average intake of methionine, folate, and vitamins B2, B6 and B12 was computed by multiplying the frequency of consumption of each food item by the nutrient content of the listed portion size (34). Intake from vitamin supplements was factored into average intake. We used food composition tables from U.S. Department of Agriculture sources (35) with supplementary information provided by collaborating laboratories and other publications. We adjusted all dietary intakes for total energy intake by using regression analysis (36). In 1987, the precision of nutrient consumption measured by the questionnaire was evaluated in a subsample of 127 participants from the Boston area (34). In brief, these men completed two 1-week dietary records and provided a fasting blood sample. Mean daily intake of nutrients, including folate, vitamins B2, B6 and B12, as assessed by the questionnaire and the dietary records were very similar, and the correlations between the two methods were 0.71 for folate, 0.85 for vitamin B2, 0.82 for vitamin B6, and 0.50 for vitamin B12. Data on methionine intake were not available from diet records. The correlation between folate intake as determined by the questionnaire and erythrocyte folate from the blood samples was 0.55 (with deattenuation for within-person variability in the blood level) (37). In a subpopulation of elderly in the Framingham study (38), in which the same dietary questionnaire was used, the correlation between folate intake and plasma folate concentration was 0.63. In that same study, the correlation between intake of vitamin B6 and plasma pyridoxal-5'-phosphate was 0.39 (39) and the correlation between vitamin B12 intake and plasma vitamin B12 was 0.35 (38).
DNA analysis for MTHFR mutation. We use the notations "+" and "–" to refer to alleles with and without the mutation, respectively. Thus, homozygosity for the mutation is denoted as +/+, heterozygosity as +/–, and homozygosity for the wild type as –/–.
We extracted DNA from the buffy coat specimens. Genotyping for MTHFR was carried out using a modification (40) of the polymerase chain reaction–restriction fragment length polymorphism assay of Frosst et al. (11). The sequence change of cytosine to thymine results in a gain of a Hinfl restriction site. Polymerase chain reaction amplification was followed by Hinfl digestion, and restriction fragments were size-fractioned on polyacrylamide gels. The 198-bp polymerase chain reaction fragment was digested into 175-bp and 23-bp fragments; the latter fragment ran off the gel. As a result, there was one 198-bp band for the –/– genotype, 198-bp and 175-bp bands for the +/– genotype and one 175-bp band for the +/+ genotype. Genotype results were scored without reference to case-control status. If there was any ambiguity, the polymerase chain reaction, Hinfl digestion, and scoring were repeated.
Approximately 10% of the samples were quality control specimens. We selected an equal number of samples known to be of the +/+ genotype, the +/– genotype and the –/– genotype (four of each) and randomly, repeatedly placed them throughout the batches to total 10% of the samples.
Statistical analysis.
Differences in coronary risk factors between case and control subjects were tested for significance by a paired Student’s t test for continuous variables and a McNemar paired chi-square test for categorical variables. The association of MTHFR genotype with risk of CHD was evaluated by calculating odds ratios (ORs) and 95% confidence intervals (CIs) for +/+ genotype, using conditional logistic regression analysis. Odds ratios were also calculated with unconditional logistic regression analysis, disregarding the pair matching for age, but with age as a covariate in the model. We present ORs for +/– and +/+ genotypes, with –/– genotype as the reference, and for +/+ subjects with both other genotypes as the reference. The unmatched risk analyses were repeated for cases of coronary artery disease (that is, all subjects with CABG or PTCA) and cases of myocardial infarction separately, because CABG and PTCA may represent atherosclerotic heart disease and myocardial infarction (athero)thrombotic heart disease. Furthermore, we performed risk analyses for CHD with presence or absence of one or more conventional risk factors for CHD (smoking, hypertension, hypercholesterolemia). In addition, we calculated ORs for +/+ subjects versus +/– or –/– subjects, within strata of dietary folate and other factors that may affect folate or homocysteine metabolism, such as age, alcohol consumption and dietary intake of methionine, and vitamins B2, B6 and B12. The three strata of each of these factors were based on quartile boundaries of the total population (
25th percentile, 26th to 75th percentile and >76th percentile). All p values are two sided.
|
Results |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
|
|
In all subsequent risk analyses, we considered –/– and +/– genotypes as one reference group, because the risk of CHD in subjects with +/– genotype was not higher and because +/– and –/– groups had plasma total homocysteine levels similar to those in other studies (11,13).
Because pathophysiologic mechanisms may differ between subjects with coronary artery disease (that is, persons who have undergone CABG or PTCA) and subjects who have had a myocardial infarction, we repeated the calculations of the ORs for +/+ genotype in these subgroups of cases, using all control subjects in both analyses and taking –/– and +/– genotypes as the reference group. The OR for coronary artery disease was 1.04 (95% CI 0.69 to 1.58) and for myocardial infarction was 0.55 (95% CI 0.32 to 0.95) after adjustment for age. When controlling for other risk factors of CHD, these ORs were 1.04 (95% CI 0.67 to 1.62) and 0.49 (96% CI 0.28 to 0.87), respectively.
Similarly, we conducted separate analyses for CHD with or without the presence of various risk factors for coronary artery disease (current smoking, hypertension, hypercholesterolemia), including all control subjects in both subanalyses. Controlling for age, the ORs for +/+ genotype versus +/– and –/– genotypes were 0.82 (95% CI 0.52 to 1.29) for nonfatal CHD without presence of current smoking, hypertension or hypercholesterolemia (256 cases) and 0.82 (95% CI 0.52 to 1.30) for nonfatal CHD with presence of one or more of the risk factors (244 cases). Additional adjustment for other factors, (body mass index, alcohol intake, diabetes, multivitamin use and family history of myocardial infarction) did not change the ORs.
Finally, we calculated ORs for the +/+ genotype versus the –/– or +/– genotypes within strata of dietary folate level and several other factors that might affect homocysteine or folate metabolism (Table 3). The OR for +/+ genotype was not higher in any of the strata where an increased risk for +/+ genotype might be expected, eg, low intake of folate or other vitamins, high intake of methionine, younger age or high alcohol consumption.
|
|
Discussion |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
A limitation of the present study was that we did not measure plasma homocysteine concentrations. It is possible that men homozygous for the MTHFR C677T mutation had normal homocysteine concentrations, which would explain the absence of an association between genotype and risk of CHD. However, in the Physicians’ Health Study (13), a population comparable to the present one in age, gender, occupation and frequency of vitamin supplement use, mean plasma total homocysteine levels were 2 to 4 µmol/l higher in subjects with +/+ genotype than the –/– or +/– genotype. Differences of this magnitude were also observed in other studies (11,14). On the basis of a large meta-analysis (2), for this difference in plasma homocysteine levels, an OR of about 1.5 for CHD would be expected for the +/+ genotype compared with the other genotypes. The CI for our overall estimate excludes this value, but it is included in the CI for the coronary artery disease end point. Thus, our study suggests that the +/+ genotype for the C677T mutation does not confer an increased risk of CHD, although we cannot completely eliminate a small increase in risk of coronary atherosclerosis.
We considered that only men who survived until 1995 could provide blood samples. If CHD in subjects with +/+ genotype was more often fatal, then these subjects might be underrepresented in our population, which could in part explain an inverse association. However, fatality because of CHD was low (18%) in case subjects. Thus, only an extremely strong relationship between genotype and case fatality could obscure a true OR of 1.5. We have no reason to believe that the relationship between increased plasma homocysteine level differs for incident cases of fatal and nonfatal CHD; in the Physicians’ Health Study results were similar for fatal and nonfatal myocardial infarction (Ma J, personal communication, July 1997).
The +/+ genotype was not associated with risk of CHD in men with no conventional risk factors, men younger than 55 years, men with low intake of folate, vitamin B2, vitamin B12 (coenzyme for remethylation of homocysteine to methionine) or vitamin B6 (coenzyme in catabolic pathway of homocysteine metabolism). The present study examined the interaction between MTHFR genotype and prospectively measured folate intake. Homozygosity for the mutation was associated with reduced plasma folate concentration (13,41); therefore, examining dietary folate is directly advantageous. The detailed food frequency questionnaire measures dietary intake of folate and other B vitamins moderately accurately (35) and correlates with measures of folate status, such as erythrocyte (37) and plasma folate levels (38), thus reducing the chance that we missed any important effect modification. However, this study population is generally well nourished, and an association might be seen with folate intake level less than the lowest quartile of this group.
Comparison with results from other studies. Several studies have investigated the relation between the MTHFR C677T mutation and risk of CVD, with most having CHD as the disease end point of interest (13,14,16–31). In five of these studies (14,17,18,24,29), a direct association was observed, with ORs for the +/+ genotype between 2 and 3, whereas the other studies found no association between this common mutation and risk of CVD. In the studies that observed direct association between the mutation and risk of CVD, patients with early-onset CVD (as expected for a genetically-determined risk factor) were more likely to be included, although other studies of premature cardiovascular disease did not find an association. Another possible explanation for the direct association in those five studies might be a lower status of vitamins involved in homocysteine metabolism, especially folate. In Europe as opposed to the United States, use of vitamin supplements and fortification of food products with folic acid is uncommon. For comparison, in a large population-based sample of Irish men (12), mean serum folate level was approximately 11 nmol/l, whereas mean plasma levels of 14 and 15 nmol/l are reported in U.S. populations (15,25). Similarly, the increased risk in the two Japanese studies (18,29) may be related to the relatively low intake of vitamin B2 (coenzyme for MTHFR) because of low consumption of milk and milk products.
Concluding remarks. There are several possible explanations for the absence of a positive association between +/+ genotype and risk of CHD. It is possible that homocysteine concentrations were not increased in subjects with the +/+ genotype. However, homocysteine concentration is generally higher among those with the +/+ genotype, as well as in well-nourished populations, such as that of the Physicians’ Health Study (13). Therefore, our findings may also indicate that higher plasma total homocysteine level is not itself a risk factor for CHD but a marker of low folate status, which may independently affect risk of CHD. However, two factors weigh against this explanation, plausible atherothrombotic effects of homocysteine (42) and the possibility that the +/+ genotype increases risk in populations with a lower folate status.
We found no evidence that men with the +/+ genotype for the C677T mutation in the MTHFR gene are at increased risk of nonfatal CHD, even among those with folate or other B-vitamin intake lower than the 25th percentile. The well-nourished condition of the study population might contribute to this finding. Our results of a decreased risk of nonfatal myocardial infarction are most likely a chance finding but introduce the possibility of a beneficial effect of the mutation. If the results of all other epidemiologic studies are considered, it is unlikely that the C677T mutation is an important risk factor for CHD in particular and cardiovascular disease in general.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
T mutation in the methylenetetrahydrofolate reductase gene: associations with plasma total homocysteine levels and risk of coronary atherosclerotic disease. Atherosclerosis 1997;132:105–3.
T mutation in 5,10-methylenetetrahydrofolate reductase gene in thrombotic vascular disease. Am J Hum Genet. 1996;59:262–264[Medline]
This article has been cited by other articles:
|
|
 |
|
  U. Lim, K. Peng, B. Shane, P. J. Stover, A. A. Litonjua, S. T. Weiss, J. M. Gaziano, R. L. Strawderman, F. Raiszadeh, J. Selhub, et al. Polymorphisms in Cytoplasmic Serine Hydroxymethyltransferase and Methylenetetrahydrofolate Reductase Affect the Risk of Cardiovascular Disease in Men J. Nutr., August 1, 2005; 135(8): 1989 - 1994. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D Castle Clinical challenges posed by new biotechnology Postgrad. Med. J., February 1, 2003; 79(928): 65 - 66. [Full Text] [PDF]
|
|
|
|
 |
|
 M. Klerk, P. Verhoef, R. Clarke, H. J. Blom, F. J. Kok, E. G. Schouten, and and the MTHFR Studies Collaboration Group MTHFR 677C->T Polymorphism and Risk of Coronary Heart Disease: A Meta-analysis JAMA, October 23, 2002; 288(16): 2023 - 2031. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. McNulty, M. C McKinley, B. Wilson, J. McPartlin, J. Strain, D. G Weir, and J. M Scott Impaired functioning of thermolabile methylenetetrahydrofolate reductase is dependent on riboflavin status: implications for riboflavin requirements Am. J. Clinical Nutrition, August 1, 2002; 76(2): 436 - 441. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Skoupy, M. Fodinger, M. Veitl, A. Perschl, H. Puttinger, C. Rohrer, K. Schindler, A. Vychytil, W. H. Horl, and G. Sunder-Plassmann Riboflavin Is a Determinant of Total Homocysteine Plasma Concentrations in End-Stage Renal Disease Patients J. Am. Soc. Nephrol., May 1, 2002; 13(5): 1331 - 1337. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M.C. Verhaar, E. Stroes, and T.J. Rabelink Folates and Cardiovascular Disease Arterioscler. Thromb. Vasc. Biol., January 1, 2002; 22(1): 6 - 13. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Roest, Y. T. van der Schouw, D. E. Grobbee, M. J. Tempelman, P. G. de Groot, J. J. Sixma, and J. D. Banga Methylenetetrahydrofolate Reductase 677 C/T Genotype and Cardiovascular Disease Mortality in Postmenopausal Women Am. J. Epidemiol., April 1, 2001; 153(7): 673 - 679. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Brattstrom and D. E. Wilcken Homocysteine and cardiovascular disease: cause or effect? Am. J. Clinical Nutrition, August 1, 2000; 72(2): 315 - 323. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. M. Ridker, J. E. Manson, J. E. Buring, J. Shih, M. Matias, and C. H. Hennekens Homocysteine and Risk of Cardiovascular Disease Among Postmenopausal Women JAMA, May 19, 1999; 281(19): 1817 - 1821. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | SUBSCRIPTIONS | CURRENT ISSUE | PAST ISSUES | CARDIOSOURCE | SEARCH | HELP | FEEDBACK |
