Homocysteine, B Vitamins, and Cardiovascular Disease
http://www.100md.com
《新英格兰医药杂志》
To the Editor: In light of the numerous observational studies that have found a positive association between plasma homocysteine levels and the risk of cardiovascular disease, the results of two homocysteine-lowering trials — the Heart Outcomes Prevention Evaluation 2 (HOPE-2) and the Norwegian Vitamin (NORVIT) trials (April 13 issue)1,2 — are disappointing. The relationship between homocysteine and dementia offers a similar paradox. Observational studies have shown positive associations, whereas homocysteine lowering with folic acid and B vitamins has revealed no cognitive benefit.3
However, the negative outcomes of these trials may not come as a complete surprise. Studies of genetic association (the so-called mendelian randomization studies) have not provided evidence of a causal relationship between functional variants of the homocysteine gene and the risk of coronary heart disease.4 Therefore, unlike patients with familial hyperhomocysteinemia (for whom a higher level of homocysteine is a causal risk factor), patients with such increased levels in the population at large may already have vascular disease or cognitive impairment.5 Thus, the aggregated data suggest that higher homocysteine levels may be a consequence rather than a cause of disease.
Anton J.M. de Craen, Ph.D.
Leiden University Medical Center
2300 RC Leiden, the Netherlands
craen@lumc.nl
David J. Stott, M.D., Ph.D.
Glasgow Royal Infirmary
Glasgow G31 2ER, United Kingdom
Rudi G.J. Westendorp, M.D., Ph.D.
Leiden University Medical Center
2300 RC Leiden, the Netherlands
References
The Heart Outcomes Prevention Evaluation (HOPE) 2 Investigators. Homocysteine lowering with folic acid and B vitamins in vascular disease. N Engl J Med 2006;354:1567-1577.
B?naa KH, Nj?lstad I, Ueland PM, et al. Homocysteine lowering and cardiovascular events after acute myocardial infarction. N Engl J Med 2006;354:1578-1588.
Stott DJ, MacIntosh G, Lowe GD, et al. Randomized controlled trial of homocysteine-lowering vitamin treatment in elderly patients with vascular disease. Am J Clin Nutr 2005;82:1320-1326.
Lewis SJ, Ebrahim S, Davey Smith G. Meta-analysis of MTHFR 677CT polymorphism and coronary heart disease: does totality of evidence support causal role for homocysteine and preventive potential of folate? BMJ 2005;331:1053-1056.
Mooijaart SP, Gussekloo J, Fr?lich M, et al. Homocysteine, vitamin B-12, and folic acid and the risk of cognitive decline in old age: the Leiden 85-Plus study. Am J Clin Nutr 2005;82:866-871.
To the Editor: The investigators in the HOPE-2 and NORVIT trials recommend against the use of folic acid, vitamin B6, and vitamin B12 as preventive treatment. However, mean levels of homocysteine, folic acid, vitamin B6, and vitamin B12 were in the normal range in both studies. If patients with hyperhomocysteinemia were included in these trials, would the results have been different? Furthermore, can vitamin supplementation impart benefits to patients with normal levels of these nutrients?
Asian Indians, who have a reduced intake of vitamin B12 and folate, are predominantly vegetarian and have higher homocysteine levels and lower levels of folate and vitamin B12 than do whites.1,2 A study assessing Asian Indians reported a mean plasma homocysteine level of 19.8 mmol per liter, with 77 percent of the subjects having hyperhomocysteinemia and more than 50 percent having a deficiency of vitamin B12.3 Hyperhomocysteinemia is an independent risk factor for coronary heart disease in India and may account for the fact that twice as many Asian Indians die from the disease as do Europeans.1 Furthermore, unlike Asian Indians, approximately 70 percent of the patients in the HOPE-2 study were exposed to folate-fortified food. Whether the study results would be applicable to Asian Indians cannot be answered until prospective, randomized trials are conducted in this population.
Amit Khare, Ph.D.
Meena Lopez, B.Pharm., M.B.A.
Jaideep Gogtay, M.D.
Cipla
Mumbai 400 008, India
amit.khare@cipla.com
All authors are employees of Cipla, an Indian pharmaceutical company that manufactures generic cardiovascular drugs.
References
Chambers JC, Obeid OA, Refsum H, et al. Plasma homocysteine concentrations and risk of coronary heart disease in UK Indian Asian and European men. Lancet 2000;355:523-527.
Misra A, Vikram NK, Pandey RM, et al. Hyperhomocysteinemia, and low intakes of folic acid and vitamin B12 in urban North India. Eur J Nutr 2002;41:68-77.
Refsum H, Yajnik CS, Gadkari M, et al. Hyperhomocysteinemia and elevated methylmalonic acid indicate a high prevalence of cobalamin deficiency in Asian Indians. Am J Clin Nutr 2001;74:233-241.
To the Editor: Despite the claim by Loscalzo in the editorial1 accompanying the reports on the HOPE-2 and NORVIT trials, the lack of benefit of lowering homocysteine concentrations with folic acid is not an "unequivocal conclusion." Relatively little is known about how homocysteine affects cardiovascular disease. Unless perturbed nutritionally or pharmacologically, homocysteine concentrations change relatively little over a five-year period2 and presumably over a longer term. The elevated homocysteine concentration found at diagnosis in patients with cardiovascular disease was probably a chronic condition. Consequently, the cardiovascular insult may have occurred over a period of many years. There is no indication as to how long it would take to reverse such damage. All three intervention studies cited by Loscalzo were of moderate duration (2, 3.5, and 5 years) and may not reflect the benefit of long-term intervention (e.g., prolonged supplementation or universal fortification, particularly for primary prevention). A case in point is the benefit of quitting smoking, since it takes more than five years after smoking cessation for the risks of cardiovascular disease,3 laryngeal cancer,4 and (in heavy smokers) lung cancer5 to diminish.
Eoin P. Quinlivan, Ph.D.
Jesse F. Gregory III, Ph.D.
University of Florida
Gainesville, FL 32611-0370
jfgy@ufl.edu
References
Loscalzo J. Homocysteine trials -- clear outcomes for complex reasons. N Engl J Med 2006;354:1629-1632.
El-Khairy L, Vollset SE, Refsum H, Ueland PM. Predictors of change in plasma total cysteine: longitudinal findings from the Hordaland Homocysteine Study. Clin Chem 2003;49:113-120.
Iso H, Date C, Yamamoto A, et al. Smoking cessation and mortality from cardiovascular disease among Japanese men and women: the JACC Study. Am J Epidemiol 2005;161:170-179.
Altieri A, Bosetti C, Talamini R, et al. Cessation of smoking and drinking and the risk of laryngeal cancer. Br J Cancer 2002;87:1227-1229.
Ebbert JO, Yang P, Vachon CM, et al. Lung cancer risk reduction after smoking cessation: observations from a prospective cohort of women. J Clin Oncol 2003;21:921-926.
To the Editor: The HOPE-2 investigators show a significant, 24 percent reduction in the relative risk of stroke among patients treated with folic acid and vitamins B6 and B12. They downplay this result by relegating a striking figure on the effect of this treatment on stroke (Figure 1) to their online Supplementary Appendix. The authors suggest that the result may be spurious, but it agrees closely with the predictions of two large meta-analyses, which suggested that the same change in homocysteine levels achieved in the HOPE-2 trial would result in a reduction in stroke of 19 to 24 percent.1,2 Their view — that a treatment benefit restricted to stroke is biologically implausible — is surprising, given the etiologic differences in coronary disease and stroke. They claim that the findings of the Vitamin Intervention for Stroke Prevention (VISP)3 and NORVIT studies support their conclusion. However, they did not refer to the reanalysis of the VISP trial, which revealed a significant effect on stroke and coronary events.4 They also did not mention that the NORVIT study was smaller, with a total of 98 strokes, as compared with 258 strokes in their own trial.
Figure 1. Kaplan–Meier Estimates of the Percentage of Patients with Stroke during a Trial of Homocysteine Lowering.
The relative risk of stroke among patients in the active-treatment group (who received folic acid and vitamins B6 and B12), as compared with patients in the placebo group, was 0.75 (95 percent confidence interval, 0.59 to 0.97; P=0.03 by the log-rank test). Data are from the HOPE-2 trial.
The message of the HOPE-2 trial should be one of cautious optimism that B vitamins may protect against stroke. Consistent with this view is a reduction in the rate of death from stroke in the United States and Canada after the introduction of folic acid fortification of food.5
Helga Refsum, M.D.
A. David Smith, D.Phil.
University of Oxford
Oxford OX1 3PT, United Kingdom
helga.refsum@physiol.ox.ac.uk
Dr. Refsum reports having received lecture fees from Diatomics, Nycomed, and Recip AB, as well as grant support from Axis-Shield and Nycomed; and Dr. Smith, lecture fees from Nycomed and Recip AB.
References
Wald DS, Law M, Morris JK. Homocysteine and cardiovascular disease: evidence on causality from a meta-analysis. BMJ 2002;325:1202-1206.
Homocysteine Studies Collaboration. Homocysteine and risk of ischemic heart disease and stroke: a meta-analysis. JAMA 2002;288:2015-2022.
Toole JF, Malinow MR, Chambless LE, et al. Lowering homocysteine in patients with ischemic stroke to prevent recurrent stroke, myocardial infarction, and death: the Vitamin Intervention for Stroke Prevention (VISP) randomized controlled trial. JAMA 2004;291:565-575.
Spence JD, Bang H, Chambless LE, Stampfer MJ. Vitamin Intervention for Stroke Prevention trial: an efficacy analysis. Stroke 2005;36:2404-2409.
Yang Q, Botto LD, Erickson JD, et al. Improvement in stroke mortality in Canada and the United States, 1990 to 2002. Circulation 2006;113:1335-1343.
To the Editor: As compared with the HOPE-2 and VISP trials, the NORVIT trial was unique in that it used a two-by-two factorial design, which allowed for an examination of the effect of each of the three treatments — folic acid plus vitamin B12, vitamin B6 alone, and folic acid plus both vitamin B6 and vitamin B12. However, the NORVIT trial did not report the analyses that we believe are important for a comprehensive assessment of the effects of B vitamins: a comparison of folic acid plus vitamin B12 with placebo, a comparison of vitamin B6 with placebo, and a test of the interaction between folic acid plus vitamin B6 and vitamin B12 in relation to the clinical outcome.
In Table 1, we present our analyses using data from the NORVIT trial. We calculated the rate ratios for each treatment using the number of observed cases and person-years and Poisson distribution and test-based methods to construct confidence intervals.1 We also calculated the rate ratio for the interaction between folic acid plus vitamin B6 and vitamin B12, and we estimated the standard error of the rate ratio by the multivariate delta method.2 We subsequently used this standard error to construct the confidence interval for the rate ratio of the interaction.
Table 1. Rate Ratios for the Three Treatment Groups and Tests of Interactions between Folic Acid plus Vitamin B6 and Vitamin B12 on Clinical Outcomes.
Our analyses showed that as compared with placebo, folic acid plus vitamin B12 has a slightly but not significantly beneficial effect on most of the clinical outcomes; the same is true for vitamin B6. We found a significant adverse interaction between folic acid plus vitamin B6 and vitamin B12 on all clinical outcomes except stroke, coronary-artery bypass surgery, and percutaneous coronary intervention. If our findings can be independently replicated, one would conclude that it was the interaction between folic acid plus vitamin B6 and vitamin B12 that led to significantly worse outcomes, whereas there was no evidence that treatment with folic acid plus vitamin B12 or with vitamin B6 alone was harmful.
The three trials raise further questions to be addressed in future research. First, all three trials used a high-dose formulation, including 2 to 6 times the recommended daily allowance (RDA) of folic acid, 166 to 416 times the RDA of vitamin B12, and 12 to 25 times the RDA of vitamin B6. The interactions among high-dose treatments may lead to undesirable clinical outcomes. Future studies will need to determine the optimal dose and combination that maximize efficacy and minimize adverse effects. Second, all three trials were conducted in patients with existing cardiovascular disease. Studies are needed to assess the role of B vitamins in the primary prevention of cardiovascular disease, especially in persons with a low intake of B vitamins or with a genetic susceptibility to hyperhomocysteinemia. Third, all three trials reported the averages of the effects in all the treatment groups. However, individual responses to the treatments may vary greatly, depending on the person's homocysteine metabolism and genetic susceptibility, as well as on the presence of other known risk factors for cardiovascular disease. Future studies will need to find a better way to identify patients who will benefit the most from interventions with B vitamins.
Xiaobin Wang, M.D., Sc.D.
Children's Memorial Hospital
Chicago, IL 60614
xbwang@childrensmemorial.org
Hakan Demirtas, Ph.D.
Xiping Xu, M.D., Ph.D.
University of Illinois at Chicago
Chicago, IL 60612
References
Sahai H, Khurshid A. Statistics in epidemiology: methods techniques and applications. Boca Raton, Fla.: CRC Press, 1996:14-8.
Rice JA. Mathematical statistics and data analysis. 2nd ed. Belmont, Calif.: Duxbury Press, 1995:149.
To the Editor: Though elevated levels of homocysteine are predictive of cardiovascular risk, the demonstration that homocysteine-lowering therapy was without benefit in the HOPE-2 and NORVIT trials provides further evidence that homocysteine represents an epiphenomenon in atherosclerosis. The pathogenesis of atherosclerosis is characterized by chronic inflammation,1 and elevated plasma homocysteine concentrations have been correlated with inflammation in conditions such as chronic renal failure, rheumatoid arthritis,2 and psoriasis and in the period after myocardial infarction.3 However, clinical studies have demonstrated that endothelial dysfunction does not improve despite effective lowering of homocysteine levels with 400 μg of folic acid orally per day.4 High-dose oral folic acid (5 to 20 mg per day) improves endothelial function in a manner largely independent of plasma homocysteine lowering, though the underlying mechanism has not been established.5 It is our view, therefore, that high-dose folate therapy (exceeding 5 mg daily) has pleiotropic effects, the benefits of which have yet to be tested in large-scale secondary-prevention trials in appropriate subjects.
David R. Tomlinson, M.R.C.P.
John Radcliffe Hospital
Oxford OX3 9DU, United Kingdom
david.tomlinson@orh.nhs.uk
Derek Lang, Ph.D.
Malcolm J. Lewis, M.B., D.Sc., F.R.C.P.
Wales Heart Research Institute
Cardiff CF14 4XN, United Kingdom
References
Ross R. The pathogenesis of atherosclerosis: a perspective for the 1990s. Nature 1993;362:801-809.
Roubenoff R, Dellaripa P, Nadeau MR, et al. Abnormal homocysteine metabolism in rheumatoid arthritis. Arthritis Rheum 1997;40:718-722.
Landgren F, Israelsson B, Lindgren A, Hultberg B, Andersson A, Brattstrom L. Plasma homocysteine in acute myocardial infarction: homocysteine-lowering effect of folic acid. J Intern Med 1995;237:381-388.
Pullin CH, Ashfield-Watt PA, Burr ML, et al. Optimization of dietary folate or low-dose folic acid supplements lower homocysteine but do not enhance endothelial function in healthy adults, irrespective of the methylenetetrahydrofolate reductase (C677T) genotype. J Am Coll Cardiol 2001;38:1799-1805.
Doshi SN, McDowell IF, Moat SJ, et al. Folic acid improves endothelial function in coronary artery disease via mechanisms largely independent of homocysteine lowering. Circulation 2002;105:22-26.
Dr. Lonn replies: De Craen et al. refer to a recent meta-analysis of mendelian randomization studies to suggest reverse causality as an explanation for the results of the HOPE-2 and NORVIT trials involving patients with coronary heart disease. Findings in patients with stroke may differ.1 Additional plausible explanations include residual confounding in epidemiologic studies and the possibility that any harmful effects of high-dose folate may offset the benefits of homocysteine lowering.
In response to the comments of Refsum and Smith, we do not dismiss the results of the HOPE-2 study in regard to stroke but caution against an overenthusiastic interpretation on the basis of multiple considerations: the overall neutral effect of treatment on the primary outcome and on most secondary and tertiary outcomes (including transient ischemic attacks), the wide confidence intervals around the estimated reduction in the risk of stroke, the apparent increase in the risk of unstable angina, and the neutral results regarding stroke in the VISP and NORVIT trials. The quoted reanalysis of the VISP trial was not prespecified and showed no reduction in stroke (the primary study outcome), death, or events associated with coronary heart disease. At best, the reanalysis showed a borderline effect on a composite cardiovascular outcome (unadjusted outcome, P=0.049; adjusted outcome, P=0.056), which is not a very convincing result to use to justify any clear treatment recommendations. A causal link between recent trends toward a lower rate of death from stroke in the United States and Canada and the fortification of food with folic acid remains speculative, since many other factors may have contributed to the decline.
With regard to the comments by Wang et al. and Tomlinson et al., the doses and combinations of B vitamins used in the large clinical trials are based on the ability of the drugs to reduce homocysteine levels in most people and on the perceived safety of the drugs.2 Whether lower or higher doses or different combinations may be useful remains unproven. A reasonable approach is to encourage people to have balanced diets, since such diets provide adequate amounts of needed macronutrients and micronutrients in most people, and to reserve the use of vitamin therapy for those with proven deficiencies.
Khare et al. and Wang et al. note that the clinical trials studied primarily white, middle-aged patients with vascular disease. Trials of primary prevention and in populations with higher homocysteine levels that are related to genetic and dietary factors are of interest, and we strongly support the completion of ongoing studies in various populations. However, we are unaware of any cardiovascular therapies that are exclusively effective for primary prevention, and in our trial, even patients in the upper fifth of the baseline homocysteine distribution (19.7 μmol per liter) derived no benefit.
Quinlivan and Gregory point out that the trials of B vitamin supplementation are of intermediate duration (two to five years). Most proven preventive therapies, such as the lowering of levels of cholesterol and blood pressure, reduce risk within months to a few years, and in several studies, such as the British Doctors Study, excess risk was halved within two to three years after the cessation of smoking.3
In summary, completed clinical trials do not provide evidence to support the preventive use of B vitamin supplements. Ongoing large trials and the planned meta-analysis of all trials4 will answer remaining relevant clinical questions.
Eva Lonn, M.D.
McMaster University
Hamilton, ON L8L 2X2, Canada
for the HOPE-2 Investigators
References
Casas JP, Bautista LE, Smeeth L, Sharma P, Hingorani AD. Homocysteine and stroke: evidence on a causal link from mendelian randomisation. Lancet 2005;365:224-232.
Homocysteine Lowering Trialists' Collaboration. Dose-dependent effects of folic acid on blood concentrations of homocysteine: a meta-analysis of the randomized trials. Am J Clin Nutr 2005;82:806-812.
Doll R, Peto R. Mortality in relation to smoking: 20 years' observation on male British doctors. Br Med J 1976;2:1525-1536.
B-Vitamin Treatment Trialists' Collaboration. Homocysteine-lowering trials for prevention of cardiovascular events: a review of the design and power of the large randomized trials. Am Heart J 2006;151:282-287.
Dr. B?naa and colleagues reply: The data from the HOPE-2, NORVIT, and VISP trials are quite consistent in showing that homocysteine lowering with folic acid and vitamin B12 (with or without vitamin B6) has no clinical benefit in patients with established vascular disease. The negative results may be interpreted in three ways.
First, homocysteine may not be a causative agent in vascular disease. High homocysteine levels may be an indicator of an unhealthy lifestyle, an epiphenomenon reflecting atherogenic processes, or a consequence of vascular disease itself, as suggested by de Craen et al. and Tomlinson et al.
Second, homocysteine-lowering therapy may still be beneficial in populations other than those studied — for example, in patients with hyperhomocysteinemia, as suggested by Khare et al. However, in the NORVIT trial, this therapy had no benefit in the 40 percent of patients with a baseline homocysteine level above 13 μmol per liter (in this subgroup, the mean homocysteine level was 17.4 μmol per liter). As pointed out by Quinlivan and Gregory and Wang et al., the trials may have been too short (mean duration, 2.5 to 5 years), and the results could possibly be different for primary prevention. However, most conventional treatments (including smoking cessation) show effects on vascular disease within five years. The results of the NORVIT trial do not preclude a protective effect of more physiologic doses of B vitamins or in primary prevention. However, it is difficult to explain biologically how a therapy that does not work in patients who have had a clinical vascular event would work well in those without a clinical event (many of whom have subclinical atherosclerosis).
Finally, as suggested by Loscalzo in his editorial, B vitamin therapy could have harmful effects that offset the homocysteine-lowering benefit. Wang et al. suggest that the trend toward a harmful effect in the combined B vitamin treatment group that was observed in the NORVIT trial was due to a significant interaction between folate plus vitamin B6 and vitamin B12. However, we believe the confidence intervals they present for the test of interaction are too narrow, and they cannot precisely estimate the rate ratios adjusted for study center from the numbers given in our article. Using rate ratios adjusted for study center estimated from Cox proportional-hazard regression and the method described by Altman and Bland,1 we obtained a test statistic (ratio of ratios) of 1.18 (95 percent confidence interval, 0.88 to 1.59) for the primary end point, indicating that there was no significant interaction.
Kaare Harald B?naa, M.D., Ph.D.
University of Troms?
N-9037 Troms?, Norway
kaare.harald.bonaa@unn.no
Aage Tverdal, Ph.D.
Norwegian Institute of Public Health
N-0403 Oslo, Norway
Per Magne Ueland, M.D., Ph.D.
University of Bergen
N-5021 Bergen, Norway
References
Altman DG, Bland JM. Interaction revisited: the difference between two estimates. BMJ 2003;326:219-219.
However, the negative outcomes of these trials may not come as a complete surprise. Studies of genetic association (the so-called mendelian randomization studies) have not provided evidence of a causal relationship between functional variants of the homocysteine gene and the risk of coronary heart disease.4 Therefore, unlike patients with familial hyperhomocysteinemia (for whom a higher level of homocysteine is a causal risk factor), patients with such increased levels in the population at large may already have vascular disease or cognitive impairment.5 Thus, the aggregated data suggest that higher homocysteine levels may be a consequence rather than a cause of disease.
Anton J.M. de Craen, Ph.D.
Leiden University Medical Center
2300 RC Leiden, the Netherlands
craen@lumc.nl
David J. Stott, M.D., Ph.D.
Glasgow Royal Infirmary
Glasgow G31 2ER, United Kingdom
Rudi G.J. Westendorp, M.D., Ph.D.
Leiden University Medical Center
2300 RC Leiden, the Netherlands
References
The Heart Outcomes Prevention Evaluation (HOPE) 2 Investigators. Homocysteine lowering with folic acid and B vitamins in vascular disease. N Engl J Med 2006;354:1567-1577.
B?naa KH, Nj?lstad I, Ueland PM, et al. Homocysteine lowering and cardiovascular events after acute myocardial infarction. N Engl J Med 2006;354:1578-1588.
Stott DJ, MacIntosh G, Lowe GD, et al. Randomized controlled trial of homocysteine-lowering vitamin treatment in elderly patients with vascular disease. Am J Clin Nutr 2005;82:1320-1326.
Lewis SJ, Ebrahim S, Davey Smith G. Meta-analysis of MTHFR 677CT polymorphism and coronary heart disease: does totality of evidence support causal role for homocysteine and preventive potential of folate? BMJ 2005;331:1053-1056.
Mooijaart SP, Gussekloo J, Fr?lich M, et al. Homocysteine, vitamin B-12, and folic acid and the risk of cognitive decline in old age: the Leiden 85-Plus study. Am J Clin Nutr 2005;82:866-871.
To the Editor: The investigators in the HOPE-2 and NORVIT trials recommend against the use of folic acid, vitamin B6, and vitamin B12 as preventive treatment. However, mean levels of homocysteine, folic acid, vitamin B6, and vitamin B12 were in the normal range in both studies. If patients with hyperhomocysteinemia were included in these trials, would the results have been different? Furthermore, can vitamin supplementation impart benefits to patients with normal levels of these nutrients?
Asian Indians, who have a reduced intake of vitamin B12 and folate, are predominantly vegetarian and have higher homocysteine levels and lower levels of folate and vitamin B12 than do whites.1,2 A study assessing Asian Indians reported a mean plasma homocysteine level of 19.8 mmol per liter, with 77 percent of the subjects having hyperhomocysteinemia and more than 50 percent having a deficiency of vitamin B12.3 Hyperhomocysteinemia is an independent risk factor for coronary heart disease in India and may account for the fact that twice as many Asian Indians die from the disease as do Europeans.1 Furthermore, unlike Asian Indians, approximately 70 percent of the patients in the HOPE-2 study were exposed to folate-fortified food. Whether the study results would be applicable to Asian Indians cannot be answered until prospective, randomized trials are conducted in this population.
Amit Khare, Ph.D.
Meena Lopez, B.Pharm., M.B.A.
Jaideep Gogtay, M.D.
Cipla
Mumbai 400 008, India
amit.khare@cipla.com
All authors are employees of Cipla, an Indian pharmaceutical company that manufactures generic cardiovascular drugs.
References
Chambers JC, Obeid OA, Refsum H, et al. Plasma homocysteine concentrations and risk of coronary heart disease in UK Indian Asian and European men. Lancet 2000;355:523-527.
Misra A, Vikram NK, Pandey RM, et al. Hyperhomocysteinemia, and low intakes of folic acid and vitamin B12 in urban North India. Eur J Nutr 2002;41:68-77.
Refsum H, Yajnik CS, Gadkari M, et al. Hyperhomocysteinemia and elevated methylmalonic acid indicate a high prevalence of cobalamin deficiency in Asian Indians. Am J Clin Nutr 2001;74:233-241.
To the Editor: Despite the claim by Loscalzo in the editorial1 accompanying the reports on the HOPE-2 and NORVIT trials, the lack of benefit of lowering homocysteine concentrations with folic acid is not an "unequivocal conclusion." Relatively little is known about how homocysteine affects cardiovascular disease. Unless perturbed nutritionally or pharmacologically, homocysteine concentrations change relatively little over a five-year period2 and presumably over a longer term. The elevated homocysteine concentration found at diagnosis in patients with cardiovascular disease was probably a chronic condition. Consequently, the cardiovascular insult may have occurred over a period of many years. There is no indication as to how long it would take to reverse such damage. All three intervention studies cited by Loscalzo were of moderate duration (2, 3.5, and 5 years) and may not reflect the benefit of long-term intervention (e.g., prolonged supplementation or universal fortification, particularly for primary prevention). A case in point is the benefit of quitting smoking, since it takes more than five years after smoking cessation for the risks of cardiovascular disease,3 laryngeal cancer,4 and (in heavy smokers) lung cancer5 to diminish.
Eoin P. Quinlivan, Ph.D.
Jesse F. Gregory III, Ph.D.
University of Florida
Gainesville, FL 32611-0370
jfgy@ufl.edu
References
Loscalzo J. Homocysteine trials -- clear outcomes for complex reasons. N Engl J Med 2006;354:1629-1632.
El-Khairy L, Vollset SE, Refsum H, Ueland PM. Predictors of change in plasma total cysteine: longitudinal findings from the Hordaland Homocysteine Study. Clin Chem 2003;49:113-120.
Iso H, Date C, Yamamoto A, et al. Smoking cessation and mortality from cardiovascular disease among Japanese men and women: the JACC Study. Am J Epidemiol 2005;161:170-179.
Altieri A, Bosetti C, Talamini R, et al. Cessation of smoking and drinking and the risk of laryngeal cancer. Br J Cancer 2002;87:1227-1229.
Ebbert JO, Yang P, Vachon CM, et al. Lung cancer risk reduction after smoking cessation: observations from a prospective cohort of women. J Clin Oncol 2003;21:921-926.
To the Editor: The HOPE-2 investigators show a significant, 24 percent reduction in the relative risk of stroke among patients treated with folic acid and vitamins B6 and B12. They downplay this result by relegating a striking figure on the effect of this treatment on stroke (Figure 1) to their online Supplementary Appendix. The authors suggest that the result may be spurious, but it agrees closely with the predictions of two large meta-analyses, which suggested that the same change in homocysteine levels achieved in the HOPE-2 trial would result in a reduction in stroke of 19 to 24 percent.1,2 Their view — that a treatment benefit restricted to stroke is biologically implausible — is surprising, given the etiologic differences in coronary disease and stroke. They claim that the findings of the Vitamin Intervention for Stroke Prevention (VISP)3 and NORVIT studies support their conclusion. However, they did not refer to the reanalysis of the VISP trial, which revealed a significant effect on stroke and coronary events.4 They also did not mention that the NORVIT study was smaller, with a total of 98 strokes, as compared with 258 strokes in their own trial.
Figure 1. Kaplan–Meier Estimates of the Percentage of Patients with Stroke during a Trial of Homocysteine Lowering.
The relative risk of stroke among patients in the active-treatment group (who received folic acid and vitamins B6 and B12), as compared with patients in the placebo group, was 0.75 (95 percent confidence interval, 0.59 to 0.97; P=0.03 by the log-rank test). Data are from the HOPE-2 trial.
The message of the HOPE-2 trial should be one of cautious optimism that B vitamins may protect against stroke. Consistent with this view is a reduction in the rate of death from stroke in the United States and Canada after the introduction of folic acid fortification of food.5
Helga Refsum, M.D.
A. David Smith, D.Phil.
University of Oxford
Oxford OX1 3PT, United Kingdom
helga.refsum@physiol.ox.ac.uk
Dr. Refsum reports having received lecture fees from Diatomics, Nycomed, and Recip AB, as well as grant support from Axis-Shield and Nycomed; and Dr. Smith, lecture fees from Nycomed and Recip AB.
References
Wald DS, Law M, Morris JK. Homocysteine and cardiovascular disease: evidence on causality from a meta-analysis. BMJ 2002;325:1202-1206.
Homocysteine Studies Collaboration. Homocysteine and risk of ischemic heart disease and stroke: a meta-analysis. JAMA 2002;288:2015-2022.
Toole JF, Malinow MR, Chambless LE, et al. Lowering homocysteine in patients with ischemic stroke to prevent recurrent stroke, myocardial infarction, and death: the Vitamin Intervention for Stroke Prevention (VISP) randomized controlled trial. JAMA 2004;291:565-575.
Spence JD, Bang H, Chambless LE, Stampfer MJ. Vitamin Intervention for Stroke Prevention trial: an efficacy analysis. Stroke 2005;36:2404-2409.
Yang Q, Botto LD, Erickson JD, et al. Improvement in stroke mortality in Canada and the United States, 1990 to 2002. Circulation 2006;113:1335-1343.
To the Editor: As compared with the HOPE-2 and VISP trials, the NORVIT trial was unique in that it used a two-by-two factorial design, which allowed for an examination of the effect of each of the three treatments — folic acid plus vitamin B12, vitamin B6 alone, and folic acid plus both vitamin B6 and vitamin B12. However, the NORVIT trial did not report the analyses that we believe are important for a comprehensive assessment of the effects of B vitamins: a comparison of folic acid plus vitamin B12 with placebo, a comparison of vitamin B6 with placebo, and a test of the interaction between folic acid plus vitamin B6 and vitamin B12 in relation to the clinical outcome.
In Table 1, we present our analyses using data from the NORVIT trial. We calculated the rate ratios for each treatment using the number of observed cases and person-years and Poisson distribution and test-based methods to construct confidence intervals.1 We also calculated the rate ratio for the interaction between folic acid plus vitamin B6 and vitamin B12, and we estimated the standard error of the rate ratio by the multivariate delta method.2 We subsequently used this standard error to construct the confidence interval for the rate ratio of the interaction.
Table 1. Rate Ratios for the Three Treatment Groups and Tests of Interactions between Folic Acid plus Vitamin B6 and Vitamin B12 on Clinical Outcomes.
Our analyses showed that as compared with placebo, folic acid plus vitamin B12 has a slightly but not significantly beneficial effect on most of the clinical outcomes; the same is true for vitamin B6. We found a significant adverse interaction between folic acid plus vitamin B6 and vitamin B12 on all clinical outcomes except stroke, coronary-artery bypass surgery, and percutaneous coronary intervention. If our findings can be independently replicated, one would conclude that it was the interaction between folic acid plus vitamin B6 and vitamin B12 that led to significantly worse outcomes, whereas there was no evidence that treatment with folic acid plus vitamin B12 or with vitamin B6 alone was harmful.
The three trials raise further questions to be addressed in future research. First, all three trials used a high-dose formulation, including 2 to 6 times the recommended daily allowance (RDA) of folic acid, 166 to 416 times the RDA of vitamin B12, and 12 to 25 times the RDA of vitamin B6. The interactions among high-dose treatments may lead to undesirable clinical outcomes. Future studies will need to determine the optimal dose and combination that maximize efficacy and minimize adverse effects. Second, all three trials were conducted in patients with existing cardiovascular disease. Studies are needed to assess the role of B vitamins in the primary prevention of cardiovascular disease, especially in persons with a low intake of B vitamins or with a genetic susceptibility to hyperhomocysteinemia. Third, all three trials reported the averages of the effects in all the treatment groups. However, individual responses to the treatments may vary greatly, depending on the person's homocysteine metabolism and genetic susceptibility, as well as on the presence of other known risk factors for cardiovascular disease. Future studies will need to find a better way to identify patients who will benefit the most from interventions with B vitamins.
Xiaobin Wang, M.D., Sc.D.
Children's Memorial Hospital
Chicago, IL 60614
xbwang@childrensmemorial.org
Hakan Demirtas, Ph.D.
Xiping Xu, M.D., Ph.D.
University of Illinois at Chicago
Chicago, IL 60612
References
Sahai H, Khurshid A. Statistics in epidemiology: methods techniques and applications. Boca Raton, Fla.: CRC Press, 1996:14-8.
Rice JA. Mathematical statistics and data analysis. 2nd ed. Belmont, Calif.: Duxbury Press, 1995:149.
To the Editor: Though elevated levels of homocysteine are predictive of cardiovascular risk, the demonstration that homocysteine-lowering therapy was without benefit in the HOPE-2 and NORVIT trials provides further evidence that homocysteine represents an epiphenomenon in atherosclerosis. The pathogenesis of atherosclerosis is characterized by chronic inflammation,1 and elevated plasma homocysteine concentrations have been correlated with inflammation in conditions such as chronic renal failure, rheumatoid arthritis,2 and psoriasis and in the period after myocardial infarction.3 However, clinical studies have demonstrated that endothelial dysfunction does not improve despite effective lowering of homocysteine levels with 400 μg of folic acid orally per day.4 High-dose oral folic acid (5 to 20 mg per day) improves endothelial function in a manner largely independent of plasma homocysteine lowering, though the underlying mechanism has not been established.5 It is our view, therefore, that high-dose folate therapy (exceeding 5 mg daily) has pleiotropic effects, the benefits of which have yet to be tested in large-scale secondary-prevention trials in appropriate subjects.
David R. Tomlinson, M.R.C.P.
John Radcliffe Hospital
Oxford OX3 9DU, United Kingdom
david.tomlinson@orh.nhs.uk
Derek Lang, Ph.D.
Malcolm J. Lewis, M.B., D.Sc., F.R.C.P.
Wales Heart Research Institute
Cardiff CF14 4XN, United Kingdom
References
Ross R. The pathogenesis of atherosclerosis: a perspective for the 1990s. Nature 1993;362:801-809.
Roubenoff R, Dellaripa P, Nadeau MR, et al. Abnormal homocysteine metabolism in rheumatoid arthritis. Arthritis Rheum 1997;40:718-722.
Landgren F, Israelsson B, Lindgren A, Hultberg B, Andersson A, Brattstrom L. Plasma homocysteine in acute myocardial infarction: homocysteine-lowering effect of folic acid. J Intern Med 1995;237:381-388.
Pullin CH, Ashfield-Watt PA, Burr ML, et al. Optimization of dietary folate or low-dose folic acid supplements lower homocysteine but do not enhance endothelial function in healthy adults, irrespective of the methylenetetrahydrofolate reductase (C677T) genotype. J Am Coll Cardiol 2001;38:1799-1805.
Doshi SN, McDowell IF, Moat SJ, et al. Folic acid improves endothelial function in coronary artery disease via mechanisms largely independent of homocysteine lowering. Circulation 2002;105:22-26.
Dr. Lonn replies: De Craen et al. refer to a recent meta-analysis of mendelian randomization studies to suggest reverse causality as an explanation for the results of the HOPE-2 and NORVIT trials involving patients with coronary heart disease. Findings in patients with stroke may differ.1 Additional plausible explanations include residual confounding in epidemiologic studies and the possibility that any harmful effects of high-dose folate may offset the benefits of homocysteine lowering.
In response to the comments of Refsum and Smith, we do not dismiss the results of the HOPE-2 study in regard to stroke but caution against an overenthusiastic interpretation on the basis of multiple considerations: the overall neutral effect of treatment on the primary outcome and on most secondary and tertiary outcomes (including transient ischemic attacks), the wide confidence intervals around the estimated reduction in the risk of stroke, the apparent increase in the risk of unstable angina, and the neutral results regarding stroke in the VISP and NORVIT trials. The quoted reanalysis of the VISP trial was not prespecified and showed no reduction in stroke (the primary study outcome), death, or events associated with coronary heart disease. At best, the reanalysis showed a borderline effect on a composite cardiovascular outcome (unadjusted outcome, P=0.049; adjusted outcome, P=0.056), which is not a very convincing result to use to justify any clear treatment recommendations. A causal link between recent trends toward a lower rate of death from stroke in the United States and Canada and the fortification of food with folic acid remains speculative, since many other factors may have contributed to the decline.
With regard to the comments by Wang et al. and Tomlinson et al., the doses and combinations of B vitamins used in the large clinical trials are based on the ability of the drugs to reduce homocysteine levels in most people and on the perceived safety of the drugs.2 Whether lower or higher doses or different combinations may be useful remains unproven. A reasonable approach is to encourage people to have balanced diets, since such diets provide adequate amounts of needed macronutrients and micronutrients in most people, and to reserve the use of vitamin therapy for those with proven deficiencies.
Khare et al. and Wang et al. note that the clinical trials studied primarily white, middle-aged patients with vascular disease. Trials of primary prevention and in populations with higher homocysteine levels that are related to genetic and dietary factors are of interest, and we strongly support the completion of ongoing studies in various populations. However, we are unaware of any cardiovascular therapies that are exclusively effective for primary prevention, and in our trial, even patients in the upper fifth of the baseline homocysteine distribution (19.7 μmol per liter) derived no benefit.
Quinlivan and Gregory point out that the trials of B vitamin supplementation are of intermediate duration (two to five years). Most proven preventive therapies, such as the lowering of levels of cholesterol and blood pressure, reduce risk within months to a few years, and in several studies, such as the British Doctors Study, excess risk was halved within two to three years after the cessation of smoking.3
In summary, completed clinical trials do not provide evidence to support the preventive use of B vitamin supplements. Ongoing large trials and the planned meta-analysis of all trials4 will answer remaining relevant clinical questions.
Eva Lonn, M.D.
McMaster University
Hamilton, ON L8L 2X2, Canada
for the HOPE-2 Investigators
References
Casas JP, Bautista LE, Smeeth L, Sharma P, Hingorani AD. Homocysteine and stroke: evidence on a causal link from mendelian randomisation. Lancet 2005;365:224-232.
Homocysteine Lowering Trialists' Collaboration. Dose-dependent effects of folic acid on blood concentrations of homocysteine: a meta-analysis of the randomized trials. Am J Clin Nutr 2005;82:806-812.
Doll R, Peto R. Mortality in relation to smoking: 20 years' observation on male British doctors. Br Med J 1976;2:1525-1536.
B-Vitamin Treatment Trialists' Collaboration. Homocysteine-lowering trials for prevention of cardiovascular events: a review of the design and power of the large randomized trials. Am Heart J 2006;151:282-287.
Dr. B?naa and colleagues reply: The data from the HOPE-2, NORVIT, and VISP trials are quite consistent in showing that homocysteine lowering with folic acid and vitamin B12 (with or without vitamin B6) has no clinical benefit in patients with established vascular disease. The negative results may be interpreted in three ways.
First, homocysteine may not be a causative agent in vascular disease. High homocysteine levels may be an indicator of an unhealthy lifestyle, an epiphenomenon reflecting atherogenic processes, or a consequence of vascular disease itself, as suggested by de Craen et al. and Tomlinson et al.
Second, homocysteine-lowering therapy may still be beneficial in populations other than those studied — for example, in patients with hyperhomocysteinemia, as suggested by Khare et al. However, in the NORVIT trial, this therapy had no benefit in the 40 percent of patients with a baseline homocysteine level above 13 μmol per liter (in this subgroup, the mean homocysteine level was 17.4 μmol per liter). As pointed out by Quinlivan and Gregory and Wang et al., the trials may have been too short (mean duration, 2.5 to 5 years), and the results could possibly be different for primary prevention. However, most conventional treatments (including smoking cessation) show effects on vascular disease within five years. The results of the NORVIT trial do not preclude a protective effect of more physiologic doses of B vitamins or in primary prevention. However, it is difficult to explain biologically how a therapy that does not work in patients who have had a clinical vascular event would work well in those without a clinical event (many of whom have subclinical atherosclerosis).
Finally, as suggested by Loscalzo in his editorial, B vitamin therapy could have harmful effects that offset the homocysteine-lowering benefit. Wang et al. suggest that the trend toward a harmful effect in the combined B vitamin treatment group that was observed in the NORVIT trial was due to a significant interaction between folate plus vitamin B6 and vitamin B12. However, we believe the confidence intervals they present for the test of interaction are too narrow, and they cannot precisely estimate the rate ratios adjusted for study center from the numbers given in our article. Using rate ratios adjusted for study center estimated from Cox proportional-hazard regression and the method described by Altman and Bland,1 we obtained a test statistic (ratio of ratios) of 1.18 (95 percent confidence interval, 0.88 to 1.59) for the primary end point, indicating that there was no significant interaction.
Kaare Harald B?naa, M.D., Ph.D.
University of Troms?
N-9037 Troms?, Norway
kaare.harald.bonaa@unn.no
Aage Tverdal, Ph.D.
Norwegian Institute of Public Health
N-0403 Oslo, Norway
Per Magne Ueland, M.D., Ph.D.
University of Bergen
N-5021 Bergen, Norway
References
Altman DG, Bland JM. Interaction revisited: the difference between two estimates. BMJ 2003;326:219-219.