Homocysteine-Lowering B Vitamin Therapy in Cardiovascular Prevention—Wrong Again?

In 1969, McCully first proposed that homocysteine, an amino acid produced during catabolism of methionine, causes arterial and venous atherothrombotic disease.¹ This theory was based on observations that children with extreme elevations of plasma homocysteine concentrations due to inborn errors of metabolism also had premature atherothrombotic disease. While the genetic disorders associated with homocystinuria are rare, these conditions provide an in vivo human model for vascular injury associated with exposure to high homocysteine concentrations. Subsequent in vitro and in vivo studies confirmed that in experimental settings, homocysteine can cause atherothrombosis by promoting oxidative stress, endothelial cell damage, endothelial dysfunction, inflammation, thrombosis, and cell proliferation.²

Epidemiological studies have in general demonstrated associations between elevated homocysteine levels and increased risk of coronary heart disease (CHD) and even stronger associations with increased risk of stroke. A meta-analysis of early cross-sectional and retrospective case-control studies suggested that plasma homocysteine levels of 3 to 4 μmol/L may increase the risk of cardiovascular disease (CVD) by about one-third.³ However, such retrospective observational studies are inevitably subject to bias and confounding and may overestimate the CVD risk associated with high homocysteine levels. Subsequent large prospective cohort studies in general supported the earlier findings, although the magnitude of the observed associations was significantly lower, especially for CHD. A meta-analysis of prospective cohort studies demonstrated that after accounting for known CVD risk factors, a 25% lower homocysteine level was associated with an 11% lower risk of CHD and a 19% lower risk of stroke.⁴ In addition, epidemiological studies suggested a graded and independent association between homocysteine levels and CVD risk extending to mild and even normal homocysteine levels.⁵ This is important because mild hyperhomocysteinemia is common and frequently caused by dietary folate and other B vitamin deficiencies.

Further supportive evidence was provided by mendelian randomization studies. Approximately 10% of the population has a TT genotype of methylene-tetrahydrofolate reductase (MTHFR), which leads to 25% higher homocysteine concentrations relative to the common CC genotype, and also leads to increased CVD risk. Such studies are not subject to reverse causality, are largely free from confounding by other risk factors, and provide some support for a causal link between elevated homocysteine concentrations and CVD risk.^{6 - 7}

Given this evidence, it is attractive to consider therapeutically altering homocysteine levels. In particular, homocysteine levels can be easily lowered by folic acid and vitamin B₁₂, which raises the prospect that simple and inexpensive dietary supplementation with B vitamins could lower CVD risk.⁸ The first indication of a possible therapeutic benefit of homocysteine lowering was provided by observations in children with homocystinuria, who generally die of CVD complications at a young age, unless treated with high doses of B vitamin supplements, sometimes with the addition of betaine and choline, and administration of a low-methionine diet. In addition, folic acid was also shown to improve vascular health by mechanisms independent of homocysteine lowering.⁹

While the experimental and epidemiological evidence does indeed support a plausible role for homocysteine lowering in CVD prevention in the population at large (as opposed to a limited role in rare genetic disorders), overzealous interpretations of such data have led to extrapolations and unjustified early enthusiasm. For example, in a document from the American Heart Association from 2005, it is stated that “The lowering of the population mean level of total homocysteine is estimated to have prevented 17 000 deaths from coronary causes each year,” and some investigators recommended the inclusion of folic acid to a polypill aimed at widespread use to prevent CHD and stroke.¹⁰

Fortunately, scientific rigor and sound judgment have prevailed and in the mid 1990s, several large trials of B vitamin supplementation in CVD prevention were initiated. These trials have collectively randomized tens of thousands of participants, most with preexistent vascular or renal disease. The trials completed to date do not provide clear evidence of any beneficial effects of B vitamin supplementation in CVD risk reduction. Thus, trials conducted in North America including the Vitamins Intervention for Stroke Prevention (VISP) trial and the Heart Outcomes Prevention Evaluation (HOPE-2) trial failed to show overall benefits.^{11 - 12} A notable exception is the 25% reduction in the risk of stroke observed in HOPE-2,¹² a finding not seen, however, in VISP,¹¹ a trial designed specifically to evaluate effects of homocysteine lowering on stroke. European studies in CHD populations, including the Cambridge Heart Antioxidant Study (CHAOS-2), Norwegian Vitamin Trial (NORVIT), and more recently the Western Norway B-Vitamin Intervention Trial (WENBIT), also failed to demonstrate benefits.¹³

The Women's Antioxidant and Folic Acid Cardiovascular Study (WAFACS) reported in this issue of JAMA by Albert and colleagues¹⁴ also found no benefit for B vitamin supplements on CHD or stroke in 5442 women, and makes an important additional contribution. The trial has unique strengths. It is focused on women, who are underrepresented in the other homocysteine-lowering trials; follow-up extended for 7.3 years, substantially longer than in previous trials; and approximately one-third of the women studied had no prior vascular events so that some data (although limited) emerge for high-risk primary prevention settings.

Several limitations should be mentioned. The trial was conducted after the introduction of policies that mandate the addition of folic acid to white flour, cereal grains, and related products in the United States, and which resulted in lower homocysteine concentrations among US women. Therefore, administration of B vitamin supplements lowered homocysteine concentrations in the trial to a lesser extent than anticipated at the time of its design and may have affected its ability to adequately test the study hypothesis. Moreover, the study was powered to detect a 20% reduction in CVD events. The epidemiological evidence that emerged after the study was under way suggests that these projections were less than realistic. It is likely that the trial was underpowered to detect a modest treatment benefit (< 10% proportional reduction in major vascular events). Homocysteine levels were measured in only 5% of study participants so that detailed analyses of potential benefits in subsets with high pretreatment homocysteine levels could not be performed. Finally, the trial evaluated a highly selective population of health professionals with relatively low CVD event rates.

Two obvious questions arise: why did the B vitamin homocysteine-lowering trials conducted to date not demonstrate clinical benefits and is there a role for additional trials or should researchers close yet another chapter, which seemed promising but failed to deliver. The answers to these questions are complex. Possible reasons for the failure of the initial trials of homocysteine lowering to demonstrate clinical benefits include (1) overestimation of a possible treatment effect based on earlier epidemiological studies resulting in underpowered trials; (2) the introduction of mandatory folate-food–fortification policies in the United States and Canada resulting in lesser effects of B vitamin supplements on homocysteine levels; (3) the vitamin doses used; and (4) potential unexpected proatherosclerotic effects of folic acid supplementation, which may have counteracted benefits associated with homocysteine lowering.¹⁵ However, it is possible that the treatment truly has no effect on vascular risk.

A number of large trials are still ongoing, including studies in populations with unfortified food supplies in Western Europe, Australia, and Asia. It is critically important to complete these studies and the preplanned meta-analysis of all large trials to provide clear answers to some of the remaining questions.¹⁶ However, until further data become available it is essential to remain firmly grounded on the available evidence and to admit that once again experimental and observational data do not always translate into therapeutic benefits. As shown with antioxidant vitamins, postmenopausal hormone therapy, and more recently, HDL-raising agents, success of cardiovascular therapies cannot always be predicted based on experimental and observational data. The need for careful and thorough evaluation of each proposed intervention in adequately powered end point trials remains essential.

In conclusion, B vitamin supplements cannot currently be recommended for the prevention of CVD events (with the exception of rare genetic disorders) and there is no role for routine screening for elevated homocysteine levels. However, ongoing clinical research should provide further evidence on whether there may be any role for homocysteine-lowering B vitamin supplements in CVD prevention and for the overall importance of homocysteine as a CVD risk factor.