Homocysteine and Cardiovascular Disease: Should We Treat?

Homocysteine was first discovered over seventy years ago and has attracted the interest of researchers ever since.¹ The small, sulfur-containing, non-protein-forming amino acid is a key intermediate in the methylation cycle and is produced by all cells during the metabolism of methionine. Excess intracellular homocysteine is exported to the circulation where approximately 1% remains in a free form and the remainder binds to albumin or forms disulfide dimers, principally with cysteine.² Total plasma homocysteine (tHcy) is measured as the combined pool of all these forms and has been linked with several diseases. However, it is the relationship between tHcy and cardiovascular disease that has received most attention in the last few decades.

McCully first reported widespread arteriosclerotic lesions in homocysteinuric children at autopsy almost forty years ago.³ Mudd et al. subsequently confirmed that individuals with cystathionine β-synthase-deficiency, a genetic defect causing severely elevated tHcy, suffered very high rates of premature vascular disease, mainly thromboembolism and stroke.⁴ These observations prompted development of a “homocysteine hypothesis”: that elevated tHcy contributes to cardiovascular risk in the general population. Epidemiological data establishing an association between elevated tHcy and an increased risk of stroke and myocardial infarction (MI) lent weight to the theory and many plausible mechanisms by which homocysteine might damage the arterial wall or increase the risk of thrombosis have been proposed, although few have been proven unequivocally.^5–10

A combination of folic acid and vitamin B₁₂ reduces tHcy by approximately 25%, even in individuals who are not overtly vitamin-deficient.¹¹ If elevated tHcy contributes to vascular damage, B-vitamin supplementation might prove a cheap and simple risk-reduction strategy. Consequently, over the last decade several large randomised placebo-controlled trials have tested whether lowering homocysteine reduces cardiovascular risk.^12–18 All trials have reported significant reductions in tHcy in vitamin-treated subjects compared with placebo.^{12,14,16,19,20} However, completed trials have largely failed to show that vitamin therapy reduces cardiovascular events or mortality.

Bazzano et al. published the first meta-analysis of trial data in late 2006.²¹ The authors analysed data from twelve studies with a combined total of almost 17,000 participants. They found no significant improvement in any of the vascular endpoints or in overall mortality in vitamin-treated subjects compared with placebo. The pooled estimates of relative risk for B-vitamin treated patients were 1.04 (95% CI 0.92, 1.17) for coronary heart disease, 0.86 (95%CI 0.71, 1.04) for stroke and 0.96 (95%CI 0.88, 1.04) for all-cause mortality. While the confidence intervals were consistent with a moderate reduction in the risk of vascular events, there was no unambiguous evidence of benefit.

A more recent meta-analysis included additional data from a large Chinese study and found that folic acid supplementation reduced the risk of stroke by 18% (RR 0.82, 95% CI 0.68, 1.00).²² Wang et al. also reported that subjects who were treated for more than 36 months or had a greater than 20% reduction in tHcy or had no prior history of stroke experienced more significant risk reduction. This paper, in combination with data showing an improvement in stroke mortality after the introduction of mandatory folic acid fortification in the United States and Canada, has generated speculation that homocysteine-lowering may prove more effective in preventing stroke than other forms of vascular disease.^23–26

Three additional trials have reported their findings recently, the Women’s Antioxidant and Folic Acid Cardiovascular Study (WAFACS, n=5442), Western Norway B-vitamin Intervention Trial (WENBIT, n=3088) and Homocysteinemia in Kidney and End Stage Renal Disease trial (HOST, n=2056). WAFACS found no difference in the composite endpoint of cardiovascular death, MI, stroke or revascularisation between treatment and placebo groups (RR 1.03, 95% CI 0.9, 1.19) or in any of the individual components of the composite.²⁷ WENBIT presented initial results on 4 September 2007 at the European Society of Cardiology Congress in Vienna. The investigators found no difference between folic acid and non-folic acid groups in a composite endpoint of death, non-fatal MI, stroke or hospitalisation with unstable angina (RR 1.09, 95% CI 0.91, 1.30), but have yet to publish detailed data. The HOST trial also reported that B-vitamin treatment in chronic kidney disease did not significantly reduce the risk of all cause mortality, MI or stroke.¹⁹ The editorial accompanying the HOST data updated the Bazzano meta-analysis, reporting an overall odds ratio of 1.00 (95% CI 0.92, 1.09) for coronary heart disease and 0.88 (95% CI 0.78, 1.00) for stroke per 3.1μmol/L reduction in tHcy.²⁸

These recent results, added to earlier data, suggest that moderately elevated tHcy is a risk marker for vascular disease rather than a causal risk factor. B-vitamin treatment modifies the marker without reducing the underlying risk. However, the homocysteine hypothesis is not yet disproved. Most large trials have enrolled subjects with existing vascular disease, with end-stage renal failure or with multiple risk factors for atherosclerosis. B-vitamins may prove more effective if treatment is begun prior to the development of clinically evident disease, although this has yet to be demonstrated.

It is also possible that long-term B-vitamin therapy has negative effects that counteract any benefit from reducing tHcy. The Heart Outcomes Prevention Evaluation (HOPE)-2 trial, in 5522 subjects with a history of vascular disease or diabetes, found that the vitamin-treated subjects were more likely to be hospitalised with unstable angina than those on placebo (RR 1.24, 95% CI 1.04, 1.49). This finding might be attributed to chance were it not for previous evidence that B-vitamin therapy increases the risk of in-stent re-stenosis and the need for revascularisation following angioplasty.^12,29 In addition, initial results from the Norwegian Vitamin Trial (NORVIT) presented at the 2005 European Society of Cardiology Congress showed a trend towards an increased cancer rate in folic acid versus non-folic acid groups (RR 1.4, 95% CI 1.0 to 2.0), although this finding was not replicated in data published subsequently.³⁰ HOPE-2, the only other large trial to publish cancer outcomes, reported a non-significant increase in the relative risk of incident cancer of 1.06 (95% CI 0.91, 1.23).¹² There are plausible biological mechanisms by which folic acid supplementation might increase cancer risk.³¹ Actively dividing cells require folate to synthesise DNA and animal studies show that excess folic acid promotes growth of in-situ tumours.^32–34 An abrupt increase in colorectal cancer rates followed the introduction of mandatory food fortification with folic acid in both the USA and Canada, and recent epidemiological data suggests that individuals who take B-vitamin supplements may be at increased risk of breast, prostate and colorectal tumours.^35–39 This new evidence may prompt homocysteine-lowering trialists to re-analyse their data for cancer outcomes.

Several large homocysteine-lowering trials are still ongoing and will provide additional data on the safety and efficacy of long-term B-vitamin supplementation.^15,17,40,41 A meta-analysis including data from more than fifty thousand trial participants is planned and will have adequate power to clearly determine whether lowering tHcy reduces vascular event rates.⁴² Currently, however, there is no evidence to support B-vitamin treatment in patients at risk of cardiovascular disease.