An Overview of the Homocysteine Lowering Clinical Trials

Abstract

Epidemiological studies have shown that cases with coronary, cerebral, or peripheral vascular disease have higher homocysteine levels than controls, but substantial uncertainty exists about the strength of any risk associations in age and sex-specific groups and in those with and without prior vascular disease. Folic acid and other B-vitamins are highly effective in delaying the vascular complications in patients with homocystinuria, who have markedly elevated levels. However, the effect on vascular risk of homocysteine reduction among those with lower levels is unknown, which has prompted calls for large-scale clinical trials to test this hypothesis. There was substantial uncertainty about the optimal regimen for lowering homocysteine levels. But many of these issues have recently been clarified by a meta-analysis of 12 randomized trials of vitamin supplements to lower homocysteine levels. The meta-analysis showed that the proportional and absolute reductions in homocysteine varied according to pre-treatment homocysteine levels, from 16% (95% CI: 11% to 20%) among those in the bottom fifth to 39% (95% CI: 36% to 43%) among those in the top fifth. After standardizing for a pretreatment homocysteine of 12 µmol/l and folate level of 12 nmol/l, there was no longer any heterogeneity in the proportional reductions in homocysteine achieved in the individual trials. Under these circumstances, the meta-analysis estimated that dietary folic acid reduced homocysteine levels by 25% (95% CI: 23 to 28%) with similar effects in a daily dosage range of 0.5 mg to 5 mg. Vitamin B12 (mean 0.5 mg) produced an additional reduction in blood homocysteine of 7%, whereas vitamin B6 had no additional effects on homocysteine levels. Vitamin B6 is effective at lowering homocysteine levels after methionine loading, which appears to have an independent effect on vascular risk. In view of the independent effects of folic acid and vitamin B6, clinical trials with factorial designs (where participants are randomly assigned to receive to receive folic acid or placebo, and individuals in each group are subsequently randomly assigned to receive either vitamin B6 or placebo) would allow assessment of the separate and combined effects of both vitamins without materially increasing the number of patients. Current and planned large-scale clinical trials to assess the effects of vitamin supplements on risk of vascular disease should provide reliable evidence for this hypothesis in almost 40000 patients with prior heart disease and several thousand patients with a prior stroke or TIA. Collaborative analysis of the post-publication follow-up of individual participants in the separate trials should provide reliable evidence of the importance of folic acid and vitamin B6 in age-specific groups, and different disease categories, and across a wide range of pre-treatment homocysteine levels. The results of these trials are required before advocating widespread screening for elevated homocysteine levels or advocating the use of vitamin supplements in high-risk individuals or changing population mean levels of folate (by fortification of flour) for the prevention of cardiovascular diseases.