Homocysteine and Cardiovascular Risk: Considering the Evidence in the Context of Study Design, Folate Fortification, and Statistical Power

Abstract

The hypothesis that moderately increased plasma total homocysteine (tHcy) concentrations are causally related to cardiovascular disease (CVD) originated from observations of vascular disease in patients with homocystinuria (1). tHcy concentrations are ∼10-fold higher in patients with untreated homocystinuria than in the general population, and these patients often suffer from CVD in early life. Homocystinuria may arise from one of several rare defects in genes involved in methionine metabolism, resulting in high tHcy concentrations, with cystathionine β-synthase ( CBS gene) deficiency being the most common. In responsive cases of homocystinuria, dietary supplementation with B-vitamins and betaine is remarkably effective at lowering plasma tHcy concentrations and decreasing the risk of CVD (2). In addition to suggesting that extremely high tHcy concentrations may be causally related to CVD in affected individuals with homocystinuria, McCully also suggested that moderately increased tHcy concentrations may be related to CVD risk in the general population (1). A single discrete mechanism of vascular injury has not been identified, but high homocysteine may have adverse effects on platelet function and clotting factors and may increase vascular smooth muscle cell proliferation. Furthermore, increased homocysteine concentrations provoke endothelial dysfunction, possibly mediated by oxidative stress or interference with nitric oxide function (3)(4). Over the last 3 decades, many observational epidemiological studies have reported associations between increased tHcy concentrations and risk of coronary heart disease (CHD) and stroke (5). Although results of prospective cohort studies (in which blood for tHcy determination was collected before the onset of disease) have been weaker and more inconsistent than those of retrospective studies (in which the blood was collected after the onset of disease), prospective studies are more reliable because they are not vulnerable to bias due to the effect of disease on tHcy (“reverse causality”) and because they control for confounding from established …