Homocyst(e)ine and Cardiovascular Disease: A Critical Review of the Epidemiologic Evidence

Cardiovascular disease remains the major cause of morbidity and death in developed countries and accounts for approximately 40% of all deaths in Canada (1). Smoking cessation and reductions in cholesterol levels and blood pressure have been shown to be effective strategies in the prevention of cardiovascular disease (2). However, these major, classic cardiovascular risk factors and such nonmodifiable risk factors as age, sex, and family history cannot fully explain why some persons develop myocardial infarction, stroke, and other cardiovascular disease but other persons do not (3-5). Other factors may also increase the likelihood of developing cardiovascular disease and contribute to atherogenesis. Pathologic and epidemiologic studies suggest that only about one half to two thirds of the variation in anatomic extent of atherosclerosis and risk for atherosclerotic vascular disease can be explained by classic risk factors (6-9). Therefore, many emerging risk factors have been investigated. Among these, elevated plasma or serum levels of homocyst(e)ine (hyperhomocyst[e]inemia) are of particular interest. Recent epidemiologic studies have shown that moderately elevated plasma homocysteine levels are highly prevalent in the general population and are associated with an increased risk for fatal and nonfatal cardiovascular disease, independent of classic cardiovascular risk factors. This association is usually consistent, strong, dose-related, and biologically plausible. Although simple, inexpensive, nontoxic therapy with folate and vitamins B6 and B12 is highly effective at reducing plasma homocyst(e)ine levels, it remains to be demonstrated that decreasing homocyst(e)ine levels reduces cardiovascular morbidity and mortality. Methods Data Sources We searched the scientific literature for all epidemiologic studies (prospective, casecontrol, cross-sectional, or geographic correlation) on cardiovascular disease (using the terms coronary heart disease, cerebrovascular disease, peripheral vascular disease, and atherosclerosis) and homocysteine or vitamins (using the terms homocysteine, B 12, B 6, and folic acid). We searched the MEDLINE database for articles published from January 1965 to January 1999 and identified additional studies by examining bibliographies of original articles, review articles, and textbooks. Study Selection We used standard definitions to define epidemiologic studies (10) and did not consider case series. The epidemiologic prospective cohort studies varied greatly in terms of patient selection, number of patients and controls, circumstances and techniques of measuring plasma homocyst(e)ine levels, definitions of elevated plasma homocyst(e)ine level, types and definitions of vascular outcome events and surrogate outcome measures, and statistical analyses. A formal meta-analysis of these studies was not performed because the results could have been misleading. We included all prospective epidemiologic studies (up to January 1999) but included only the largest retrospective studies (those that involved at least 150 cases) because prospective studies generally provide a robust estimate of association, whereas small retrospective studies are often subject to various biases. We further restricted the retrospective studies to those published after the meta-analysis by Boushey and colleagues (11) because that article provides a good critical review of the studies done up until 1995. Homocysteine Metabolism Homocysteine is a sulfur-containing amino acid produced during catabolism of the essential amino acid methionine. Homocysteine can be metabolized by two major pathways. When methionine is in excess, homocysteine is directed to the transsulphuration pathway, where it is irreversibly sulfoconjugated to serine by cystathionine -synthase in a process requiring vitamin B6 as a cofactor. However, under conditions of negative methionine balance, homocysteine is primarily metabolized through a methionine-conserving remethylation pathway. In most tissues, homocysteine is remethylated in a process that requires methionine synthase, vitamin B12 as a cofactor, and methyltetrahydrofolate as a cosubstrate. This pathway requires an adequate supply of folic acid and the enzyme methylene tetrahydrofolate reductase (MTHFR) (12). Genetic and acquired abnormalities in the function of these enzymes or deficiencies in folic acid, vitamin B6, or vitamin B12 cofactors can lead to elevated homocysteine levels. In the plasma, approximately 70% of homocysteine circulates in a protein-bound form; approximately 25% combines with itself to form the dimer homocystine; and the remainder (<5%) combines with other thiols, including cysteine, to form disulphide (a homocysteine-cysteine mix) or circulates as the free thiol compound (13). In North America, the term homocyst(e)ine is often used to refer to the total pool of circulating plasma homocysteine, whereas the term tHcy is more common in Europe. Homocyst(e)ine Theory of Atherosclerosis Severe hyperhomocyst(e)inemia associated with homocystinuria can be caused by several rare inherited disorders, including homozygous deficiency of cystathionine -synthase, MTHFR, or methionine synthase or defects in vitamin B12 metabolism (12, 14). These distinct genetic conditions share the following features: extreme elevations of plasma homocyst(e)ine levels and premature atherothrombotic disease with typical histopathologic features of endothelial injury, proliferation of vascular smooth-muscle cells, progressive arterial stenosis, and hemostatic changes suggestive of a prothrombotic state (15). The characteristic clinical and pathologic features of these genetically diverse conditions support the hypothesis that elevated plasma homocyst(e)ine levels are responsible for the vascular damage and led McCully and Wilson (16) to propose the homocyst(e)ine theory of atherosclerosis. Although these genetic errors of metabolism are extremely rare (homozygous cystathionine -synthase deficiency occurs in approximately 1 in 150 000 live births and is associated with plasma homocyst[e]ine levels as high as 400 mol/L), they provide a useful in vivo human model for vascular injury associated with high homocyst(e)ine levels . Laboratory Measurement of Homocyst(e)ine Levels Most assays for measuring homocysteine concentrations are based on chromatographic techniques; high-performance liquid chromatography is still the most widely used (13). However, a simple and relatively inexpensive immunoassay has become commercially available and may soon enable widespread measurement of plasma homocyst(e)ine levels in the clinical laboratory (17). Plasma homocyst(e)ine levels are usually measured in the fasting state and can be measured before or after methionine loading. Methionine loading, a method of stressing the homocyst(e)ine metabolic pathways, may be more sensitive than measurement of fasting homocyst(e)ine levels for detecting mild disturbances in the transsulfuration pathway that may be caused by vitamin B6 deficiency or partial cystathionine -synthase deficiency (18, 19). The procedure involves measuring the baseline fasting plasma homocyst(e)ine level, administering a standard oral dose of methionine, and measuring the plasma homocyst(e)ine level again 4 to 6 hours later. Methionine loading may help to discriminate between defects involving the transsulfuration and remethylation pathways (20); it may also help to identify patients who have impaired homocysteine metabolism despite a normal fasting total plasma homocysteine level and who may, therefore, be at increased risk for vascular disease (19). Reliable measurements of plasma homocyst(e)ine levels require the use of accurate assays as well as optimal procedures for collection and handling of blood samples (13). Patient characteristics (such as fasting state and posture) and recent vascular events may also affect measured total homocysteine levels (20-22). Definition and Prevalence of Hyperhomocyst(e)inemia Hyperhomocyst(e)inemia is usually defined by using arbitrary cut-off pointsfor example, above the 95th percentile or more than two SDs above the mean of values obtained from fasting, healthy controls. This is similar to the way in which high plasma cholesterol levels were originally defined. Normal plasma homocyst(e)ine levels usually range from 5 to 15 mol/L (17). However, the definition of elevated homocyst(e)ine levels is not standardized, and substantial differences exist in the normal reference levels used in the literature. Higher fasting values are arbitrarily classified as mild and moderate hyperhomocyst(e)inemia (16 to 100 mol/L) and severe hyperhomocyst(e)inemia (>100 mol/L). The prevalence of hyperhomocyst(e)inemia depends on the way in which the condition is defined and measured. When the common definition of hyperhomocysteinemialevels of total homocysteine exceeding the 95th percentile of the distribution in a healthy sample of controlsis used, 5% of the normal population will necessarily be defined as having an elevated homocyst(e)ine level (23). Between 13% and 47% of patients with symptomatic atherosclerotic vascular disease have been reported to have hyperhomocystein(e)mia (24). However, little evidence suggests a sudden increase in risk for vascular disease above a certain threshold level of plasma homocyst(e)ine; the relation between plasma homocyst(e)ine levels and risk for cardiovascular disease seems to be graded and linear (25). Causes of Mild and Moderate Hyperhomocyst(e)inemia One or a combination of genetic, physiologic, pathologic, and nutritional factors (Table 1) causes modest elevations in homocyst(e)ine levels without associated homocystinuria. Therefore, MTHFR mutations (for example, thermolabile MTHFR); older age; male sex; postmenopausal status; smoking; sedentary lifestyle; dietary factors, including increased intake of animal proteins (which have a higher methionine content); low intake of folic acid, vitamin B6, and vit