Homocysteine is a key substance in the methionine cycle, which is involved in one-carbon methyl group-transfer metabolism. The methylenetetrahydrofolate reductase (
MTHFR) C677T (rs1801133) polymorphism is a well-characterized genetic variant. C677T of the
MTHFR gene results in amino-acid substitution (Ala222Val), and causes a reduction of enzyme activity and higher plasma total homocysteine levels [
1]. The association of this variant with the plasma total homocysteine was confirmed by a recent meta-analysis of genome-wide association studies [
2]. Previous meta-analyses of association studies between total homocysteine, which includes plasma and serum total homocysteine, and schizophrenia suggest that an elevated total homocysteine level is a risk factor for schizophrenia [
3,
4]. However, observational studies have potential limitations, such as confounding, reverse causation, and selection bias. In fact, several findings of observational studies have been shown to be spurious causes by subsequent randomized controlled trials, such as hormone replacement therapy in coronary heart disease, β carotene in lung cancer, and vitamin E and vitamin C in cardiovascular disease [
5]. In addition to genetic variants [
2], many determinants that affect plasma total homocysteine concentrations, including physiologic determinants, such as age and sex, lifestyle determinants, such as vitamin intake, smoking and coffee, and clinical conditions, such as folate deficiency and renal failure, have been reported [
6]. Moreover, whether hyper-homocysteine itself causes schizophrenia or schizophrenia causes hyper-homocysteine has not been clarfied. For examples, the high prevalence of smoking and decreased folate levels in patients with schizophrenia have been reported [
7‐
10], both of which are known to be associated with increased plasma total homocysteine concentrations.
Mendelian randomization refers to the random allocation of alleles at the time of gamete formation. A specific genotype carried by a person results from two such randomized transmission, one from the paternally inherited allele and the other from the maternally allele. As a consequence of these randomizations, genotypes are not expected to be associated with known or unknown confounders for any outcome of interest, except those lying on the causal pathway between the genotype and the outcome. This allows analyzing the genotype-risks factor association (in this case, the genotype-the plasma total homocysteine) and the genotype-outcome association (in this case, the genotype-schizophrenia) in an unconfounded manner [
13]. By combining the results of the genotype-risk factor association and the genotype-outcome association, one can get an estimate of the risk factor-outcome association (in this case, the plasma total homocysteine-schizophrenia). In addition, genetic variants are equivalent to lifetime differences in risk factor (in this case, the plasma total homocysteine), and indicate the long term effects of risk factor on disease (in this case, schizophrenia). They therefore generate more realistic estimates of causal effects between risk factor and disease [
5]. The Mendelian randomization approach has similar properties to the analysis of the intention to treat in randomized controlled trials [
5,
17], and this approach has provided new insights into the pathology of several diseases, such as cardiovascular disease, diabetes, and Parkinson disease [
18‐
26].
In this study, we evaluated a causal relationship between plasma total homocysteine and schizophrenia by conducting a Mendelian randomization analysis based on the MTHFR C677T polymorphism as an instrumental variable.