Editorial
Alcohol Consumption and Cerebral Small- and Large-Vessel Diseases: A Mendelian Randomization Analysis
2024 年 31 巻 2 号 p. 119-121
詳細
2024 年 31 巻 2 号 p. 119-121
See article vol. 31: 135-147
Habitual alcohol drinking affects cardiovascular health. However, there are a variety of confounding factors, e.g., age, gender, lifestyle and socioeconomic status, for the relationship between alcohol intake and cardiovascular events. In addition, former drinkers are sometimes included in the abstainer group in analysis due to individual health issues. Genotypes of alcohol-metabolizing enzymes are robustly associated with individual alcohol consumption; thus Mendelian randomization analysis using single-nucleotide polymorphisms (SNPs) of genes of alcohol-metabolizing enzymes is a useful method for investigating causal effects of alcohol on health problems. The SNP rs671 in the aldehyde dehydrogenase 2 gene (ALDH2) is a frequent SNP, especially found in East Asian populations1). The SNP rs1229984 in alcohol dehydrogenase (ADH1B) is also known to affect the behavior of alcohol intake, although the frequency of inactive type of ADH1B- rs1229984 is much lower than that of ALDH2-rs671 in Asians2). In the study by Hisamatsu et al.3), the relation between alcohol intake and cerebral small- and large-vessel diseases in Japanese population was investigated by using Mendelian randomization analysis regarding ALDH2-rs671 (inactive A allele [GA and AA, 45.9%] vs. non-A allele [GG, 54.1%]). The assessment of cerebral vessel diseases was carefully performed by image analysis using magnetic resonance imaging to diagnose a large-vessel disease (intracranial artery stenosis) and small-vessel diseases (lacunar infarcts, white matter hyperintensities and cerebral microbleeds).
In inactive A carriers, alcohol consumption is greatly restricted because of low ALDH2 activity and consequent greater accumulation of acetaldehyde, which causes uncomfortable symptoms such as palpitation and flushing through the vasodilatory action of acetaldehyde. As expected, the amount of alcohol consumption and percentage of drinkers were much higher in non-A allele carriers compared to A allele carriers. Hypertension-related variables including systolic and diastolic blood pressure and frequencies of hypertension and anti-hypertensive medication were higher in non-A allele carriers than in A allele carriers, while LDL cholesterol levels were lower in the former than in the latter. The major findings of this interesting study are vessel size-dependent associations of ALDH2-rs671 genotype with cerebral vessel diseases demonstrated by age-adjusted logistic regression and Poisson regression: non-A allele carriers of ALDH2-rs671 had a higher burden of small-vessel diseases and lower burden of a large-vessel disease than did inactive A allele carriers. Additionally, dose-dependent relationships between alcohol consumption and cerebral vessel diseases were found in logistic regression. The above associations were attenuated in multivariable analyses with adjustment for other confounding factors including smoking, body mass index, hypertension, LDL cholesterol, HDL cholesterol, statins, diabetes mellitus, C-reactive protein and amount of alcohol intake. There were no significant relationships between ALDH2-rs671 genotype and cerebral vessel diseases when the amount of alcohol intake was adjusted in addition to adjustment for other possible confounding factors in multivariable analysis. This finding supports the notion that the associations between ALDH2-rs671 genotype and cerebral vessel diseases are mediated by alcohol intake.
Habitual alcohol drinking has diverse effects on cardiovascular health: Its harmful effect is mainly due to alcohol-induced hypertension4), while its beneficial effect is explained by modification of blood lipid profile including increase in HDL cholesterol and decrease in LDL cholesterol and suppression of blood coagulation5). Consequently, as shown in previous epidemiological studies, the risks of coronary heart disease and ischemic types of stroke were lower and the risks of hemorrhagic types of stroke (cerebral hemorrhage and subarachnoid hemorrhage) were higher in drinkers than in nondrinkers6). In the study by Hisamatsu et al., it was excellent that the main confounding factors for the associations between ALDH2-rs671 genotype and different cerebral vessel diseases were determined by using each factor one by one, in addition to age, as an explanatory variable for adjustment in multivariable analyses. Namely, strong attenuation of an association by one confounding factor indicates its main involvement in the association. In fact, adjustment for hypertension and LDL cholesterol as explanatory variables in multivariable analyses most strongly attenuated the associations of the SNP of ALDH2-rs671 with small-vessel diseases and a large-vessel disease, respectively. These findings led the authors to a speculation that alcohol-induced hypertension and decrease in LDL cholesterol level are major intermediaries in the causal pathway from alcohol consumption to cerebral small-vessel diseases and a large-vessel disease, respectively. These explanations are reasonable since the arteriosclerotic process and the atherosclerotic process (accumulation of lipid-rich plaques) are mainly involved in the pathogenesis of small and large-vessel diseases, respectively, (Fig.1).
ALDH2, acetaldehyde dehydrogenase 2.
HDL cholesterol level is known as an indicator of alcohol consumption7). Unexpectedly, no association was found between the SNP of ALDH2-rs671 and HDL cholesterol levels despite the fact that there was a large difference in alcohol intake between the A allele and non-A allele carrier groups. However, general effects of alcohol were confirmed by the findings of higher levels of liver enzymes including γ-glutamyl transferase, which is also a sensitive indicator of habitual alcohol intake8), in non-A allele carriers than in A allele carriers. A limitation of this study is indirect measurement using the Friedewald formula of LDL cholesterol concentrations9) and according exclusion of subjects with hypertriglyceridemia of 400 mg/dl or higher (n=16, 2.3%). Thus, LDL cholesterol concentrations depended on total cholesterol, triglycerides and HDL cholesterol concentrations, which were not significantly different between non-A allele carriers and A allele carriers. Another limitation of this study is no inclusion of variables related to blood coagulation in the analyses. Since antithrombotic effects of alcohol are involved in the inverse associations between alcohol intake and ischemic arterial diseases10) and antiplatelet drugs are used for primary and secondary prevention therapies for stroke11), it would be interesting to know whether ALDH2-rs671 genotype is associated with blood coagulation markers and whether they confound the associations between the genotype and cerebral artery diseases. Proximal and distal portions of the coronary circulation are more prone to be affected by blood lipids and diabetes, respectively, as risk factors of ischemic heart diseases. Therefore, it would also be interesting to investigate in future studies whether and how the degrees of atherosclerosis in different loci of coronary arteries are affected by genotypes of alcohol-metabolizing enzymes and whether and how those relationships are modified by coronary risk factors including blood lipids and diabetes.
None.
