Introduction

Hypertension is a major public health concern worldwide1, which is influenced by both genetic and environmental factors2. Epidemiological studies have revealed that multiple environmental factors are associated with hypertension, such as obesity, alcohol consumption, sodium intake, and physical inactivity3,4,5,6. A better understanding of modifiable risk factors for hypertension is useful for early detection and prevention of hypertension, which contributes to reducing this disease and its associated complications.

Homocysteine (Hcy) is a non-proteinogenic α-amino acid produced by the demethylation of methionine. Hyperhomocysteinemia is known as an established risk factor for cardiovascular disease (CVD) and stroke7,8. To date, attention has been focused on the fact that this adverse effect caused by hyperhomocysteinemia might be partially mediated by a positive association between Hcy levels and blood pressure9. Increased plasma Hcy causes oxidative stress and endothelial dysfunction, which leads to vascular constriction, stiffness, and decrease in the vasodilation by nitric oxide, resulting in an increase in blood pressure9,10,11. Some previous observational studies have demonstrated positive associations between Hcy levels and blood pressure or the risk of hypertension12,13,14,15. However, other studies including large prospective cohort studies do not support these associations16,17,18,19,20,21. Thus, the association between Hcy levels and risk of hypertension remains inconclusive. A recent meta-analysis of relevant randomized trials also revealed that folate supplementation is effective in reducing blood pressure and Hcy levels among patients with hypertension and hyperhomocysteinemia22. In addition, observational studies showed that dietary folate intake and folate level in blood were inversely associated with hypertension23,24,25. These findings suggest that it is important to consider dietary intake and plasma levels of folate typically involved in the metabolism of Hcy in relation to the association between Hcy levels and risk of hypertension. Interestingly, a previous study suggested that vitamin B12 as related to Hcy metabolism is also inversely associated with blood pressure26. Nevertheless, few studies have examined the association with these nutrients. It is also of interest to examine the hypothesis that folate and vitamin B12 are independently associated with the risk of hypertension in a general population. The evidence for this hypothesis is weak, although these vitamins are essential in the metabolism of methionine and are major nutritional determinants of plasma Hcy levels27. Furthermore, the association of Hcy with hypertension may be dependent on folate and/or vitamin B12. Therefore, to consider these nutrients involved in Hcy metabolism in addition to Hcy itself should be important.

Thus, in the present study, we aimed to simultaneously examine the association of plasma levels of Hcy, folate, and vitamin B12, and dietary folate intake with the prevalence of hypertension, after controlling for multiple confounding factors, using a relatively large sample size in a cross-sectional study conducted among a general Japanese population.

Methods

Participants

Participants in the present study were enrolled in the Japan Multi-Institutional Collaborative Cohort (J-MICC) Study28. The J-MICC Study is a large cohort study in Japan, launched in 2005 to identify interactions between genetic and lifestyle factors for lifestyle-related diseases, including any cancer. The details and rationale of the J-MICC Study have been described elsewhere29. Briefly, participants completed a self-administered questionnaire on lifestyle and medical status, and then donated a peripheral blood sample during the baseline survey. The J-MICC Study enrolled residents in the community, health check examinees, and patients at a cancer hospital. All participants in the study provided their written informed consent. The study protocol was approved by the ethics committees of Nagoya University Graduate School of Medicine, Aichi Cancer Center, Chiba Cancer Center, Nagoya City University Graduate School of Medical Sciences, Shiga University of Medical Science, Kyoto Prefectural University of Medicine, Kyushu University Graduate School of Medical Sciences, Saga University Faculty of Medicine, Kagoshima University Graduate School of Medical and Dental Sciences, Tokushima University Graduate School, and University of Shizuoka participating in the J-MICC Study. The present study was conducted according to the principles expressed in the World Medical Association Declaration of Helsinki.

A total of 92,631 participants were recruited from 14 different areas throughout Japan between 2004 and 2014 (the dataset used in the present study was fixed on February 15, 2019). Of the total population, 2845 participants in eight study areas underwent measurement of plasma concentrations of Hcy, folate, and vitamin B12. Among the total, we excluded all participants from two areas (n = 763) because of no available data for blood pressure. We further excluded those with missing data for blood pressure and use of antihypertensive medication (n = 1 and 2, respectively). Thus, the present study finally included 2079 individuals (1046 men and 1033 women) from six study areas (Okazaki, Shizuoka, Takashima, Kyoto, Saga, and Kagoshima) with complete data for the analysis.

Assessment of lifestyle factors, dietary folate intake, and blood pressure

We used a self-administered questionnaire that included the following demographic characteristics and lifestyle and medical factors: alcohol consumption, smoking status, education level, physical activity, self-reported psychological stress, sleeping hours, family and personal medical history, current medication, and menstruation in women.

Physical activity was estimated as metabolic equivalent hours per day, based on the frequency and duration of daily and leisure time activities30,31. Ethanol intake (g/day) was estimated for current drinkers (defined as those who consumed alcohol at least once a week during the last year), based on the reported consumption frequency and amount consumed per one time for six alcoholic beverages (Japanese sake, shochu, shochu-based cocktails, beer, whisky, and wine)32. Dietary folate and energy intakes were estimated using a validated short food frequency questionnaire33,34,35,36; Spearman correlation coefficients between estimated intakes according to the questionnaire and 3-day weighed dietary records were 0.41 in men and 0.36 in women for folate, and 0.36 in men and 0.37 in women for total energy. Dietary folate intake per day was adjusted for total energy intake using the nutrient density method37, which computes the dietary folate intake per 1000 kcal of daily total energy intake. We did not consider the loss of dietary folate intake due to the cooking with heat for foods because of limited data. In addition, this study did not include participants who reported folate supplement use, and thus we did not consider folate supplementation in examining the associations.

Height and weight were measured directly on the day of the survey; body mass index (BMI) was calculated as weight in kilograms divided by the square of height in meters (kg/m2). Each participant’s blood pressure was measured by a nurse or trained staff during health checks, using an automated blood pressure measurement monitor, with the patient in a seated position38.

Blood samples and measurements of plasma Hcy, folate, and vitamin B12

Fasting peripheral blood was drawn from participants at health checks, which was collected in a 7-mL EDTA-2Na-containing vacuum tube for plasma. Plasma separated from whole blood by using a centrifuge was stored at approximately − 80 °C until analysis. Plasma concentration of Hcy was determined using high-performance liquid chromatography. Concentrations of folate and vitamin B12 were determined using chemiluminescent enzyme immunoassay. All measurements were conducted at SRL Co., Ltd., Hachioji, Japan.

Statistical analyses

Participant characteristics were presented as mean ± standard deviation (SD) for continuous variables and as number and proportion for categorical variables, according to quartiles of plasma Hcy level by sex. Differences in the mean or proportion between quartiles of Hcy level were tested using analysis of variance or the chi-squared test, respectively. Difference in the mean of Hcy level between those with hypertension and those without was tested using the Mann–Whitney U test. The Spearman correlation coefficient was calculated between plasma folate levels and energy-adjusted dietary folate intake. We performed all analyses stratified by sex, as Hcy levels differ greatly by sex39.

Hypertension was defined according to any of the following criteria: systolic blood pressure ≥ 140 mmHg, diastolic blood pressure ≥ 90 mmHg, or use of antihypertensive medication40,41. Crude, age-adjusted, and multivariate-adjusted odds ratios (ORs) and 95% confidence intervals (CIs) for hypertension were estimated according to quartile levels of plasma Hcy, folate, and vitamin B12, and dietary folate intake using unconditional logistic regression models. The multivariate-adjusted model included the following covariates: age (as a continuous variable), alcohol consumption (never, former, current drinker who consumed < 23 g/day ethanol, current drinker who consumed ≥ 23 g/day ethanol), smoking status (never, former, current), education level (9, 10–15, ≥ 16 years), BMI (< 18.5, 18.5 to < 25.0, ≥ 25.0 kg/m2), physical activity (as a continuous variable), psychological stress (not at all, not much, a little, a lot), sleeping hours (< 6, 6 to < 8, ≥ 8 h), family history of parental hypertension (yes, no for both father and mother), medical history of diabetes mellitus (yes, no), medical history of dyslipidemia (yes, no), medical history of CVD (yes, no), medical history of stroke (yes, no), menstruation for women (premenopausal, perimenopausal, postmenopausal), and study area (Okazaki, Shizuoka, Takashima, Kyoto, Saga, Kagoshima). In an additional multivariate-adjusted model, we further controlled for plasma concentrations of Hcy, folate, and vitamin B12, and dietary folate intake each other (as a continuous variable, respectively). Participants with missing data for covariates were included as additional categories in the analysis. The linear trend for risk was evaluated using a continuous variable for plasma levels of Hcy, folate, and vitamin B12, and dietary folate intake, respectively.

A two-tailed P value of < 0.05 was considered statistically significant. All statistical analyses were performed using SAS 9.4M5, which runs on SAS University Edition (SAS Institute Inc., Cary, NC, USA).

Results

Baseline characteristics of 2079 study participants (1046 men and 1033 women) according to the quartiles of plasma Hcy level by sex are shown in Table 1. Mean age ± SD was 56.0 ± 8.9 years, and the proportion of men was 50.3%. Current drinkers accounted for 79.6% of men and 34.8% of women. There was no difference in the mean of age among quartiles of Hcy level. Plasma Hcy concentration was inversely associated with plasma levels of folate and vitamin B12, dietary folate intake (in men), and energy intake. Spearman correlation coefficients between plasma folate level and energy-adjusted dietary folate intake were 0.105 in men and 0.129 in women (P < 0.001, respectively). The distributions of alcohol consumption, education level, BMI, physical activity, psychological stress, sleeping hours, and family history of parental hypertension did not significantly differ among quartiles of Hcy level. Men with higher plasma Hcy levels were more likely to be current smokers; this association was not observed in women. There were no differences in the distribution of medical histories of diabetes mellitus, dyslipidemia, and CVD among quartiles of Hcy level, and that of stroke differed only in women. Differences in the distribution of study areas were observed among quartiles of Hcy level.

Table 1 Participant characteristics according to quartiles (Q1–Q4) of plasma Hcy level, by sex.
Full size table

Table 2 presents the crude, age-adjusted and multivariate-adjusted ORs and 95% CIs for hypertension according to quartile levels of plasma Hcy, folate, and vitamin B12, and dietary folate intake. A total of 734 participants (404 men and 330 women, representing 35.3% of participants) had hypertension. Plasma Hcy level in those with hypertension was significantly higher than that in those without hypertension; the means ± SD (nmol/mL) were 8.8 ± 3.2 and 8.3 ± 3.6, respectively (P < 0.001). Plasma Hcy level was positively and significantly associated with the prevalence of hypertension, after controlling for multiple covariates including plasma levels of folate and vitamin B12, and dietary folate intake. The multivariate-adjusted ORs for the highest quartile group of plasma Hcy were 2.36 (95% CI 1.41–3.96) in men and 1.86 (95% CI 1.11–3.11) in women, as compared with the lowest group (P for trend = 0.014 and 0.005, respectively). Dietary folate intake was not correlated with hypertension in both men and women. The multivariate-adjusted OR for the highest quartile group of dietary folate intake was 0.65 (95% CI 0.41–1.05) in men and 0.93 (95% CI 0.58–1.51) in women, as compared with the lowest group (P for trend = 0.099 and 0.703, respectively). The multivariate-adjusted ORs for the second and highest quartile groups of plasma folate in men were significantly higher than unity; 2.22 (95% CI 1.40–3.51) and 1.98 (95% CI 1.23–3.21), respectively, as compared with the lowest group (P for trend = 0.231). The linear trend, however, was not statistically significant. The corresponding ORs were not significant in women. Plasma vitamin B12 level was positively associated with hypertension, especially in women. The multivariate-adjusted OR for the highest quartile group of plasma vitamin B12 was 1.49 (95% CI 0.93–2.37), as compared with the lowest group (P for trend = 0.027).

Table 2 The ORs and 95% CIs for hypertension according to quartile levels (Q1–Q4) of plasma Hcy, folate, and vitamin B12, and dietary folate intake, by sex.
Full size table

Discussion

In the present study, we found that increased plasma Hcy level was significantly associated with the prevalence of hypertension in both men and women after controlling for multiple covariates including plasma levels of folate and vitamin B12, and dietary folate intake. Dietary folate intake was not associated with the prevalence of hypertension in both men and women.

Our finding for the association between Hcy level and hypertension is consistent with those from previous cross-sectional studies12,13,14,15. As a large previous cross-sectional study, the Third National Health and Nutrition Examination Survey in the United States demonstrated that those with the highest quintile group of plasma Hcy had a two- to three-fold OR of hypertension, in comparison with the lowest group14, which is similar to our findings. In men, the multivariate-adjusted OR of hypertension for the highest quintile group (range for plasma Hcy concentration: 13.1–132 μmol/L) was 1.9 (95% CI 0.71–5.14) in comparison with the lowest group (3.3–7.5 μmol/L) (P for trend = 0.12). In women, the OR for the highest quintile group (11.0–118 μmol/L) was 3.0 (95% CI 1.70–5.39) as compared with the lowest group (3.0–6.1 μmol/L) (P for trend = 0.0001).

There are several potential mechanisms by which increased plasma Hcy can cause hypertension. Hcy has been shown to be associated with oxidative stress, endothelial dysfunction, increased inflammation, and decreased bioavailability of endothelium-derived nitric oxide, resulting in increased blood pressure9,10,11. Additionally, Hcy activates metalloproteinase and induces collagen synthesis, and causes imbalances of elastin/collagen ratio. Interestingly, Hcy also promotes angiotensin-converting enzyme activity that may lead to upregulation of angiotensin II and subsequently hypertension42. Nevertheless, several cross-sectional and large prospective studies showed no evidence for an association between Hcy level and hypertension, after controlling for confounding factors16,17,18,19,20,21. The Mendelian randomization (MR) approach is a method of using the measured variation in genes to examine the causal effect of modifiable exposures for disease. The MR has recently been applied to evaluate this association. In that study, Hcy was reported not to be a causal factor for blood pressure43, suggesting that increased plasma Hcy may be concomitant with hypertension. However, the evidence from prospective studies and MR analysis is insufficient for this association; thus, further investigation will be needed to verify this relationship.

Although folate plays an important role in the metabolism of Hcy, limited data are available on the association between plasma folate levels and dietary folate intake and the risk of hypertension. In the present study, dietary folate intake was not associated with the prevalence of hypertension, whereas the ORs of hypertension were significantly higher than unity for the second and highest quartiles of plasma folate in men. The reason for this discrepancy is unclear. Cross-sectional and prospective studies have shown an inverse association between serum folate level and blood pressure23,24. In addition, two prospective studies have demonstrated that dietary folate intake is inversely associated with the incidence of hypertension24,25. Our findings are inconsistent with these results. When we calculated the Spearman correlation coefficient between plasma folate level and energy-adjusted dietary folate intake, the positive correlation appeared to be weak, which is similar to results from a previous study in Japan44. The biological mechanisms for the weak correlation between plasma folate level and dietary folate intake remained unclear. Measuring folate intake with a questionnaire and the plasma folate assay are prone to a different type and magnitude of systematic or random errors.

It is possible that unknown confounders brought significantly high ORs of hypertension for the second and highest quartiles of plasma folate level in men, although we controlled for multiple confounders to evaluate this association. As the trend was not significant in men and we observed no clear association of either dietary intake or plasma levels of folate with hypertension in women, further studies are warranted to examine the role of plasma folate.

We also found no clear association between plasma vitamin B12 levels and hypertension, although there was a positive correlation only in women. These results are consistent with those from a previous case–control study45. Interestingly, a cross-sectional study showed that dietary vitamin B12 intake is inversely associated with blood pressure among preschool children26, suggesting that vitamin B12 may play a role in reducing the risk of hypertension. However, we did not consider this issue owing to no available data for dietary vitamin B12 intake. Further studies using data for both dietary intake and plasma levels of vitamin B12 will elucidate the inconsistent associations with hypertension.

We found a significant positive association between Hcy level and the prevalence of hypertension, with consideration for folate and vitamin B12 as related to Hcy level. It is meaningful that the strong positive association remained even when controlling for dietary folate intake, and plasma levels of folate and vitamin B12. The association may be independent of folate and vitamin B12.

The strengths of the present study include its relatively large sample size, use of a validated FFQ for dietary folate and energy intake, and controlling for several confounders obtained from the self-administered questionnaire. Several potential limitations, however, should be mentioned. First, we were unable to determine the causal relationship between Hcy level and risk of hypertension because of the cross-sectional design of this study. Second, because most participants analyzed in the present study were Japanese, the generalizability of our findings to other populations remains to be explored. Third, although our FFQ was well validated using dietary record33,34,35,36, the validity correlation coefficient for dietary folate intake was not so high, so that the dietary folate intake is a rough estimate to rank the participants. Fourth, we did not consider the loss of dietary folate intake due to the cooking with heat for foods because of limited data. Fifth, we did not examine the associations with consideration for folate supplement use because no participant reported its use in the present analysis. Lastly, we could not consider the genetic background of the included individuals in relation to the association with risk of hypertension because it requires too much analysis to be included in one study. Methylenetetrahydrofolate reductase (MTHFR) C677T (rs1801133) is one of the important polymorphisms strongly associated with Hcy levels in the human body, whose T allele significantly reduces MTHFR activity, resulting in increased plasma Hcy levels. However, this variant itself has been reported not to be associated with blood pressure43. In addition, other genetic background regarding blood pressure appeared to be unrelated to Hcy levels in the present study because we observed no clear differences in family history of parental hypertension among the quartiles of plasma Hcy level, by sex.

In conclusion, we found a significant positive association between plasma Hcy levels and the prevalence of hypertension when considering plasma folate and vitamin B12, and dietary folate intake typically involved in the metabolism of Hcy. Further studies are warranted to explore the causal inference for this association and to clarify the roles of folate and vitamin B12 in the risk of hypertension as related to Hcy.