Association between serum 5-methyltetrahydrofolate and homocysteine in Chinese hypertensive participants with different MTHFR C677T polymorphisms: a cross-sectional study

Study participants and baseline characteristics

As shown in the flow chart (Supplemental Fig. 1), 2328 participants were included in the final analysis of this study. Baseline characteristics of study participants stratified by MTHFR C677T are shown in Table 1. Genotype CT participants and genotype TT participants had higher levels of Hcy concetrations, DBP, BMI, TC and fasting glucose but lower levels of 5-MeTHF concentrations, vitamin B12 and folate compared with genotype CC participants at baseline. Similar trends were found in males and females (Supplemental Table 1, Supplemental Table 2).

Table 1 Baseline characteristics of the study participants stratified by MTHFR C677TAssociation of serum 5-MeTHF and Hcy and stratification by MTHFR C677T

Overall, there existed an apparent inverse association between serum 5-MeTHF and Hcy during the baseline period, whether among all participants (Fig. 1A) or stratified by MTHFR C677T (Fig. 1B; solid, dotted and two-dash lines represent for genotype CC, CT and TT, respectively), but the association plateaued when the 5-MeTHF concentration was over 10 ng/mL (Supplemental Table 3). Thus, the 2328 participants were divided into two subsets, one of which had concentrations of 5-MeTHF ≤ 10 ng/mL, while the other subset had concentrations over 10 ng/mL. Multiple linear regression was performed on both subsets (Supplemental Table 4). The results of the former subset were displayed in Table 2.

Fig. 1figure 1

Association of Hcy levels and 5-MeTHF levels among all participants (A) and stratified by MTHFR C677T (B). Panel A was adjusted for sex, age, study site, MTHFR C677T, BMI, systolic blood pressure, diastolic blood pressure, estimated glomerular filtration rate, total cholesterol, triglycerides, high-density lipoprotein cholesterol, vitamin B12, vitamin D3, fasting glucose, folate, smoking and drinking at baseline. Panel B was adjusted for the variables in panel A but removed MTHFR C677T

Table 2 Multiple linear regression model of Hcy on level of 5-MeTHF in all participants and stratified by MTHFR C677T genotype (5-MeTHF ≤ 10 ng/mL only)

A 1-ng/mL increment in 5-MeTHF concentration was associated with a 0.50- μ mol/L decrease in Hcy concentration (95% CI: − 0.69, − 0.31; P <  0.001) among all participants. Consistently, when 5-MeTHF concentration was assessed as quartiles, the adjusted β s for participants in quartiles 2, 3 and 4 were − 2.18 (95% CI: − 3.32, − 1.03; P <  0.001), − 2.51 (95% CI: − 3.71, − 1.31; P <  0.001) and − 3.62 (95% CI: − 4.87, − 2.38; P <  0.001), respectively, when compared with those in quartile 1 (P for trend < 0.001) (Table 2).

Similarly, a significant, negative association was also found in genotype CT and TT participants, however, it was slightly not significant in genotype CC participants. A 1-ng/mL increase in 5-MeTHF concentration was associated with a 0.14- μ mol/L decrease (95% CI: − 0.30, 0.02; P = 0.087), a 0.20- μ mol/L decrease (95% CI: − 0.35, − 0.05; P = 0.011) and a 1.19- μ mol/L decrease (95% CI: − 1.77, − 0.62; P <  0.001) in Hcy concentration among CC, CT and TT genotype participants, respectively (Table 2). Moreover, the results showed that among participants whose serum 5-MeTHF was ≤10 ng/mL, the decline in trend in genotype TT participants was stronger than that in genotype CC and genotype CT participants, and there was no significant difference in the decline in trend between genotype CC and genotype CT participants. However, there were no significant differences in the decline in trend between genotype CC, genotype CT and genotype TT participants, whose serum 5-MeTHF concentrations were over 10 ng/mL (Supplemental Table 5).

Decrements in Hcy among CC genotype participants in quartile 2 (P = 0.662), quartile 3 (P = 0.353) and quartile 4 (P = 0.070) were not significant compared with quartile 1 when 5-MeTHF was assessed as quartiles. While a declining trend existed, this trend was not significant (P for trend = 0.061). Genotype CT participants showed a significant decrement in Hcy in quartile 4 (P = 0.013), but in genotype TT participants, when compared with quartile 1, all quartiles (all of the P values were ≤ 0.001) and the decline in trend were of statistical significance (P for trend < 0.001) (Table 2).

Considering the potential heterogeneity between participants with and without outcomes (incident stroke, cancer or all-cause mortality), we performed a sensitivity analysis on these two sub-groups respectively. The overall association was consistent with the above results regardless of the subgroup (Supplemental Table 6, Supplemental Table 7).

There were no declining trend results found among the participants whose 5-MeTHF concentration was over 10 ng/mL, either in all participants or stratified by MTHFR C677T (Supplemental Table 4).

Modification of MTHFR C677T on the association between 5-MeTHF and Hcy

Stratified analyses were performed to assess the association between serum 5-MeTHF (≤ 10 ng/mL, as a continuous variable) and Hcy (Supplemental Table 8) in various subgroups including MTHFR C677T. A significantly stronger negative association between serum 5-MeTHF and Hcy was observed in genotype TT participants (per increment, β: -1.19; 95% CI: − 1.77, − 0.62; P <  0.001) than in genotype CC participants (per increment, β: -0.14; 95% CI: − 0.30, 0.02; P = 0.087) and genotype CT participants (per increment, β: -0.20; 95% CI: − 0.35, − 0.05; P = 0.011). However, there was no significant difference between genotype CC and genotype CT participants (Supplemental Table 5). Further, in participants with lower vitamin B12 ≤ 381.4 pg/mL (per increment, β: -0.75; 95% CI: − 1.08, − 0.41; P <  0.001) a significantly stronger negative association was observed than in those with vitamin B12 > 381.4 pg/mL (per increment, β: -0.27; 95% CI: − 0.46, − 0.07; P = 0.007; P for interaction = 0.007).

None of the other variables, including age, SBP, eGFR, vitamin D3 and fasting glucose over the baseline period showed significant effect modification on the association between serum 5-MeTHF and Hcy (P for interaction > 0.05 for all of these stratified variables).

Interaction effect analysis of 5-MeTHF and MTHFR C677T on Hcy concentration

Participants whose 5-MeTHF concentration was ≤ 10 ng/mL, were quartered according to their 5-MeTHF concentration (i.e., Q1, < 2.7 ng/mL; Q2, 2.7 – < 4.4 ng/mL; Q3, 4.4 – < 6.4 ng/mL; Q4, ≥ 6.4 ng/mL) and were stratified by MTHFR C677T. Moreover, genotype CC participants in quartile 1 were set as reference. Genotype TT participants in Q1 ~ Q4 (all of the adjusted P values were <  0.05) showed significant positive coefficients. In particular, when comparing Q4 in genotype TT participants with Q1 in genotype CC participants (reference), the coefficient was 3.46 (95% CI: 1.17, 5.75; P = 0.003). However, there were no significant differences in Q2 ~ Q4 among genotype CC participants and Q1 ~ Q4 among genotype CT participants when compared with Q1 among genotype CC particpants (all of the adjusted P values were > 0.05), and the P value for the trend test was < 0.001 (Table 3).

Table 3 Interaction effect between serum 5-MeTHF (≤ 10 ng/mL) concentration and MTHFR C677T on Hcy concentration

留言 (0)

沒有登入
gif