Evidence on the causal link between homocysteine and hypertension from a meta‐analysis of 40 173 individuals implementing Mendelian randomization

Abstract

Numerous researchers have investigated the associations among methylenetetrahydrofolate reductase gene (MTHFR) C677T polymorphism, homocysteine (Hcy) concentration, and hypertension. However, the results are controversial. Thus, a meta‐analysis implementing Mendelian randomization approach was conducted to examine the hypothesis that elevated Hcy concentration plausibly contributes to increased risk of hypertension. Based on several inclusion and exclusion criteria, eligible studies were selected to explore the correlation between MTHFR C677T and hypertension risk, MTHFR C677T and Hcy concentration in hypertension, and Hcy concentration and hypertension, and they were evaluated by odds ratios (ORs), effect size (ES), and standard mean difference with their corresponding 95% confidence intervals (95% CIs), respectively. Moreover, Mendelian randomization was implemented to evaluate the relationship between Hcy and hypertension. Consequently, 14 378 cases and 25 795 controls were involved in this study and the results showed that MTHFR C677T led to an elevated risk of hypertension (for T vs C: OR = 1.27, 95% CI = 1.17‐1.37; for TT vs CC: OR = 1.53, 95% CI = 1.30‐1.79). Additionally, in hypertensive subjects, the pooled Hcy concentration in individuals of TT genotype was 7.74 μmol/L (95% CI: 5.25‐10.23) greater than that in individuals of CC genotype. Moreover, the pooled Hcy concentration in hypertensive was 0.69 μmol/L (95% CI: 0.50‐0.87) greater than that in controls. The estimated causal OR associated with hypertension was 1.32 for 5 μmol/L Hcy increment. Via MTHFR C677T polymorphism, the findings in the present study demonstrated that there exists evidence on causal link between Hcy concentration and the risk of hypertension.

1. INTRODUCTION

According to the previous research reports, hypertension is influencing more than 1 billion people worldwide, and over 29.2% of adults are estimated to be recognized as hypertensive in 2025.1 Hypertension is a crucial risk factor for cardiovascular disease and the pivotal source of global health burden, which has contributed to 9.4 million deaths each year.2 Although the etiology of this disease has not been fully elucidated and numerous lifestyle risk factors lead to the development of hypertension, genes, environmental factors, and their interactions are indicated to have a substantial effect on the pathological process of hypertension.3, 4 Therefore, identifying genetic factors which led to the presence of hypertension and understanding the process affecting this disease could be of great importance not only in comprehending its pathogenesis, but also in giving more choices to take effective intervention for reducing the incidence of hypertension.

Over the past several decades, the candidate gene approach has been extensively applied to detect the deleterious genes efficiently in complex human disease.5 Methylenetetrahydrofolate reductase (MTHFR) serves as an important rate‐limiting enzyme playing a role in the altering of 5,10 methylenetetrahydrofolate to 5‐methylenetetrahydrofolate, which is regarded as a pivotal remethylated process of homocysteine (Hcy) to methionine.6 MTHFR C677T indicates the change in cysteine to thymine nucleotide, which signifies the substitution of alanine to valine. As a result, the subjects with TT genotype express a heat‐sensitive enzyme with reduced activity, which results in the elevated plasma Hcy concentration eventually.6, 7 The increasing level of plasma Hcy may contribute to certain pathological changes, such as vascular endothelial injury, nitric oxide production inhibition, and vascular smooth muscle proliferation,8 which might result in the development of hypertension.9

Recently, a large number of studies investigated the association between MTHFR C677T polymorphism and hypertension. For example, a study suggested that MTHFR C677T is a protective predictor of hypertension in a Maya‐Mestizo women population.10 However, another study announced an opposite result completely.11 Because of the contradicted results, it is difficult to come to an incontestable conclusion to date. The relationship of MTHFR C677T polymorphism, Hcy, and hypertension is still elusive.

In order to provide more solid evidence for this issue, a meta‐analysis of the published studies related to the risk of hypertension associated with an elevated Hcy level and the MTHFR C677T polymorphism was conducted. Here, we integrated the previous published studies with regard to the associations among MTHFR C677T, Hcy concentration, and hypertension risk to obtain pooled results of their reported associations. Furthermore, Mendelian randomization approach, which is recognized as an epidemiological method based on the fact that individuals randomly inherit their genetic variation from their parents, was carried out to test the hypothesis that elevated Hcy level is plausibly associated with the enhanced risk of hypertension.

2. MATERIALS AND METHODS

2.1. Selection of studies

Researches that estimated the association between the MTHFR gene 677 C>T polymorphism and Hcy level with the development of hypertension were encompassed in this study. To identify potentially relevant researches, a comprehensive literature retrieval was performed independently by two investigators from PubMed, Embase, and Web of Science with the following terms: “MTHFR,” “rs1801133,” “MTHFR C677T,” “homocysteine,” “Hcy,” “hypertension” and “blood pressure,” up to August 1, 2019.

The following criteria were proposed for the inclusion of eligible researches: (a) case‐control, cross‐sectional, or case‐cohort designed researches; and (b) descripting the distribution of the MTHFR C677T genotypes or sufficient data for calculating it in cases with hypertension and in controls free of hypertension. Reviews or letters, duplicated studies, abstracts, and editorials were excluded. The languages were limited to English and Chinese.

2.2. Data extraction

Two investigators extracted the following information from each study independently: (a) first author's surname, year of publication, subjects’ country, and ethnicity; (b) number of cases and controls, the diagnostic standard of cases and controls, and genotype distribution in both of groups. Data were extracted, respectively, for studies involving more than one population. Discrepancies were resolved after discussion with a third reviewer.

2.3. Statistical analysis

Chi‐square test was performed to estimate whether Hardy‐Weinberg equilibrium (HWE) was significantly deviated in controls of each study. Moreover, sensitivity analysis by removing one study with controls not in HWE was carried out to evaluate the stability of the results.

Under five genetic models including homozygous codominant model (TT vs CC), heterozygous codominant model (TT vs CT), allelic model (T vs C), dominant model (TT + TC vs CC), and recessive model (TT vs TC + CC), this meta‐analysis estimated the overall association of MTHFR C677T with the risk of hypertension. All associations for these five genetic models were represented as odds ratios (OR) with their corresponding 95% confidence interval (CI). Then, a pooled OR was estimated based on the ORs of separate study for each of those five models. Metan command in Stata, version 12.0 (StataCorp LP), was used to estimate the mean difference between MTHFR 677TT group and MTHFR 677CC group in hypertensive subjects. We also pooled the Hcy levels to identify the standardized mean difference (SMD) and its corresponding 95% CI for comparisons between hypertensive and healthy subjects. Between‐study heterogeneity testing was performed by Cochrane's Q test (significance at I ² > 50.0% and P < .10).12, 13 Once the effects were assessed to be absent from heterogeneity, the fixed effects model was fitted; otherwise, the random effects model was employed.14 In this meta‐analysis, subgroup analyses by ethnicity and age range (children/adults) were also carried out. The publication bias was measured by Begg's funnel plot and Egger's regression test (P < .05 was considered as statistically significant) if too many studies were included.15 Extracted data for all statistical analyses were conducted by using the Stata.

Mendelian randomization analysis is a method to make unbiased intermediated phenotype‐disease plausible association by integrating the information of genotype‐intermediate phenotype and genotype‐disease association into an analytical framework. A valid instrumental variable in the Mendelian randomization needs to satisfy three conditions:16 (a) associated with Hcy strongly; (b) not associated with confounders, which could bias the correlation between Hcy and hypertension; and (c) affects hypertension only through its effect on the specific intermediate Hcy. MTHFR C677T may satisfy all these conditions well based on the existing studies.9, 17, 18, 19 Thus, the Mendelian randomization coefficient estimated by applying MTHFR C677T as an instrument would provide the causal link free of bias between Hcy and hypertension. Compared with MTHFR 677 CC, the genotype of MTHFR 677 TT elevates the hypertension risk and the effect is evaluated by the odds ratio (OR_{TT vs CC}). Compared with CC, TT is associated with the mean difference (Δ) of Hcy concentration. Further OR₁ = (OR_{TT vs CC})^1/Δ would be an unconfounded evaluation of hypertension effect caused by one unit change in the Hcy. In this study, we employed the formula OR_k = (OR_{TT vs CC})^k/Δ for an increment of k units,20 and we analyzed 5 μmol/L increase in Hcy to evaluate the corresponding OR.21

3. RESULTS

3.1. Characteristics of the studies

Figure 1 describes the comprehensive information of screening progress in this meta‐analysis. As a result, fifty‐seven studies evaluated the association between MTHFR C677T and hypertension, which provided 14 378 cases and 25 795 controls.9, 10, 11, 17, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73 In this pooled data, the frequencies of the CC genotype were the highest in pooled cases and in pooled controls, respectively, while that for genotype TT was the lowest. Ten studies only described the association between MTHFR C677T and Hcy level in hypertensive patients.66, 74, 75, 76, 77, 78, 79, 80, 81, 82 However, in eight studies,30, 44, 47, 54, 56, 66, 71, 72 the genotype distribution in the control subjects was not in line with HWE (P < .05; Table 1).

Figure 1.

PRISMA flow diagram for selection of studies in the meta‐analysis

Table 1.

The genotypic and allelic distribution of MTHFR C677T for case and control

First author	Year	Country	Ethnicity	Children/adults	Diagnose		Genotype distribution						Allele frequency				P‐value HWE	Number of cases/controls
							Cases			Controls			Cases		Controls
					SBP	DBP	CC	CT	TT	CC	CT	TT	C	T	C	T
Nishio22	1996	Japan	Asian	Adults	>140 mm Hg	>90 mm Hg	29	44	9	16	26	5	102	62	58	36	.24	82/47
Nakata23	1998	Japan	Asian	Adults	>160 mm Hg	>95 mm Hg	63	91	19	65	83	36	217	129	213	155	.31	173/184
Powers24	1999	United States	American	Adults	≥140 mm Hg	≥90 mm Hg	46	58	18	54	46	14	150	94	154	74	.39	122/114
Zusterzeel29	2000	Netherlands	White	Adults	—	≥90 mm Hg	32	33	11	205	162	36	97	55	572	234	.62	76/403
Zhan28	2000	China	Asian	Adults	≥140 mm Hg	≥90 mm Hg	44	68	15	62	84	24	156	98	208	132	.60	127/170
Rajkovic27	2000	United States	American	Adults	≥140 mm Hg	≥90 mm Hg	142	28	1	151	32	0	312	30	334	32	.19	171/183
Li26	2000	China	Asian	Adults	≥140 mm Hg	≥90 mm Hg	14	31	17	46	32	12	59	65	124	56	.11	62/90
Kobashi25	2000	Japan	Asian	Adults	≥140 mm Hg	≥90 mm Hg	64	98	22	83	99	33	226	142	265	165	.70	184/215
Benes30	2001	Czech	White	Adults	≥140 mm Hg	≥90 mm Hg	73	93	27	86	106	17	239	147	278	140	.05	193/209
Kahleova31	2002	Czech	White	Children and Adolescents	≥90a	—	82	55	27	86	69	18	219	109	241	105	.46	164/173
Wang32	2002	China	Asian	Adults	≥140 mm Hg	≥90 mm Hg	17	51	37	14	23	9	85	125	51	41	.94	105/46
Rodriguez‐Esparragon33	2003	Spain	White	Adults	>140 mm Hg	>90 mm Hg	83	115	34	95	100	20	281	183	290	140	.39	232/215
Sun34	2003	China	Asian	Adults	≥140 mm Hg	≥90 mm Hg	6	22	27	14	23	9	34	76	51	41	.94	55/46
Yilmaz35	2004	Turkey	White	Adults	>140 mm Hg	>110 mm Hg	29	28	7	24	17	6	86	42	65	29	.30	64/47
Heux9	2004	Australia	White	Adults	>140 mm Hg	>95 mm Hg	87	125	35	105	119	25	299	195	329	169	.30	247/249
Frederiksen17	2004	Denmark	White	Adults	—	—	576	555	136	3853	3406	712	1707	827	11 112	4830	.30	1267/7971
Cesari36	2005	Italy	White	Adults	—	—	40	39	11	42	38	10	119	61	122	58	.75	90/90
Liu37	2005	China	Asian	Adults	≥140 mm Hg	≥90 mm Hg	29	45	26	31	50	19	103	97	112	88	.88	100/100
Lwin42	2006	Japan	Asian	Adults	>140 mm Hg	>90 mm Hg	39	58	18	64	117	38	136	94	245	193	.22	115/219
Li41	2006	China	Asian	Adults	≥140 mm Hg	≥90 mm Hg	18	6	2	21	7	2	42	10	49	11	.23	26/30
Hu39	2006	China	Asian	Adults	≥140 mm Hg	≥90 mm Hg	55	39	16	61	42	12	149	71	164	66	.25	110/115
Kalita40	2006	India	Asian	Adults	≥140 mm Hg	≥90 mm Hg	18	7	3	21	9	2	43	13	51	13	.46	28/32
Demir38	2006	Turkey	White	Adults	>140 mm Hg	>90 mm Hg	33	53	14	43	47	12	119	81	133	71	.88	100/102
Nagy46	2007	Hungary	White	Adults	>160 mm Hg	>90 mm Hg	49	43	9	32	35	6	141	61	99	47	.40	101/73
Markan45	2007	India	Asian	Adults	>140 mm Hg	>90 mm Hg	105	40	8	105	28	0	250	56	238	28	.17	153/133
Marinho44	2007	Portugal	White	Adults	—	—	15	49	0	57	67	4	79	49	181	75	2.87 × 10—3	64/128
Hui43	2007	Japan	Asian	Adults	>160 mm Hg	>100 mm Hg	83	129	49	104	123	44	295	227	331	211	.45	261/271
Xing47	2007	China	Asian	Adults	≥140 mm Hg	≥90 mm Hg	202	300	184	182	222	105	704	668	586	432	.02	686/509
Soares51	2008	Brazil	American	Adults	≥140 mm Hg	≥90 mm Hg	13	15	2	9	5	2	41	19	23	9	.36	30/16
Lin50	2008	China	Asian	Adults	>140 mm Hg	>90 mm Hg	19	27	4	73	44	6	65	35	190	56	.85	50/123
Ilhan49	2008	Turkey	White	Adults	>140 mm Hg	>90 mm Hg	36	32	10	72	26	2	104	52	170	30	.84	78/100
Fridman48	2008	Argentina	American	Adults	≥140 mm Hg	≥90 mm Hg	15	21	4	39	38	9	51	29	116	56	.95	40/86
Canto10	2008	Mexico	White	Adults	≥140 mm Hg	≥90 mm Hg	36	66	23	61	131	82	138	112	253	295	.53	125/274
Ng11	2009	Australia	White	Adults	≥140 mm Hg	≥90 mm Hg	14	14	10	40	32	8	42	34	112	48	.67	38/80
Fakhrzadeh53	2009	Iran	Asian	Adults	≥160 mm Hg	≥95 mm Hg	99	44	17	36	31	9	242	78	103	49	.56	160/76
Deshmukh52	2009	United States	American	Adults	>140 mm Hg	>90 mm Hg	22	16	4	52	48	18	60	24	152	84	.22	42/118
Liu54	2011	China	Asian	Adults	≥140 mm Hg	≥90 mm Hg	58	70	27	74	47	19	186	124	195	85	.01	155/140
Fowdar55	2012	Australia	White	Adults	≥140 mm Hg	≥90 mm Hg	170	174	33	175	183	35	514	240	533	253	.19	377/393
Zhang57	2012	China	Asian	Adults	≥140 mm Hg	≥90 mm Hg	128	53	8	117	41	7	309	69	275	55	.18	189/165
Yin56	2012	China	Asian	Adults	≥140 mm Hg	≥90 mm Hg	244	358	68	322	309	51	846	494	953	411	.05	670/682
Fridman58	2013	Argentina	American	Adults	≥140 mm Hg	≥90 mm Hg	29	40	6	71	64	15	98	52	206	94	.92	75/150
Yao59	2013	China	Asian	Adults	≥140 mm Hg	≥90 mm Hg	32	69	49	61	67	22	133	167	189	111	.61	150/150
Cai60	2014	China	Asian	Adults	≥140 mm Hg	≥90 mm Hg	39	99	62	61	89	50	177	223	211	189	.13	200/200
Husemoen61	2014	Denmark	White	Adults	≥140 mm Hg	≥90 mm Hg	2231	2012	451	3790	3208	699	6474	2914	10 788	4606	.59	4694/7697
Vazquez‐Alaniz62	2014	Mexico	American	Adults	>140 mm Hg	>90 mm Hg	62	93	39	54	97	43	217	171	205	183	.96	194/194
Wen68	2015	China	Asian	Adults	≥140 mm Hg	≥90 mm Hg	45	53	76	258	291	85	143	205	807	461	.84	174/634
Perez‐Razo(Adults)65	2015	Mexico	American	Adults	≥140 mm Hg	≥90 mm Hg	112	174	87	90	200	101	398	348	380	402	.64	373/391
Perez‐Razo(Children)65	2015	Mexico	American	Children and Adolescents	≥95thb	—	34	98	67	35	108	56	166	232	178	220	.17	199/199
Wei67	2015	Malaysia	Asian	Adults	≥140 mm Hg	≥90 mm Hg	143	82	21	260	78	10	368	124	598	98	.17	246/348
Nassereddine64	2015	Morocco	African	Adults	≥140 mm Hg	≥90 mm Hg	47	40	14	54	45	3	134	68	153	51	.07	101/102
Bayramoglu63	2015	Turkey	White	Adults	>140 mm Hg	>90 mm Hg	65	38	22	56	38	5	168	82	150	48	.65	125/99
Wang66	2015	China	Asian	Adults	≥140 mm Hg	≥90 mm Hg	96	60	34	144	91	52	252	128	379	195	6.78 × 10−7	190/287
Fan70	2016	China	Asian	Adults	≥140 mm Hg	≥90 mm Hg	37	102	75	119	234	141	176	252	472	516	.26	214/494
Amrani‐Midoun69	2016	Algerian	African	Adults	≥140 mm Hg	≥90 mm Hg	37	36	9	44	25	3	110	54	113	31	.81	82/72
Dwivedi71	2017	India	Asian	Adults	—	—	71	24	5	184	34	5	166	34	402	44	.03	100/223
Rios72	2017	Brazil	American	Adults	≥160 mm Hg	≥110 mm Hg	13	29	54	20	20	45	55	137	60	110	7.82 × 10−6	96/85
Arina73	2019	Indonesia	Asian	Adults	—	—	32	16	5	43	10	0	80	26	96	10	.45	53/53

The P‐value for HWE testing for controls is shown.

Abbreviations: DBP, diastolic blood pressure; HWE, Hardy‐Weinberg equilibrium test; MTHFR, methylenetetrahydrofolate reductase; SBP, systolic blood pressure.

^aAccording to developmental distribution tables and curves: “Anonymous: Update on the 1987 Task Force Report on High Blood Pressure in Children and Adolescents: A working group report from the National High Blood Pressure Education Program. National High Blood Pressure Education Program Working Group on Hypertension Control in Children and Adolescents.”

^bAccording to the criteria: “The fourth report on the diagnosis, evaluation, and treatment of high blood pressure in children and adolescents.”

3.2. The mean Hcy concentration difference between MTHFR genotypes in hypertensive

According to the inclusion criteria, nineteen studies (9931 hypertensive patients) 26, 32, 35, 36, 45, 49, 53, 58, 66, 71, 74, 75, 76, 77, 78, 79, 80, 81, 82 were selected to report the plasma Hcy concentration of different genotypic groups with the arithmetic mean and the matching standard deviation in hypertensive patients. The mean Hcy level was higher in MTHFR 677TT subjects than that in the other genotypes. The pooled mean Hcy level in MTHFR 677TT subjects was 7.74 μmol/L (95% CI: 5.25‐10.23) greater than that in MTHFR 677CC subjects (P < .001; Figure 2). Moreover, the MTHFR 677TT subjects had 5.58 μmol/L (95% CI: 4.15‐7.01) higher Hcy level than MTHFR 677CT subjects (P < .001; Figure 3).

Figure 2.

Forest plot of the evaluation for the effect size (ES) in Hcy level between the MTHFR genotypes (TT vs CC) in hypertensive patients

Figure 3.

Forest plot of the evaluation for the effect size (ES) in Hcy level between the MTHFR genotypes (TT vs CT) in hypertensive patients

3.3. The association between MTHFR C677T and the risk of hypertension

The prime outcome for investigating the association of MTHFR 677 TT and hypertension risk compared with MTHFR 677 CC showed significant heterogeneity (I ² = 69.5%, P < .001) among the involved studies; the pooled OR was statistically significant: OR = 1.53 (95% CI: 1.30‐1.79; P < .001; Figure 4). Similarly, the T allele in the MTHFR C677T conferred a higher risk of hypertension (OR = 1.27 [95% CI: 1.17‐1.37; P < .001], I ² = 73.4%, P < .001; Figure 5). Moreover, compared with MTHFR 677 CC, MTHFR 677 CT indicated a significantly higher hypertension risk (OR = 1.22 [95% CI: 1.12‐1.32; P < .001], I ² = 42.2%, P = .001). MTHFR 677 TT showed a significantly greater risk of hypertension compared with CC + CT genotype (OR = 1.37 [95% CI: 1.19‐1.58; P < .001], I ² = 67.7%, P < .001). MTHFR 677 TT + CT also revealed a significantly higher risk of hypertension compared with CC genotype (OR = 1.31 [95% CI: 1.19‐1.43; P < .001], I ² = 58.2%, P < .001). Stratification by ethnicity implicated the correlations under all the genetic models in Asian and European. What's more, subgroup of the age range stratification showed the associations under all studied genetic models in adults with the risk of developing hypertension (Table 2).

Figure 4.

Forest plot of the MTHFR C677T associated with hypertension risk (under homozygous codominant model: TT vs CC)

Figure 5.

Forest plot of the MTHFR C677T associated with hypertension risk (under allelic model: T vs C)

Table 2.

Stratified analysis of associations of MTHFR C677T polymorphisms with hypertension

Subgroup	Dominant		Recessive		Homozygous codominant		Heterozygous codominant		Allelic model
	OR (95% CI)	P h	OR (95% CI)	P h	OR (95% CI)	P h	OR (95% CI)	P h	OR (95% CI)	P h
Overall	1.31 (1.19‐1.43)	<.001	1.37 (1.19‐1.58)	<.001	1.53 (1.30‐1.79)	<.001	1.22 (1.12‐1.32)	.001	1.27 (1.17‐1.37)	<.001
Ethnicity
Asian	1.45 (1.26‐1.67)	<.001	1.47 (1.16‐1.85)	<.001	1.70 (1.32‐2.19)	<.001	1.34 (1.18‐1.52)	.021	1.39 (1.22‐1.58)	<.001
European	1.18 (1.06‐1.33)	.025	1.34 (1.09‐1.65)	.002	1.43 (1.14‐1.79)	.001	1.12 (1.01‐1.23)	.158	1.18 (1.07‐1.30)	.001
American	1.05 (0.84‐1.32)	.101	0.98 (0.81‐1.20)	.825	0.95 (0.73‐1.22)	.373	1.06 (0.84‐1.35)	.111	1.01 (0.89‐1.15)	.281
African	1.53 (1.01‐2.32)	.364	4.01 (1.69‐9.56)	.481	4.45 (1.83‐10.82)	.654	1.29 (0.78‐2.13)	.252	1.63 (1.17‐2.26)	.632
Age range
Adults	1.32 (1.20‐1.45)	<.001	1.37 (1.18‐1.58)	<.001	1.54 (1.30‐1.81)	<.001	1.23 (1.14‐1.34)	<.001	1.27 (1.18‐1.38)	<.001
Children and Adolescents	1.01 (0.72‐1.40)	.892	1.41 (0.99‐2.01)	.489	1.37 (0.88‐2.13)	.59	0.88 (0.62‐1.25)	.762	1.13 (0.92‐1.40)	.963

Abbreviations: CI, confidence interval; MTHFR, methylenetetrahydrofolate reductase; OR, odds ratio; P _h, P‐value for heterogeneity test.

3.4. The associations of plasma Hcy level with hypertension

The SMD of Hcy between hypertensive group and control group indicated that Hcy has an effect on hypertension. A forest plot revealed the SMD outcomes in all the included studies, which was shown in Figure 6. In this meta‐analysis, there existed significant heterogeneity (I ² = 97.8%, P < .001) among the included studies. In these 42 studies,11, 26, 31, 32, 33, 34, 35, 49, 50, 51, 53, 57, 58, 59, 60, 68, 71, 83, 84, 85, 86, 87, 88, 89, 90, 91, 92, 93, 94, 95, 96, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106 the pooled mean Hcy level in hypertension group was 0.69 μmol/L (95% CI: 0.50‐0.87) greater than that in control group in random effect model (P < .001). We further performed a subgroup analysis for ethnicity, and all the results from European, Asian, and others indicated significant difference in Hcy level between hypertensive subjects and control subjects. Additionally, both Begg's and Egger's tests were conducted to assess the potential publication bias. There was no indication of any substantial publication bias for the Hcy‐hypertension association (data not shown).

Figure 6.

Forest plot of standardized mean difference (SMD) in Hcy levels between hypertensive patients and control subjects in included studies

3.5. Plausible relationship between Hcy level and hypertension via Mendelian randomization

By virtue of MTHFR C677T as an instrument variable for Hcy level, the Hcy per unit increase associated with the predicted OR of hypertension by direct or indirect measurement was shown in Figure 7. The Hcy concentrations were positively associated with the risk of developing hypertension. The estimated plausible OR was 1.32 (95% CI: 1.22‐1.49) for 5 μmol/L Hcy increment.

Figure 7.

The potential causal association between Hcy and hypertension risk was assessed via Mendelian randomization

4. DISCUSSION

Overall, this study suggested that MTHFR 677T allele was positively associated with plasma Hcy concentration. Furthermore, the mean Hcy concentration in hypertensive patients was higher than those without hypertension. The findings via Mendelian randomization method reinforced the assumption that elevated Hcy level had a plausible effect on the elevated risk of hypertension.

MTHFR C677T signifies the change in cysteine into thymine nucleotide at its corresponding position, which leads to the transition of alanine to valine in the MTHFR enzyme.107 Furthermore, the thermostability of the enzyme would reduce due to the variant in the MTHFR gene with the decreasing MTHFR enzyme activity, especially when the temperature is 37 ℃ or higher. The MTHFR enzyme activity in homozygous individuals is lower approximately 50%‐60% at 37°C and 65% at 46°C compared with the normal non‐mutated controls.108, 109 The decrease in the enzyme activity accelerates the conversion of 5,10‐methylenetetrahydrofolate to 5‐methylenetetrahydrofolate, which results in the increase in Hcy concentration in the mutated subjects. Therefore, the Hcy levels of homozygous subjects are greater than those of heterozygous mutated subjects, and the heterozygous subjects have slightly increased Hcy concentrations compared with the non‐mutated controls.109 The findings of our study demonstrated the assumption that the MTHFR C677T was strongly associated with Hcy concentrations in hypertensive patients. The homozygous subjects confer significantly higher Hcy levels than the heterozygous subjects in hypertensive group, which are in consistent with previous report in hypertensive patients.110

Recently, many studies have reported inconclusive results concerning the altering Hcy levels in hypertensive compared with normal subjects.97, 98, 100, 101 However, other studies suggested that the levels of Hcy in hypertensive were higher than in subjects without hypertension.71, 99, 102 This study analyzed the weighted mean difference of Hcy concentrations between hypertensive cases and normal controls in all included studies, which showed significant heterogeneity among different studies. Therefore, a random effects model was employed to estimate the pooled mean Hcy. Subsequently, the absolute pooled mean Hcy level in hypertensive was significantly higher than that in controls under the random effects model. The presence of the heterogeneity in these studies might be explained by the ethnicity of different regions and the coexistence of hypertension‐related disease.71, 97, 98, 99, 100, 101, 102, 103, 104, 105, 106 According to the fact that plasma Hcy could not only induce arteriolar constriction, increase sodium reabsorption, but also increase arterial stiffness,58, 111 Hcy may be involved in the pathogenesis of hypertension. Our previous findings have demonstrated that the MTHFR C677T is significantly associated with childhood obesity, and a strong relationship between Hcy level and obesity was emphasized by virtue of MTHFR C677T polymorphism.112, 113 However, the underlying role by MTHFR C677T polymorphism in leading to hypertension remains unclear, and further researches on genetics and animal experiments are needed to elucidate this hypothesis.

To the best of our knowledge, MTHFR C677T was first identified as a significant variant associated with hypertension in cardiovascular disease subjects in a Japanese population.23 Subsequently, another study indicated that MTHFR 677TT homozygosity is a predictor of essential hypertension in a Spanish male population.33 Although a large number of studies have investigated the association between MTHFR C677T and hypertension, the results are inconsistent in different population.10, 11, 54, 65 A recent meta‐analysis comprising thirty studies with 5207 patients and 5383 control subjects examined the association between MTHFR C677T polymorphism and hypertension, which revealed that carriers of the T allele and TT genotype significantly elevated the hypertension risk in all genetic models.114 Compared with that study, the present meta‐analysis had several advantages in the following aspects: First, the present study investigated 57 eligible studies with a total of 14 378 hypertensive cases and 25 795 controls, which provides more reliable conclusions. Second, we assessed not only the association between MTHFR C677T and hypertension, but also the relationship between MTHFR C677T and plasma Hcy concentration, which could provide more extra information about the effect of MTHFR C677T on hypertensive risk. Finally, by means of Mendelian randomization method, the effect size of MTHFR C677T on hypertension may be estimated more accurately. The reasons why inconsistent results exist in studies regarding MTHFR C677T and hypertension are still unclear, but a major reason for equivocal results might be the racial difference in the included studies. The frequency distribution of the 677TT genotype was highest in the Italian and Hispanics.115 Nevertheless, the homozygous frequency was very low for Black in Brazil and American.116, 117, 118 Furthermore, the study design flaws, small sample size, or other biases seem to be more common factors for the discrepancies comprising in different studies with regard to genetic studies.119, 120 Based on case‐control, cross‐sectional, or case‐cohort designed studies, the overall results of the present meta‐analysis indicated that MTHFR C677T is associated with hypertension and MTHFR TT genotype has a great effect on increasing the risk of hypertension. Additionally, sensitivity analysis indicated that an omission of studies that departure from HWE did not alter the magnitude of present effect, suggesting that the results were generally reliable and robust.

5. STRENGTHS AND LIMITATIONS

To date, this is the first meta‐analysis aggregating all the data available for evaluating the effect of MTHFR‐linked Hcy on hypertension. Therefore, it has provided the most sufficient data on this issue. However, several limitations should be mentioned: First, although applying Mendelian randomization approach could obtain an unbiased result because of avoiding the advent of confounding factors, the gene‐environment and gene‐gene interactions have affected the development of hypertension unavoidably. The result would be more precise if these factors are taken into account. Second, by virtual of different high‐performance liquid chromatography methods, the Hcy measurement showed discrepancies among different laboratories.121 Thus, Hcy measurement seems to contribute to variation among various studies applying different approaches. Third, the method of genotyping was used by different approaches, which could influence the results. Finally, only English and Chinese articles were included, which might result in publication bias, although Egger's and Begg's tests revealed no sign of the existence of pivotal publication bias with respect to the small sample size.

6. CONCLUSIONS

In summary, our results provided sufficient data supporting the assumption that the TT genotype in MTHFR C677T contributes to the susceptibility of hypertension. By means of Mendelian randomization approach, the findings are in support of the hypothesis that enhanced Hcy level is plausibly linked to increase the risk of hypertension. To some extent, the presence of gene‐environment interactions may result in the discordance of results comprising in the genetic association studies. Thus, prospective studies that investigate gene‐environment with large sample size are needed to further elucidate the genetic pathogenesis of hypertension.

DISCLOSURES

The authors declare there are no conflicts of interests.

AUTHOR CONTRIBUTIONS

LF and YQH conceptualized and designed the study. LF, YQH, and YL acquired the data. LF and YL analyzed and interpreted the data and drafted the manuscript. LF, YL, DL, SD, BW, and YQH involved in critical revision of the manuscript for important intellectual content. All authors have read and approved the final version of manuscript.

Fu L, Li Y‐N, Luo D, Deng S, Wu B, Hu Y‐Q. Evidence on the causal link between homocysteine and hypertension from a meta‐analysis of 40 173 individuals implementing Mendelian randomization. J Clin Hypertens. 2019;21:1879–1894. 10.1111/jch.13737