BR Biomedical Reports 2049-9434 2049-9442 D.A. Spandidos 10.3892/br.2014.302 br-02-05-0699 Articles Methylenetetrahydrofolate reductase C677T gene polymorphism and essential hypertension: A meta-analysis of 10,415 subjects YANGKE-MING1* JIAJIAN1* MAOLI-NA1 MENCHEN1 TANGKANG-TING2 LIYAN-YAN1 DINGHAI-XIA1 ZHANYI-YANG1 Department of Geriatrics, First Affiliated Hospital of Nanjing Medical University, Nanjing, Jiangsu 210029, P.R. China Department of Cardiology, First Affiliated Hospital of Nanjing Medical University, Nanjing, Jiangsu 210029, P.R. China Correspondence to: Dr Yi-Yang Zhan, Department of Geriatrics, First Affiliated Hospital of Nanjing Medical University, 300 Guangzhou Road, Nanjing, Jiangsu 210029, P.R. China, E-mail: yiyangzhan@sina.com

Contributed equally

 9 2014 25 06 2014 2 5 699 708 16 02 2014 05 06 2014 Copyright © 2014, Spandidos Publications 2014

The methylenetetrahydrofolate reductase (MTHFR) C677T gene polymorphism has been suggested to be associated with the risk of essential hypertension (EH), however, results remain inconclusive. To investigate this association, the present meta-analysis of 27 studies including 5,418 cases and 4,997 controls was performed. The pooled odds ratio (OR) and its corresponding 95% confidence interval were calculated using the random-effects model. A significant association between the MTHFR C677T gene polymorphism and EH was found under the allelic (OR, 1.32; 95% CI, 1.20–1.45; P=0.000), dominant (OR, 1.39; 95% CI, 1.25–1.55; P=0.000), recessive (OR, 1.38; 95% CI, 1.18–1.62; P=0.000), homozygote (OR, 1.59; 95% CI, 1.32–1.92; P=0.000), and heterozygote (OR, 1.32; 95% CI, 1.20–1.45; P=0.000) genetic models. A strong association was also revealed in subgroups, including Asian, Caucasian and Chinese. The Japanese subgroup did not show any significant association under all models. Meta-regression analyses suggested that the study design was a potential source of heterogeneity, whereas the subgroup analysis additionally indicated that the population origin may also be an explanation. Another subgroup analysis revealed that hospital-based studies have a stronger association than population-based studies, however, the former suffered a greater heterogeneity. Funnel plot and Egger’s test manifested no evidence of publication bias. In conclusion, the present study supports the evidence for the association between the MTHFR C677T gene polymorphism and EH in the whole population, as well as in subgroups, such as Asian, Caucasian and Chinese. The carriers of the 677T allele are susceptible to EH.

 methylenetetrahydrofolate reductase meta-analysis essential hypertension gene polymorphism Introduction

Essential hypertension (EH), as a polygenic disease with a high prevalence rate, is considered to develop from a complex interaction of diverse environmental conditions with multiple genetic factors in which the underlying genetic mechanism remains unknown (1,2).

A number of studies have been conducted to investigate the genetic components of EH. The genes involved in classical systems, including Renin-angiotensin-aldosterone system (RAAS), the central nervous system and the vascular-endothelial system, have been previously focused on (3,4) and the homocysteine (Hcy) pathway is an emerging target. The methylenetetrahydrofolate reductase (MTHFR) C677T mutation is one of the most common gene polymorphisms. Mapped to chromosomal region 1p36.3, the MTHFR gene spans a 2.2-kb length with 11 exons and 10 introns. The product, known as MTHFR, is a critical enzyme in Hcy metabolism (5). The C677C to T mutation in the catalyzing region of the MTHFR gene may induce the displacement of alanine by valine. This change may lead to the thermolability of the enzyme and the inhibition of MTHFR activity, thus decreasing the transformation from 5,10-methyltetrahydrofolate to 5-methyltetrahydrofolate, which acts as a co-substrate for the conversion from Hcy to methionine (6). The level of plasma Hcy, as a result, increases and may lead to certain pathological changes, including vascular endothelial injury, vascular smooth muscle proliferation and nitric oxide production inhibition (7), which may lead to the development of hypertension, as well as other diseases (8–11).

Studies on the MTHFR C677T gene polymorphism and EH association have been extensively performed previously, but the results obtained remain controversial. In 2003, a study conducted in Spain suggested that the MTHFR TT genotypes can predict the development of EH in males, and the association may be mediated by the elevation of the Hcy level (12). In 2008 and 2012, Lin et al (13) and Yin et al (14) investigated the association among the Chinese population in the Taichung and Guangxi areas, respectively, and the studies indicated that the MTHFR 677C to T genotypes may be significant risk factors for EH. However, in 2012, a case-control study regarding the association between four Hcy pathway gene variants, including MTHFR 677T and EH in Caucasians, indicated no significant association (15). More contrary results were found in a Japanese study by Nakata et al (16), which showed that the risk of hypertension significantly increased in subjects with the 677CC genotype, whereas the 677T gene polymorphism contributed to lower blood pressure.

Therefore, considering that the results of the studies carried out in the same ethnicity and different populations vary widely, the present meta-analysis of 10,415 subjects was conducted from 27 eligible studies to obtain a comprehensive conclusion on this association. The study demonstrated the association between the MTHFR C677T gene polymorphism and EH in the whole population. The association in different subgroup populations with reasonable clarification of the heterogeneity was shown clearly.

 Materials and methods Publication search and inclusion criteria

Electronic databases, including PubMed, Web of Science, China National Knowledge Infrastructure, WanFang and WeiPu were searched for relevant studies written in English or Chinese using the keywords: ‘Methylenetetrahydrofolate reductase or MTHFR’, ‘polymorphism’ and ‘hypertension’ (the last search was updated on April 1, 2013). Included studies were required to meet the following major criteria: i) Case-control or cross-sectional studies investigating the association between the MTHFR C677T polymorphism and EH; ii) present genotype counts of the patient and control groups; iii) diagnosis of EH patients based on the criteria of systolic blood pressure ≥140 mmHg or diastolic blood pressure ≥90 mmHg. Secondary hypertension was excluded; and (iv) for overlapping studies, the studies with the larger sample size were included.

 Data extraction

The following data from each eligible study were independently extracted in duplicate by two investigators: First author’s name, publication year, population origin, genotype counts of cases and controls, study design, sample size, genotyping method, male percentage and age of patients and controls. The two extractions were compared and any discrepancy was resolved through discussion until a consensus was reached.

 Statistical analysis

In the current meta-analysis, the association between the MTHFR C677T gene polymorphism and EH was revealed and analyzed by the odds ratio (OR) corresponding to the 95% confidence interval (CI) between cases and controls under five genetic models, which were the allelic (T/C), dominant (TT + CT/CC), recessive (TT/CT + CC), homozygote (TT/CC) and heterozygote models (CT/CC).

Between-study heterogeneity was estimated by χ2-based Q analysis, and significance was set at P<0.05. The variation caused by heterogeneity was also evaluated by the inconsistency index, I2 (I2=0–25%, no heterogeneity; 25–50%, moderate; 50–75%, large; and 75–100%, extreme heterogeneity). The DerSimonian and Laird random-effects model was applied whether or not heterogeneity among studies was observed. The Z test was used to determine the pooled OR, and significance was set at P<0.05. In addition, to examine the degree to which an individual study affects the overall estimate, sensitivity analyses were conducted by removing one study at a time and analyzing the change of the pooled effect. To explore the potential sources of heterogeneity, a random-effects meta-regression was performed. Additionally, subgroup analyses were implemented subsequent to stratifying the included studies according to population origin and study design.

Fisher’s exact test was used to assess the Hardy-Weinberg equilibrium (HWE). Potential publication bias was estimated by Begg’s funnel plot together with Egger’s linear regression test. Statistical analysis was performed using STATA 11.0 (StataCorp, College Station, TX, USA).

 Results Search results and study characteristics

A total of 53 potentially relevant papers were obtained by the literature search, of which 27 met the inclusion criteria (8,12–37). Of the 26 excluded studies, four studies were reviews or meta-analyses, five were overlapping studies, six lacked controls groups, three did not provide genotype counts, two were unrelated to MTHFR C677T and the EH risk and four studies deviated from the HWE. One family study and one prospective cohort study were also excluded (Fig. 1). The present meta-analysis involved 5,418 EH patients and 4,997 controls of Asian (Chinese, Japanese, Indian and Turks) and Caucasian (Australian, South Welsh, Argentinean, Austrian, Polish, Spanish and Czech) origin. Specifically, 6,188 subjects of 15 studies were Chinese from various parts of China, including Beijing, Tianjin, Shanghai, Fujian, Shenzhen, Yunnan, Ningxia, Guangxi, Mongolia, Xinjiang and Taiwan (Table I).

 Pooled effect under different genetic models

Among the 27 included studies, 14 separate studies showed a significant association between the MTHFR 677T gene polymorphism and EH risk (8,12–14,17–26), whereas the remaining 13 studies did not (15,16,27–37). Through meta-analysis, a significant association was revealed between the MTHFR 677T gene polymorphism and EH under the allelic (OR, 1.32; 95% CI, 1.20–1.45; P=0.000), dominant (OR, 1.39; 95% CI, 1.25–1.55; P=0.000), recessive (OR, 1.38; 95% CI, 1.18–1.62; P=0.000), homozygote (OR, 1.59; 95% CI, 1.32–1.92; P=0.000), and heterozygote (OR, 1.32; 95% CI, 1.20–1.45; P=0.000) genetic models (Figs. 2–5 and Table II). The overall pooled effect remained almost unchanged following the sensitivity analysis, which indicated the robustness of the results in the meta-analysis.

A subsequent subgroup analysis stratified by ethnicity was conducted in which a significant association between the MTHFR C677T gene polymorphism and EH was revealed in the Asian and Caucasian subgroups. The association was evident in all genetic models in the Asian subgroup (allelic: OR, 1.37; 95% CI, 1.21–1.54; P=0.000; dominant: OR, 1.47; 95% CI, 1.28–1.68; P=0.000; recessive: OR, 1.37; 95% CI, 1.12–1.68; P=0.002; homozygote: OR, 1.62; 95% CI, 1.27–2.05; P=0.000; and heterozygote: OR, 1.40; 95% CI, 1.25–1.56; P=0.000), whereas in the Caucasian subgroup an association was apparent under all models, except the heterozygote model (allelic: OR, 1.19; 95% CI, 1.05–1.36; P=0.007; dominant: OR, 1.19; 95% CI, 1.02–1.40; P=0.031; recessive OR, 1.42; 95% CI, 1.10–1.84; P=0.008; homozygote: OR, 1.53; 95% CI, 1.15–2.04; P=0.004; and heterozygote: OR, 1.13; 95% CI, 0.96–1.34; P=0.152). The association detected in the Asian group was stronger compared to the Caucasian group under all genetic models, with the exception of the recessive model (Figs. 2–5 and Table II).

 Heterogeneity analysis

The heterogeneity test was significant under the allelic (I2=54.7%, P=0.000), dominant (I2=35.6%, P=0.036) and homozygote (I2=39%, P=0.021) models. Following the subgroup analysis, there was no significant heterogeneity observed in the Caucasian subgroup, whereas it remained apparent in the Asian subgroup (Figs. 2–5 and Table II). A further subgroup analysis in the Asian group was considered by stratification of Chinese, Japanese and others (Indian and Turk). Heterogeneity was not evident in any subgroup under the different genetic models, indicating that population origin may be one explanation for the between-study heterogeneity (Figs. 6–7 and Table III).

The further subgroup analysis also revealed a strong association between the MTHFR C677T gene polymorphism and EH in the Chinese population (allelic: OR, 1.39; 95% CI, 1.26–1.55; P=0.000; dominant: OR, 1.52; 95% CI, 1.34–1.71; P=0.000; recessive: OR, 1.45; 95% CI, 1.24–1.70; P=0.000; and homozygote: OR, 1.71; 95% CI, 1.44–2.03; P=0.000). None of these genetic models suggested any association between MTHFR 677T and EH among the Japanese population (Figs. 6–7 and Table III).

In addition, to explore other potential sources of heterogeneity, a random-effects meta-regression analysis including all 27 studies was conducted under the allelic, dominant and homozygote models. The publication year, population origin (Chinese as 1; Japanese as 2; other Asian, including Indian and Turk, as 3; and Caucasian as 4), case sample size, control sample size, total sample size, the ratio of case and control sample, the percentage of males, age of cases and controls, study design [hospital-based (HB) as 1 and population-based (PB) as 2] and genotyping method [polymerase chain reaction (PCR)-restriction fragment length polymorphism as 1; others, including Sequenom and real-time fluorescence quantitative-PCR, as 2] were considered as independent variables. In contrast to the subgroup analysis, no association of population origin was observed under any genetic model. However, a significant and marginal association of study design was revealed under the dominant (P=0.035) and allelic genetic models (P=0.068), respectively. Therefore, another stratified analysis was conducted by organizing studies into HB or PB studies under the dominant and allelic genetic models. There was no significant heterogeneity in the PB subgroup under the genetic models (Table IV), whereas the HB subgroup yielded a significance under the allelic genetic model indicating that study design may play a role in heterogeneity. Additionally, an association between the MTHFR C677T gene polymorphism and EH was found to be stronger in the HB studies compared to the PB studies.

 Publication bias

Publication bias was assessed by Begg’s funnel plot and Egger’s test. The visible symmetrical funnel shape indicated no apparent evidence of publication bias in the meta-analysis, which is in accordance with the results of Egger’s test (allelic: P=0.122; dominant: P=0.378, recessive: P=0.078, homozygote: P=0.104; and heterozygote genetic model: P=0.550). There was no significant difference revealed, which suggests a low probability of publication bias (Fig. 8).

 Discussion

Meta-analysis is an efficient method to obtain a pooled effect of various studies with relatively small sample sizes and to achieve more comprehensive conclusions. Several meta-analyses that have been conducted in our Department of Geriatrics (First Affiliated Hospital of Nanjing Medical University, Nanjing, China) recently provided reliable results for the association between different gene polymorphisms and cardiovascular diseases (38,39). The present meta-analysis, which involved 5,418 cases and 4,997 controls from 27 studies carried out worldwide, indicated a strong association between the MTHFR C677T gene polymorphism and EH in the whole population under the allelic (OR, 1.32; 95% CI, 1.20–1.45; P=0.000), dominant (OR, 1.39; 95% CI, 1.25–1.55; P=0.000), recessive (OR, 1.38; 95% CI, 1.18–1.62; P=0.000), homozygote (OR, 1.59; 95% CI, 1.32–1.92, P=0.000), and heterozygote (OR, 1.32; 95% CI, 1.20–1.45; P=0.000) genetic models.

A significant association was detected in the Asian and Caucasian groups (P<0.05) suggesting that carriers of 677T or 677TT in these populations may suffer a significantly increased risk of developing EH. In the further subgroup analysis of the Asian population, stratified as Chinese, Japanese and others (Indian, Turk), the significant heterogeneity no longer existed in any subgroup suggesting that the population origin may contribute to heterogeneity. Future studies should pay increasing attention to the possible differences in genetic background of various populations. In addition, the pooled analysis of 6,188 Chinese subjects from 15 studies showed a strong association between the MTHFR C677T gene polymorphism and EH, revealing that it was a possible risk factor for EH in the Chinese population. There was no apparent association between the MTHFR C677T gene polymorphism and EH among the Japanese population. Considering that only three studies were included in the subgroup, a meta-analysis including more Japanese studies and future studies with larger samples could be conducted to confirm this result. Additionally, the study design was identified as an explanation for between-study heterogeneity in the meta-regression analysis, as well as the subgroup analysis. A stronger association together with greater heterogeneity was observed in HB studies compared to PB studies. This difference may be explained by the fact that control subjects from hospitals tend to have more complicated health states and confounding factors, whereas the controls selected from the general population would be more representative. The poor comparability between patients and controls in HB studies may contribute to the heterogeneity and exaggerate the association of the MTHFR 677T gene polymorphism and EH to a certain degree (40).

Two meta-analyses, by Qian et al (41) and Niu et al (42), were conducted previously in the whole population and in the Chinese population, respectively. Compared to these studies, the present meta-analysis had substantial advantages in the following aspects: First, only nine studies with 1,520 patients and 1,334 controls in the study by Niu et al (42) and 12 out of 25 studies in the study by Qian et al (41) were studies regarding the association between the MTHFR C677T gene polymorphism and EH. The remaining studies in the two meta-analyses assessed hypertension in pregnancy (HIP). Considering the multiple pathogenesis and complicated physiopathological changes in HIP cases, it is not suitable to analyze the pooled effect of the association between the MTHFR C677T gene polymorphism and hypertension with HIP patients. Therefore, with an increase of 18 studies from the latest meta-analysis regarding the association between MTHFR C677T and EH (42), the present study assessed 27 eligible studies with a total of 5,418 EH cases and 4,997 controls. This provides more reliable results with less clinical heterogeneity. One overlapping study and particular studies of secondary hypertension, which were included previously, were excluded. Second, compared to the unexplained heterogeneity of the Asian subgroup in the Qian et al (41) study, the heterogeneity of the Asian population no longer existed in the Chinese, Japanese and other (Indian and Turk) subgroups following stratified analysis in the present study. This suggested that the population origin may be one reason for heterogeneity. Third, in terms of meta-regression, while only a marginal association (P=0.082) of study design was observed in the Niu et al (42) study, the association in the present study was significant (P=0.035). This may be due to the relatively few included studies in the study by Niu et al, and studies involved in meta-regression should usually be ≥10 (43). Therefore, with stronger evidence, the present meta-analysis demonstrated the effect of study design in heterogeneity. Finally, the present meta-analysis involved a wide range of various populations and the pooled effect of 15 Chinese studies with 3,420 cases and 2,768 controls was also revealed in the subgroup analysis. For this reason, the study renewed the two previous meta-analyses with additional supplements and appropriate adjustments, and thus, to the best of our knowledge, becoming the most comprehensive meta-analysis regarding the association between MTHFR C677T and EH.

The present meta-analysis, although notably having advantages, such as large sample sizes, reliable statistical results and sound heterogeneity management, has certain limitations. First, large-scale studies on the association between the MTHFR gene polymorphism and EH are inadequate. Second, as discussed previously, the MTHFR C677T gene mutation is postulated to play a role in developing EH through generating hyperhomocysteinemia. However, the level of plasma Hcy is also influenced by certain environmental factors, including lack of folate and vitamin B6 in the diet, and is associated with plasma folate levels (8,26). The interaction between gene and environment was not taken into account. Therefore, future prospective cohort studies with large sample sizes should be performed to confirm these results and studies should include these considerations.

In conclusion, the present meta-analysis, including 27 studies with 10,415 subjects, indicates that the MTHFR C677T gene polymorphism is associated with increased EH risks in different populations as a whole, as well as in separate subgroups, such as Asian, Caucasian and Chinese. Carriers of the 677T allele are susceptible to EH. MTHFR C677T should therefore be tested for as a predictive screening for the identification of the individuals that may develop EH in the future. As an improvement on the previous studies with noteworthy reinforcement and development, the present results may potentially provide a stronger foundation for the identification of the mechanism through which the MTHFR gene may play a role in developing EH. More studies should be conducted to confirm the conclusion, and polymorphisms within the same linkage disequilibrium area should be focused on as well.

 Acknowledgements

The authors would like to thank all the subjects participating in the present study. The study was supported by the ‘six elite peaks’ project, funded by the Jiangsu Province Government, China (grant no. 2008-329).

 References MeinCACaulfieldMJDobsonRJMunroePBGenetics of essential hypertensionHum Mol Genet13R169R1752004 DominiczakAFNegrinDCClarkJSBrosnanMJMcBrideMWAlexanderMYGenes and hypertension: from gene mapping in experimental models to vascular gene transfer strategiesHypertension351641722000 LuptonSJChiuCLLindJMA hypertension gene: are we there yet?Twin Res Hum Genet142953042011 MenCTangKLinGLiJZhanYENOS-G894T polymorphism is a risk factor for essential hypertension in ChinaIndian J Biochem Bio481541572011 GoyettePPaiAMilosRGene structure of human and mouse methylenetetrahydrofolate reductase (MTHFR)Mamm Genome96526561998 FrosstPBlomHJMilosRA candidate genetic risk factor for vascular disease: a common mutation in methylenetetrahydrofolate reductaseNat Genet101111131995 TawakolAOmlandTGerhardMWuJTCreagerMAHyperhomocyst(e)inemia is associated with impaired endothelium-dependent vasodilation in humansCirculation95111911211997 HeuxSMorinFLeaRAOvcaricMTajouriLGriffithsLRThe methylentetrahydrofolate reductase gene variant (C677T) as a risk factor for essential hypertension in CaucasiansHypertens Res276636672004 LiYYMethylenetetrahydrofolate reductase C677T gene polymorphism and coronary artery disease in a Chinese Han population:a meta-analysisMetabolism618468522012 AlamMAHusainSANarangRChauhanSSKabraMVasishtSAssociation of polymorphism in the thermolabile 5, 10-methylene tetrahydrofolate reductase gene and hyperhomocysteinemia with coronary artery diseaseMol Cell Biochem3101111172008 Homocysteine Studies CollaborationHomocysteine and risk of ischemic heart disease and stroke: a meta-analysisJAMA288201520222002 Rodríguez-EsparragónFHernández-PereraORodríguez- PérezJThe effect of methylenetetrahydrofolate reductase C677T common variant on hypertensive risk is not solely explained by increased plasma homocysteine valuesClin Exp Hypertens252092202003 LinPTChengCHWeiJCHuangYCLow plasma pyridoxal 5′-phosphate concentration and MTHFR 677C-->T genotypes are associated with increased risk of hypertensionInt J Vitam Nutr Res7833402008 YinRXWuJZLiuWYAssociation of several lipid-related gene polymorphisms and blood pressure variation in the Bai Ku Yao populationAm J Hypertens259279362012 FowdarJYLasonMVSzvetkoALLeaRAGriffithsLRInvestigation of homocysteine-pathway-related variants in essential hypertensionInt J Hypertens20121909232012 NakataYKatsuyaTTakamiSMethylenetetrahydrofolate reductase gene polymorphism relation to blood pressure and cerebrovascular diseaseAm J Hypertens11101910231998 LiuHYMaPXuQBThe correlation between polymorphisms of N5,10 methylenetetrahydrofolate reductase and essential hypertension in Han population in NingxiaGuangdong Med J32197719802011 LiuHChenSMaPXuQThe association between gene polymorphisms of N5,10-methylenetetrahydrofolate reductase and essential hypertension in patients of Ningxia Hui nationalityTianjin Med J39109510982011 CaiYMGongWXLinkage study on methylenetetrahydrofolate reductase single nucleotide polymorphisms and hypertension in the elderly with rheumatoid arthritisChin J Birth Health and Heredity1714172009 LuoJWTangYChenHWuXYWuYNDengYLStudy on MTHFR C677T polymorphism in hypertensive subjects with blood stasis syndromeJ Beijing Univ Tradit Chin Med313513542008 XingXHuaQRelationships between the polymorphism of methylenetetrahydrofolate reductase gene C677T and hypertension, cardiac structure and functionMed J Chin PLA327417442007 TangYChenHWuXYLuoJWThe C677T point mutation of N5,10-methylenetetrahydrofolate reductase (MTHFR) and essential hypertensionMol Cardiol China72052072007 WangLGuoHLiYMTHFR gene C 677 T polymorphisms and variation of plasma homocysteine level in primary hypertensionTianjin Med J305795822002 IlhanNKucuksuMKamanDIlhanNOzbayYThe 677 C/T MTHFR polymorphism is associated with essential hypertension, coronary artery disease, and higher homocysteine levelsArch Med Res391251302008 MarkanSSachdevaMSehrawatBSKumariSJainSKhullarMMTHFR 677 CT/MTHFR 1298 CC genotypes are associated with increased risk of hypertension in IndiansMol Cell Biochem3021251312007 NgXBoydLDufficyLFolate nutritional genetics and risk for hypertension in an elderly population sampleJ Nutrigenet Nutrigenomics2182009 FridmanOPorcileRVanascoVJuncoMNGariglioLStudy on homocysteine levels and methylenetetrahydrofolate reductase gene variant (C677T) in a population of Buenos Aires CityClin Exp Hypertens305745842008 ZhanSGaoYYYinXElevated serum homocysteine, MTHFR gene mutataion and essential hypertension in ChineseChin J Hypertens821252000 LwinHYokoyamaTYoshiikeNPolymorphism of methylenetetrahydrofolate reductase gene (C677T MTHFR) is not a confounding factor of the relationship between serum uric acid level and the prevalence of hypertension in Japanese menCirc J7083872006 ZhangYWangHZhangXWangLWuGRelationship between homocysteine, methylene tetrahydrofolate reductase C677T polymorphisms and essential hypertension in Kazak nationality in XinjiangJ Clin Cardiol (China)285705732012 CaoZThe Association between methylenetetrahydrofolate reductase gene polymorphism and hypertensive elderly patient with acute myocardial infarctionChin J Gerontol32511851202012 LiXHuangWThe analysis of MTHFR gene polymorphism in patients with renal damage caused by hypertension and patients with renal parenchymal hypertensionJ Capital Univ Med Sci274975002006 HuRNiuGMZhaoSGZhangCYHuRLWangZGJiangMFThe association between gene polymorphisms of N5, 10-methylenetetrahydrofolate reductase and Mongolian patients with drimary hypertensionChin J Hypertension142742762006 LiuJWYeLLiuJLiXYStudy on homocysteine metabolism-related enzymes gene polymorphisms in elderly essential hypertension patients with peripheral arterial occlusive diseaseChin J Geriatr243323352005 HuiPNakayamaTMoritaACommon single nucleotide polymorphisms in Japanese patients with essential hypertension: aldehyde dehydrogenase 2 gene as a risk factor independent of alcohol consumptionHypertens Res305855912007 TylickiLFödingerMPuttingerHMethylenetetrahydrofolate reductase gene polymorphisms in essential hypertensionAm J Hypertens18144214482005 BenesPKankováKMuzíkJGrochLBenedikJMethylenetetrahydrofolate reductase polymorphism, type II diabetes mellitus, coronary artery disease, and essential hypertension in the Czech populationMol Genet Metab731881952001 LiYYα-Adducin Gly460Trp gene mutation and essential hypertension in a Chinese population: a meta-analysis including 10,960 subjectsPLoS One7e302142012 LiYPlasminogen activator inhibitor-1 4G/5G gene polymorphism and coronary artery disease in the Chinese Han population: a meta-analysisPLoS One7e335112012 Ruano-RavinaAPérez-RíosMBarros-DiosJMPopulation-based versus hospital-based controls: are they comparable?Gac Sanit226096132008 QianXLuZTanMLiuHLuDA meta-analysis of association between C677T polymorphism in the methylenetetrahydrofolate reductase gene and hypertensionEur J Hum Genet15123912452007 NiuWQYouYGQiYStrong association of methylenetetrahydrofolate reductase gene C677T polymorphism with hypertension and hypertension-in-pregnancy in Chinese: a meta-analysisJ Hum Hypertens262592672012 HigginsJThompsonSDeeksJAltmanDStatistical heterogeneity in systematic reviews of clinical trials: a critical appraisal of guidelines and practiceJ Health Serv Res Policy751612002

Flow diagram of the study selection process in the meta-analysis. Electronic databases, including PubMed, Web of Science, China National Knowledge Infrastructure (CNKI), WanFang and WeiPu, were searched for relevant studies that were written in English or Chinese, using the keywords ‘methylenetetrahydrofolate reductase or MTHFR’, ‘polymorphism’ and ‘hypertension’. A total of 53 potentially relevant studies were obtained by the literature search (last study was updated on April 1, 2013), of which 27 met the inclusion criteria. Of the 26 excluded studies, four were reviews or meta-analyses, five were overlapping studies, six lacked a controls group, three did not provide genotype counts, two were unrelated to MTHFR 677T and essential hypertension (EH) risk and four studies deviated from the Hardy-Weinberg equilibrium (HWE). One family study and one prospective cohort study were also excluded.

Forest plot of essential hypertension associated with the MTHFR C677T gene polymorphism under the allelic model (T/C). A significant association was revealed in the whole population (OR, 1.32; 95% CI, 1.20–1.45), Asian subgroup (OR=1.37; 95% CI, 1.21–1.54) and Caucasian subgroup (OR=1.19; 95% CI, 1.05–1.36). The heterogeneity test was significant in the whole population (I2=54.7%, P=0.000) and Asian subgroup (I2=60.4%, P=0.000), whereas the Caucasian subgroup showed no significant heterogeneity (I2=12.7%, P=0.333). MTHFR, methylenetetrahydrofolate reductase; OR, odds ratio; CI, confidence interval.

Forest plot of essential hypertension associated with the MTHFR C677T gene polymorphism under the dominant model (TT + CT/CC). A significant association was revealed in the whole population (OR, 1.39, 95% CI, 1.25–1.55), Asian subgroup (OR, 1.47, 95% CI, 1.28–1.68) and Caucasian subgroup (OR, 1.19, 95% CI, 1.02–1.40). A moderate heterogeneity was found in the whole population (I2=35.6%, P=0.036) and Asian subgroup (I2=39.9%, P=0.034), but did not existed in Caucasian subgroup (I2=0.0%, P=0.654). MTHFR, methylenetetrahydrofolate reductase; OR, odds ratio; CI, confidence interval.

Forest plot of essential hypertension associated with the MTHFR C677T gene polymorphism under the recessive model (TT/CT + CC). A significant association was revealed in whole population (OR, 1.38; 95% CI, 1.18–1.62), Asian subgroup (OR, 1.37; 95% CI, 1.12–1.68) and Caucasian subgroup (OR, 1.42; 95% CI, 1.10–1.84). Only a marginal heterogeneity was apparent in the Asian subgroup (I2=37.6%, P=0.047), whereas no significant heterogeneity was found in the whole population (I2=29.2%, P=0.079) and Caucasian subgroup (I2=2.6%, P=0.405). MTHFR, methylenetetrahydrofolate reductase; OR, odds ratio; CI, confidence interval.

Forest plot of essential hypertension associated with the MTHFR C677T gene polymorphism under the homozygote model (TT/CC). A significant association was revealed in the whole population (OR, 1.59; 95% CI, 1.32–1.92), Asian subgroup (OR, 1.62; 95% CI, 1.27–2.05) and Caucasian subgroup (OR, 1.53; 95% CI, 1.15–2.04). The heterogeneity test was significant in the whole population (I2=39.0%, P=0.021) and Asian subgroup (I2=47.1%, P=0.011), whereas the Caucasian subgroup showed no significant heterogeneity (I2=10.2%, P=0.351). MTHFR, methylenetetrahydrofolate reductase; OR, odds ratio; CI, confidence interval.

Forest plot of essential hypertension associated with the MTHFR C677T gene polymorphism in the Asian population under the allelic model (T/C) stratified by different populations, including Chinese, Japanese and others (Indian and Turk). A significant association was revealed in the Asian population (OR, 1.37; 95% CI, 1.21–1.54), as well as the Chinese (OR, 1.39; 95% CI, 1.26–1.55) and the others (OR, 2.3; 95% CI, 1.56–3.40) subgroups. No significant association was found in the Japanese subgroup (OR, 0.97; 95% CI, 0.76–1.25). A significant heterogeneity was detected in the Asian group (I2=60.4%, P=0.000), but no longer existed in the Chinese (I2=31.4%, P=0.118), Japanese (I2=55.2%, P=0.107) and others (I2=17.9%, P=0.270) subgroup, through subgroup analysis. MTHFR, methylenetetrahydrofolate reductase; OR, odds ratio; CI, confidence interval.

Forest plot of essential hypertension associated with the MTHFR C677T gene polymorphism in the Asian population under the dominant model (TT + CT/CC) stratified by different populations; Chinese, Japanese and others (Indian and Turk). A strong association was revealed in the Asian population (OR, 1.47; 95% CI, 1.28–1.68), as well as the Chinese (OR, 1.52; 95% CI, 1.34–1.71) and others (OR, 2.22; 95% CI, 1.28–3.83) subgroups. No significant association was found in the Japanese subgroup (OR, 1.05; 95% CI, 0.78–1.41). Moderate heterogeneity was apparent in the Asian population (I2=39.9%, P=0.034), whereas no significant heterogeneity was detected in the Chinese (I2=13.7%, P=0.300), Japanese (I2=32.6%, P=0.227) and others (I2=43.6%, P=0.183) subgroups. MTHFR, methylenetetrahydrofolate reductase; OR, odds ratio; CI, confidence interval.

Begg’s funnel plot for studies of the association between the MTHFR C677T polymorphism and EH under the dominant genetic model (CT + TT/CC). Begg’s funnel plot was used to assess the publication bias. The vertical axis is log OR and the horizontal axis is the standard error (s.e.) of the log OR. The visible symmetrical funnel shape shows no apparent evidence of the publication bias in the meta-analysis. MTHFR, methylenetetrahydrofolate reductase; EH, essential hypertension; OR, odds ratio.

Characteristics of 27 included studies of the association between the methylenetetrahydrofolate reductase C677T gene polymorphism and essential hypertension (EH).

First author (Refs.) Year Origin Case Control EH/control Study design Geno-typing Male, % EH age, years Control age, years
CC CT TT CC CT TT
Yin (14) 2012 Chinese 244 358 68 322 309 51 670/682 PB case-control PCR-RFLP 0.494 M49.9±16.2F48.3±15.8 M48.8±11.8F47.1±11.2
Zhang (30) 2012 Chinese 128 53 8 117 41 7 189/165 PB case-control PCR-RFLP 0.729 50.3±9.7 49.9±7.7
Cao (31) 2012 Chinese 65 105 53 49 68 30 223/147 HB case-control PCR-RFLP 0.608 78.9±7.1 74.5±8.2
Liu (17) 2011 Chinese 58 70 27 74 47 19 155/140 HB case-control PCR-RFLP 0.695 44.1±11.3 41.7±11.4
Liu (18) 2011 Chinese 54 59 33 61 39 12 146/112 HB case-control PCR-RFLP 0.531 48.3±16.0 45.1±16.2
Cai (19) 2009 Chinese 77 44 9 31 7 1 130/39 HB case-control PCR-RFLP - Total 61.8±9.8
Luo (20) 2008 Chinese 260 151 31 138 51 6 442/195 HB case-control PCR-RFLP 0.538 62.8±10.9 62.0±10.3
Lin (13) 2008 Chinese 19 27 4 73 44 6 50/123 HB case-control PCR-RFLP 0.584 60.1±10.8 59.0±8.7
Xing (21) 2007 Chinese 202 309 184 182 222 105 695/509 HB case-control PCR-RFLP 0.489 48.9±12.6 48.5±13.1
Tang (22) 2007 Chinese 139 93 20 138 51 6 252/195 HB case-control PCR-RFLP 0.539 63.0±12.8 61.0±10.2
Li (32) 2006 Chinese 18 6 2 21 7 2 26/30 HB case-control PCR-RFLP 0.536 60.3±11.3 55.5±13.5
Hu (33) 2006 Chinese 55 39 16 61 42 12 110/115 PB case-control Sequenom 0.458 56.7±10.6 55.3±10.3
Liu (34) 2005 Chinese 29 45 26 31 50 19 100/100 PB case-control PCR-RFLP 0.500 67.2±4.2 66.3±4.4
Wang (23) 2002 Chinese 17 51 37 14 23 9 105/46 HB case-control PCR-RFLP 0.165 59.0~73.0 56.0~62.0
Zhan (28) 2000 Chinese 44 68 15 62 84 24 127/170 PB case-control PCR-RFLP 0.468 57.9±9.9 50.6±10.5
Hui (35) 2007 Japanese 83 129 49 104 123 44 261/271 PB case-control RTFQ-PCR 0.662 51.1±5.6 51.5±8.6
Lwin (29) 2006 Japanese 39 58 19 64 117 38 116/219 PB cross-sectional PCR-RFLP 1.000 Total 53.1±8.9
Nakata (16) 1998 Japanese 63 91 19 65 83 36 173/184 HB case-control PCR-RFLP - - -
Markan (25) 2007 Indian 105 40 8 105 28 0 153/133 HB case-control PCR-RFLP 0.531 47.7±12.4 46.2±10.8
Ilhan (24) 2008 Turk 36 32 10 72 26 2 78/100 HB case-control RTFQ-PCR 0.573 57.2±10.0 57.5±11.1
Fowdar (15) 2012 Australian 170 174 33 175 183 35 377/393 PB case-control PCR-RFLP 0.452 63.1±10.9 61.0±10.5
Ng (26) 2009 South Welsh 14 14 10 40 32 8 38/80 PB case-control PCR-RFLP 0.432 66.0~83.0 65.0~90.0
Fridman (27) 2008 Argentine 15 21 4 39 38 9 40/86 HB cross-sectional PCR-RFLP 0.468 Total 45.2±16.8
Tylicki (36) 2005 Austrian, Polish 40 39 11 42 38 10 90/90 HB case-control PCR-RFLP 0.367 59.4±1.3 52.1±1.6
Heux (8) 2004 Australia 87 125 35 105 119 25 247/249 HB case-control PCR-RFLP 0.562 Total 55.0±11.0
Rodriguez (12) 2003 Spanish 83 115 34 95 100 20 232/215 PB case-control PCR-RFLP 0.734 58.6±9.0 57.5±8.1
Benes (37) 2001 Czech 73 93 27 86 106 17 193/209 HB case-control PCR-RFLP 0.761 55.6±11.5 54.9±11.6

Age is expressed as means ± standard deviation or the range. PCR-RFLP, polymerase chain reaction-restriction fragment length polymorphism; RTFQ-PCR, real-time fluorescence quantitative PCR; Sequenom, Sequenom MassARRAY system; PB, population-based studies; HB, hospital-based studies; M, male; F, female; Total, total sample size; --, information unavailable.

Summary of the meta-analysis results of the association between the MTHFR C677T gene polymorphism and essential hypertension under the different genetic models.

Genetic model Group Pool OR (95% CI) P-value Heterogeneity Studies number Case size, n Control size, n
I2, % P-value
Allelic Whole population 1.32 (1.20–1.45) 0.000a 54.7 0.000a 27 5418 4997
Asian 1.37 (1.21–1.54) 0.000a 60.4 0.000a 20 4201 3675
Caucasian 1.19 (1.05–1.36) 0.007a 12.7 0.333 7 1217 1322
Dominant Whole population 1.39 (1.25–1.55) 0.000a 35.6 0.036a 27 5418 4997
Asian 1.47 (1.28–1.68) 0.000a 39.9 0.034a 20 4201 3675
Caucasian 1.19 (1.02–1.40) 0.031a 0.0 0.654 7 1217 1322
Recessive Whole population 1.38 (1.18–1.62) 0.000a 29.2 0.079 27 5418 4997
Asian 1.37 (1.12–1.68) 0.002a 37.6 0.047a 20 4201 3675
Caucasian 1.42 (1.10–1.84) 0.008a 2.6 0.406 7 1217 1322
Homozygote Whole population 1.59 (1.32–1.92) 0.000a 39.0 0.021a 27 5418 4997
Asian 1.62 (1.27–2.05) 0.000a 47.1 0.011a 20 4201 3675
Caucasian 1.53 (1.15–2.04) 0.004a 10.2 0.351 7 1217 1322
Heterozygote Whole population 1.32 (1.20–1.45) 0.000a 9.7 0.321 27 5418 4997
Asian 1.40 (1.25–1.56) 0.000a 12.7 0.296 20 4201 3675
Caucasian 1.13 (0.96–1.34) 0.152 0.0 0.878 7 1217 1322

P<0.05.

MTHFR, methylenetetrahydrofolate reductase; OR, odds ratio; CI, confidence interval.

Summary of the subgroup analysis results under different genetic models in different Asian populations.

Genetic model Group Pool OR (95% CI) P-value Heterogeneity Studies number Case size, n Control size, n
I2, % P-value
Allelic Chinese 1.39 (1.26–1.55) 0.000a 31.4 0.118 15 3420 2768
Japanese 0.97 (0.76–1.25) 0.838 55.2 0.170 3 550 674
Others (Indian and Turk) 2.30 (1.56–3.40) 0.000a 17.9 0.270 2 231 233
Dominant Chinese 1.52 (1.34–1.71) 0.000a 13.7 0.300 15 3420 2768
Japanese 1.05 (0.78–1.41) 0.755 32.6 0.227 3 550 674
Others (Indian and Turk) 2.22 (1.28–3.83) 0.004a 43.6 0.183 2 231 233
Recessive Chinese 1.45 (1.24–1.70) 0.000a 0.0 0.737 15 3420 2768
Japanese 0.85 (0.51–1.40) 0.523 60.3 0.081 3 550 674
Others (Indian and Turk) 8.58 (2.20–33.53) 0.002a 0.0 0.632 2 231 233
Homozygote Chinese 1.71 (1.44–2.03) 0.000a 0.0 0.512 15 3420 2768
Japanese 0.88 (0.50–1.56) 0.668 61.8 0.073 3 550 674
Others (Indian and Turk) 11.30 (2.85–44.79) 0.001a 0.0 0.742 2 231 233

P<0.05.

OR, odds ratio; CI, confidence interval.

Summary of the subgroup analysis results of all included studies stratified by study design.

Subgroup Allelic genetic model Dominant genetic model Studies number Case size, n Control size, n
OR (95% CI), P-value Heterogeneity OR (95% CI), P-value Heterogeneity
HB 1.45 (1.26–1.66), P=0.000a I2=57.9%, P=0.002a 1.55 (1.34–1.79), P=0.000a I2=31.6%, P= 0.104 17 3198 2587
PB 1.17 (1.05–1.31), P=0.006a I2=31.6%, P=0.156 1.22 (1.06–1.42), P=0.007a I2=24.0%, P=0.222 10 2220 2410
Overall 1.32 (1.20–1.45), P=0.000a I2=54.7%, P=0.000a 1.39 (1.25–1.55), P= 0.000a I2=35.6%, P=0.036a 27 5418 4997

P<0.05.

OR, odds ratio; CI, confidence interval; HB, hospital-based studies; PB, population-based studies.