Methylenetetrahydrofolate Reductase C677T Polymorphism and Recurrent Pregnancy Loss Risk in Asian Population: A Meta-analysis

Abstract

The C677T polymorphism of the methylenetetrahydrofolate reductase (MTHFR) gene was implicated to be associated with thrombophilia due to its role in catalyzing the formation of 5-methylenetetrahydrofolate, a co-substrate for the conversion of homocysteine to methionine. Several case–control studies were investigated MTHFR C677T polymorphism as risk for recurrent pregnancy loss (RPL). These studies rendered contradictory results, some indicating that the polymorphism is associated with the risk of RPL whereas others concluded there is no association. To shed light on these inconclusive findings, a meta-analysis of all available studies published from Asian population relating the C677T polymorphism to the risk of RPL was conducted. The following electronic databases were searched without language restrictions: PubMed, Google Scholars, Elsevier and Springer Link up to December, 2015. Meta-analysis was performed using MetaAnalyst and Mix version 1.7. Meta-analysis results suggested that MTHFR C677T polymorphism contributed to the increased RPL risk in Asian population using all five genetic models (for T vs. C: OR 1.35, 95 % CI 1.09–1.68, p = 0.009; for TT + CT vs. CC: OR 1.44, 95 % CI 1.14–1.82, p = 0.006; for CT vs. CC: OR 1.39, 95 % CI 1.07–1.8, p = 0.01; for TT vs. CC: OR 1.79, 95 % CI 1.23.2.6, p = 0.007; for TT vs. CT + CC: OR 1.61, 95 % CI 1.02–2.56, p = 0.04). In conclusion, this meta-analysis demonstrates a strong association between the MTHFR C677T variant and RPL in Asian population and raising the importance of the use of folate in its treatment and prevention.

Introduction

Recurrent pregnancy loss (RPL) or spontaneous abortions (SA) is defined as three or more consecutive miscarriages [1–3]. RPL is a major concern in gynecology, affecting about 1–5 % of couples [2, 4, 5] and frequently accompanied by maternal morbidity as well as a considerable psychological burden. The risk of recurrence increases with the maternal age and number of successive losses [6, 7]. It is a multifactorial disorder caused very often by genetic abnormalities (gene mutations and abnormal embryonic karyotypes), endocrine disorders, uterine anatomy anomalies, infectious or immunologic factors, alcohol use and chemical exposure [8–11]. Despite intense anatomic, endocrinologic, and immunologic screening efforts, up to 50 % of RPL remain unexplained [12].

Published studies showed that the inherited thrombophilic polymorphisms are significant risk factors for obstetric complications, such as pre-eclampsia, placental abruption, stillbirth and fetal growth restriction [13–16]. RPL is also speculated to be associated with inherited thrombophilia that encompass diverse conditions including the thermolabile variant of the methylenetetrahydrofolate reductase (MTHFR) [17, 18]. MTHFR is a key enzyme in folate/homocysteine pathway, which catalyzes the reduction of 5,10-methylenetetrahydrofolate to 5-methyltetrahydrofolate (5-THF), and then methionine synthase catalyzed the conversion of 5-THF and homocysteine to methionine and tetrahydrofolate. Under the condition of folate deficiency and/or hypo functional MTHFR facilitate the conversion of 5,10-methylene THF to less 5-methyl THF, and causes less conversion of homocysteine to methionine, which may result in abnormal DNA methylation and DNA stand breaks etc. The gene encoding MTHFR has been mapped to chromosomal region 1p36.3. There were about 40 different genetic polymorphisms of MTHFR and out of which C677T variant is most studied and clinically important polymorphism. C677T missense mutation (rs 1,801,133; Ala 222 Val) at nucleotide 677 results in an enzyme that is thermolabile and exhibits reduced activity compared with the wild type. This mutation is associated with hyperhomocysteinemia [19–21]. TT MTHFR homozygotes are predisposed to increased plasma homocysteine levels, particularly in individuals with low folate [22, 23]. Hyperhomocysteinemia has been implicated in premature vascular disease [24], venous thrombosis [25] and unexplained early pregnancy loss [23, 26]. Hyperhomocysteinemia caused by the C677T polymorphism has been associated with coronary artery disease, venous thrombosis and complications of pregnancy i.e. RPL.

Numerous studies have focused on the relationship between MTHFR C677T polymorphism and RPL risk [27–30], but the conclusions remain controversial. The discrepancies among studies may be ascribed to the relatively small sample size in each investigation as well as ethnicity difference. Therefore, present meta-analysis was carried out by using genotype data from all eligible investigations to provide a more precise evaluation of the association of MTHFR C677T polymorphisms with RPL susceptibility in Asian population.

Methods

The articles were identified by searching PubMed, Google Scholar, Elsevier and Springer Link databases up to December, 2015 using following terms: ‘‘methylenetetrahydrofolate reductase’’, “MTHFR”, “C677T” and “Recurrent pregnancy loss”, “RPL”. A cited reference search of the retrieved articles was carried out, and publications were also identified by reviewing their bibliographies.

Data Extraction

Following data from each publication were extracted: author name; country of origin; selection and characteristics of cases and controls; demographic information; racial descent of the study population; numbers of eligible and genotyped cases and controls; and numbers of cases and controls for each MTHFR genotype.

Inclusion–Exclusion Criteria

The following criteria were used to include published studies: (a) Studies must have a case–control and must be published as full papers, (b) Authors must investigate RPL patients and healthy control subjects, (c) Authors must provide information on genotype/allele numbers of the MTHFR C677T polymorphism or sufficient data to calculate these. The major reasons for exclusion of studies were (1) only case studied, (2) review papers, editorial, letter to editor and (3) containing overlapping data and (4) no enough data to estimate OR with 95 % CI.

Statistical Analysis

The meta-analysis examined the overall association for the allele contrast (T vs. C), homozygotes (TT vs. CC), heterozygote/co-dominant (CT vs. CC), recessive (TT vs. CT + CC) and dominant (TT + CT vs. CC) models. The effect of association was indicated as odds ratio (OR) with the corresponding 95 % confidence interval (CI). The pooled OR was estimated using fixed effects (FE) [31] and random effects (RE) [32] models [33]. Sensitivity analysis performed by exclusion of the studies in which control population was not in Hardy–Weinberg equilibrium, studies with small sample sizes and higher p value.

For the assessment of publication bias the Begg’s test (funnel plot method) and the Egger regression asymmetry test was used. The significance of the intercept was determined with the t-test suggested by Egger. p < 0.05 was considered representative of statistically significant publication bias [34, 35]. All analyses were performed using the computer program MIX version 1.7 [36]. A p value less than 0.05 was considered statistically significant, and all the p values were two sided.

Results

Characteristics of Included Studies

One hundred two (102) articles were retrieved after search of PubMed, Google Scholar, Elsevier and Springer Link databases. After screening the titles and abstracts of all retrieved articles, 37 articles were excluded. Then 65 full texts were reviewed and 12 articles were further excluded. Another 24 articles from remaining 53 articles were again excluded because studied population was not Asian. Finally, 29 studies were included in present meta-analysis [11, 15, 16, 21, 27–30, 37–57] (Fig. 1; Table 1).

Fig. 1.

Flow chart shows study selection procedure. Twenty-five case–control studies were included in present meta-analysis

Table 1.

Characteristics of twenty-five studies included in the present meta-analysis

Study	Country	Control	Case	Reference
Brener et al. (1999)	Israel	106	76	Thromb. Haemost. 82, 6–9
Lissak et al. (1999)	Israel	18	41	Am J Obstet Gynecol 181, 126–130
Wang et al. (2002)	China	119	62	Lancet 18, 291–293.
Kumar et al. (2003)	India	24	24	J Obstet Gynaecol 23, 55–58
Li et al. (2004)	China	50	57	Zhonghua Yi Xue Yi Chuan Xue Za Zhi 21, 39–42
Makino et al. (2004)	Japan	76	85	Am J Reprod Immunol 52, 60–66
Wang et al. (2004)	China	82	147	Zhonghua Fu Chan Ke Za Zhi 39, 238–241
Guan et al. (2005)	China	117	127	Zhonghua Yi Xue Yi Chuan Xue Za Zhi 22, 330–333
Kobashi et al. (2005)	Japan	174	38	Semin Thromb Hemost 31, 266–271
Song et al. (2005)	China	56	50	Zhonghua Wei Chan Yi Xue Za Zhi 8, 160–164
Mtiraoui et al. (2006)	Behrain	200	200	Reproduction 131, 395–401
Wang et al. (2006)	China	82	147	International Journal of Gynecology and Obstetrics (2006) 92, 264–265
Vettriselvi et al. (2008)	India	120	104	J Obstet Gynaecol Res 34, 301–306
Govindaiah et al. (2009)	India	140	140	Clin Biochem 42, 380–386
Mukhopadhyay et al. (2009)	India	80	84	Genet Test Mol Biomarker
Abu-Asab et al. (2011)	Palestine	402	329	Volume 13, Number 6, 2009
Jeddi-Tehrani et al. (2011)	Iran	100	100	American Journal of Reproductive Immunology 66 (2011) 149–156
Settin et al. (2011)	Egypt	136	70	Am J Reprod Immunol 67, 251–255
Dissanayke et al. (2012)	Srilanka	171	200	Genetic testing and Molecular Biomarkers, 15, 887–892
Nair et al. (2012)	India	140	106	J Obstet Gynaecol Res Vol. 38, No. 9: 1168–1176
Ozdemir et al. (2012)	Turkey	106	327	Reproductive Sciences, 19(2), 210–215.
Torabi et al. (2012)	Iran	100	100	Genetic testing and Molecular Biomarker, 16, 279–28
Zonouzi et al. (2012)	Iran	50	89	ISRN Obst Gynec. Article ID 945486, 6
Kaur et al. (2012)	India	593	107	J Reprod Infertil 13(2), 89–94
Parveen et al. (2013)	India	300	200	ISRN Obstet Gynecol, 2012;94
Cao et al. (2014)	China	166	82	Genes Nutr 402–407
Yousefian et al. (2014)	Iran	204	116	Iran Red Crescent Med J. 16(7), e16763
Farahmand et al. (2015)	Iran	330	350	J Matern Fetal Neonatal Med. doi:10.3109/14767058.2015.1044431
Vanill et al. (2015)	India	15	15	J Clin Diagn Res 9(2), 15–18

All included studies were published between 1999 and 2013. All these twenty-five studies were performed in different countries like-Behrain [47], China [38–40, 42, 44, 46, 54], Egypt [52], India [11, 21, 29, 30, 41, 48, 49, 57], Iran [28, 51, 53, 55, 56], Israel [15, 37], Japan [43, 45], Palestine [50], SriLanka [16], Turkey [27]. Smallest sample size was 24 [41] and largest sample size was 329 [50]. Seven studies did not show any association between C677T polymorphism and RPL risk [21, 28, 29, 37, 43, 47, 52], remaining twenty-two studies showed significant association. In twenty-nine studies, total cases were 3725 with CC (1971), CT (1325) and TT (429), and controls were 4105 with CC (2545), CT (1218), and TT (342). In controls, genotypes percentage of CC, CT and TT were 61.99, 29.67 and 8.33 % respectively. In total cases, percentage of CC, CT and TT genotypes were 52.91, 35.57 and 11.51 % respectively. Frequencies of CC and CT genotypes were highest in both cases and controls (Table 2). Number of C and T alleles were also calculated and presented in Table 2. Control population of ten studies was not in HWE [15, 28, 29, 38, 39, 44, 46, 47, 51, 52].

Table 2.

The distributions of MTHFR C677T genotypes and allele frequencies of RPL disease cases and controls in Asian studies

Study ID	Genotype						Alleles
	CC		CT		TT		C		T
	Case	Control	Case	Control	Case	Control	Case	Control	Case	Control
Brener et al. (1999)	54	86	8	9	14	11	116	36	181	31
Lissak et al. (1999)	17	7	20	7	4	4	54	28	21	15
Wang et al. (2002)	13	43	33	71	16	5	59	65	157	81
Kumar et al. (2003)	18	22	6	2	0	0	42	6	46	2
Li et al. (2004)	16	25	32	20	9	5	64	50	70	30
Makino et al. (2004)	33	29	42	32	10	15	108	62	90	62
Wang et al. (2004)	49	43	78	16	20	23	176	118	102	62
Guan et al. (2005)	13	19	59	73	55	25	85	169	111	123
Kabashi et al. (2005)	5	67	30	82	3	25	40	36	216	132
Song et al. (2005)	36	40	2	12	12	4	74	26	92	20
Mtiraoui et al. (2006)	156	92	30	47	14	61	342	58	231	169
Wang et al. (2006)	49	43	78	34	20	5	176	118	120	44
Vettriselvi et al. (2008)	86	98	15	19	3	3	187	21	215	25
Govindaiah et al. (2009)	111	112	25	28	4	0	247	33	252	28
Mukhopadhyay et al. (2009)	75	78	6	2	3	0	156	12	158	2
Abu-Asab et al. (2011)	145	182	151	177	33	43	441	217	541	263
Jeddi-Tehrani et al. (2011)	43	66	42	25	15	9	128	72	157	43
Settin et al. (2011)	40	67	26	68	4	1	106	34	202	70
Dissanayke et al. (2012)	158	142	39	27	3	2	355	45	311	31
Nair et al. (2012)	75	118	26	21	5	1	176	36	257	23
Ozdemir et al. (2012)	145	79	130	27	52	0	420	234	185	27
Torabi et al. (2012)	43	66	42	25	15	9	128	72	157	43
Zonouzi et al. (2012)	53	27	30	22	6	1	136	42	76	24
Kaur et al. (2013)	86	463	16	109	5	21	188	26	1035	151
Parveen et al. (2013)	110	196	70	90	20	14	290	110	482	118
Cao et al. (2014)	53	29	83	43	30	10	189	101	143	63
Yousefian et al. (2014)	96	63	90	43	18	10	282	169	126	63
Farahmand et al. (2015)	180	230	114	85	36	35	474	545	186	155
Vanill et al. (2015)	13	13	2	2	0	0	28	28	2	2

Meta-analysis

Significant association was detected between the MTHFR C677T polymorphism and the susceptibility to RPL in Asian population in all the genetic models using random effect model (for T vs. C: OR 1.38, 95 % CI 1.08–1.75, p = 0.009; CT vs. CC: OR 1.39, 95 % CI 1.07–1.8, p = 0.01; for TT + CT vs. CC: OR 1.47, 95 % CI 1.10–1.9, p = 0.006; for TT vs. CC: OR 1.95, 95 % CI 1.2–3.2, p = 0.007; for TT vs. CT + CC: OR 1.61, 95 % CI 1.02–2.56, p = 0.04) (Table 3; Figs. 2, 3, 4). Significant association was also found in fixed effect models using all genetic models (for T vs. C: OR 1.28, 95 % CI 1.17–1.4,, p < 0.0001; for TT + CT vs. CC: OR 1.32, 95 % CI 1.18–1.47, p < 0.0001; for TT vs. CC: OR 1.43, 95 % CI 1.18–1.7, p = 0.0002; for TT vs. CT + CC: OR 1.29, 95 % CI 1.08–1.5; for CT vs. CC: OR 1.32, 95 % CI 1.17–1.5, p = 0.004) (Table 3).

Table 3.

Summary estimates for the odds ratio (OR) of MTHFR C677T in various allele/genotype contrasts, the significance level (p value) of heterogeneity test (Q test), and the I² metric and publication bias p value (Egger Test)

Genetic Models	Fixed effect OR (95 % CI), p	Random effect OR (95 % CI), p	Heterogeneity p value (Q test)	I2 (%)	Publication Bias (p of Egger’s test)
Allele Contrast (T vs. C)	1.28 (1.17–1.4), <0.0001	1.35 (1.09–1.68), 0.009	<0.0001	85.07	0.24
Co-dominant (CT vs. CC)	1.32 (1.17–1.5), <0.0001	1.39 (1.07–1.8), 0.01	<0.0001	72.43	0.45
Homozygoote (TT vs. CC)	1.43 (1.18–1.7), 0.0002	1.79 (1.23.2.6), 0.007	<0.0001	77.98	0.02
Dominant (TT + CT vs. CC)	1.32 (1.18–1.47), <0.0001	1.44 (1.14–1.82), 0.006	<0.0001	80.8	0.16
Recessive (TT vs. CT + CC)	1.29 (1.08–1.5), 0.004	1.61 (1.02–2.56), 0.04	<0.0001	78.1	0.06

Fig. 2.

Forest plot for the association between MTHFR C677T polymorphism and RPL for allele contrast model (T vs. C) with random effect model in Asian population

Fig. 3.

Forest plot for the association between MTHFR C677T polymorphism and RPL for homozygote model (TT vs. CC) with random effect model in Asian population

Fig. 4.

Forest plot for the association between MTHFR C677T polymorphism and RPL for dominant model (TT + CT vs. CC) with random effect model in Asian population

Heterogeneity and Sensitivity Analysis

A true heterogeneity existed between studies for allele contrast (p_{heterogeneity} < 0.0001, Q = 160.70, I² = 85.07 %, t² = 0.30, z = 2.59), genotype homozygote (p_{heterogeneity} < 0.0001, Q = 104.46, I² = 77.98 %, t² = 1.0, z = 2.65), dominant (p_{heterogeneity} < 0.0001, Q = 1274.97, I² = 80.8 %, t² = 0.337, z = 2.72) and recessive (p_{heterogeneity} < 0.0001, Q = 105.05, I² = 78.1 %, t² = 0.87, z = 2.05) comparisons.

Control population of ten studies [15, 28, 29, 38, 39, 44, 46, 47, 51, 52] were not in HW equilibrium and exclusion of these ten studies decreased heterogeneity (p < 0.0001, I² = 72.84 %) and increased OR (OR 1.54, 95 % CI 1.2–1.96). However, exclusion of three studies with small sample size, less than 50 [37, 41, 45] did not decreased heterogeneity (p_{heterogeneity} < 0.0001, I² = 86.65 %). Similarly exclusion of seven studies with very high p value [21, 28, 29, 37, 39, 48, 50] did not decrease heterogeneity (p_{heterogeneity} < 0.0001, I² = 88.38 %) but increased odds ratio (OR 1.59, 95 % CI 1.14–2.22).

Publication Bias

Except homozygote model, p values of Begg’s and Egger’s tests were more than 0.05 (Begg’s p = 0.84, Egger’s p = 0.24 for T vs. C; Begg’s p = 0.02, Egger’s p = 0.02 for TT vs. CC; and Begg’s p = 0.27, Egger’s p = 0.45 for CT vs. CCA; Begg’s p = 0.27, Egger’s p = 0.16 for TT + CT vs. CC; Begg’s p = 0.04, Egger’s p = 0.06 for TT vs. CT + CC) (Table 3). The funnel plots were also symmetrical (Fig. 5).

Fig. 5.

a Forest plot for the association between MTHFR C677T polymorphism and RPL for allele contrast model (T vs. C) with fixed effect model, b funnel plot precision versus OR (T vs. C), c standard error versus OR (T vs. C) in Asian studies

Discussion

Malnutrition and malabsorption of folate and vitamin B₁₂ or inherited MTHFR deficiency, may result in hyperhomocysteinemia. C677T polymorphism in MTHFR gene was associated with elevated plasma homocysteine level, increased risk of arterial stiffness [58] and women with elevated total homocysteine concentrations showed a significant association with defective chorionic villous vascularization [59–61]. In embryonic development during pregnancy, the embryo survives and grows by stimulating its own blood supply through angiogenesis. A good exchange between fetus and mother is necessary to ensure normal fetal growth; therefore, impaired chorionic villous vascularization may result in embryonic death leading to miscarriage [62]. It has been also suggested that independent of homocysteinemia, association between C677T polymorphism and RPL was due to interference with red cell folate metabolism [18, 63].

Hyperhomocysteinemia is known to cause direct endothelial injury through increased oxidative stress to induce increased blood pressure and impairment in endothelial synthesis of vasodilatory substances, to increase the expression of procoagulants and to increase platelet aggregation [51]. This may cause thrombophilia, which is an important factor in increasing the risk of RPL in mothers. Both MTHFR polymorphism and hyperhomocysteinemia have been reported to predispose to placental vasculopathy associated with intrauterine growth retardation, abruption placentae and pre-eclampsia [64].

Meta-analysis is the statistical analysis of a large collection of analysis results for the purpose of integrating the findings and it is a powerful tool for systematic review of a focused topic in the literature that provides a quantitative estimate for the effect of a treatment intervention or exposure [65]. Because of the large sample sizes, meta-analysis has more statistical power than a single study to obtained reliable result. Several large-scale meta-analyses combining data from multiple studies have been published investigating the association between MTHFR C677T polymorphism and various disease/disorders such as—Down syndrome [66], Neural Tube defects [67], cleft lip with/without palate [68], congenital heart defects [69], stroke [70], diabetes mellitus [71], Alzheimers disease [72], schizophrenia [73] and cancer [74].

Three meta-analysis studies have been reported in an effort to draw conclusions on the association of MTHFR C677T polymorphism with RPL [18, 75, 76] but the information is incomplete on the Asian population, hence present meta-analysis was conducted on previously published case–control reports on Asian population. Ren and Wang [75] found in their meta-analysis that MTHFR C677T mutation was not related with RPL, except in Chinese population. They covered 28 studies in their meta-analysis, and most of the studies were conducted in the Caucasian population, among these only five studies conducted in China resulted in positive relation of that mutation with RPL. This large meta-analysis clearly showed the importance of the ethnicity in single nucleotide mutations. Cao et al. [76] conducted a meta-analysis (3559 RPL cases and 5097 healthy controls) and reported overall random-effects odds ratios (ORs) as 1.68 (95 % CI, 1.32–2.13) for TT versus CC genotypes, and 1.35 (95 % CI, 1.04–1.76) for TT + CT genotype combined versus total CC genotypes.

The quality of meta-analysis is compromised by presence of heterogeneity. However to minimize this limitation, author tried to use appropriate inclusion and exclusion criteria, performed sensitivity analysis and included samples only from single ethnic population (Asian) to reduce selection bias and to lower heterogeneity [76, 77] but failed to minimize the heterogeneity. The heterogeneity might be due to different sampling method and variations in genetic background of the subjects etc.

The current meta-analysis has few limitations to be addressed. First, the sample size of cases from some eligible studies is relatively limited (<100). The relative limited cases may have compromised statistical power. Second, the overall results were based on unadjusted ORs; while a more precise evaluation should be adjusted by potentially confounding factors, including age, gender, body mass index, smoking status, drink abuse, and environmental factors. Third, heterogeneity was observed, so the results should be interpreted cautiously. Fourth, only one gene was considered, other genes involved in folate metabolism should be considered for a more comprehensive understanding of the exact role of the folate pathway in RPL susceptibility. Finally, the effect of gene–gene and gene–environment interactions was not fully addressed in the meta-analysis due to the lack of sufficient data. Along with limitations, present meta-analysis had strengths also like-absence of publication bias and inclusion of larger number of studies of single ethnic population.

In conclusion, results of present meta-analysis suggested that the women having MTHFR C677T polymorphism may have an increased risk of RPL. This finding supports the hypothesis that folic acid may play a role in the etiology of RPL. Large and rigorous case–control studies that investigate gene–gene and gene–environment interactions need to be performed before conclusive claims about the genetics of RPL.

Compliance with Ethical Standards

Conflict of interest

None.