Methylenetetrahydrofolate reductase C677T polymorphism and the risk of unexplained recurrent pregnancy loss: A meta-analysis

RECURRENT PREGNANCY LOSS Methylenetetrahydrofolate reductase C677T polymorphism and the risk of unexplained recurrent pregnancy loss: A meta-analysis Aiguo Ren, M.D, Ph.D., and Juan Wang, M.D. Institute of Reproductive and Child Health, Peking University Health Science Center, Beijing, China Objective: To investigate the association between methylenetetrahydrofolate reductase (MTHFR) C677T polymorphism and the risk of unexplained recurrent pregnancy loss (URPL). Design: Meta-analysis of published case-control studies of the MTHFR C677T variant and URPL risk. Setting: A research institution in China. Patient(s): Women with URPL. Intervention(s): None. Main Outcome Measure(s): Odds ratios (ORs) and 95% confidence intervals (CIs) for TT versus CC genotype, CT versus CC genotype, TT versus CT and CC genotype combined, and for T versus C allele. Result(s): Twenty-six studies with 2120 URPL cases and 2949 controls were included. Overall random-effects ORs of 1.49 (95% CI, 1.12 2.00) for TT versus CC genotype, 1.40 (95% CI, 1.11 1.77) for TT versus CT and CC genotype combined, and 1.21 (95%CI, 1.04-1.40) for T versus C allele were found. Stratified analysis showed that significant strong associations between MTHFR C677T polymorphism and URPL were present only in the five Chinese studies (OR 2.96 for TT versus CC genotype; OR 2.30 for TT versus CT CC genotype; OR 1.73 for T versus C allele), but not in any other studies, including the studies conducted in the European countries. Conclusion(s): The MTHFR C677T mutation is not a genetic risk factor for URPL except in a Chinese population. (Fertil Steril 2006;86:1716 22. 2006 by American Society for Reproductive Medicine.) Key Words: Methylenetetrahydrofolate reductase, polymorphism, unexplained recurrent pregnancy loss, miscarriage, abortion, meta-analysis Recurrent pregnancy loss occurs in about 1 2% of all reproductive-aged women (1). Among women with this problem, specific causes cannot be identified in 50% or more of them by any evaluation, including parental karyotypes, hysterosalpingography or hysteroscopy, and antiphospholipid antibody testing, and these cases are therefore considered as unexplained recurrent pregnancy loss (URPL) (2). An elevated plasma homocysteine concentration has been suggested as a risk factor for URPL, but the results of epidemiological studies have been inconsistent (3 6). The common polymorphism (677C T) in the gene encoding methylenetetrahydrofolate reductase (MTHFR), which converts dietary folate to its active cofactor in homocysteine catabolism, results in reduced enzyme activity and, subsequently, in elevated homocysteine concentrations (7, 8). Received January 26, 2006; revised and accepted May 12, 2006. Reprint requests: Aiguo Ren, M.D., PhD., Institute of Reproductive and Child Health, Peking University Health Science Center, 38 Xueyuan Rd, Haidian District, Beijing 100083, China (FAX: 86-10-8280-1141; E-mail: renag@sohu.com). Associations between the genetic variant MTHFR C677T and URPL have been reported. In a previous meta-analysis, which included six studies with a total number of 599 cases and 498 controls, Nelen et al. (9) obtained an overall odds ratio (OR) of 1.40 (95% confidence interval [CI], 1.0 2.0) for the MTHFR TT versus CT and CC genotype combination. Since the publication of this meta-analysis in 2000, many more new studies have appeared. Adding these and studies already published but presumably missed by the search strategy used by previous investigators more than triples the number of cases available for analysis compared with the previous meta-analyses. We have therefore carried out an updated meta-analysis using a comprehensive literature search. MATERIALS AND METHODS Identification and Eligibility of Relevant Studies PubMed MEDLINE and EMBASE searches (last updated December 2005) used combinations of the terms ((miscarriage OR abortion OR (pregnancy loss) OR (fetal loss) OR (fetal death) OR (pregnancy wastage)) AND ((methylenetet- 1716 Fertility and Sterility Vol. 86, No. 6, December 2006 0015-0282/06/$32.00 Copyright 2006 American Society for Reproductive Medicine, Published by Elsevier Inc. doi:10.1016/j.fertnstert.2006.05.052

rahydrofolate reductase) OR mthfr)). CBMDisc, the Chinese Biomedical Literature Database, which is the main Chinese medical literature retrieval system, was also used to search pertinent literature in the Chinese language. The search was complemented with a review of the bibliographies of retrieved papers and review articles. To minimize heterogeneity and facilitate the interpretation of our results, the following inclusion criteria were specified: [1] case-control design with women both with and without URPL being genotyped; [2] recurrent pregnancy loss, that is, two or more consecutive pregnancy losses; [3] unexplained pregnancy losses, that is, pregnancy losses with known causes being excluded. If a study included both cases with primary and secondary recurrent pregnancy loss and provided subgroup analysis, we included in the analysis only cases with primary pregnancy loss. Whenever the complete pertinent information required for the quantitative synthesis of the data was unavailable, we contacted study investigators to obtain additional information and clarifications. When the study populations overlapped, we generally retained for the meta-analysis only the study with the most extensive data to avoid duplication. We set no language restrictions. TABLE 1 Characteristics of the studies included in the meta-analysis. Author Publication year Country No. of cases No. of controls No. of pregnancy losses, gestation Nelen et al. (10) 1997 The Netherlands 185 113 2, 17 weeks Grandone et al. (11) 1998 Italy 94 150 2, 17 weeks Quere et al. (4) 1998 France 100 100 3, early Brenner et al. (12) 1999 Israel 76 106 3 in first trimester or 2 in second trimester or 1 intrauterine fetal death Holmes et al. (13) 1999 United Kingdom 173 67 3, 74% 12 weeks, 1% 12 weeks, 25% both Kutteh et al. (14) 1999 United States 50 50 3, 92% in first trimester Lissak et al. (15) 1999 Israel 41 18 2, 16 weeks Foka et al. (16) 2000 Greece 80 100 2, first or second trimester Murphy et al. (17) 2000 Ireland 40 540 2, at any point Wramsby et al. (18) 2000 Sweden 36 a 69 3, 22 weeks Pihusch et al. (19) 2001 Germany 102 128 2, 25 weeks Carp et al. (21) 2002 Israel 108 82 3, 26 weeks Dilley et al. (22) 2002 United States 60 92 3, at any point Wang et al. (25) 2002 China 62 119 2, 28 weeks Hohlagschwandtner 2003 Austria 145 101 3, 20 weeks et al. (27) Kumar et al. (6) 2003 India 24 24 3, 22 weeks Pauer et al. (28) 2003 Germany 69 b 122 2, 12 weeks Dossenbach-Glaninger 2004 Austria 49 48 2, early et al. (29) Li et al. (30) 2004 China 57 50 2, 28 weeks Makino et al. (31) 2004 Japan 85 b 76 2, 10 weeks Wang et al. (32) 2004 China 147 82 3, 20 weeks Couto et al. (33) 2005 Brazil 88 88 3, not stated Gerhardt et al. (34) 2005 Germany 34 b 277 3, 12 weeks Guan et al. (35) 2005 China 127 117 3, 12 weeks Kobashi et al. (36) 2005 Japan 38 174 2, not stated Song et al. (38) 2005 China 50 56 Not stated, not stated a Cases with primary pregnancy loss. b Cases with two (or three) or more early pregnancy losses. Fertility and Sterility 1717

Eligible Studies We identified 31 published reports of potentially eligible studies: 26 in English and five in Chinese (4, 6, 10 38). Of the 31 potentially eligible studies, we excluded two studies where no genotype frequencies were given for cases and/or for controls (20, 37). We further excluded two studies (24, 26) because the study population was probably also included in subsequent studies. In two reports, the study subjects were overlapped, but the earlier one (12) gave more extensive genotype data so the latter one (23) was then excluded. Thus, 26 reports were included in the meta-analysis (Table 1). Three studies included women with cases with two (or three) or more early fetal losses ( 10 or 12 weeks gestation) and cases with at least one late fetal losses ( 10 or 12 gestation weeks), and we kept only cases with two (or three) or more early fetal losses to make case definitions consistent (28, 31, 34). In an additional study (18), both cases with primary pregnancy loss (no pregnancy had gone beyond 22 weeks) and cases with secondary pregnancy loss (patient had given birth to at least one child) were included, and we used only the cases with primary pregnancy loss. Data Extraction For each study, we extracted information on authors, year of publication, country where the study was conducted, number of pregnancy losses, gestation weeks used to define recurrent pregnancy loss, and criteria to exclude known causes. Data extraction was performed by the two authors independently. In case of disagreement, consensus was reached by joint review of the study. Analysis We estimated the OR for URPL for various genetic contrasts. We evaluated the contrasts of TT homozygotes and CT heterozygotes separately against wild-type (CC homozygotes). We also compared the TT genotype with CT and CC genotype combined and TT and CT combined with the CC genotype. Finally, we examined a contrast of alleles (T versus C). For each eligible study group, we estimated the OR for URPL between the contrasted genotypes and the 95% CI. Between-study heterogeneity for the OR was evaluated using the 2 -based Q statistic. We then synthesized the data across studies using both fixed-effects and random-effects methods. Fixed-effects modeling assumes that the OR is the same across the considered studies and the observed variability is due to chance (random error) alone. The OR of each study group is weighted by the inverse of its variance. Random effects assume that there may be genuine diversity in the results of various studies and thus they estimate and incorporate in the calculations a between-study variance. When there is no observed between-study heterogeneity, the fixed- and random-effects estimates coincide. In the presence of betweenstudy heterogeneity, random effects usually provide wider confidence intervals. When appropriate, stratified analyses by ethnical-geographical regions were performed to explore possible ethnic predispositions to URPL. We also analyzed the data stratified by number of pregnancy losses used to define URPL. All analyses were performed in RevMan (version 4.2) (39) and in SPSS (Chicago) (version 11.5). A two-sided P.05 was considered statistically significant. RESULTS The 26 studies were conducted in 15 countries: five in China, three each in Germany and Israel, two each in Austria, Japan, and the United States, and one each in other nine countries. All studies were published between 1997 and 2005. The number of cases included in the studies varied from 24 to 185, with a mean of 82, and the number of controls varied from 18 to 540, with a mean of 113. In 11 studies, three or more pregnancy losses were used to define recurrent pregnancy loss, and in another 13 studies, two or more pregnancy losses were used. In one study, cases were defined as three or more first trimester fetal losses, two or more second trimester fetal losses, or at least one intrauterine fetal death. Another study did not clearly mention the number of pregnancy losses used to define recurrent pregnancy loss. Gestation in defining pregnancy loss also varied. One study used pregnancy losses within 10 weeks of gestation, some other studies used 12 weeks, 16 weeks, 17 weeks, 20 weeks, 22 weeks, 25 weeks, or 28 weeks. Two studies did not have limitations on gestational weeks or trimesters. Five studies did not provide information on gestational weeks in selecting women with recurrent pregnancy loss. The majority of the studies excluded known causes for recurrent pregnancy losses. However, exclusion criteria varied. Almost all studies excluded women with uterine anomalies, abnormal chromosomes in either partner, positive antiphospholipid antibodies, hormone problems, thromboembolic diseases, endocrine disorders (diabetes, thyroid dysfunction), and/or genital tract infection. Meta-Analysis: Main Results A total of 2120 URPL cases and 2949 healthy controls were included in the analyses. Overall, there was large and significant between-study heterogeneity in the magnitude of the observed association between the presence of the C677T polymorphism and URPL in the TT versus CC (P.04), CT versus CC (P.007), TT versus CT CC (P.05), TT CT versus CC (P.009), and T versus C (P.004) comparisons, Thus, random-effects estimates would be more appropriate for data synthesis, and fixed-effects estimates are not shown. In the overall analysis, women with the TT genotype had a 49% (OR 1.49; 95% CI, 1.12 2.00) and 40% (OR 1.40; 95% CI, 1.11 1.77) higher risk of developing URPL compared with women with the CC genotype (Fig. 1) and the CT and CC genotype combined (Fig. 2), respectively. Heterozygous (CT) genotype was not associated with an elevated risk of URPL 1718 Ren and Wang MTHFR and recurrent pregnancy loss Vol. 86, No. 6, December 2006

FIGURE 1 MTHFR C677T polymorphism and the risk of URPL: contrast of TT genotype against CC genotype. Each study is shown by an OR estimate with the corresponding 95% CI. The random-effects pooled OR is shown. The horizontal axis is plotted on a log scale. Studies are ordered according to year of publication. (OR 1.13; 95% CI, 0.91 1.41) compared with wild-type (CC). When TT and CT genotypes were combined, the association became weaker and statistically insignificant (OR 1.21; 95% CI, 0.99 1.48). In addition, the T allele was associated with a 20% increase in risk compared with the C allele (OR 1.21; 95% CI, 1.04 1.40). Analysis by Ethnic-Regional Groups Stratified analysis by geographical region (China, all other countries including European countries, or just European countries) removed much of the heterogeneity. Analysis with the five studies conducted in China showed strong and significant associations between the TT genotype and the risk of URPL (OR 2.96, for TT versus CC; OR 2.30 for TT versus CT CC; and OR 1.91 for TT CT versus CC genotype) and T allele (OR 1.73), except for the CT versus CC comparison (OR 1.48) (Table 2). However, when all other 21 studies that were conducted in 14 countries or just the 12 studies conducted in nine European countries were analyzed, neither of the comparisons showed an asso- FIGURE 2 MTHFR C677T polymorphism and the risk of URPL: contrast of TT genotype against CT and CC genotype combination. Each study is shown by an OR estimate with the corresponding 95% CI. The random-effects pooled OR is shown. The horizontal axis is plotted on a log scale. Studies are ordered according to year of publication. Fertility and Sterility 1719

TABLE 2 Summary estimates for the odds ratio (OR) of C677T polymorphism in various genotype contrasts. Genotype or allele Cases/controls OR (95% CI) Five Chinese studies: TT vs. CC 239/232 2.96 (1.88 4.66) CT vs. CC 331/362 1.48 (0.83 2.67) TT vs. CT CT 443/424 2.30 (1.59 3.31) TT CT vs. CC 443/424 1.91 (1.39 2.62) T vs. C 886/848 1.73 (1.41 2.12) Twenty-one non-chinese studies: TT vs. CC 790/1087 1.20 (0.88 1.64) CT vs. CC 1,177/1,745 1.03 (0.83 1.27) TT vs. CT CT 1,677/2,525 1.22 (0.95 1.57) TT CT vs. CC 1,356/1,968 1.07 (0.87 1.31) T vs. C 2,712/3,936 1.08 (0.92 1.26) Twelve European studies: TT vs. CC 547/712 1.25 (0.86 1.82) CT vs. CC 815/1239 0.87 (0.72 1.07) TT vs. CT CT 1,107/1,815 1.30 (0.94 1.79) TT CT vs. CC 944/1,390 0.94 (0.78 1.14) T vs. C 1,888/2,780 1.04 (0.89 1.21) ciation between TT genotype and the risk of URPL or between T allele and the risk of URPL, although the sample sizes were more than twice the sample size of the five Chinese studies (Table 2). Analysis by Number of Pregnancy Losses When the five Chinese studies were excluded, analysis of the 11 studies that defined recurrent pregnancy loss as two or more pregnancy losses and of the nine studies that defined current pregnancy loss as three or more pregnancy losses showed no effect of the C677T polymorphism on URPL except for the TT versus CT CT genotype comparison for the three or more pregnancy loss subgroup (OR 1.55; 95% CI, 1.09 2.21). DISCUSSION Our meta-analysis of 26 studies gives an estimate of about a 40% (TT versus CT CC genotype) to 50% (TT versus CC genotype) greater risk of URPL associated with the MTHFR polymorphism. However, there was statistically significant heterogeneity in the results of different studies. The heterogeneity could be largely explained by race or by the geographical region where the study was conducted. Further analysis of the five Chinese studies showed much stronger associations between MTHFR C677T polymorphism and the risk of URPL (OR 2.96 for TT versus CC genotype; OR 2.30 for TT versus CT CC genotype). In contrast, no association was observed in any of the contrasts from the pooled 21 other studies conducted in 14 countries or the subset of 12 studies that wre conducted in nine European countries. This means that the observed overall association was largely attributable to the five Chinese studies. It has been reported that the Chinese population has higher TT genotype and T allele frequencies compared with other ethnic groups (40). In this analysis, the overall TT genotype frequencies for the five Chinese studies were 25% (112/443) and 15% (62/424) among the URPL cases and controls, respectively, whereas the corresponding figures for the other 21 studies conducted in 14 other countries were 13% (218/ 1,677) and 11% (280/2,525). Most noticeably, in one of the Chinese studies by Guan et al. (35), the TT genotype frequency among URPL cases reached as high as 43%. Even the control group had an incidence of 21% of the TT genotype. The genotype frequency for the cases is higher than that in one of our recent studies on cleft lips and palate, in which 32% of mothers of babies with cleft lips and palate carried the TT genotype (41). The higher frequency of the TT genotype may reflect the genetic predisposition to URPL among (mostly northern) Chinese women, who are at higher risk for neural tube defects (42). The strong positive studies reported from China could also represent publication bias, especially given the high levels of scientific activity around MTHFR variants in the country. Publication bias is an important problem, particularly in relation to meta-analyses of genetic association studies. Negative results, especially smaller ones, may not be submitted for publication, let alone accepted, rendering any systematic review of published results misleading. All five 1720 Ren and Wang MTHFR and recurrent pregnancy loss Vol. 86, No. 6, December 2006

Chinese studies gave strong positive associations, with ORs ranging from 2.30 to 3.51 for the TT and CC contrast, and no negative reports have been seen. This supports the speculation on publication bias. Although we invested considerable effort in searching for published studies of genetic associations and found considerably more than previous metaanalyses, our estimates of effect are still likely to represent overestimation through publication bias. Lower blood folate levels have been reported to be associated with the risk of URPL (5, 43), and blood homocysteine concentrations associated with MTHFR polymorphism depend on blood folate levels (44). Thus folate status may confound the association between MTHFR polymorphism and the risk of UPRL. However, most of the studies included in this analysis do not provide blood folate data, hindering further analysis stratified by blood folate status. As with all meta-analyses, there are limitations that are present in this analysis, which mainly relate to the lack of clinical homogeneity between the subjects in the included studies. Although our inclusion criteria ensured that the selected studies were broadly similar, it was not possible to eliminate all sources of heterogeneity. For instance, studies were inconsistent in defining recurrent pregnancy loss. Some studies included women with three or more pregnancy losses; some included women with two or more pregnancy losses. Further, the number of gestational weeks used in defining pregnancy loss was different among studies, ranging from 10 weeks to third trimester. In addition, although virtually all studies excluded women with known causes of recurrent pregnancy loss, the capacity to screen those causes and the techniques used to do the screening may differ. However, we consider it unlikely that these differences introduced significant bias into the analysis since the selection is not based on genotypes. In conclusion, although we observed an overall 50% increase in the risk of URPL associated with the MTHFR C677T polymorphism, the association was only present in a Chinese population, not in other populations as a whole. Therefore, no conclusion on a causal relationship can be drawn from the available data. REFERENCES 1. Stirrat GM. Recurrent miscarriage. I. Definition and epidemiology. Lancet 1990;336:673 5. 2. ACOG Committee on Practice Bulletins. ACOG practice bulletin. Management of recurrent pregnancy loss. Int J Gynaecol Obstet 2002;78: 179 90. 3. Wouters MG, Boers GH, Blom HJ, Trijbels FJ, Thomas CM, Borm GF, et al. Hyperhomocysteinemia: a risk factor in women with unexplained recurrent early pregnancy loss. Fertil Steril 1993;60:820 5. 4. Quere I, Bellet H, Hoffet M, Janbon C, Mares P, Gris JC. A woman with five consecutive fetal deaths: case report and retrospective analysis of hyperhomocysteinemia prevalence in 100 consecutive women with recurrent miscarriages. Fertil Steril 1998;69:152 4. 5. Nelen WL, Blom HJ, Steegers EA, den Heijer M, Thomas CM, Eskes TK. Homocysteine and folate levels as risk factors for recurrent early pregnancy loss. Obstet Gynecol 2000;95:519 24. 6. Kumar KS, Govindaiah V, Naushad SE, Devi RR, Jyothy A. Plasma homocysteine levels correlated to interactions between folate status and methylene tetrahydrofolate reductase gene mutation in women with unexplained recurrent pregnancy loss. J Obstet Gynaecol 2003;23: 55 8. 7. Frosst P, Blom HJ, Milos R, Goyette P, Sheppard CA, Matthews RG, et al. A candidate genetic risk factor for vascular disease: a common mutation in methylenetetrahydrofolate reductase. Nat Genet 1995;10:111 3. 8. Harmon DL, Woodside JV, Yarnell JW, McMaster D, Young IS, McCrum EE, et al. The common thermolabile variant of methylene tetrahydrofolate reductase is a major determinant of mild hyperhomocysteinaemia. Q J Med 1996;89:571 7. 9. Nelen WL, Blom HJ, Steegers EA, den Heijer M, Eskes TK. Hyperhomocysteinemia and recurrent early pregnancy loss: a meta-analysis. Fertil Steril 2000;74:1196 9. 10. Nelen WL, Steegers EA, Eskes TK, Blom HJ. Genetic risk factor for unexplained recurrent early pregnancy loss. Lancet 1997;350:861 [letter]. 11. Grandone E, Margaglione M, Colaizzo D, d Addedda M, D Andrea G, Pavone G, et al. Methylene tetrahydrofolate reductase (MTHFR) 677T C mutation and unexplained early pregnancy loss [letter]. Thromb Haemost 1998;79:1056 7. 12. Brenner B, Sarig G, Weiner Z, Younis J, Blumenfeld Z, Lanir N. Thrombophilic polymorphisms are common in women with fetal loss without apparent cause. Thromb Haemost 1999;82:6 9. 13. Holmes ZR, Regan L, Chilcott I, Cohen H. The C677T MTHFR gene mutation is not predictive of risk for recurrent fetal loss. Br J Haematol 1999;105:98 101. 14. Kutteh WH, Park VM, Deitcher SR. Hypercoagulable state mutation analysis in white patients with early first-trimester recurrent pregnancy loss. Fertil Steril 1999;71:1048 53. 15. Lissak A, Sharon A, Fruchter O, Kassel A, Sanderovitz J, Abramovici H. Polymorphism for mutation of cytosine to thymine at location 677 in the methylenetetrahydrofolate reductase gene is associated with recurrent early fetal loss. Am J Obstet Gynecol 1999;181:126 30. 16. Foka ZJ, Lambropoulos AF, Saravelos H, Karas GB, Karavida A, Agorastos T, et al. Factor V Leiden and prothrombin G20210A mutations, but not methylenetetrahydrofolate reductase C677T, are associated with recurrent miscarriages. Hum Reprod 2000;15:458 62. 17. Murphy RP, Donoghue C, Nallen RJ, D Mello M, Regan C, Whitehead AS, et al. Prospective evaluation of the risk conferred by factor V Leiden and thermolabile methylenetetrahydrofolate reductase polymorphisms in pregnancy. Arterioscler Thromb Vasc Biol 2000;20:266 70. 18. Wramsby ML, Sten-Linder M, Bremme K. Primary habitual abortions are associated with high frequency of factor V Leiden mutation. Fertil Steril 2000;74:987 91. 19. Pihusch R, Buchholz T, Lohse P, Rubsamen H, Rogenhofer N, Hasbargen U, et al. Thrombophilic gene mutations and recurrent spontaneous abortion: prothrombin mutation increases the risk in the first trimester. Am J Reprod Immunol 2001;46:124 31. 20. Raziel A, Kornberg Y, Friedler S, Schachter M, Sela BA, Ron-El R. Hypercoagulable thrombophilic defects and hyperhomocysteinemia in patients with recurrent pregnancy loss. Am J Reprod Immunol 2001; 45:65 71. 21. Carp H, Salomon O, Seidman D, Dardik R, Rosenberg N, Inbal A. Prevalence of genetic markers for thrombophilia in recurrent pregnancy loss. Hum Reprod 2002;17:1633 7. 22. Dilley A, Benito C, Hooper WC, Austin H, Miller C, El-Jamil M, et al. Mutations in the factor V, prothrombin and MTHFR genes are not risk factors for recurrent fetal loss. J Matern Fetal Neonatal Med 2002;11: 176 82. 23. Sarig G, Younis JS, Hoffman R, Lanir N, Blumenfeld Z, Brenner B. Thrombophilia is common in women with idiopathic pregnancy loss and is associated with late pregnancy wastage. Fertil Steril 2002;77: 342 7. 24. Unfried G, Griesmacher A, Weismuller W, Nagele F, Huber JC, Tempfer CB. The C677T polymorphism of the methylenetetrahydrofolate reductase gene and idiopathic recurrent miscarriage. Obstet Gynecol 2002;99:614 9. Fertility and Sterility 1721

25. Wang YW, Li F, Li YP, Xue MZ, Yu XW, Li XC. The study on the relationship between the methylenetetrahydofolate reductase 677C T mutation and unexplained recurrent pregnancy loss [in Chinese]. Zhongguo Shi Yong Fu Ke Yu Chan Ke Za Zhi 2002;18:291 3. 26. Dossenbach-Glaninger A, van Trotsenburg M, Dossenbach M, Oberkanins C, Moritz A, Krugluger W, et al. Plasminogen activator inhibitor 1 4G/5G polymorphism and coagulation factor XIII Val34Leu polymorphism: impaired fibrinolysis and early pregnancy loss. Clin Chem 2003;49:1081 6. 27. Hohlagschwandtner M, Unfried G, Heinze G, Huber JC, Nagele F, Tempfer C. Combined thrombophilic polymorphisms in women with idiopathic recurrent miscarriage. Fertil Steril 2003;79:1141 8. 28. Pauer HU, Voigt-Tschirschwitz T, Hinney B, Burfeind P, Wolf C, Emons G, et al. Analyzes of three common thrombophilic gene mutations in German women with recurrent abortions. Acta Obstet Gynecol Scand 2003;82:942 7. 29. Dossenbach-Glaninger A, van Trotsenburg M, Krugluger W, Dossenbach MR, Oberkanins C, Huber J, et al. Elevated coagulation factor VIII and the risk for recurrent early pregnancy loss. Thromb Haemost 2004;91:694 9. 30. Li XM, Zhang YZ, Xu YX, Jiang S. Study on the relationship of MTHFR polymorphisms with unexplained recurrent spontaneous abortion [in Chinese]. Zhonghua Yi Xue Yi Chuan Xue Za Zhi 2004;21: 39 42. 31. Makino A, Nakanishi T, Sugiura-Ogasawara M, Ozaki Y, Suzumori N, Suzumori K. No association of C677T methylenetetrahydrofolate reductase and an endothelial nitric oxide synthase polymorphism with recurrent pregnancy loss. Am J Reprod Immunol 2004;52:60 6. 32. Wang XP, Lin QD, Ma ZW, Zhao AM. C677T and A1298C mutation of the methylenetetrahydrofolate reductase gene in unexplained recurrent spontaneous abortion [in Chinese]. Zhonghua Fu Chan Ke Za Zhi 2004;39:238 41. 33. Couto E, Barini R, Zaccaria R, Annicchino-Bizzacchi JM, Passini Junior R, Pereira BG, et al. Association of anticardiolipin antibody and C677T in methylenetetrahydrofolate reductase mutation in women with recurrent spontaneous abortions: a new path to thrombophilia? Sao Paulo Med J 2005;123:15 20. 34. Gerhardt A, Scharf RE, Mikat-Drozdzynski B, Krussel JS, Bender HG, Zotz RB. The polymorphism of platelet membrane integrin alpha2beta1 (alpha2807tt) is associated with premature onset of fetal loss. Thromb Haemost 2005;93:124 9. 35. Guan LX, Du XY, Wang JX, Gao L, Wang RL, Li HB, et al. Association of genetic polymorphisms in plasminogen activator inhibitor-1 gene and 5, 10-methylenetetrahydrofolate reductase gene with recurrent early spontaneous abortion [in Chinese]. Zhonghua Yi Xue Yi Chuan Xue Za Zhi 2005;22:330 3. 36. Kobashi G, Kato EH, Morikawa M, Shimada S, Ohta K, Fujimoto S, et al. MTHFR C677T polymorphism and factor V Leiden mutation are not associated with recurrent spontaneous abortion of unexplained etiology in Japanese women. Semin Thromb Hemost 2005;31:266 71. 37. Krause M, Sonntag B, Klamroth R, Heinecke A, Scholz C, Langer C, et al. Lipoprotein (a) and other prothrombotic risk factors in Caucasian women with unexplained recurrent miscarriage. Results of a multicentre case-control study. Thromb Haemost 2005;93:867 71. 38. Song LY, Qi QH, She DX, Zhang HY, Chen SQ, Li JY. Relationship between genetic polymorphism of homocysteine metabolism enzyme and unexplained repeated spontaneous abortion [in Chinese]. Zhonghua Wei Chan Yi Xue Za Zhi 2005;8:160 4. 39. Review Manager. Version 4.2 for Windows. Oxford: Cochrane Collaboration, 2002. 40. Hasbargen U, Lohse P, Thaler CJ. The number of dichorionic twin pregnancies is reduced by the common MTHFR 677C T mutation. Hum Reprod 2000;15:2659 62. 41. Zhu JH, Ren A, Hao L, Pei LJ, Liu JM, Zhu HP, et al. Variable contribution of the MTHFR C677T polymorphism to non-syndromic cleft lip and palate risk in China. Am J Med Genet 2006;140:551 7. 42. Berry RJ, Li Z, Erickson JD, Li S, Moore CA, Wang H, et al. Prevention of neural-tube defects with folic acid in China. China-U.S. Collaborative Project for Neural Tube Defect Prevention [corrected; erratum to be published]. N Engl J Med 1999;341:1485 90. 43. Li XM, Xu YX, Jiang S. The correlation study on the homocysteine metabolic abnormality and unexplained repeated spontaneous abortion [in Chinese]. Sheng Zhi Yi Xue Za Zhi 2003;12:206 10. 44. Vaughn JD, Bailey LB, Shelnutt KP, Dunwoody KM, Maneval DR, Davis SR, et al. Methionine synthase reductase 66A- G polymorphism is associated with increased plasma homocysteine concentration when combined with the homozygous methylenetetrahydrofolate reductase 677C- T variant. J Nutr 2004;134:2985 90. 1722 Ren and Wang MTHFR and recurrent pregnancy loss Vol. 86, No. 6, December 2006