Our Meta analysis showed that except for the overdominant model, the 4G/5G polymorphism of PAI-1 gene was closely associated with recurrent abortion, and 4G allele would increase the risk of recurrent abortion. Subgroup analysis suggested that the continuity of race and abortion did not affect the genetic susceptibility of 4G alleles, but with the increase of the number of abortions (≥ 3 versus ≥ 2), under specific genetic models. The risk of recurrent abortion caused by this gene polymorphism changed from significant to insignificant.
The first Meta analysis of PAI-1 gene 4G/5G polymorphism and recurrent abortion by Mei-Tsz Su et al in 2013 did not show a risk association with recurrent abortion [51]. Subsequently, Meta analysis, which covered 22 studies by Xuejiao Li et al., suggested that the two were only associated with Caucasians [52]. A recent paper from Zhan Huang et al., including 31 studies, suggested that there was a correlation among yellow, white and black people, but not with the number of miscarriages [12]. On the basis of previous work, this paper includes the newly published literature, which is the most comprehensive Meta analysis of the relationship between PAI-1 gene 4G/5G polymorphism and recurrent abortion, and takes the continuity of abortion as a characteristic factor classification for the first time to explore the correlation between them. More sample size, more genetic models and more detailed grouping will help to reduce heterogeneity and test the robustness of the model. Therefore, the conclusion of our Meta analysis is more credible.
In general, in terms of genetic susceptibility to recurrent abortion, we believe that the 4G4G genotype of PAI-1 is a risk factor, 5G5G genotype is a protective factor, and 4G5G heterozygote is between the two, and this conclusion is also applicable to Caucasians and yellow races and whether abortion is continuous. Different from the results of previous studies[12], our Meta analysis showed that the number of abortions changed the effect of PAI-1 gene 4G/5G polymorphism on the risk of recurrent abortion: with the increase of abortion times, the pathogenic effect of 4G allele on recurrent abortion was gradually weakened under allele model, recessive gene model and homozygous model. The frequency of abortion is an important factor in the diagnosis of recurrent abortion. it is obvious that the diagnostic criteria of ≥ 3 times are more stringent and the number of populations is less than ≥ 2 times, and the causes of recurrent abortion may be more complicated. among them, the proportion of genetic related factors may be reduced, while advanced age and the decline of ovarian function have become new risk factors. To a certain extent, this can explain the paradox that the polymorphism of the gene loses its pathogenicity to more recurrent abortions. in addition, the heterogeneity among the original studies is also an important factor that can not be ignored. When high heterogeneity is allowed, except for allele model and overdominance model, the more times of abortion, the greater the risk value of 4G allele contribution to recurrent abortion in the remaining four genetic models. When we try to reduce the heterogeneity, the significance also disappears, which suggests that the heterogeneity of our unknown sources affects the stability of Meta analysis results, and we need to explore the source of heterogeneity from more levels.
Compared with the previous Meta analysis, this paper discusses the relationship between PAI-1 gene 4G/5G polymorphism and recurrent abortion under the overdominant model for the first time, and we are confused that the correlation between 4G/5G polymorphism and other genetic models is not consistent with other genetic models: 4G5G heterozygotes have a higher risk of recurrent abortion than 4G4G and 5G5G homozygotes. This genetic model was proposed by Shall C H and East E M in 1918 respectively. It emphasizes the interaction between genes and believes that there is a complementary effect between the two members of the heterozygous allele, which is stronger than any homozygous type. This breaks our understanding of the traditional mode of single gene inheritance and explains why the 4G5G genotype has the highest risk of recurrent abortion in this genetic model.
Plasminogen activator inhibitor-1 encoded by PAI-1 gene is the main physiological inhibitor of fibrinolytic system. The increase of plasminogen activator inhibitor-1 level will lead to the decrease of fibrinolytic activity and promote thrombosis, which is one of the independent risk factors of thrombotic diseases. 4G allele can up-regulate the transcription and translation of PAI-1 gene. The plasma PAI-1 level of people with 4G / 4G genotype reaches the peak, while the PAI-1 level of 5G / 5G homozygote is the lowest. Overexpression of PAI-1 protein can reduce the hydrolysis of plasma fibrin and restrict the development of trophoblast, resulting in early placental protein deposition in spiral arterioles, resulting in placental microthrombosis. It eventually leads to miscarriage [53][54]. Therefore, for recurrent abortion patients with high level of PAI-1 protein, early use of anticoagulants may be an effective measure to improve prognosis [55].
In addition to being directly related to recurrent abortion, PAI-1 is also related to other obstetrical and gynecological diseases, including preeclampsia (PE), gestational diabetes mellitus (GDM), fetal growth restriction (FGR), polycystic ovary syndrome (PCOS) and endometriosis. Diseases such as preeclampsia, gestational diabetes and polycystic ovary syndrome are themselves high risk factors for abortion. It can be seen that the secondary harm of PAI-1 gene polymorphism can not be underestimated, therefore, recurrent abortion patients with the above history of recurrent abortion are more recommended for genetic detection, in order to more comprehensively prevent the occurrence of subsequent abortion from the etiology [57].
Our Meta analysis also has some limitations. First of all, the heterogeneity of the overall data is high, and there is no significant decrease in heterogeneity after we try to do subgroup analysis, which indicates that population genetic balance, race, number of miscarriages and continuity are not the sources of heterogeneity. In fact, when we extracted the original data, we also extracted the cause of abortion in the original study, the primary or secondary of abortion, and planned to take this as a subgroup for further analysis to find out whether the association between PAI-1 gene 4G/5G polymorphism and recurrent abortion was related, but in the end, we found that many studies did not mention the relevant information in the article. After all, the definition of recurrent abortion does not address etiology and previous deliveries. At present, most of the recurrent abortion patients who have given birth to live fetuses have a good prognosis and are less affected by genetic factors, so it is suggested that such studies can select primary patients in the future to reduce the heterogeneity of the subjects, and the results may be more convincing.
In addition, the publication bias under the homozygote model will also affect the stability of our results, there may be unpublished grey literature not included, look forward to more related studies can be published in the future to eliminate publication bias.
In summary, this Meta analysis study showed that the 4G/5G polymorphism of the PAI-1 gene was associated with recurrent abortion, and the 4G allele increased the risk of recurrent abortion. Therefore, when pregnant women have two miscarriages, they should attract enough attention. In clinical work, for women of childbearing age who have a previous history of abortion with 4G allele, relevant strategies should be taken to prevent the occurrence of recurrent abortion, so as to reduce the physical, psychological and financial burden of women caused by abortion.