The 2003 Rotterdam Conference on Polycystic Ovarian Syndrome (PCOS) diagnostic criteria are currently in clinical use. The Rotterdam criteria state that PCOS is diagnosed by the presence of any two of the following clinical and laboratory tests: ( 1) anovulation or ovulatory disturbance. ( 2) Hyperandrogenic manifestations and biochemical changes. ( 3) Ultrasound examination of polycystic ovaries. ( 4) Exclusion of other endocrine disorders. This criterion recommends an ultrasound picture of PCOS with polycystic ovarian changes.1 The association of PCOS with recurrent abortion (RSA) has been observed according to this diagnostic criterion, but the relationship between PCOS and recurrent abortion as described by many studies is full of uncertainty. However, clinical observations have shown that the rate of miscarriage in pregnancies with PCOS is higher than in women with spontaneous pregnancies. The possible mechanisms of PCOS leading to RSA are generally considered to be hyperandrogenemia, hyperinsulinemia, hyperluteinizing hormone, and obesity. The prevalence of PCOS in RSA is generally thought to be increased in PCOS pregnancies, but many authors have suggested that this is still an open question, meaning that the incidence of spontaneous abortion in PCOS is difficult to determine. This is because: (1) most patients with PCOS require ovulatory therapy (anti-estrogens or gonadotropins) and the incidence of spontaneous abortion is higher in women with ovulatory treatment pregnancies than in spontaneous pregnancies. Clomiphene is the first-line drug for ovulation in PCOS, and some authors have reported a 25% average incidence of miscarriage after ovulation-promoting pregnancies with this drug, which is significantly higher than in the normal population. Clomiphene has an antagonistic effect on the endometrial estrogen receptor, which leads to an increase in FSH and LH, which is detrimental to pregnancy. ( 2) Patients with PCOS who conceive during consultation and treatment tend to be monitored more closely in early pregnancy and therefore may show a higher incidence of miscarriage compared to those who conceive spontaneously. ( 3) Patients with PCOS tend to be obese, and obesity itself has a higher incidence of spontaneous abortion than women with normal body mass index, so it is not known whether the causative factor is PCOS per se or obesity. ( 4) PCOS is a complex syndrome with different subgroups of patients (e.g. high LH type, high androgen type, insulin resistant type, etc.) and there is no information on the abortion rate in different types of PCOS patients. (5) Not all previously published articles refer to the PCOS criteria recommended by the Rotterdam Conference in 2003, and some articles limit the diagnosis to ultrasonographic signs. This is why some authors have looked at the diagnosis of polycystic ovary syndrome in patients with RSA based mostly on ultrasonography and not using the Rotterdam criteria. ( 2) Most studies reported transvaginal ultrasound, but some studies were transabdominal ultrasound. It is generally considered that the former is better than the latter in diagnosing polycystic ovary syndrome, especially in patients who are obese and have deep ovaries with higher accuracy of examination. ( 3) The inclusion criteria were inconsistent, with some studies including early and mid-trimester miscarriages, some including only late miscarriages, and more studies not specifying the gestational week of miscarriage. In addition, some reports excluded other causes of RSA, while some articles did not specify whether other causes were excluded.2 Some authors have suggested that in studying the prevalence of PCOS in patients with RSA, ultrasound alone cannot be used to diagnose PCOS, since 28% of the normal population has polycystic changes in the ovaries. /L) to diagnose PCOS and calculated the incidence of PCOS in RSA to be 23%. In contrast, among 1246 patients with recurrent miscarriage investigated by Yang et al4, 49 cases of PCOS were diagnosed by a combination of clinical manifestations, ultrasound and biochemical tests, accounting for only 3. 93% of the total number of cases. 2. Possible causes of recurrent miscarriage due to polycystic ovary syndrome 2. 1 Obesity PCOS is closely related to obesity, which occurs in about 35% of PCOS patients and is associated with metabolic disorders of hyperandrogenemia and hyperinsulinemia.5. Studies have shown that among women with PCOS, a high body mass index ( bodymass index (BMI) is associated with an increased risk of miscarriage. In a large cohort study of women with assisted reproduction, there was no significant difference in miscarriage rates between the PCOS and non-PCOS groups when both BMI and treatment modality were screened out as confounders, suggesting that the high rate of miscarriage in PCOS patients can be attributed to obesity. This study also suggests that obesity is an independent risk factor for miscarriage.6 In a retrospective analysis of 28,538 women, Boots et al6 showed that 13.6% of the 3800 women with a single obesity factor had a miscarriage, compared to 10.7% of the 17,146 women with a normal BMI. Among women with habitual miscarriages, obese women had a higher miscarriage rate compared to non-obese women (46% vs. 43%; OR 1. 71). Based on a retrospective study, the authors concluded that obesity is associated with the rate of spontaneous abortion in women. The authors used the following criteria to count obese (BMI ≥ 30), overweight (BMI 25-29), and normal women (BMI < 25). Data from a prospective study of a recurrent miscarriage clinic showed a significant increase in the incidence of post-pregnancy miscarriage in obese and low weight patients compared to normal body mass index women, but no significant increase in miscarriage in overweight women. Linear regression analysis showed that the most important predictor of miscarriage occurrence was advanced age (P = 0. 01), followed by body mass index BMI (P = 0. 04).7 A recent prospective study on obesity and miscarriage, which looked at the incidence of miscarriage in 3000 pregnant women confirmed by ultrasound, found a miscarriage rate of 3. 9% (n = 117), with moderately severe obesity (BMI > 34. 9) in primigravid women. The miscarriage rate was 11.8% (n = 8) for first-time pregnant women with BMI > 34.9, 2.7% (n = 24) for those with normal BMI, and 3.7% (n = 5) for those with mild obesity (BMI > 25 and < 34.9). In menstruating mothers, there was no significant increase in miscarriage rate in moderately and severely obese patients with normal BMI.8 Analysis of mean body composition values showed that the ratio of adipose to non-adipose components in first-time mothers was significantly higher in the miscarriage group than in the non-miscarriage group, whereas there was no significant difference between the two groups in menstruating mothers. This article concludes that the rate of miscarriage is increased in first-time pregnant women with moderate to severe obesity. The mechanism of obesity-induced miscarriage is unclear, and it may be that obesity-related metabolic disorders, including hyperandrogenemia, hyperinsulinemia, and hyperleptin, produce an environment that is detrimental to the oocyte and embryo. Therefore, for women with recurrent miscarriage, further research on the possible relationship and mechanism of obesity is still needed. 2. 2 hyperinsulinemic women with PCOS exhibit hyperinsulinemia and insulin resistance, the mechanism of which is not entirely clear. there is evidence that hyperinsulinemia is closely associated with elevated plasma fibrinogen activator inhibitor 1 (PAI-1), which is elevated in patients with PCOS. PAI-1 plays an important role in the fibrinolytic system, inhibiting fibrinogen formation during fibrinogen activation and promoting thrombosis leading to placental function. It promotes thrombosis leading to placental malfunction and thus miscarriage. Another possible mechanism is that elevated insulin concentrations in early pregnancy in PCOS patients cause a decrease in blood glucose and insulin growth factor binding protein 1 (IGFBP-1), and a decrease in IGFBP-1 indicates endometrial damage during the peripartum period.10 Craig et al.10 investigated the prevalence of insulin resistance in patients with recurrent miscarriage. The authors compared 74 patients with RSA and 74 normal controls, matched for age and BMI in both groups. The criteria for hyperinsulinemia and insulin resistance were: elevated fasting insulin (FI) level ≥ 20 μU/mL or fasting glucose (FG)/insulin (FI) < 4.5. The results showed that there were significantly more cases of insulin resistance in the RSA group than in the control group. There was no difference in age and body mass index between the two groups, and FG and FI were measured, and no significant difference was found between the FG and FG/FI groups (P > 0. 05), but a statistically significant difference was found between the FI groups (P = 0. 0119). Diejomaoh et al.12 analyzed 35 patients with recurrent miscarriage who were not pregnant and 30 non-pregnant menstrual non-RSA patients were selected as controls, and the case and control groups were matched for age, race, and BMI, and FG and FI were measured. There was no significant difference in fasting glucose between the case and control groups, and the difference between the two groups of insulin resistance calculated by FG/FI < 4. 5 was not statistically significant. The authors concluded that there was no significant correlation between RSA and insulin resistance and concluded that the relationship between the two should be further investigated. 2. 3 Hypersecretion of LH Early suggestions suggested that high LH levels in blood during the early follicular phase are a cause of recurrent miscarriage.13-14 These studies used an LH above 10 mU/L in the early follicular phase as a cut-off. There are two reasons to explain the association of high LH with recurrent miscarriage. First, high LH concentrations cause premature follicular maturation, resulting in decreased follicular quality, and cause follicular development to be out of sync with the endometrium. Another possible mechanism is that high LH has a detrimental effect on the endometrium, thus affecting embryo implantation and growth, which has been shown to be associated with endometrial HCG and LH receptors.15-16 One study found that the incidence of miscarriage in patients with PCOS treated with gonadotropin ovulation promotion was 33%, whereas the incidence of miscarriage in hypopituitary patients also treated with this ovulation promotion was only 10.6%. One of the reasons for the high rate of miscarriage in the former may be the presence of high LH, whereas the opposite is true in hypopituitary patients.13 Homburg et al.17 treated recurrent miscarriages in PCOS with pituitary agonists and obtained better results, which is also corroborative of high LH causing miscarriages. The incidence of miscarriage was lower in the buserelin and gonadotropin combination group (15/74, 20%) than in the clomiphene and gonadotropin combination group (51/108, 47%). This is because buserelin decreases endogenous LH. However, subsequent studies have reported inconsistent effects of blood LH in RSA patients and the above findings, concluding that there was no significant difference in serum LH values between RSA patients and normal controls.19 Cocksedge et al.19 summarized different studies by eight authors and concluded that high LH did not correlate with RSA. In conclusion, even if there may be a relationship between LH and miscarriage rate, further studies are needed to confirm it. 2. 4 Hyperandrogenemia Tulppala et al. 201993 first studied blood androgen concentrations in patients with recurrent miscarriage and found that mean androgen levels were generally normal in patients with RSA, whereas during subsequent pregnancies, total testosterone, free testosterone and dehydroepiandrosterone were significantly higher in the blood of patients with recurrent miscarriage than in those with successful pregnancies. Following this, Watson et al21 found that mean total testosterone levels during pregnancy were significantly higher in patients with RSA than in controls. mulders et al22 and van Wely et al23 concluded that ovulation promotion in patients with anovulatory infertility (without a history of RSA) had a higher rate of successful conception with low levels of total testosterone and free testosterone indices. However, Liddell et al24 compared free testosterone levels in the blood of patients with repeat pregnancy with RSA, and no significant difference in testosterone was found between the group with miscarriage and the group with successful pregnancy. In these studies on the possible relationship between hyperandrogenemia and miscarriage, there is considerable variation in the indicators used to measure androgens, with some researchers measuring only total testosterone, some measuring free testosterone, and some calculating free testosterone or free androgen index from total testosterone (T) and sex hormone binding globulin (SHBG) values using the formula: free T = T/100 ( 2. 28 - Of all the androgen indicators, the calculation of free testosterone and FAI is currently considered to be the most sensitive for determining hyperandrogenemia. An investigation from the Recurrent Miscarriage Research Centre used the Free Androgen Index ( FAI) in patients with RSA to determine the presence of hyperinsulinemia as a predictor of pregnancy outcome. The study recruited 571 women from the Sheffield Recurrent Miscarriage Study Centre and measured total early follicular phase serum testosterone, sex hormone binding globulin and derived FAI. 11% of RSA patients had plasma hyperandrogenemia and the rate of miscarriage was significantly higher in the high FAI group (68% in the FAI > 5 and 40% in the FAI ≤ 5 group, p = 0. 002). Therefore, the authors suggest that elevated FAI may be a more important predictor of subsequent miscarriage in RSA patients than advanced age (≥40 years) or previous multiple miscarriages (n≥6).25 If there is indeed a correlation between elevated androgens and miscarriage, the possible mechanism is that elevated androgens lead to abnormal endometrial growth. Alternatively, hyperandrogenism may adversely affect oocyte quality and embryonic viability; another explanation is that hyperandrogenemia is often associated with hyperinsulinemia in PCOS, which may indirectly affect the endometrium by altering insulin-like growth factor receptors. It is a competitive inhibitor of 17β-estradiol, which causes an increase in hypothalamic GnRH, which in turn promotes an increase in the secretion of FSH and LH by the pituitary gland to achieve a pro-ovulatory effect. Some studies on clomiphene have reported miscarriage rates ranging from 14% to 25% with clomiphene, and it is difficult to directly compare the miscarriage rates of PCOS patients with and without clomiphene. However, the observed rate of miscarriage is similar to or slightly higher than the rate of spontaneous abortion. The efficacy of clomiphene in PCOS patients with recurrent miscarriages is difficult to determine. In this group of patients, ovulation disorders and difficulties in conception are common and therefore ovulation promotion therapy is required. Gonadotropins are the second line of treatment for patients with anovulatory PCOS in whom clomiphene is not effective. Although the success rate of ovulation induction with gonadotropins is close to 30% per cycle, the high rate of multiple pregnancies and ovarian hyperstimulation increase the adverse pregnancy outcome. In order to reduce complications, the “small-dose incremental approach” is used, i.e., starting with a small dose and slowly increasing it before ovulation is induced, and the “decremental approach” starts with a higher dose and gradually decreases it later. Both methods significantly reduce the incidence of multiple births and ovarian hyperstimulationsyndrome (OHSS). Recent studies have shown that immunosuppressive glycoproteins and IGFBP-1 are significantly reduced in early pregnancy in pregnant women with PCOS compared with normal controls, that reductions in these proteins are associated with delayed endometrial development and recurrent miscarriage, and that treatment of the underdeveloped endometrium improves pregnancy outcomes in PCOS-associated RSA. The authors hypothesized that stimulation of the ovaries with gonadotropins in women with delayed endometrial development would increase follicular estrogen production and improve the endometrial basal environment, allowing for normal luteal phase endometrial growth. The results of this study are encouraging: the endometrial response to ovarian stimulation and the rate of miscarriage were significantly reduced. However, large-scale studies are needed to verify this. Early results were encouraging when GnRH analogs were used to suppress pituitary function in an attempt to reduce the rate of miscarriage in PCOS patients, but later studies found no reduction in the incidence of OHSS and no statistically significant incidence of miscarriage. 3. 2 Ovarian perforation Surgical treatment of anovulation in patients with PCOS consists of laparoscopic electroacupuncture or laser perforation and is the second-line treatment for anovulatory PCOS in which clomiphene treatment has failed. This approach reduces the production of ovarian primary androgens and the conversion of circulating androgens to estrogens, and also reduces LH/FSH values, mean LH, testosterone, FAI and mean ovarian volume over the longer term, although the exact mechanism is unknown, it does improve ovulation and conception rates in PCOS. The advantage of the surgical approach is that it does not cause OHSS or multiple pregnancies and does not require additional treatment to resume normal cycles. The disadvantage is that it is an invasive procedure with a risk of adhesions.26 Several randomized controlled trials comparing the effectiveness of ovarian perforation with gonadotropin therapy have shown similar pregnancy and miscarriage rates, with a significantly lower rate of multiple pregnancies with ovarian perforation.27 Farquhar et al. The rate of pregnancy miscarriage after ovarian perforation in PCOS patients has been reported to be between 11% and 15%, which is significantly less than in untreated PCOS patients and almost proportional to the incidence of miscarriage in the normal population.29 However, the outcome of PCOS patients with RSA who are then treated with perforation has not been reported. 3. 3 Application of metformin Metformin is a biguanide oral hypoglycemic agent. It is able to reduce the concentration of insulin without affecting normal blood glucose levels. In addition, it can enhance the blood supply to the uterus, reduce the level of plasma endothelin-1, increase the level of serum glycodelin (a lipid carrier protein, one of the biochemical markers reflecting the progress of pregnancy) during the luteal phase, reduce the level of androgens and LH, and reduce the weight of some patients. The above pharmacological effects suggest a possible role of metformin in the treatment of PCOS occurrence and prevention of early miscarriage in PCOS patients.30 This is reinforced by the analysis of previous studies in which 328 pregnant women who took oral metformin before and during pregnancy had a 20% early miscarriage rate in a large sample, compared to 319 pregnant women who did not take metformin and had a 65% early miscarriage rate in the previous study. 65%. In the previous study, 65 pregnant women treated with metformin during pregnancy had a miscarriage rate of 8. 8%, compared to 42% in the control group of 31 pregnant women. A comparison between metformin and laparoscopic ovarian febrile therapy in patients with overweight clomiphene-resistant PCOS has been made and metformin treatment was found to be superior to the latter. These findings suggest that metformin not only improves the live birth rate and significantly reduces the incidence of early miscarriage, but also that it is safe to take and has no significant effect on fetal development (congenital disorders, malformations). metformin is classified as a class B drug by the FDA. Current studies have shown that metformin during pregnancy not only reduces the incidence of early miscarriage in PCOS, but also reduces the incidence of gestational diabetes, preeclampsia, and macrosomia. A recently published systematic review article evaluating the effects of metformin found inconsistent effects of metformin on pregnancy outcomes in prospective, retrospective, and multicenter randomized double-blind controlled studies.30 This suggests that more in-depth multicenter studies are needed. 3. 4 Reducing LH levels The use of gonadotropin-releasing hormone agonists in patients with IVF suppresses LH levels. high LH levels may cause harm in patients with PCOS. A large proportion of patients with PCOS may have chronically elevated LH levels during ovulation. By suppressing LH levels, gonadotropin-releasing hormone agonists not only eliminate premature luteinization, but also reduce the rate of miscarriage. However, gonadotropin-releasing hormone agonists are not a standardized treatment for ovulation in patients with PCOS. This is because although experience and studies have shown that the combination of gonadotropin-releasing hormone agonists and gonadotropins can reduce miscarriage rates, the combination of gonadotropin-releasing hormone agonists and gonadotropins is cumbersome, has long cycles, requires more gonadotropins to promote ovulation, has more follicle development, and has an increased incidence of ovarian hyperstimulation syndrome and multiple pregnancy. In addition, the endogenous feedback mechanism is disrupted with gonadotropin-releasing hormone agonists, requiring more gonadotropins to promote follicle development, which is why gonadotropin-releasing hormone agonists are not an appropriate regimen to reduce miscarriage rates in patients with PCOS. There is no evidence to report a role for gonadotropin-releasing hormone antagonists in reducing the rate of early miscarriage in PCOS, and the cost of treatment is a consideration for clinicians.31 3. 5 Weight loss Overweight, obesity, and insulin resistance in patients with PCOS may all have an impact on early miscarriage. Losing weight before pregnancy through lifestyle changes may reduce the incidence of early miscarriage. With or without insulin resistance, fertility can be improved by controlling weight. This has been mentioned in the paragraph on etiology and will not be repeated here. To reduce the incidence of early miscarriage in PCOS, it is recommended to avoid pre-pregnancy overweight and obesity, and metformin has great potential for use. More studies need to focus on the near and long term effects on the offspring. In conclusion, polycystic ovary syndrome is a common endocrine disorder typically characterized by ovulatory disturbances, hyperandrogenemia, and polycystic ovarian changes. Although PCOS is associated with recurrent miscarriage, the prevalence of PCOS in patients with recurrent miscarriage is still not fully established, and most studies examining the prevalence of PCOS use only ultrasound indicators to define PCOS; future studies should use the Rotterdam criteria. The mechanism of PCOS causing RSA remains unclear, although there are many possibilities, but LH hypersecretion, hyperandrogenemia, obesity, and hyperinsulinemia may be risk factors for RSA, either independently or in combination with each other to cause miscarriage. In cases where the cause is not completely clear, pre-pregnancy treatment in patients with PCOS is necessary to ameliorate these possible adverse factors, and weight loss and metformin application may reduce the incidence of miscarriage, as reported in previous studies.