Effects of Low Molecular Weight Heparin on Early Pregnancy Loss in Women With Polycystic Ovary Syndrome.

Polycystic ovary syndrome (PCOS) is the most common endocrine condition in women of reproductive age, with a prevalence of 6 to 10 percent according to different diagnostic criteria. 1 The main characteristics of PCOS include functional androgen excess (AE), ovulatory dysfunction, and/or polycystic ovaries, menstrual disorders, and metabolic dysfunction. 2 At present, most expert groups use the Rotterdam criteria to make the diagnosis of PCOS. 3 , 4 In order to achieve better-individualized treatment, a growing body of research and clinical work has begun to make a clear distinction between different subtypes of PCOS. There are four different subtypes: Subtype A (also known as “full PCOS” or “classic PCOS”) includes biochemical or clinical hyperandrogenism, oligoovulation, and polycystic ovarian morphology; Subtype B (also known as “classic PCOS”) includes hyperandrogenism and oligoanovulation; Subtype C (also known as “ovulatory PCOS”) includes hyperandrogenism and polycystic ovarian morphology; Subtype D (also known as “non-hyperandrogenic PCOS”) includes oligoanovulation and polycystic ovarian morphology. 5 PCOS is also the most common cause of anovulatory infertility in women of reproductive age, accounting for 70%–85% of infertility. Infrequent ovulation makes it more difficult to conceive in women with PCOS. 6 The first-line treatment for anovulatory women with PCOS is lifestyle modification and ovulation induction with letrozole or clomiphene citrate. Meanwhile, it is recommended to take a combination of oral contraceptives to reduce hyperandrogenism and restore menstrual cycles, and take anti-androgens to treat hirsutism. If combined with obesity or insulin resistance, metformin could also be used. 7 Oral contraceptive pretreatment may increase the cumulative clinical pregnancy rate of the oocyte retrieval cycle. 8 Successive and cyclic oral contraceptive pill pretreatment could also improve the outcomes of in vitro fertilization (IVF) and intracytoplasmic sperm injection (ICSI) in women with PCOS, especially in the live birth rate per embryo transfer. 9 Pregnancy losses taking place before 12 weeks of gestation are defined as early pregnancy loss (EPL). 10 Some studies have shown that 30%-50% of women with PCOS who conceived would suffer from EPL, a rate five times higher than that of the general population. 11 12 13 14 Improvement in live birth rates for women with PCOS is required. Women with PCOS often have a prothrombotic state, with changes in their coagulation and fibrinolytic systems, and dysregulation of the coagulation, anticoagulant, and fibrinolytic systems associated with thrombosis. 15 Low molecular weight heparin (LMWH) is one of the most widely used anticoagulants and antithrombotic agents. Moreover, LMWH can regulate the function of intravascular cells, improve the condition of placental microcirculation and promote normal growth and development of the fetus. 16 In recent years, LMWH has been widely used to treat recurrent pregnancy loss (RPL) in gynecological and reproductive disorders. 17 , 18 Pre-pregnancy use of LMWH in conjunction with aspirin may be beneficial for women with RPL and PCOS. 19 But its effectiveness in preventing EPL in women with PCOS has not been directly investigated. Therefore, the aim of this study was to explore whether LMWH treatment can reduce the rate of EPL in pregnant women with PCOS. We conducted a retrospective study of all women with PCOS who visited our gynecologic endocrine clinic from June 2019 to March 2022 and became pregnant during that time. Specifically, we compared pregnancy outcomes (EPL) in PCOS women with or without LMWH administration during early pregnancy.

Li Li and Gendie E Lash designed the study and supervised all research. Fanglan Luo collected and organized the data. Fanglan Luo and Qinsheng Lu prepared the manuscript. Fanglan Luo, Wei Wei and Qinsheng Lu analyzed the data. Gendie E Lash and Qinsheng Lu revised the manuscript. Yingmei Cen, Yinchun Huang, Shuang Qin and Chunjiao Wei participated in the experiments and patients’ management. All authors reviewed the manuscript.

Gendie E Lash is the Deputy Editor-in-Chief of Maternal-Fetal Medicine. The article was subject to the journal’s standard procedures, with peer review handled independently of this editor and the associated research groups.

This study was supported by the Guangzhou Science and Technology Bureau (Grant Nos. 2024A03J1175, awarded to Qinsheng Lu, and SL2022A03J01127, awarded to Li Li), as well as the Basic and Applied Basic Research Fund Committee of Guangdong Province (Grant Nos. 2023A1515010216, awarded to Li Li, and 2022B1515120074, awarded to Gendie E. Lash).

This retrospective study included women with PCOS who became pregnant and were treated at the Gynecological Endocrine Clinic of Guangzhou Women and Children’s Medical Center between June 2019 and March 2022. All PCOS patients were diagnosed according to the Rotterdam diagnostic criteria. 3 , 4 Eligible participants were between 24 and 40 years of age. Women with systemic diseases, such as diabetes, hypertension, thyroid abnormalities, and chromosome structure abnormalities, or more than two spontaneous miscarriages, were excluded. Participants were assigned to either the control group or the LMWH treatment group. This was a non-randomized trial: after pregnancy, the PCOS patients voluntarily chose whether to receive LMWH treatment. Those in the control group received standard care, while patients in the LMWH group received standard care plus additional LMWH therapy, as described below. All patients were managed by the same team of physicians and their baseline characteristics were recorded. All women in the study were treated with combined oral contraceptives for at least three months to adjust the menstrual cycle and decrease androgen levels. For patients who were obese or had insulin resistance, 2.5 mg metformin per day was given until they successfully conceived. In addition, overweight and obese patients were recommended to control their diet and increase exercise levels to achieve weight loss. Anthropometry measurements, a simple and commonly used method for evaluating risk factors for non-communicable diseases such as diabetes, hypertension and dyslipidemia, 11 were used to assess metabolic risk. Given the well-documented dyslipidemia patterns in women with PCOS, 10 , 12 blood lipid profiles were included in the evaluation. After at least three months of individualized treatment, repeat tests of blood hormone and lipid levels were conducted. Tested hormones included prolactin (PRL), luteinizing hormone (LH), follicle-stimulating hormone (FSH), estradiol (E2), progesterone (P), sex hormone binding protein (SHBG), dehydroepiandrosterone (DHEAS) and serum free testosterone (T), and the free testosterone index (FTI) was calculated. Following the completion of the oral contraceptive, some patients attempted natural conception for one cycle before starting ovulation induction. Patients under ovulation induction were administered letrozole (2.5 mg/day) orally from day 5 to day 10 of the menstrual cycle. From day 8 onwards, the follicular development, endometrial thickness, and morphology were monitored by ultrasound once every other day. Monitoring became daily when a follicle diameter of ≥ 18 mm was observed in at least one follicle. At this point, an intramuscular injection of human chorionic gonadotropin (hCG, 5000-10,000 IU) was administered to trigger ovulation. Timed intercourse was arranged for 24-36 hours after hCG injection when the follicles were mature. Then urine serum hCG, and ultrasound were used to determine whether women were pregnant. The positive outcome was based on biochemical pregnancy, and after 2 weeks, transvaginal ultrasound was used to determine the location of the pregnancy. Clinical pregnancy was diagnosed if the gestational sac could be seen. Patients in the LMWH group received LMWH 2500 IU daily from the diagnosis of clinical pregnancy to 12 weeks of pregnancy. Women in both control and LMWH groups were treated with progesterone 10 mg twice daily after ovulation until a fetal heartbeat could be detected. Early pregnancy ultrasound evaluations were repeated every 2 weeks for all pregnant women. Whenever the ultrasonographic evaluation suggested a poor prognosis for pregnancy outcome, or women had symptoms suggestive of miscarriage (e.g., vaginal bleeding, abdominal pain), additional hCG and progesterone level determinations were performed. Miscarriage was documented by a negative pregnancy test or hCG less than 50 IU/L and confirmed by uterine ultrasonography. The women who continued their pregnancy were followed to the time of delivery and recorded their delivery status (gestational age, mode of delivery, blood loss at delivery, fetal birth weight, and gender of newborn). Venous blood samples were collected in the fasting state, for the tests including insulin, total cholesterol, triglycerides, high-density lipoprotein cholesterol, and low-density lipoprotein cholesterol. Blood samples for E2, LH, FSH, and T, etc., were collected two days before menstruation. For the measurement of insulin and sex hormones, blood samples were collected in an anticoagulant-free tube, centrifuged and analyzed by an Abbott I2000 automated chemiluminescence immunoanalyzer (Abbott Laboratories, USA). For blood lipid tests, blood samples were collected in an anticoagulant-free tube and centrifuged for serum, which was determined using a Hitachi 7600 automatic biochemical analyzer (Hitachi Co., Ltd., Tokyo, Japan). All demographic, clinical and paraclinical information and pregnancy outcomes were collected and analyzed using SPSS software version 18 (SPSS Inc., Chicago, IL, USA). Continuous variables were tested for normality using the Shapiro-Wilk test. Normally distributed data were presented as mean ± standard deviation ( SD ) and compared using the two-tailed unpaired Student’s t test. Non-normally distributed data were reported as median with interquartile range and analyzed using the Mann-Whitney U test. Categorical variables were expressed as frequencies and percentages. Differences between groups were assessed using the Pearson Chi-squared test or Fisher’s exact test where appropriate. For multiple categorical variables, Z -test with adjusted P value by Bonferroni method was applied to the Chi-square test for the post-hoc analysis. Odds ratios ( OR s) with 95% confidence intervals ( CI s) were calculated to assess the association between treatment and outcome variables. Kaplan-Meier survival analysis was performed to evaluate the impact of LMWH treatment and other variables on pregnancy outcomes. A P value < 0.050 was considered statistically significant. The study was conducted in accordance with the Declaration of Helsinki and was approved by the Medical Ethics Committee of Guangzhou Women and Children’s Medical Center (No. 236A01) on September 27, 2022. Written informed consent was obtained from all participants prior to enrollment.

In the present study, 166 women were involved who had PCOS-related infertility and conceived successfully after gynecological endocrine intervention treatment. Among them, the following patients were excluded: 15 patients with hypothyroidism, 5 patients with hyperprolactinemia and 6 cases with endometriosis. Finally, 140 records met the inclusion criteria. Of the 140 participants, 38 cases received LMWH treatment during early pregnancy, and 102 patients without LMWH treatment were in the control group (Fig. 1 ). CONSORT flowchart of the study. The diagram illustrates the number of PCOS cases enrolled in the study and their early pregnancy outcomes, stratified by whether LMWH treatment was administered during the first trimester. LMWH: Low molecular weight heparin; PCOS: Polycystic ovary syndrome. The baseline characteristics of the PCOS patients in the LMWH and control groups were analyzed. The LMWH group had an older mean age than that in the control group (Z = –2.797, P = 0.005) (Table 1 ). Both of the groups had 4 subtypes of PCOS women involved, and it was significantly different in the two groups ( P = 0.012). There were 16, 3, 4, 6 and 60, 7, 1, 38 cases for A, B, C and D subtypes in the LMWH and control groups, respectively. Characteristics of women with PCOS before conception in the LMWH and control groups. P < 0.050 was considered statistically significant. * Mann-Whitney U test. † Chi-square test. ‡ Fisher’s exact test. IQR: Interquartile range; LMWH: Low molecular weight heparin; PCOM: Polycystic ovary morphology; PCOS: Polycystic ovary syndrome; –: Not applicable. To discover more potential interfering factors on EPL in LMWH treatment, we explored the differences in infertility types, miscarriage history, hormone levels, and anthropometric parameters, before conception, in the LMWH and control groups. However, these results showed that there were no significant differences in these clinical parameters, including infertility type, recurrent pregnancy loss, ovulatory induction cycle, polycystic ovarian morphology and androgen excess. Among the 140 patients, we also recorded their hormone levels after lifestyle adjustment and endocrine therapy. PRL, LH, FSH, E2, P, SHBG, DHEAS, T, FTI, and the ratio of LH/FSH, were not statistically different between the two groups (Table 2 ). Hormone levels before conception of PCOS women in the LMWH and control groups. Non normally distributed data were reported as median (interquartile range) and were analyzed using Mann-Whitney U test. DHEAS: Dehydroepiandrosterone; E2: Estradiol; FSH: Follicle-stimulating hormone; FTI: Free testosterone index; LH: Luteinizing hormone; LMWH: Low molecular weight heparin; P: Progesterone; PCOS: Polycystic ovary syndrome; PRL: Prolactin; SHBG: Sex hormone binding protein; T: Serum free testosterone. Previous studies have shown that women with PCOS have an increased risk of metabolic dysfunction including insulin resistance and dyslipidemia. 20 , 21 Therefore, we further analyzed the differences in fasting serum glucose, blood lipids (total cholesterol, triglyceride, high-density lipoprotein (HDL) and low-density lipoprotein) (Table 3 ) and body fat characteristics including maternal body mass index (BMI), basal metabolic rate, visceral fat distribution, body fat percentage and waist-to-hip ratio. Results revealed that between the control and LMWH groups the total cholesterol levels (4.89 ± 0.90 vs . 5.56 ± 0.83, P = 0.003) and HDL (1.54 (1.27–1.69) vs . 1.74 (1.43–1.98), Z = –2.373, P = 0.018) showed a significant difference, while other parameters were not significantly different ( P > 0.050). There was no significant difference in maternal BMI, basal metabolic rate, visceral fat distribution, body fat percentage or waist-to-hip ratio between the two groups ( P > 0.050) (Table 4 ). Blood lipids characteristics of women with PCOS in the LMWH and control groups. Data were reported as median (IQR) or mean ± SD . * Mann-Whitney U test. † Student’s t test. HDL: High-density lipoprotein; IQR: Interquartile range; LDL: Low-density lipoprotein; LMWH: Low molecular weight heparin; PCOS: Polycystic ovary syndrome; SD : Standard deviation. Anthropometric parameters of women with PCOS in the LMWH and control groups. Data were reported as median (IQR) or mean ± SD . * Mann-Whitney U test. † Student’s t test. BMI: Body mass index; IQR: Interquartile range; LMWH: Low molecular weight heparin; PCOS: Polycystic ovary syndrome; SD : Standard deviation. The early pregnancy outcomes of both groups are shown in Table 5 . Women using LMWH during early pregnancy showed a lower EPL rate than those who did not (Table 5 ). Of the 38 women who received LMWH during early pregnancy, only 2 cases (5.3%) ended in EPL. In contrast, 27 cases out of the 102 women in the control group (26.5%) ended in EPL ( P = 0.006). The OR value for the effect of LMWH treatment was 0.154, with 95% CI : 0.035 - 0.685, which indicated that LMWH treatment in PCOS patients during early pregnancy had a protective effect. Rates of EPL among women with PCOS in the LMWH and control groups. Data were reported as n (%). LMWH: Low molecular weight heparin; PCOS: Polycystic ovary syndrome; OR : Odds ratio; CI : Confidence interval. Patients with a continuing pregnancy were further followed up until delivery. In the LMWH group, 23 cases were successfully delivered and 13 cases lost to follow-up; in the control group, 50 cases were successfully delivered and 25 cases lost to follow up. There were no significant differences in terms of gestation length, mode of delivery, gender of the newborn, bleeding during delivery or fetal birth weight between the two groups (Supplementary file 1, http://links.lww.com/MFM/A99 ). Kaplan-Meier survival analysis was performed to evaluate the effect of LMWH treatment on EPL rate in PCOS patients. Pregnancy survival in the LMWH group was greater than that in the control group during early pregnancy (12 weeks) (Log Rank (Mantel-Cox) χ 2 = 4.144, P = 0.040) (Fig. 2 ). In addition, we continued to follow up the pregnancy status to the end of pregnancy in the two groups and found that there were no cases and 2 cases with pregnancy loss in the LMWH group and the control group after the first trimester, respectively. The pregnancy status became relatively stable in both groups in the second and third trimesters. The survival rate of patients using LMWH early in pregnancy was significantly higher than that of the control group, suggesting that, overall, the use of LMWH early in pregnancy played an important role in improving early pregnancy outcomes in patients with PCOS infertility. Kaplan-Meier curve comparing the trends of pregnancies between the two groups of the study. Kaplan-Meier survival analysis was performed to evaluate the effect of LMWH treatment in PCOS patients. LMWH may decrease the risk of EPL compared with the control group in PCOS patients in the first trimester ( χ 2 = 4.144, P = 0.040). +: censored observations (lost to follow-up or ongoing). EPL: Early pregnancy loss; LMWH: Low molecular weight heparin; PCOS: Polycystic ovary syndrome.

PCOS seriously affects women’s reproductive health. Most PCOS women find it difficult to get pregnant, and the risk of miscarriage remains very high, especially in the first trimester. 22 , 23 Clinical studies have found that infertile PCOS women have different degrees of a pre-thrombotic state, which mainly manifests as abnormal coagulation-fibrinolysis activity and blood hypercoagulability, resulting in endometrial local blood abnormalities, increasing the risk of infertility and miscarriage. Improvement of the pre-thrombotic state of PCOS patients may be beneficial to their pregnancy and reduce the risk of EPL. LMWH is an anticoagulant and LMWH has been widely used to improve the outcome of pregnancy in women with unexplained recurrent pregnancy loss in recent years. 24 However, the role of LMWH in PCOS patients’ pregnancy has not been fully investigated. In the current study, we found that the rate of EPL was significantly lower in patients who used LMWH in the first trimester than that in women who did not use LMWH. LMWH has many biological activities and pharmacological mechanisms, such as anti-inflammatory effects, anti-fibrosis activity, anti-lipid metabolism disorders, and anti-viral, anti-tumor and anti-inflammatory activities. 25 , 26 Another study showed that in patients with thrombophilia, giving preventive LMWH from the 8th week was more effective than low-dose aspirin. 27 Based on these data, in our study the EPL rate in the control group (26.5%) is comparable with the 20%-40% rate described in the literature for women with PCOS. 11 12 13 14 28 In contrast, the EPL rate of 5.3% in the LMWH group is lower than the rate of 10%-15% reported for clinically recognized pregnancies in normal women. 29 Moreover, we further analyzed the predisposing factors affecting pregnancy loss chance and their interaction with the study intervention in patients with early pregnancy loss and found that the use of LMWH in early pregnancy can reduce the EPL. Therefore, LMWH may reduce the rate of EPL in women with PCOS by the intervention in the first trimester. In our study, total cholesterol was statistically higher in the LMWH group than in the control group. One of the previous studies has shown that dyslipidemia can lead to the formation of a prothrombotic state, 30 which is mostly associated with PCOS patients and affects their pregnancy outcome. 31 LMWH, as a commonly used anticoagulant, has been widely used in the prevention of thrombosis, so we suspect that the use of LMWH in early pregnancy may improve early pregnancy outcome by modifying dyslipidemia in patients with PCOS. In addition, we also recorded the effect of LMWH on pregnancy outcomes of the second and third trimesters, and delivery in patients with ongoing pregnancy. We found significant differences in survival curves between the two groups in early pregnancy outcomes, which gradually flattened out in the second and third trimesters. LMWH did not alter gestation length, bleeding during delivery, or mode of delivery in those non-EPL women, nor their fetuses’ birthweight or gender of the newborn, in the two groups. Therefore, we further speculated that the use of LMWH in early pregnancy is more beneficial for improving pregnancy outcomes in infertile patients with PCOS. We found that the PCOS subtypes were significantly different in the LMWH and control groups. The data were not balanced in the four subtypes of PCOS, in which most cases belonged to subtypes A and D, and subtypes B and C had fewer cases. A larger clinical cohort is needed to evaluate the effect of LMWH on the EPL in different types of PCOS women in the future. To date, there are a few studies reporting the effects of LMWH on pregnancy outcomes. However, most of these studies have focused on women with RPL, and those results show that LMWH could not decrease the EPL rate or increase a higher live birth rate in women with RPL. 32 Though there may be significant heterogeneity between PCOS patients and patients with recurrent miscarriage, resulting in different outcomes of EPL after LMWH treatment. The effects of LMWH on decreasing the risk of preeclampsia and other placenta-related pregnancy complications have been confirmed by other studies. 33 , 34 Therefore, further research is required on defined clinical cohorts, with extrapolation from one group of women to another not being possible. There are several other limitations in this study. Firstly, it was a retrospective design, not a prospective randomized controlled study; the sample size was relatively small, and there were some differences in patient characteristics that might have affected the study outcome. Secondly, the population of this study comprised a single ethnicity (the Chinese Han population), which may not be sufficiently representative. Thirdly, we have not investigated the mechanism of LMWH treatment on PCOS patients’ pregnancy outcomes. In addition, a karyotype assessment of the miscarried fetuses was not performed; this information would provide better information on the types of miscarriage being suffered by this group of women. Finally, because treatment or control groups were selected based on patient choice with no clear inclusion or exclusion criteria, bias may exist, with factors such as patient age, lipid profiles and PCOS subtype potentially impacting the study outcomes. While the data presented here provides proof of concept that LMWH may prevent EPL in women with PCOS, a larger clinical cohort in a randomized controlled trial is needed in the future to evaluate the effect of LMWH on pregnancy outcomes in PCOS patients.

LMWH usage among pregnant women with PCOS may improve early pregnancy outcomes. However, a larger multicenter prospective clinical trial is required before changes in clinical care can be considered.

None.

The authors thank the Departments of Obstetrics and Gynecology, Guangzhou Women and Children’s Medical Center for their assistance with sample collection and clinical tests, and thank all subjects participating in this project.

The data and materials that support the findings of this study are available from the corresponding author upon reasonable request and with permission of Guangzhou Women and Children’s Medical Center.