The association between serum fatty acids and pregnancy in PCOS women undergoing ovulation induction

Abstract Background Long-term dietary fat intake is thought to affect metabolism and pregnancy of polycystic ovary syndrome (PCOS) patients, and the type of fatty acids one consumes plays an important role. Previous studies mostly used questionnaires to analyze the type and proportion of fatty acids. Methods This prospective study included 91 PCOS patients. Serum fatty acids were measured by the gas chromatograph-mass spectrometry method before ovulation induction. We compared the fatty acids between the pregnancy group and the nonpregnancy group and explored the influence of the fatty acids on live births and pregnancy loss. Results Nervonic acid was lower in the pregnancy group than in the nonpregnancy group (0.25% vs. 0.30%, p = .017). The following trans-fatty acids were significantly lower in the pregnancy group than in the nonpregnancy group: trans-10-heptadecenoic acid, trans-vaccenic acid, trans-11-eicosenoic acid, and brassidic acid. The level of polyunsaturated fatty acids in the live birth group was significantly higher than the pregnancy loss group (16.95% vs. 15.10%, p = .039). Among individual PUFAs, the levels of linoleic acid (p = .043), docosapentaenoic acid (p = .024), alpha-linolenic acid (p = .042), and eicosapentaenoic acid (p = .035) were higher in the live birth group than in the pregnancy loss group. After adjusting for infertility duration, age, and body mass index, our findings suggested an inverse association between pregnancy and nervonic acid, trans-10-heptadecenoic acid, trans-vaccenic acid, trans-11-eicosenoic acid, and brassidic acid and pregnancy. Conclusions Our findings indicate that polyunsaturated fatty acids are associated with live birth in PCOS patients. Serum trans-fatty acids and nervonic acid might be risk factors for nonpregnancy. The mechanism of the influence of different fatty acids on pregnancy and live birth merits further exploration.


Background
Polycystic ovary syndrome (PCOS) is a common endocrine disorder in women of reproductive age, and the incidence rate is 4% to 18% [1]. Women with PCOS are at greater risk for developing metabolic dysfunction and account for nearly 80% [2] of women with anovulatory infertility. In guidelines for diagnosis and treatment of PCOS, ovulation induction is the main treatment option for PCOS infertility patients [3]. However, the association between metabolic changes and ovulation induction outcomes in PCOS has not been fully explored.
Fatty acids, which provide an essential energy resource, are crucial to reproduction as they affect ovulation and embryo quality [4]. Depending on molecular structure, fatty acids are classified as saturated fatty acids (SFAs), monounsaturated fatty acids (MUFAs), and polyunsaturated fatty acids (PUFAs). According to the position of the first carbon double bond from methyl carbon, unsaturated fatty acids are divided into n -3, n -6, n -7, and n -9 series. According to the different spatial structure of fatty acids, unsaturated fatty acids are divided into cis fatty acids and trans-fatty acids (TFAs). Studies of the relationship between fatty acids and reproductive diseases have received increasing attention. However, the influence of different types of fatty acids on reproductive diseases is still controversial. Although most studies suggest that supplementing n -3 PUFAs, such as eicosapentaenoic acid (EPA), could significantly improve hyperandrogenism, obesity, chronic inflammation, and oocyte quality [5] as well as have beneficial effects on pregnancy achievement in PCOS patients [6]. An retrospective study of 1228 women attempting pregnancy found that a higher level of serum total PUFAs was associated with lower probability of pregnancy [7].
Long-term dietary fat intake is thought to affect metabolism and pregnancy of PCOS patients, and the type of fatty acids one consumes plays an important role. Previous studies mostly used questionnaires to analyze the type and proportion of fatty acids. In this study, serum fatty acids were measured by the gas chromatograph-mass spectrometry method (GC-MS) before ovulation induction to explore the influence of serum fatty acids on pregnancy in PCOS when undergoing ovulation induction.

Patient selection
This study was approved by Tianjin Medical University General Hospital. All of the participants signed informed consent forms to be included in the study.
The study enrolled infertility PCOS patients, diagnosed according to Rotterdam's criteria, who underwent ovulation induction from January 2020 to December 2020 at the Tianjin Medical University General Hospital. All included patients were ≤35-year old. Selection required the diagnosis of two of the three following conditions: ovulatory dysfunction, hyperandrogenism or polycystic ovaries by ultrasonography [2]. Exclusion criteria were as follows: (i) male factor infertility; (ii) tubal factor infertility; (iii) endometriosis; (iv) endometrial abnormalities; (v) uterine malformations ( Figure 1). All patients did not take hormonal drugs for 2 months before ovulation induction.

Interventions
Patients received letrozole (LE 2.5 mg) or clomiphene citrate (CC 50 mg) for ovarian stimulation from Day 3 to Day 7 of the menstrual cycle. Ovarian stimulation was continued with a personalized gonadotropin from Day 9 of the menstrual cycle. Follicular monitoring by trans-vaginal ultrasonography (TVS) was performed until a mature follicle was detected. Dosage of the gonadotropins was adjusted according to ovarian response in each patient. A single injection of 8000 IU human chorionic gonadotropin (hCG) was given, if at least one follicle attained 18 mm. TVS was performed after 48 h of hCG injection to determine follicle rupture. Corpus luteum support was maintained for fourteen days after ovulation and patients with pregnancy continued corpus luteum support. Those patients without pregnancy stopped corpus luteum support and received ovulation induction at the next menstrual cycle. The total number of ovulation induction cycles that an individual patient received did not exceed six.

Clinical measurements
Anthropometric parameters included height and weight. Body mass index (BMI) was calculated to assess body fat using the formula BMI = weight (kg)/height (m 2 ). Biochemical assays of reproduction and metabolism were tested at baseline. The basal serum levels of sex hormone binding globulin (SHBG), sex hormone and insulin were measured by chemiluminescence immunoassay (ARCHITECT i2000SR, USA). Fasting plasma glucose was tested by the glucose oxidase method (VITROS 5600, USA). The free androgen index (FAI) was calculated using the following equation: FAI = T (nmol/L)/SHBG (nmol/L) × 100. The homeostatic model of insulin resistance (HOMA-IR) was calculated by the formula HOMA-IR = fasting plasma insulin (mIU/L) × fasting plasma glucose (mmol/L)/22.5 [8].

GC-MS method
Fasting blood was drawn from Day 2 to Day 4 before ovulation induction treatment for fatty acid determination. Fatty acids were measured only before the first ovulation induction cycle. We collected whole blood from patients using serum separation gel promoting coagulation tube. Immediately after blood collection, gently invert the blood collection tube 4-5 times to mix the specimens, wait for the specimens to fully coagulate and need to stand for 30 min, the centrifugation radius is 8 cm, and the centrifugation speed is maintained at 3500-4000 rpm for 10 min. Serum and blood clots were completely separated by separating gel and we stored the serum at −80 °C until GC-MS analyses were conducted.
First, 2 mL of 1% sulfuric acid methanol solution (Thermo, USA) was added to the 200 μL samples and placed in an 80 °C water bath for 30 min. Then, 1 mL hexane (Yong Hua, China) and water were added to the samples. The samples were then centrifuged at 3500 rpm 4 °C for 10 min, and 100 mg of anhydrous sodium sulfate powder was then added to remove excess water. Last, the samples were diluted with 200 μL hexane, and 15 μL 500 ppm methyl salicylate (TCI) was added as an internal standard. Furthermore, 200 μL of supernatant was collected for GC-MS.
A GC-MS system (Thermo Trace 1310 GC-Thermo ISQ 7000, Thermo-Fisher Scientific Corp, Fairlawn, USA) was used to quantitate fatty acids. The system utilized a Thermo TG-Fame capillary column, 50 m × 0.25 mm ID × 0.20 μm. Hydrogen was used as the carrier gas, and the inlet, ion source, and transfer line temperatures were set at 250 °C, 200 °C, and 280 °C, respectively. Forty-three fatty acids were measured.

Follow-up
Ninety-three infertility PCOS patients accepted ovulation induction treatment. Two patients withdrew, and 91 patients remained with complete follow-up data ( Figure 1). Pregnancies were confirmed by ultrasound at six-to-seven week gestation. Clinical pregnancy was defined as a gestational sac observed by TVS. Biochemical pregnancy was diagnosed as hCG ≥ 25 mIU/mL. Pregnancy loss included loss of intrauterine pregnancy <12 weeks as well as biochemical pregnancy losses. Live birth was defined as parturition of newborns weighing ≥ 1000 g after at least 28 weeks of gestation [9].

Statistical analysis
Data were initially tested for normal distribution using the Kolmogorov-Smirnov test. Data with normal distribution were analyzed using the t test to compare groups. Data not normally distributed were analyzed using the Mann-Whitney U test. Data were presented as mean ± SD or median (interquartile range). Categorical variables were analyzed using the Chi-square test [10]. Binary logistic regression analysis was used to adjust [11,12]. The value of hosmer was >0.05 in our study. To better visualize changes in the proportions of particular fatty acids, the results were calculated with respect to the percentage of fatty acids in serum. SPSS version 23 was used for statistical analyses and p < .05 was considered statistically significant.

Results
Overall, 172 ovarian cycles were studied in 91 PCOS patients. Among the 91 infertility PCOS patients, 37 patients (40.66%) achieved clinical pregnancy. Of the 37 patients who successfully became pregnant, 29 patients delivered 30 healthy infants (one set of twins). The live birth rate was 78.38%. Eight patients (21.62%) miscarried. Patients were divided into two groups: the pregnancy group (n = 37) and the nonpregnancy group (n = 54).

Discussion
Baseline characteristics including metabolism and endocrine and endometrial thicknesss were not statistically different between the pregnancy group and the nonpregnancy group in our results. We found that levels of SFAs and MUFAs were not associated with pregnancy and live birth. We observed that nervonic acid was negatively associated with pregnancy. Nervous acid is a monounsaturated fatty acid. It is mainly derived from the oilseed trees which mainly distributed in limited regions of Guangxi and Yunnan province in china. For many years, the seeds of M. oleifera have been used for making edible oils. We found that serum nervonic acid is about 0.3%, which is indeed lower than the figures in other research. We speculate that it may be due to different dietary habits in different regions. Xiaojing et al. proposed nervonic acid as a potential biomarker of PCOS patients, and there correlation analysis showed that nervonic acid was positively correlated with the level of free testosterone [13]. Szczuko et al. also found that the level of nervonic acid in PCOS patients was higher than in other women [10]. We speculate that PCOS patients with higher nervonic acid may exhibit a heavier phenotype and thus find it more difficult to become pregnant. However, more research is needed to explore its possible mechanism.
Bhardwaj et al. showed that serum TFAs, including PCOS, were related to ovulatory infertility in women [14]. In our study, the levels of TFAs were not statistically different between the pregnancy group and the nonpregnancy group, but there was a decreasing trend in the pregnancy group. Meanwhile, trans-10-heptadecenoic acid, trans-vaccenic acid, trans-11-eicosenoic acid, and brassidic acid were negatively associated with pregnancy. Our results illustrated that serum TFAs not only affect ovulation but also adversely affect pregnancy. TFAs are known to cause changes in membrane enzyme functions and in certain cellular reactions because they block the oxidation of cis fatty acids and alter membrane fluidity as a component of membrane phospholipids [15]. They were thought to have an adverse effect on ovum quality as a result of changing membrane lipid composition [15]. TFAs might influence clinical pregnancy by affecting the quality of the ovum. A retrospective study comparing dietary data on TFAs among 104 women with insulin resistance showed a significant relationship between TFAs and fetal loss (OR = 0.84, p = .017) [16]. Keenwan et al. included 1228 women attempting pregnancy with one-to-two previous pregnancy losses, and the results suggested that TFAs were not associated with pregnancy outcomes [7]. Perhaps the number of pregnancy loss patients was small, but our study found no statistical difference in serum TFAs between the live birth group and the pregnancy loss group.
In our study, levels of total PUFAs and several individual PUFAs, -including LA, docosapentaenoic acid, ALA, and EPAin the live birth group were significantly higher than in the pregnancy loss group. ALA and EPA belong to the n -3 series of PUFAs. On one hand, during pregnancy, there is a higher metabolic requirement for n -3 PUFAs. PUFA levels in the maternal circulation reflect the major determinant of fetus EPA supply [17]. In our study, higher levels of preconception serum EPA in the live birth group illustrated the importance of adequate EPA reserves for infertility PCOS patients. On the other hand, n -3 PUFAs inhibit the n -6 PUFAs' pathway and inhibit the n -6 PUFAs' pathway prostaglandins (PGs), which are generally categorized as anti-inflammatory [18]. Increasing n -3 PUFA levels could improve the inflammatory response and potentially sustain continued pregnancy. In a prospective cohort study, Mirabi et al. analyzed the influence of fatty acids in serum and follicular fluid on the outcome of intracytoplasmic sperm injection (ICSI) and found that the average serum EPA level of pregnant women was significantly higher than that of nonpregnant women [19]. Chiu et al. found that serum EPA was positively associated with the live birth rate among infertility patients with assisted reproductive technology treatment. For every 1% increase of n -3 PUFA in serum, both the clinical pregnancy rate and the live birth rate increased by 8% [20]. LA derivatives are inflammatory mediators and inflammation is associated with the normal course of embryo implantation, the course of pregnancy, and childbirth [21]. Previous research found serum n -6 PUFAs were associated with a lower probability of pregnancy [7], which was contrary to our result. Serum n -6 PUFAs play a role in promoting inflammation, but inflammation in the presence of LA derivatives is necessary during pregnancy. For example, it has been found that in cyclooxygenase-2 (COX-2) deficiency conditions, incorrect embryo implantation and decidual reaction occur [22]. Our results revealed that the average level of serum LA was lower in the pregnancy loss group which showed that n -6 PUFAs are necessary to maintain pregnancy. Further investigation might examine the trend of n -6 PUFAS at different gestational weeks and explore the significance of n -6 PUFAs in depth.
The literature suggests adherence to the Mediterranean diet could improve metabolic abnormalities and infertility in PCOS patients [23]. A prospective study that recruited 93 women undergoing in vitro fertilization treatment suggested that the embryo fragmentation score was significantly positively correlated with oleic acid concentration [24]. Mirabi et al. proposed that the number of mature oocytes was positively correlated with the levels of oleic acid [19]. However, in terms of implantation, clinical pregnancy and live birth, our study did not find oleic acid to be beneficial.
The study's primary limitation is that it was conducted at a single center with a small number of cases. Hence, further large-scale trials are necessary to clarify our findings.

Conclusion
In conclusion, serum fatty acids were associated with pregnancy. Our findings indicate that polyunsaturated fatty acids are associated with live birth in PCOS patients. Serum trans-fatty acids and nervonic acid might be risk factors for nonpregnancy. The mechanism of the influence of different fatty acids on pregnancy and live birth merits further exploration.