Reproductive function and pregnancy outcomes in women treated for idiopathic hyperprolactinemia: A non-randomized controlled study

Abstract Background Few studies have focused to determine the peculiarities of the course of pregnancy and pregnancy outcomes after treatment in women with idiopathic hyperprolactinemia. Objective To determine the peculiarities of the course of pregnancy and pregnancy outcomes in women treated for idiopathic hyperprolactinemia, with history of infertility and/or recurrent pregnancy loss. Materials and Methods A non-randomized controlled study was conducted at the Center for Reproductive Medicine “Universe" and Medical Clinic “Medhealth” during 2016-2018, involving 96 women with idiopathic hyperprolactinemia, aged 20-44 yr with infertility and/or a history of recurrent pregnancy loss. Prolactin (PRL), follicle-stimulating hormone, luteinizing hormone, estradiol (E2), free testosterone, and progesterone were studied in blood serum using immunoassay analysis method. Before the occurrence of pregnancy, hyperprolactinemia was treated with bromocriptine. Dydrogesterone was used to support the luteal phase. Results PRL levels decreased significantly and normalized within two-five months, regular menstrual cycle was restored in two-four months, ovulation was restored in three-seven months, and pregnancy was achieved in three-fourteen months. E2 and progesterone levels increased significantly (p < 0.001). Prior to the treatment, significant negative correlation between PRL and E2 (r = -0.386, p = 0.007), PRL and progesterone (r = -0.420, p = 0.003) was detected. Threatened early abortion prevailed among pregnancy complications. Pregnancy loss in the first trimester was recorded in 3.12% of cases. Conclusion Pregnancy outcomes in women with idiopathic hyperprolactinemia are improved by prolonged and continuous treatment with bromocriptine before pregnancy and administration of dydrogesterone in support of the luteal insufficiency.


Introduction
Prolactin (PRL) is anterior pituitary hormone, which significantly influences woman's reproductive function (1). In cases of abnormal increase of blood PRL levels, often an ovulation disorder, luteal phase deficiency, and chronic anovulation are present that lead to fertility disorder (2,3). Moreover, the prevalence of hyperprolactinemia in women with reproductive disorders varies between 9 and 17% (4).
Besides infertility, recurrent pregnancy loss (RPL) is also a significant problem, which ranges from 2-5% among the couples (5). Endocrine factors are among important causes of RPL, which also include hyperprolactinemia (6), though its role in RPL is not completely specified. One of the mechanisms of negative impact of hyperprolactinemia on reproductive function is considered to be a luteal phase deficiency developed on its background (7). Luteal phase deficiency is considered to be a condition of decreased progesterone secretion, which is essential for both the secretory transformation of endometrium and normal embryo implantation and growth (8). There are three types of hyperprolactinemia: mild hyperprolactinemia, where PRL level varies from 25-50 ng/ml, moderate hyperprolactinemia, where PRL level is between 50 and 100 ng/ml, and higher PRL levels > 100 ng/ml.
Besides, in about 40% of moderate hyperprolactinemia, the cause of PRL hypersecretion cannot be determined. In such cases, it is qualified as idiopathic hyperprolactinemia (9)(10)(11). Medication treatment with dopamine agonists is considered as a firstline treatment for idiopathic hyperprolactinemia. Currently, the three most widely used medications in this group are bromocriptine (non-selective agonist), cabergoline (selective agonist), and quinagolide (selective agonist). They restore ovulation and fertility (12). Bromocriptine is preferred drug among women who want to give birth. It improves pregnancy outcomes in women with history of RPL and idiopathic hyperprolactinemia (13). Although bromocriptine is considered as safe at early pregnancy stages, its safety for the whole term of pregnancy is not yet determined (2). Treatment with bromocriptine has good outcomes; however, in 75% of cases the PRL levels increase again after the discontinuation of treatment (14). It is important for infertility to maintain low PRL levels from 10-12 months (2).
To date there are very few studies (2,6,12,13) focused on determining the peculiarities of the course of pregnancy and pregnancy outcomes after treatment in women with hyperprolactinemia. There are no sufficient evidences on the effectiveness of dopamine agonists in relation to improved pregnancy outcomes in women with idiopathic hyperprolactinemia and RPL (12). It is important to conduct research in this direction. Therefore, the objective of our study is to determine the peculiarities of the course of pregnancy and pregnancy outcomes in women treated for idiopathic hyperprolactinemia, with history of infertility and/or RPL.

Materials and Methods
A non-randomized controlled study was conducted at the Center for Reproductive Medicine "Universe" and Medical Clinic "Medhealth" between September 2016 and October 2018 and included 96 women aged 20-44 yr (average age 29.94 ± 4.59 yr), diagnosed with infertility and/or two or more pregnancy loss before their inclusion in the study, and those who had mild and moderate increase in PRL level (39.0 ng/ml -68.0 ng/ml; Average 49.46 ± 8.77). The inclusion criteria of the study were women with infertility and/or RPL with idiopathic hyperprolactinemia, who had not received treatment for six months prior to inclusion in the study. However, other etiological factors (such as anatomical, genetic, immunological, infectious, and other endocrinological factors) of infertility and/or RPL were considered as the criteria for non-inclusion in the study. Additionally, those with hyperprolactinemia due to organic damages, polycystic ovary syndrome and congenital adrenal hyperplasia, primary hypothyroidism, or due to pharmacological medications were also excluded.
Prior to inclusion in the study, all women were tested on thyroid-stimulating hormone and free thyroxine levels in blood serum and underwent ultrasound examination of thyroid gland. Based on the results of these studies, hyperprolactinemia developed due to primary hypothyroidism was excluded. PCOS, as one of the most prevailed mild and moderate causes of hyperprolactinemia, was excluded based on the Rotterdam 2003 Consensus Diagnostic Criteria for PCOS. Congenital adrenal hyperplasia was excluded on the basis of hormonal investigation of 17-hydroxyprogesterone and dehydroepiandrosterone sulfate. Organic causes of hyperprolactinemia were excluded with the study of hypothalamic-pituitary area with the use of magnetic resonance imaging with contrast agent.
Women involved in the study were divided into three groups: group (I) 63 women with primary infertility; Group (II) 15 women with RPL; and Group (III) 18 women with secondary infertility. Group III was further divided into subgroups: subgroup (A): 7 women with the history of physiological delivery and subgroup (B): 11 women with the history of one spontaneous abortion. Archival material were considered for controlmedical records of 78 women with idiopathic hyperprolactinemia, who were not treated with bromocriptine continuously before pregnancy, stopped taking it immediately as the ovulation was restored, and did not receive dydrogesterone neither before nor during pregnancy. During the study of the history, attention was focused on peculiarities of menstrual and reproductive functions. Among the objective data, we focused on body mass index, galactorrhea, and dermatopathies (acne, seborrhea, and hirsutism). All women underwent gynecological examinations.
Using the immunoassay analysis method (ELISA) (with the use of Rayto Rt-2100 C Microplate Rider), the following hormones in the blood serum were studied: before starting the treatment, the serum PRL levels were determined twice, on the 2nd and 3rd days of menstrual cycle: PRL (initially once a month before the restoration of normal levels and then once every three months), follicle-stimulating hormone (FSH), luteinizing hormone (LH), estradiol (E2), free testosterone (FT) (initially before treatment and after PRL levels were normalized), progesterone level was determined on the 22 nd and 23 rd days of cycle (initially before treatment and after PRL levels were normalized). Prior to the treatment, in cases of amenorrhea and oligomenorrhea, hormonal study was conducted on the days mentioned earlier, following the menstrual induction. With the use of study device (Voluson E8) ultrasound examination of small pelvic cavity organs was performed. Ovulation was identified with ultrasound examination through follicle monitoring and determination of progesterone on the 22 nd and 23 rd days of cycle.
Further, women included in the study were treated with bromocriptine continuously until the occurrence of pregnancy. Treatment was started with 1.25 mg daily. Dose was increased gradually by 1.25 mg every week, until PRL reached normal levels (to a median of 5.0-7.5 mg/day in two intakes). In order to maintain long-term normal PRL levels, bromocriptine was prescribed in minimal doses (1.25 mg/day in two intakes) for three menstrual cycles after restoration of ovulation against the background of barrier contraception. In order to support luteal phase, 10 mg dydrogesterone was prescribed for peroral administration for at least three menstrual cycles from 14 th through 25 th day. After the discontinuation of barrier contraception treatment was continued with both bromocriptine and dydrogesterone in the same doses until the occurrence of pregnancy. For the prevention of luteal deficiency, women were given dydrogesterone 10 mg daily during the first trimester of pregnancy. Women with the history of RPL, dydrogesterone 10 mg was given twice a day until the 20 th week of pregnancy with gradual dose decrease. In cases of threatened abortion, they were given a single dose (40 mg) of dydrogesterone, followed by 10 mg every 8 hr until symptoms disappeared. All pregnancy complications and pregnancy outcomes were accurately recorded.

Ethical consideration
The study was approved by the Ethics Committee of the Center for Reproductive Medicine "Universe", Tbilisi, Georgia (No3/2016). Written informed consent to participate in the study were obtained from all patients.

Statistical analysis
While the continuous variables are expressed as Mean ± SD, the categorical variables are expressed as frequencies and percentages. Paired Samples t test was used to compare the continuous variables, and a correlation analysis between continuous variables of PRL and other hormones (FSH, LH, FT, E2, and progesterone) was performed using the Pearson's correlation test. All statistical analyses were performed using the Statistical Package for the Social Sciences, version 22.0, SPSS Inc., Chicago, Illinois, USA. Idiopathic hyperprolactinemia and pregnancy outcomes increased significantly (p < 0.001). While the FT levels decreased significantly (p < 0.001), it was within the normal range before and after treatment. FSH and LH levels were within the normal range before treatment and did not change significantly after the treatment (Table  I).

Results
Correlations between PRL and other hormones were detected. Before treatment, FSH, LH, E2, and progesterone showed reliable negative correlation with PRL (Table II).
Among the studied women, threatened early abortion prevailed among the complications of pregnancy -11.45% (n = 11). Its highest rate was recorded in women with RPL and reached 20% (n = 3). While among women with secondary infertility and a history of one spontaneous abortion, threatened abortion was observed in 18.2% (n = 2), it was observed in 9.5% (n = 6) of the women with primary infertility. In women with secondary infertility with the history of physiological delivery, no case of threatened early abortion was detected. Threatened preterm labor was recorded in 4.17% (n = 4) of women involved in the study and all these occurred in the group of women with primary infertility (6.3%) on week 25-28 of pregnancy. Placental insufficiency occurred in 2.08% (n = 2) and both cases were recorded in the group of women with primary infertility (3.17%). Preeclampsia occurred in 2.08% (n = 2). Both cases occurred in the group of women with primary infertility and made up for 3.17%. In both cases, pregnancy was carried to full term and ended in vaginal delivery. Pregnancy loss occurred in 3.12% (n = 3). All the three cases were recorded in group I: two in the fifth week and one in the seventh week of pregnancy. In all other cases, pregnancy ended in a live healthy newborn. About 85.42% (n = 82) of women involved in the study had a vaginal delivery. Cesarean section was performed in 11.46% (n = 11). Cesarean delivery mainly prevailed in group I and the reasons for cesarean section were: abnormal fetal position in four cases, narrow pelvic outlet in four cases, in pelvic varicose veins in one case, and myopia in two cases.
Pregnancy loss rate was significantly low in study group than in the control group. All cases of pregnancy loss in both the study and control groups were reported in the first trimester of pregnancy. Timely delivery rate was significantly higher in the study group than in the control group and preterm labor was reported in the control group only (Table  III).  Data presented as n (%); categorical variables were compared using the Pearson's 2

Discussion
PRL is known to directly affect the ovaries. It inhibits secretion of estrogen and progesterone. Hyperprolactinemia causes inhibition of luteinization and steroidogenesis in the ovaries (15)(16)(17). In our study, patients had a moderate increase in PRL at initial stage. The higher the PRL level, the greater the decrease in E2 and progesterone levels in the blood serum, FSH and LH levels remained within the norm. On the background of moderate hyperprolactinemia 73.96% of women included in our study had oligomenorrhea, 4.17% had oligomenorrhea, 14.58% secondary amenorrhea, and 7.29% regular menstrual cycle, of which 71.43% had anovulation and 28.57% luteal phase deficiency. It should be mentioned that at higher PRL levels more profound menstrual disorders were observed. In the study reported by other authors, the rate of amenorrhea was 35% (17), which is probably due to the higher PRL levels compared to our study.

Conclusion
Prolonged and continuous treatment with bromocriptine before the pregnancy and with dydrogesterone before and in the first trimester of the pregnancy led to a significantly high rate of timely delivery in the study group compared to the control group. The rate of miscarriage and preterm labor was significantly low in the study group than in the control group. In both the groups miscarriages were reported in the first trimester of the pregnancy.
Pregnancy outcomes in women with idiopathic hyperprolactinemia were improved by prolonged and continuous treatment with bromocriptine before the occurrence of pregnancy and use of dydrogesterone to support the luteal phase before pregnancy and during the first trimester of pregnancy.