The Effect of Maternal Body Mass Index on In Vitro Fertilization-Intracytoplasmic Sperm Injection Treatment: A Prospective Comparative Study


 Purpose: To evaluate whether or not maternal body mass index (BMI) has an effect on pregnancy rates following in vitro fertilization-intracytoplasmic sperm injection (IVF-ICSI) treatment.Methods: A total of 869 patients who had undergone IVF-ICSI treatment between 2012 and 2017 were included in this study. The participants were stratified according to maternal BMI as Group 1 (BMI<25 kg/m2; n=394), Group 2 (25 kg/m225 kg/m2; n=172). Basal parameters and IVF-ICSI outcomes were compared between the groups.Results: While there were no differences between the groups in terms of age, smoking status, etiology of infertility, thyroid-stimulating hormone, prolactin levels, antral follicle count, and stimulation protocol (p>0.05), there was significant statistical difference (p<0.05) in terms of BMI, duration of infertility, baseline follicle-stimulating hormone, luteinizing hormone, estradiol (E2), duration of stimulation, total gonadotropin dose required, peak E2 levels, progesterone levels, endometrial thickness on hCG administration, and cycle cancellation rate. In addition, the numbers of MII and 2PN oocytes retrieved and the rates of clinical pregnancy, live births, and miscarriages were also different between the groups (p<0.05).Conclusion: Our data suggest that there is an inverse impact of increased BMI on laboratory and reproductive outcome parameters of IVF-ICSI treatment. Taking cost-effectiveness into consideration, weight loss should be suggested before ovulation is induced.


Introduction
Obesity is a serious public health problem in developed and developing countries. Historically more prevalent in more advanced age groups, it is now seen with increasing frequency in reproductive ages where endocrinological effects such as hypothalamic-pituitary ovarian axis disorder can cause changes in the secretion of pulsatile gonadotropin, sex hormonebinding globulin, and ovarian androgen. This can prompt menstrual irregularity, anovulation, insulin resistance, have negative psychological and social effects, and can increase the risk of infertility by three times (1)(2)(3). It has also been demonstrated that weight loss can return women to spontaneous ovulation and pregnancy without any additional treatment (4-6).
Elsewhere, it has been reported that obesity reduces fecundity and increases the rate of miscarriage through a negative effect on endometrial receptivity (7). The adverse effects of pre-pregnancy obesity on perinatal (e.g., preeclampsia, gestational diabetes mellitus, preterm labor, and surgical delivery) and neonatal (e.g., macrosomic fetus) outcomes have also been reported (2). Additionally, obesity has been found to have a negative effect on serum testosterone and estrogen in men and to decrease sperm motility and quality (4,5) There are con icting results in the literature regarding the effects of maternal obesity on the success of assisted reproductive techniques (ART). Although some studies have found that maternal body mass index (BMI) has no negative effect on ART outcomes (8-11), others show that a higher BMI increases the amount of gonadotropin required, produces fewer oocytes, increases IVF-ICSI cancellation rate, decreases clinical pregnancy and live birth rates, and increases miscarriage rate (12)(13)(14). Since the effects of obesity on ART outcomes have not been fully elucidated, the current study sought to evaluate whether or not maternal BMI has an effect on pregnancy rates following in vitro fertilization-intracytoplasmic sperm injection (IVF-ICSI) treatment.

Materials And Method
Study participants and data collection This prospective study was carried out at Ali Kemal Belviranlı Women's Health and Children's Hospital, IVF Unit. Outcomes of 757 fresh ICSI cycles were reviewed between January 2012 and December 2017. Inclusion criteria were participants aged ultrasound, and normal baseline hormonal levels. Participants were excluded from the study if they were >45 years, any diseases that affect the outcome of IVF/ICSI, such as hydrosalpinx and endometriosis. The ethical board approval was given from the institutional review board (2012/57). Written and oral informed agrement was given from the participants.
Ovarian stimulation and oocyte retrieval Controlled ovulation stimulation was negotiated using the gonadotropin-releasing hormone agonist (GnRHa) or the exible gonadotropin-releasing hormone antagonist (GnRHant) protocol.

ET Procedure
Two senior physicians performed the ETs accompanied ultrasonographic appearance (Logiq 200 Pro, General Electric, Seoul, South Korea) using an embryo transfer catheter system. A sterile speculum was introduced to the vagina in the lithotomy position and the vagina and the cervix were cleared using sterile cotton swabs.
An embryologist loaded the embryos into a soft transfer catheter which was advanced to the ET physician who deposited the embryos approximately 10 mm from the uterine fundus under USG. The catheter was gently removed after 5 seconds. In cases of ET with external guidance, an initial catheter with inner sheath was inserted into the external cervical os, and then advanced through the cervical canal and internal os to 10 mm of the uterine fundus using USG. The internal sheath was withdrawn, and a second catheter loaded with embryos was introduced in its place and advanced to approximately 10 mm from the uterine fundus where the embryos were deposited. Di cult transfers required the use of a stylet in addition to this form of external guidance.
All catheters were immediately checked for retained embryos, blood, and the patient remained in the Trendelenburg position for about 10 minutes. Patients in whom tenaculum were excluded from the study. Luteal phase support was provided with progesterone in the form of Crinone 8% gel (Serono, Istanbul, Turkey) at a daily dose of 90 mg. Baseline parameters and IVF-ICSI outcomes were compared between the groups. Biochemical pregnancy was detected with a by hCG levels in venous blood tests performed 12-14 days after embryo transfer, and clinical pregnancy was accepted as those with a gestational sac accompanying fetal heart-beart on ultrasound examination at 4-5 weeks after embryo transfer. Live birth was de ned as the birth of a live fetus after 22 weeks of gestational age. The subjects were strati ed according to the maternal BMI as Group 1 (BMI<25 kg/m 2 ; n= 394), Group 2 (25 kg/m 2 < BMI <30 kg/m 2 ; n=303), and Group 3 (BMI >25 kg/m 2 ; n=172). Basal parameters, clini cal and laboratory IVF-ICSI outcomes, and reproductive outcome parameters were compared between the groups.

Statistical analysis
The statistical analyses were performed using SPSS 15.0 for Windows (SPSS, Chicago, IL, USA). The Kolmogorov-Smirnov test was used for examining the continuous variables with normal and non-normal distributions. The one-way analysis of variance (ANOVA) for normally distributed variables and the Kruskal-Wallis test for not-normally distributed variables were used to compare groups. Categorical data were examined by Pearson's chi-square test, and Fisher's exact test was applied if the expected frequency was less than 5 in >20% of all cells. The continuous variables were presented as the mean±standard deviation (SD) and the categorical variables were demonstrated as the number of cases and percentages. The Bonferroniadjustment was used to control the type I errors for all possible multiple comparisons. A p<0.05 value was established as statistically signi cant.
A comparison of the sociodemographic and stimulation characteristics of the participants is provided in Table 1. While there were no differences between the groups in terms of age, smoking status, etiology of infertility, thyroid-stimulating hormone, prolactin levels, antral follicle count, and stimulation protocol (p>0.05), there was signi cant statistical difference (p<0.05) in terms of BMI, duration of infertility, baseline follicle-stimulating hormone, luteinizing hormone, estradiol (E 2 ), duration of stimulation, total gonadotropin dose required, peak E 2 levels, progesterone levels, endometrial thickness on hCG administration, and cycle cancellation rate.
The laboratory and reproductive outcomes of the participants are summarized in Table 2. The numbers of MII oocytes and 2PN retrieved and the rates of clinical pregnancy, live births, and miscarriages were also different between the groups (p<0.05).

Discussion
We found that overweight and obese patients with higher BMIs had worse responder rates, lower peak E 2 levels, and less endometrial thickness and required higher gonadotropin doses than the normal weight group. In addition, lower numbers of MII and 2PN oocytes were retrieved, the clinical pregnancy and live birth rates were lower, and the miscarriage rat was higher.
Obesity has historically been observed more frequently in adult and advanced age groups. It is now a global epidemic and has become an important public health problem in younger age groups, too (2). Infertile women, including obese patients, in the reproductive age group bene t from ART to ful ll their fertility requirements, and so the possible effects of increased BMI on ART are of great importance for the clinician, the patient, and public health (3).
Endocrinological and paracrinological factors play a role in the interaction between embryo and endometrium for successful implantation and live birth (15). Hyperandrogenemia, insulin resistance, and abnormal hormone levels that occur with increased BMI can negatively affect this process (1,16). A higher BMI also affects the levels of in ammatory markers such as insulin-like growth factors, tumor necrosis factor-alpha, and interleukine-6 which play roles in cell differentiation and differentiation, folliculogenesis, oocyte maturation, and embryo development. As a result, embryo implantation can be negatively affected and the risk of miscarriage can increase (17)(18)(19). Previous studies have shown that increased BMI is associated with poor IVF-ICSI outcomes through the effect of these endocrinological factors (2,20-21) with one study reporting that a reduction of one BMI unit can increase the chance of pregnancy by 19% (22). It has also been found that advanced maternal age and smoking negatively affect live birth rates (23), although the mean age and smoking rates were similar between the groups in our study.
In the literature, con icting results exist regarding the effects of increased BMI on ART outcomes. found that obese and overweight patient groups had poorer outcomes compared to normal weight women and that the obese group was worse than the overweight patient group (1). These ndings were in agreement with our results. Sartorius et al. (25) demonstrate that an increased BMI can have other negative perinatal outcomes such as preeclampsia, preterm birth, and surgical delivery, as well as reducing live birth rates. Relatedly, Pinborg et al. (6) evaluate 1,417 IVF-ICSI cycles and show that the cancellation rate increases with increased BMI, the two key reasons being that obesity makes the oocyte pick-up procedure more di cult and insu cient follicles are developed despite the use of high gonadotropin doses.
The possible negative effects of an increased BMI on ART should be explained to overweight and obese women who are scheduled for IVF-ICSI treatment. Before the process begins, it should also be explained that weight loss can increase the chance of success in terms of pregnancy and live birth. Overweight and obese women should consequently be encouraged to lose weight, and clinicians planning ART should implement weight loss programs involving diet and exercise. In addition, it should be understood that the gonadotropin dose required will increase with higher BMIs meaning that treatment costs will also grow despite the increasingly negative perinatal outcomes.
The strong point of the current study consist of its prospective arrangement, the adequate number of subjects in each group, and the prototypical sample from central Turkey. The results can be generalized to most of the country's population. However, the potential limitations of the study are that it was conducted in a tertiary single care institution and that the cumulative CPR was not evaluated because no frozen ETs were included.
In conclusion, this study found that an increased BMI has a negative effect on ART outcomes as shown in decreasing clinical pregnancy and live birth rates and increasing miscarriage rates. Further studies with more participants are needed to elucidate this effect.    Figure 1 Flowchart of the Study. Enrollment and follow-up of the study subjects.