The Body Composition of Women Undergoing ART, and its Relationship with Sex Hormones and Biochemical Indicators

The Body of Undergoing and its Relationship with Hormones and Biochemical Indicators. Abstract This study displays a retrospective cohort analysis in the Reproductive and Genetic Health Center of Peking University First Hospital. Data were collected from 411 women receiving ART. The body composition, sex hormones and biochemical indicators were assessed by bioelectrical impedance analysis (BIA). Subjects were grouped by body mass index (BMI) and body fat percentage (BFP), and differences of body composition among groups were analyzed. Multiple linear regression analysis was used to evaluate the relationship between sex hormones, biochemical indicators and body composition. Compared with the non-overweight group (BMI < 24kg/m 2 ), the lean body mass and fat mass of the overweight group (BMI≥24kg/m 2 ) increased significantly (P < 0.001). Notably, 43.7% of non-overweight subjects have increased body fat as distinguished by PBF. The visceral adipose tissue increased by 50.9% in subjects with more fat. Fasting blood glucose (FBG), triglyceride, low density lipoprotein cholesterol (LDL-C) and testosterone were positively correlated with trunk fat, while high density lipoprotein cholesterol (HDL-C) and estradiol were negatively correlated with trunk fat. Moreover, total cholesterol was positively correlated with bone mineral content (BMC). Therefore, this research demonstrated significant differences in body composition between overweight and non-overweight women receiving ART. Combining PBF with BMI, obesity can be identified more accurately. Biochemical metabolism and sex hormones were related to body composition. Therefore, women receiving ART requires detailed measurements of body composition, not BMI. Further studies are needed to confirm the relationship between pregnancy outcomes and body composition in women receiving ART treatment.


Introduction
The prevalence of obesity is increasing worldwide, almost tripling in the last four decades. In 2016, over 650 million adults were obese, 15% of whom were women (WHO, 2017). Meanwhile infertility has become a crucial health issue in recent decades. In 2010, as many as 48.5 million couples worldwide suffered from being unable to conceive [1]. More and more obese women are being evaluated for infertility [2]. It is confirmed that an increase in BMI can lead to ovulation failure [3]. Moreover, higher BMI was associated with worse ART outcomes, including lower implantation rates and increased risk of abortion [4][5][6][7]. It is assumed that the effects of obesity on sex hormones and metabolism may partly explain the differences in treatment outcomes [8]. Obesity is always characterized by a high BMI, which means not only changes in fat content, but also other physical parameters, such as total body water (TBM), muscle and bone mineral content (BMC). Therefore, for infertile women with the high BMI, more attention should be paid to the changes of body composition. It is necessary to explore In the analysis of body composition, PBF seems to be a new index of obesity compared with BMI. Unlike BMI, PBF directly reflects the content and distribution of body fat. Whether BMI or PBF is more suitable for assessing obesity remains to be determined.
The purpose of this study was to assess the body composition of women receiving ART with abnormal BMI, and to evaluate the effectiveness of BMI and PBF to identify fat elevation. The second goal is to determine the relationship between sex hormones, metabolic markers and body composition.

Measures and variables
Venous blood samples were collected after 10-12 hours of overnight fasting. FBG, triglyceride, total cholesterol, HDL-C, LDL-C, urea and creatinine were analyzed by automatic analyzer. Blood samples for sex hormone analysis were collected around 9:00 am on the second or third day of menstruation.
The patient's height was measured by an ultrasonic altimeter in meters with a measurement accuracy of 1 cm. The nude weight was measured with an electronic scale accurate to 100 grams.
The patients needed to wear underwear only, fast overnight, and empty urine and stool. BMI was calculated as weight (kg)/height (m) 2 . According to the criteria of China, subjects were divided into overweight (BMI ≥ 24kg/m 2 , including overweight and obesity) and non-overweight (BMI < 24kg/m 2 ) [13]. Waist circumference was measured with soft elastic tape midway between the 12 th rib and iliac crest. When the subjects were standing, hip circumference was determined as the widest horizontal plane of the hip. WHR was calculated as waist circumference (cm)/hip circumference (cm).
The bioelectrical impedance analysis device detects the human body as five cylinders (four limbs and trunk) and calculates the whole and part of the human body by using eight electrodes and multi-frequency current. 411 eligible subjects were measured.
Those with metal objects such as pacemakers, defibrillators, coronary stents and artificial joints were excluded from physical analysis. Subjects were asked to have at least one diet and stressfree activity within six hours before the trial (2016).

Demographic, sex hormonal and biochemical characteristics
In this study, the majority of women receiving ART treatment  Table 1.
Anthropometric and physical measurement of overweight and non-overweight women. The categorical data were shown as n (%), the parametric variables were presented as mean±SD, and the non-parametric variables were presented as median (IQR). p<0.001) of the overweight women. In overweight subjects, lean body mass and fat mass increased. In addition, the basal metabolic rate (BMR) of overweight women was significantly higher than that of non-overweight women (1326 vs. 1220, P < 0.001) ( Table 2).

Sex hormones, biochemical indicators related to body composition
The more trunk fat mass, the higher FBGn. While triglyceriden,

Discussion
In this study of women receiving ART, we found that with obese group compared to those in normal weight group [15,16].
Therefore, there are significant differences in body composition between overweight and non-overweight groups. On the other hand, it was also confirmed that BMI cannot determine the overall weight gain caused by fat or lean mass. As a result, well-muscled people with normal body fat can be misdiagnosed as overweight or obese because of their high BMI. In addition, 43.7% of subjects were considered as non-overweight, but as overweight when combined with PBF because of the increase in fat content, which indicated that BMI cannot distinguish between normal-weight people who maintained a large amount of fat and those who did not have excess fat. Therefore, a more accurate estimation of body composition can be provided when combined BMI with PBF.
In the current study, apparently normal weight subjects with elevated fat mass had a higher VAT, rather than leg fat or arm fat mass. For fat distribution, VAT plays a vital role in metabolic disorders. Previous studies have detected that VAT were closely related to cardiometabolic risk factors and remarkably influenced the development of insulin resistance [17,18]. PBF seems to be a valid parameter reflecting total fat mass and VAT, and a larger PBF means a greater risk of metabolic diseases as a major increase in VAT.
In this study, results of multiple linear regression showed that testosterone was positively correlated with trunk fat mass, while estradiol was negatively correlated with testosterone, similar to previous studies on women [19][20][21][22]. Interestingly, there was a negative correlation between trunk fat mass and testosterone, while a positive correlation between estradiol and trunk fat mass in man [8,[23][24][25][26]. This difference between sex hormones and body composition needs further study. Current study has shown that FBG was positively correlated with trunk fat mass, which can be explained by the positive correlation between fat mass, especially VAT, and insulin resistance [19,27]. Furthermore, trunk fat mass but HDL-C had a positive effect on lipid distribution. Lipid metabolism and body fat have been studied in recent years. It is concluded that lipid metabolism is disturbed by adipocytokines fat mass, leading to hypertriglyceridemia and changes in HDL-C and LDL-C [28][29][30].
In particular, total cholesterol was related to BMC rather than fat mass in this study. However, previous studies have shown that total cholesterol is correlated with bone mineral density, thus it is required more studies to confirm BMC as a novel predictor for total cholesterol [31][32][33][34]. Considering the effects of body composition on sex hormones and metabolism, body composition analysis is superior to BMI in assessing women receiving ART. A favorable body composition can promote hormones and metabolic status, which is beneficial to the acquisition of fertility.
There were some limitations in this study. First of all, the retrospective design cannot make cause-effect inferences. Secondly, subjects from a single reproductive center were less representative.
Thirdly, we have not adjusted for recognized determinants, such as dietary habits, physical activities, socioeconomic status and history of drinking and smoking [35,36]. Moreover, the data of FBG, triglyceride, HDL-C, estradiol and testosterone were not normal distributed, and were transformed to normal scores using Blom's formula when analyzing with multiple linear regression, which reduces the clinical significance of multiple linear regression.

Conclusion
In conclusion, our results showed that there are significant differences in body composition and BMI in women under ART treatment. It was noticed that non-overweight subjects may have elevated body fat mass, combining PBF can provide more accurate body composition estimation due to BMI cannot distinguish between fat and lean mass. Biochemical metabolism and sex hormones were associated with body composition. Hence measurement methods for detailed body composition analysis rather than BMI are needed to examine women receiving ART treatment, and provide a basis for a self-management. Further studies are needed to determine the relationship between pregnancy outcomes and body composition in women receiving ART.

Conflicts of Interest
The authors declare no conflict of interest.