Does obesity compromise ovarian reserve markers? A clinician’s perspective
- First Online:
- Cite this article as:
- Malhotra, N., Bahadur, A., Singh, N. et al. Arch Gynecol Obstet (2013) 287: 161. doi:10.1007/s00404-012-2528-7
The aim of the study was to ascertain if increasing body mass index (BMI) adversely affects ovarian reserve among infertile women of Asian origin undergoing in vitro fertilization (IVF).
Materials and methods
This prospective study on 183 women was carried out in the infertility clinic of All India Institute of Medical Sciences, New Delhi, India. Blood hormonal assay in all patients including follicle-stimulating hormone (FSH), luteinizing hormone (LH) and inhibin B was performed on day 2/3 of a spontaneous cycle. A transvaginal ultrasonographic examination on day 2–5 of the menstrual cycle was done for antral follicle count (AFC) and ovarian volume. A correlation between BMI and ovarian reserve parameters like FSH, LH, inhibin B, antral follicle count and ovarian volume was noted.
Age was comparable in the three BMI groups. The mean duration of infertility was 8.38 years. Compared to the normal weight, the overweight and obese women had a statistically significantly low inhibin B (p < 0.0259). The AFC when taken together on both sides was not statistically significant between the groups; however, the overweight and obese women had a significantly low AFC (p < 0.0129) on the right side.
Incorporating anti-mullerian hormone, a newer marker for ovarian reserve, may benefit these obese infertile women. Further work is required to elucidate the mechanisms underlying the effect of obesity on inhibin B as a marker of ovarian reserve in infertile women. The best marker to check the ovarian reserve is perhaps the woman’s performance during an IVF cycle. However, considering the psychological and financial stress of the procedure, it may seem wise to consider counseling of obese women on their expected performance in the first cycle of IVF through such studies.
Obesity has been defined by the World Health Organization (WHO) as a body mass index of (BMI) ≥30 kg/m2. In recent years, a change in lifestyle in the resource-rich countries has witnessed an increase in the overweight (BMI 25–30 kg/m2) and obese population. Also, the developing world is now facing this greatest public health challenge. An imbalance between physical inactivity and caloric intake gradually tilts the balance toward obesity. The burden of this chronic disease is alarming with over 400 million adults worldwide being obese and another 1.6 million overweight in the year 2005 and the figures will only rise further . Obesity has a profound effect on the health of an individual, predisposing him to cardiovascular disease, diabetes, osteoarthritis and malignancies particularly of the colon, breast and endometrium .
There is a well-established connection between obesity and adverse reproductive outcome in a woman of reproductive age; the mechanism, however, is complex and multifactorial. A high body mass index (BMI) has a direct correlation with elevated levels of leptin in the serum and follicular fluid and decreased levels of serum adiponectin [3, 4]. Leptin, a hormone produced by adipocytes, inhibits ovarian steroidogenesis by acting through the receptors on theca and granulosa cells and lower adiponectin levels are associated with increased levels of circulating insulin [5–7]. Insulin resistance is thus a consequence of obesity and is inextricably linked to obesity-induced anovulation. High levels of insulin leads to low levels of sex hormone-binding globulin (SHBG), hyperandrogenemia and high levels of free insulin-like growth factor 1. Hyperandrogenemia in obese women leads to granulosa cell apoptosis and this may have a pathophysiological effect on ovarian function leading to disturbances in menstrual cycle and oligo-ovulation. Adipose tissue is the site for production and metabolism of sex steroids. The secretion of gonadotropin is inhibited by the peripheral conversion of androgens to estrogen (aromatase activity) in the adipose tissue, thus altering follicular development. Obesity per se impairs reproduction in women with increased rates of miscarriage and further decreases successful pregnancy outcome. Even a 5–10 % weight loss increases ovulation rates, reduces these biochemical abnormalities and improves fertility outcomes in a woman .
Obese women have a lower chance of conception following assisted reproductive techniques (ART), requiring higher dosage of gonadotropin, respond poorly to ovarian stimulation and have a higher chance of miscarriage. This could be attributed to poor quality oocytes or a defect in the endometrial receptivity and an insulin resistance.
The objective of this prospective study was to ascertain whether increasing body mass index (BMI) adversely affects ovarian reserve among infertile women of Asian origin undergoing in vitro fertilization (IVF).
Materials and methods
A total of 278 women infertile women who were attending the infertility clinic at All India Institute of Medical Sciences, New Delhi, India between Jan 2007 and December 2009 were seen prospectively. We selected 183 women who had both the ovaries intact and excluded 95 others who had either endometriosis polycystic ovaries and/or a single ovary and in patients with a prior surgical procedure done on the ovary/ovaries, such as cystectomy, ovarian drilling or endometriosis, and tubo-ovarian masses were excluded from this study. The study was approved by the institute’s ethics committee. The patients were divided into three groups according to their body mass index (BMI): normal weight group (18.5–24.9 kg/m2, n = 93), overweight group (25–29.9 kg/m2, n = 69) and obesity group (≥30 kg/m2, n = 21). The patients had infertility due to tubal factor, male factor and unexplained infertility. After a complete evaluation of both partners, they were planned for an IVF cycle subsequently.
All patients had a blood hormone assay including follicle-stimulating hormone (FSH), luteinizing hormone (LH) and inhibin B on day 2/3 of a spontaneous cycle. A transvaginal ultrasonographic examination on day 2–5 of the menstrual cycle was done by two independent clinicians for antral follicle count and ovarian volume.
The results were tabulated and presented as mean ± for quantitative variables and frequency (percentage) for qualitative variables.
Data were presented as number (%) or mean ± SD/median (range) as appropriate. The difference in means of various ovarian reserve markers among the different categories of body mass index (BMI) were compared using one-way analysis of variance (ANOVA). As inhibin B did not follow normal distribution, it was tested using Kruskal–Wallis test. The p value <0.05 was considered statistically significant. The statistical analysis was carried out using STATA 9.0 (College Station, Tx, USA).
Baseline demographic characteristics
Age in years
32.13 ± 4.04 (20–42)
Duration of infertility in years
8.38 ± 4.26 (2–22)
Body mass index (kg/m2)
25.35 ± 3.92 (17.15–36.5)
Normal (18.5–24.9 kg/m2)
93 (50.8 %)
Overweight (25–29.9 kg/m2)
69 (37.7 %)
Obese (≥30 kg/m2)
21 (11.5 %)
112 (61.2 %)
71 (38.8 %)
Primary and secondary outcome variables
Body mass index category
Antral follicle count, right ovary
Antral follicle count, left ovary
Ovarian volume, right ovary in cm3
Ovarian volume, left ovary in cm3
Follicle-stimulating hormone in IU/l
Luteinizing hormone in IU/l
Normal (18.5–24.9 kg/m2)
7 ± 2.64 (1–12)
6 ± 2.37 (2–12)
4.9 ± 3.9 (1.3–29.5)
5.1 ± 3.68 (1.6–25)
5.9 ± 1.85 (2.62–11.82)
4.9 ± 2.02 (1.13–10.71)
Overweight (25–29.9 kg/m2)
6 ± 2.46 (2–10)
6 ± 2.45 (1–10)
5.6 ± 2.8 (1.6–16.2)
4.8 ± 2.75 (2.4–14.9)
6.3 ± 2.48 (2.67–13.79)
3.8 ± 2.27 (1.14–11)
Obese (≥30 kg/m2)
5 ± 2.66 (0–10)
5 ± 2.36 (0–10)
4.3 ± 2.4 (1.6–10.8)
3.8 ± 3.56 (1.5–13.7)
6.4 ± 2.01 (2.91–10.56)
4.1 ± 1.34 (1.8–7.06)
The impact of obesity has been extensively explored and it is known to have sub-optimal reproductive performance [9, 10]. Recent reports suggest that obesity impairs oocyte quality by a mechanism known as lipotoxicity . It is well documented that overweight and obese women require higher amounts of gonadotropins and more days to achieve follicular maturation. Despite higher dosages of drugs, the number and quality of oocytes retrieved are significantly less with lower fertilization rates and higher cancelation rates in these morbidly obese women. There is a progressive loss of follicles between the ages of 30–40 years despite normal menstrual cycles with a subsequent challenge of identifying women with a compromised ovarian reserve. Considering the modern trend of postponement of pregnancy by women along with a rising trend in obesity, there is a need to evaluate the ovarian reserve in order to counsel couples seeking assisted reproductive technique (ART) and decide the stimulation protocol. Our study is unique, as we attempted to find a correlation between obesity and ovarian reserve in infertile women of Asian origin.
In our study, compared to normal weight, overweight and obese women had a statistically significantly low inhibin B (p < 0.0259) (Fig. 1). Inhibin B is a glycoprotein secreted by granulosa and theca cells and is selectively responsible for pituitary inhibition of FSH secretion. In women with normal ovulatory cycles, the serum concentration of inhibin B is inversely proportional to the FSH secretion and increases steadily up to the midpoint of the follicular phase and then gradually decreases to low concentrations that persist in the luteal phase. Due to this pattern, it is assumed that inhibin B plays a role in the follicular development and thus reflects the ovarian function and serves as a marker for ovarian reserve. While investigating infertile women in the age group of 24–40 years, Tinkanen et al.  observed a statistically significant negative correlation between inhibin B and FSH and a positive correlation between inhibin B and antral follicle count. Its role as an ovarian reserve marker remains controversial. Several other studies have been unable to reproduce similar results. However, Franchimont et al.  and Fowler et al.  did propose a possible correlation between the levels of inhibin B and the quality of oocytes in assisted reproductive cycles. In a cross-sectional study by Pergola et al. , the authors concluded that in the early follicular phase, fertile overweight and obese women have lower levels of FSH, LH, estradiol and inhibin B, which may have a direct inhibitory effect of body mass on gonadotropins and estradiol independent of age, insulin and other hormones. In another prospective study on 134 consecutive patients with polycystic ovarian syndrome (PCOS) and 78 control women, the authors elucidated a negative effect of obesity on inhibin B levels which was not specific to women with PCOS .
Our results did not show a correlation between increased BMI with ovarian volume and other hormonal assays like FSH and LH. Studies have shown that during IVF treatment, other ovarian reserve markers like antral follicle count, ovarian volume and Doppler flows predict a better outcome than age and basal FSH alone . Despite a small sample size, the effect of age with increasing BMI was analyzed for each of the groups and it was not significant statistically.
Some authors state that obesity is a chronic inflammatory state with an increased expression of pro-inflammatory factors and a reduced expression of anti-inflammatory factors [24, 25]. In obese women undergoing assisted reproductive techniques, this inflammatory process is potentiated and results in adverse reproductive outcome like still birth and fetal loss [26–28]. Balen et al.  have suggested an upper limit of BMI as 35 kg/m2 before initiating treatment for assisted reproduction. Others have proposed that weight loss be incorporated into the infertility program before conception occurs .
Anti-mullerian hormone (AMH) has become popular for assessing ovarian reserve and also in determining the starting dose of gonadotropin during an in vitro fertilization (IVF) cycle. Anti-mullerian hormone is a glycoprotein produced by the granulosa cells within the pre-antral and early antral follicles. Serum levels of AMH are directly proportional to the growing cohort of small follicles sensitive to gonadotropins and thus are an ideal predictor of ovarian reserve . However, we did not measure serum AMH levels in our infertile women as it was unavailable at the time of initiation of our study.
Since there is a scarcity of strong evidence on this topic, future studies with a larger sample size are required to clarify the predictive value of obesity on ovarian reserve in infertile women. Thus, the search for the perfect biochemical or ultrasonographic ovarian reserve marker that can predict the response to assisted reproductive cycle remains unfulfilled. The gold standard to check ovarian reserve is the woman’s performance during an IVF cycle. However, considering the psychological and financial stress of the procedure, it may seem wise to consider counseling obese women on their expected performance in the first cycle of IVF through such studies.
Our study showed a negative correlation between elevated BMI, AFC and inhibin B among infertile women. However, our study was limited by the small sample size in the overweight and obese groups. Perhaps, a larger sample size of obese patients is necessary to intrinsically explore the effect of obesity and ovarian reserve. Incorporating AMH, a newer marker for ovarian reserve, may benefit these obese infertile women. Further work is required to elucidate the mechanisms underlying the effect of obesity on inhibin B as a marker of ovarian reserve in infertile women. The best marker to check ovarian reserve is perhaps the woman’s performance during an IVF cycle. However, considering the psychological and financial stress of the procedure, it may seem wise to consider counseling the obese women on their expected performance in the first cycle of IVF through such studies. Since there is a scarcity of strong evidence on this topic, future studies with a larger sample size are required to clarify the predictive value of obesity on ovarian reserve in infertile women.
The authors have not received any financial support for this study and wish to acknowledge Mr. Ashok Bhatt and Mr. Yogeshwar Prasad for their technical support.
Conflict of interest
The authors do not have any conflict of interest.