Obesity and Reproduction

The purpose of this guideline is to provide a framework for the delivery of fertility care to women with obesity. In Canada, as in many other countries, the prevalence of obesity in adults is increasing. More than half of Canadian men and women are overweight, and from 1978 to 2011 the rate of obesity increased from 14 to 26%. Furthermore, the proportion of Canadians who are obese now exceeds the proportion of Canadians who regularly smoke (11%) or drink alcohol to excess (16%).

The World Health Organization (WHO) defines obesity as abnormal or excessive fat accumulation that impairs health. Obesity is most commonly defined using body mass index (BMI). BMI is the weight in kilograms divided by the square of height in metres (kg/M²). The international cut-off values for defining underweight, normal weight, overweight and obese using BMI are listed in Table 1. It should be noted that other measures of obesity such as waist circumference, waist-to-hip ratio and the Edmonton obesity scoring system have merit,^,  but that this guideline focuses primarily on BMI as this is the measure most commonly used in the obesity and reproductive medicine literature.

Excessive caloric intake is the fundamental cause of obesity. Globally we are seeing an increase in the consumption of energy-dense foods, and a reduction in physical activity. The health consequences of this imbalance include cardiovascular disease, diabetes, musculoskeletal disorders, sleep apnea and an increased risk of certain malignancies, such as breast, endometrial and colon cancer.^,  It has been estimated that in young adults a BMI >45 reduces life expectancy by as much as 5 to 20 years.

Obesity also has a profound impact on reproductive health. Women who have obesity are at increased risk for menstrual dysfunction, anovulatory infertility and pregnancy-related complications. In this guideline we will highlight the current literature related to obesity and reproduction, and present evidence-based recommendations using GRADE (Table 2).

The impact of obesity on fecundity is complex.^,  Perhaps the best understood association is that between obesity and anovulation. Obesity induces a hormonal milieu consisting of insulin resistance, hyperinsulinemia, low sex hormone-binding globulin, elevated androgens, increased peripheral conversion of androgens to estrogens, increased free insulin-like growth factor 1 and high leptin.^,  The combined effect of these changes causes hypothalamic dysfunction, aberrant gonadotropin secretion, reduced folliculogenesis and lower luteal progesterone levels.^{,  ,} 

As a result, menstrual irregularity is more common in women with obesity, and increases in frequency with the level of obesity.^,  Several cohort studies have shown that the incidence of anovulatory infertility is more than twice as likely in women with obesity compared to normal BMI controls.^{,  ,}  Interestingly, the distribution of body fat is also important. It has been shown that women with higher waist circumference (excess abdominal fat) are more likely to suffer from anovulation than obese women with the same BMI who have less abdominal fat.^, 

However, even in ovulatory women, obesity appears to inhibit natural fecundity and prolong the time to conception.^{,  ,  ,  ,}  A study of 3029 Dutch women with ovulatory cycles, at least one patent fallopian tube and whose partner had normal semen parametres revealed that the likelihood of pregnancy within a 12 month period was reduced by 4% for each kg/M² increase in BMI over 29. In that study, women with a BMI of 35 had a 26% lower likelihood of spontaneous pregnancy, and women with a BMI of 40 had a 43% lower likelihood of spontaneous pregnancy than women with a BMI between 21 and 29. The authors commented that frequency of intercourse may be one potential cofounder. Obesity has been associated with decreased sexual desire, erectile dysfunction and decreased frequency of sexual intercourse. However, a Danish study of 1651 women that controlled for frequency of sexual intercourse, cycle regularity and waist circumference confirmed a progressive increase in the time to pregnancy among both nulliparous and parous women as the BMI increased from 25 to 30 to over 35.

Obesity may also alter the endometrium. There is evidence of altered endometrial gene expression during the implantation window of natural cycles in obese women. Similarly, there is evidence of lower implantation and clinical pregnancy rates in obese donor egg recipients.^, 

Obesity often affects both partners in a relationship.^,  For men with obesity the average temperature of the scrotum may be higher due to closer contact with surrounding tissues.^,  Moreover, the combination of insulin resistance, reduced sex hormone binding globulin and increased peripheral conversion of androgens to estrogens may also disrupt gonadotropin release.^{,  ,}  Consequently, men with obesity have decreased testosterone levels, that correlate negatively with both their fasting insulin and leptin levels.^, 

Despite these changes research regarding the relationship between male obesity and semen parametres has yielded conflicting results. Some cohort studies have reported an association between male obesity and oligospermia,^{,  ,  ,}  while others have found no such association.^{,  ,  ,  ,  ,  ,}  Likewise, meta-analyses have reported conflicting results with one reporting an increased risk of oligospermia and azoospermia, while two others reported no impact of male obesity on semen parametres.^,  Similarly, data regarding the impact of obesity on sperm DNA fragmentation has been mixed, with some studies suggesting a negative effect,^{,  ,}  while others report no differences.^{,  ,}  Thus, although it is clear that obese couples have a higher incidence of infertility, it remains unclear to what extent changes in sperm quality contribute to this association.

In contrast the literature clearly shows that obese men have a higher incidence of erectile dysfunction.^{,  ,}  Several randomized controlled trials have demonstrated that lifestyle changes associated with weight loss may improve erectile dysfunction in obese men.^,  Similarly, improvement in erectile function have been documented after bariatric surgery.^{,  ,  ,  ,} 

Obesity has been associated with a reduced response to gonadotropins.^,  In a large retrospective cohort study of 1189 gonadotropin/intrauterine insemination (IUI) cycles, BMI over 30 was associated with significantly higher gonadotropin requirements, prolonged gonadotropin stimulation, lower peak estradiol levels, and fewer large and medium size follicles. The reduced responsiveness of obese women to gonadotropins is likely due to the increased volume of distribution. A randomized, cross-over study showed that there was no difference in gonadotropin absorption between subcutaneous and intramuscular routes, but that a decline in area under the curve occurred in women with elevated BMI. The same decline in area under the curve is true for injections of HCG, but in this case two separate cross-over studies have reported better serum levels if HCG is given intramuscularly compared to subcutaneously.^, 

With regard to clinical pregnancy rates in women with obesity undergoing gonadotropin IUI study findings are mixed. Some studies report no difference^{,  ,}  in the clinical pregnancy rates in obese patients compared to non-obese controls, while several other report a paradoxical increase.^{,  ,  ,}  Possible reasons for an increased effectiveness of gonadotropin/IUI in women with obesity include correction of anovulation, and compensation for erectile dysfunction and decreased frequency of intercourse.

In IVF patients, female obesity is associated with increased gonadotropin requirements (both increases starting dose and duration of gonadotropins), higher cycle cancellation rates, decreased peak estradiol levels and decreased oocyte yield.^{,  ,  ,  ,  ,  ,  ,  ,}  However, there has been little consensus regarding the impact of female obesity on IVF success rates. Some studies have reported reductions of clinical pregnancy and live birth rates on the order of 15% to 30% in obese women undergoing IVF compared to non-obese controls.^{,  ,  ,  ,  ,}  Other studies have reported reductions in clinical pregnancy and live birth rates of more than 50%.^,  In contrast, at least nine studies have reported no discernable impact of female obesity on IVF pregnancy rates.^{,  ,  ,  ,  ,  ,  ,  ,} 

The most comprehensive study to date regarding the impact of female obesity on IVF outcomes is from the Society for Assisted Reproductive Technology (SART) registry. This analysis involved 239 127 fresh IVF cycles from 2008 to 2010 and included over 6000 cycles in women with a BMI 30 to 40 and almost 1000 cycles in women with a BMI over 40. The results of this study are summarized in Table 3. In brief, there is a slight decline in the number of oocytes retrieved and the number of high quality embryos as the BMI rises over 40. Implantation, clinical pregnancy and live birth rates all decline gradually with increasing severity of obesity. However, the absolute decline in pregnancy rates is small. Based on the data reported, implantation rates declined by approximately 0.2% to 0.25% and live birth rates declined by approximately 0.3% to 0.4% for each 1 kg/M² increase in BMI over 25. Thus the overall likelihood of a live birth per cycle start declined from 31.4% in women with a normal BMI, to 28% in women with BMI 30 to 34.9, to 24.3% in women with a BMI 40 to 44.5, and down to 21.2% in women with a BMI >50.

The exact mechanism by which obesity lowers IVF success rates is unclear. Some studies have demonstrated alterations in embryo development^,  and day-3 spent culture media metabolomics while others have not detected any changes of indicators of embryo quality between obese women and non-obese controls.^{,  ,  ,}  Alternatively, obesity may alter endometrial receptivity.^, 

Perhaps the best model to help elucidate the impact of obesity on reproduction is oocyte donation. Several studies have suggested that obesity does not impact donor egg recipient implantation or live birth rates^{,  ,}  ; while other studies have found a negative association.^{,  ,} 

Again, the most comprehensive data comes from the SART registry. In an analysis of 22 317 fresh donor/recipient cycles performed between 2008 and 2010 the implantation rate (49%) and live birth rate (51%) in normal BMI recipients declined to 41% in women with a BMI over 40. Moreover, the pregnancy loss rate before 24 weeks gestation increased from 8.6% in normal BMI recipients to 13.5% in recipients with a BMI over 40.This finding is consistent with that of a meta-analysis of miscarriage risk after oocyte donation in obese recipients. Since the embryos came primarily from young, healthy donors there is no reason to suspect that the differences were related to an embryo factor. Indeed, an observational cohort study of 372 women with recurrent pregnancy loss showed that euploid miscarriages were significantly more common in women with obesity (58%) compared to non-obese controls (37%).

This increase in pregnancy loss rate (8.6% with normal BMI to 13.5% with BMI over 40) in oocyte recipients is comparable to the change in pregnancy loss reported by SART in women using their own eggs: 11.3% with normal BMI to 14.8% with BMI 40 to 45, 17.6% with BMI 45–50 and 20.3% with BMI over 50 suggesting that changes in embryo quality are probably not the primary driver for the BMI related increase in pregnancy loss rates after IVF. Obesity clearly increases miscarriage risk. However, the absolute risk of pregnancy loss in women with obesity undergoing IVF is still lower than the reported risk of spontaneous pregnancy loss in women with 2 or more prior pregnancy losses (25%) or women age over 40 (≥35%).

Aside from IVF outcomes, it is important to also recognize that obesity complicates the delivery of assisted reproductive technologies. In obese women undergoing controlled ovarian hyperstimulation, the ovaries may shift to a higher position in the pelvis, making them more difficult to visualize with transvaginal scanning and increasing the risk of complications with oocyte retrieval such as bleeding, infection, and injury to surrounding tissue. In addition, the risks of providing anesthesia to obese patients is well described, and makes management of these patients through non-hospital centres a challenge. In a recent survey of obesity policies at IVF facilities in the United States, 62% of respondents cited anesthesia concerns as the primary reason for their BMI cut-off.

The risk of miscarriage is increased in women with obesity who conceive with IVF, irrespective of whether they use their own eggs or donor eggs. Miscarriage risks are also increased for women with obesity who conceive spontaneously. A recent Danish cohort study involving over 5000 women reported a hazard ratio for miscarriage of 1.23 in obese women compared to non-obese controls. Similar results were reported in a British population study involving 1644 women with obesity and 3288 controls (odds ratio 1.2 for miscarriage if obese). Among women with recurrent miscarriages the association with obesity may be even greater. Two studies have suggested that among women with recurrent miscarriages the odds of the next pregnancy resulting in miscarriage is elevated 3- to 4-fold in obese patients.^, 

Women with obesity who conceive are also at greater risk for gestational diabetes. The baseline risk of gestational diabetes in over one million pregnancies in Ontario from 1996 to 2010 was documented as 4.1%. Similarly, a study of over 400 000 pregnancies in Alberta found that the incidence of gestational diabetes increased from 3.1% in 2000 to 4.6% in 2009. In British Columbia, a study of 226 000 singleton pregnancies from 2004 to 2012 reported the incidence of gestational diabetes as 7.9%. In that study the risk of gestational diabetes increased steadily with increasing BMI (Table 4). The risk of gestational diabetes was doubled with a BMI over 30 and more than tripled with a BMI over 40.

The risk of preeclampsia is doubled in women that are overweight, and tripled in those that are obese.^,  The risk further increases with the severity of obesity (Table 4). It has been suggested that increasing physical activity in pregnancy may reduce the incidence of both gestational diabetes and pre-eclampsia. However, a recent multicentre randomized control trial in over 1500 women with obesity found no difference in the incidence of gestational diabetes or macrosomia in women randomized to behavioural intervention in pregnancy, versus routine care.

A recent Canadian meta-analysis revealed that there is a 31% increased risk of delivery prior to 37 weeks in obese women.^,  This may be related to the higher incidence of preterm premature rupture of membranes (PPROM) and need for medical induction of labour. PPROM might be more common in obese patients with increased circulating adipokines and inflammatory proteins. Increased rates of sleep apnea and antenatal depression have also been reported.^, 

A systematic review of the risks associated with obesity in pregnancy demonstrated that pregnant women with obesity have a longer duration of labour, particularly the first stage (4.7 hours compared with 4.1 hours in non-obese women).^{,  ,}  Women who with obesity have less successful trials of labour and vaginal birth after Caesarean section (VBAC): 54% to 68% success if BMI >29, 13% if weight >136 kg/300 lbs. There is increased fetal distress, instrumental delivery, and shoulder dystocia in women with obesity going through labour (Table 4).^,  The Caesarean section rate increases significantly with increasing BMI (Table 4).^{,  ,}  There are also increases in wound infection and dehiscence, post-partum hemorrhage and DVT.^{,  ,} 

Women with obesity are also at an increased risk of fetal macrosomia (Table 4), an association compounded by the presence of gestational diabetes. A recent Swedish population-based cohort study involving 1 249 908 singleton births between 1998 and 2012 documented an incidence of macrosomia of 7.7% for non-obese women without gestational diabetes, compared to 14% in gestational diabetics. The odds ratio for macrosomia in women with obesity was 2.55 (95% CI 2.51–2.59) without gestational diabetes compared to 6.50 (95% CI 6.10–6.92) with. Macrosomia increases the need for obstetrical interventions like induction of labour and Caesarean section, and increases the incidence of a number of complications such as shoulder dystocia, nerve palsy, and NICU admission. Maternal obesity also increases the risk of obesity in the child, with a higher risk of diabetes and cardiovascular disease later in life.^{,  ,  ,  ,} 

The National Birth Defect Prevention Study found that the mothers of offspring with anomalies were more likely to have obesity, with an odds ratio of 1.3 to 2.1, suggesting that maternal adiposity alters development in the embryonic period. In the 10 249 cases compared to the 4065 controls, there was a 39% increase in neural tube defects, with 50% increase in spina bifida, and 20% increase in oral clefts. Hydrocephaly (OR 1.68), anorectal atresia (OR 1.48), limb reduction anomalies (OR 1.34) and cardiovascular anomalies were also increased. The metabolic basis for those anomalies is not known, but poor glycemic control may play a role. Some have advocated for increased folic acid supplementation but evidence to support this practice is unclear. Obesity can also prevent visualization of congenital defects on ultrasound possibly leading to their discovery later in pregnancy or lack of prenatal detection. A large retrospective study demonstrated 20% less detection of anomalous fetuses in women with high BMI compared to normal BMI.

The 2006 Canadian clinical practice guidelines on the management and prevention of obesity in adults and children contain recommendations for assessment of obesity and its complications. Obesity induces insulin resistance and a baseline inflammatory state and increases the risk of diabetes, hypertension, dyslipidemia and cardiovascular disease.^{,  ,  ,  ,}  The prevalence of diabetes in the United States general population increases with BMI (8% risk if normal BMI, 43% risk if BMI ≥40 kg/m²) and obesity is the most important modifiable risk factor for the prevention of type 2 diabetes.^,  An analysis of the United States National Health and Nutrition Examination Survey (NHANES) found the prevalence of diabetes in 2011–2012 among adults aged 22 to 44 years was 5.0% (95% CI 3.8–6.7), and when stratified by BMI, the increase in prevalence of diabetes from 1988 to 2012 was seen only among those with BMI ≥30, regardless of age. NHANES also found that regardless of age, ethnicity, education or smoking habits, lipid profile worsened and the risk of hypertension increased with increasing BMI. In this study, woman age 20–39 with a BMI ≥30 were 7 times more likely to have high blood pressure, 3.4 times more likely to have high cholesterol, and 13.2 times more likely to have low HDL cholesterol than women with a BMI <25. Not surprisingly obesity, and particularly central obesity, is associated with increased risk of cardiovascular disease.^{,  ,} 

Abundant literature demonstrates that elevated BMI is associated with an increased risk for complex endometrial hyperplasia and endometrial cancer in premenopausal women.^{,  ,  ,  ,}  In a recent review the pooled odds ratio for endometrial cancer was 5.3 in women with a BMI >30 and 19.8 in women with a BMI >40. Similarly, a prospective cohort study followed 495 477 women in the United States for 16 years showed a significant increase in risk of mortality from cancers according to BMI. Compared to non-obese controls, women with a BMI of 40 or more had a relative risk of 2.1 for dying from breast cancer and a relative risk of 6.3 of dying from endometrial cancer.

Women with obesity are also at risk for obstructive sleep apnea: periodic, partial or complete obstruction of the upper airway during sleep which leads to repetitive arousal from sleep. The airway obstruction may also cause episodic sleep-associated oxygen desaturation, episodic hypercapnia, significant negative intrathoracic pressure and cardiovascular dysfunction. Approximately 35% of women with a BMI ≥35 report symptoms compatible with obstructive sleep apnea.^,  Sleep apnea exacerbates the cardio-metabolic risk attributed to obesity and metabolic syndrome. Self-reported sleep apnea appears to be an independent prognostic marker of all-cause mortality.

Canadian guidelines on managing obesity have focused primarily on eating habits and exercise. The 2015 Canadian Task Force on Preventive Health Care recommended structural behavioural interventions focused on diet and exercise as first-line treatment for patients with a BMI 30 to 39.9. Pharmacotherapy and surgery are options for more severe cases of obesity (BMI ≥40). It should be noted that body weight is tightly regulated by a complex homeostatic system and that powerful neuroendocrine mechanisms defend the body against weight loss, thus accounting for the over 95% recidivism (weight regain) generally associated with behavioural obesity management.^,  The same is seen when pharmacologic treatments are discontinued or bariatric surgery is reversed.

Obesity has been associated with poor body image, low self-esteem, depression and anxiety, all of which may influence binge eating.^,  One of the challenges of lifestyle modification is breaking this pattern and introducing behavioural strategies that facilitate weight loss and prevent weight regain.

Lifestyle modification with diet and exercise yields modest weight loss. Dietary interventions typically focus on reduced energy uptake from fat and carbohydrates, and increased uptake of fruits, vegetables, whole grains and nuts. The Diabetes Prevention Program (DPP, 3234 participants)^,  and Diabetes Prevention Study (DPS, 522 participants) showed that weight loss of approximately 4 to6 kg could be achieved following an intensive program of lifestyle changes (counselling, education, support, changes in exercise and diet) and that this was sufficient to reduce the incidence of type 2 diabetes, metabolic syndrome and dyslipidemia in individuals at risk. Unfortunately, weight lost in the first year of lifestyle intervention is gradually gained back, and in the DPP study, total weight loss in the intervention group was 2 kg after 10 years, which was only 1 kg more than in the control group. Indeed, the meta-analysis on which the current Canadian weight loss guidelines are based noted an average weight loss of only 3.1 kg (95% CI 2.4–3.9) with behavioural interventions.

Current Canadian guidelines recommend adding pharmacological treatment only when lifestyle modification is not sufficient to achieve or maintain clinically significant weight loss. The rationale for this approach is that behavioural modification has better benefit-to-harm ratio than pharmacotherapy. In Canada, two drugs are currently approved by Heath Canada for the treatment of obesity- Orlistat and Liragutide. Orlistat (Xenical®) is a selective absorption fat inhibitor.^{,  ,}  It exerts an effect in the digestive tract by binding to gastric and pancreatic lipases and prevents absorption of at least 30% of dietary fat. However, side effects include flatulence, greasy stool, fecal urgency and abdominal cramps. Liragutide (Saxenda®) is an analogue of human glucagon-like-peptide-1.^,  Liragutide leads to weight loss by decreasing appetite. The most common side effects are nausea, diarrhea and hypoglycaemia by stimulating insulin secretion and inhibiting glucagon. Liragutide 1.2 to 3 mg subcutaneously per day may lead to greater weight loss (4.8–7.2 kg after 20 weeks) than Orlistat 120 mg by mouth 3 times per day (4.1 kg after 20 weeks). The safety of both of these drugs in pregnancy is unknown. Although metformin is not specifically approved by Health Canada for weight loss, its properties in this regard have been well studied.^{,  ,  ,  ,  ,  ,  ,  ,}  Current data suggests that metformin induces weight loss of an average of 1.9 kg, while Orlistat on average 3.1 kg. Unfortunately, even in controlled trials with fixed timelines, drop-out rates with weight loss medications typically exceed 30%.

Bariatric surgery is considered when the BMI is over 40, or if BMI exceeds 35 with comorbidities, and if other weight loss attempts have failed. Bariatric surgery modifies the architecture of the digestive tract in order to limit caloric intake. There are restrictive procedures such as vertical banded gastroplasty, laparoscopic adjustable gastric banding or sleeve gastrectomy designed to accelerate the feeling of “being full,” and there are malabsorptive procedures such as biliopancreatic diversion with or without duodenal switch that are designed to reduced nutrient absorption. Bariatric surgery may also be a mixed restrictive/malabsorptive procedure as Roux en Y gastric bypass. In malabsorptive or mixed procedures, the greater the excluded intestinal portion, the greater the decrease in caloric intake.^, 

Bariatric surgery typically induces significant weight loss. The Swedish Obese Subjects (SOS) study included over 2000 bariatric surgery patients and showed that maximal weight loss was achieved after 1 year (gastric bypass 45 kg, vertical banded gastroplasty 30.8 kg and gastric banding 24.9 kg) but that significant weight reduction was maintained even after 10 years of follow-up (gastric bypass 29.6 kg, vertical banded gastroplasty 19.5 kg and gastric banding 15.6 kg). Other studies have confirmed profound weight loss 1 to 2 years after gastric bypass surgery (29–45 kg), biliopancreatic diversion (48 kg) and sleeve-gastrectomy (25.1 kg).^, 

Furthermore, bariatric surgery improves metabolic profile and significantly reduces all-cause mortality.^,  In fact, bariatric surgery has been shown to be more effective than usual care in the prevention of type 2 diabetes, and more effective than medical therapy in the treatment of type 2 diabetes in obese patients. Finally, among obese patients with uncontrolled type 2 diabetes, three years of intensive medical therapy plus bariatric surgery results in significantly better glycemic control and weight reduction than medical therapy alone.

Mortality associated with bariatric surgery is less than 1%. The most common postoperative complications in the SOS study were pulmonary complications (6.1%). Other complications (<5%) are bleeding, venous thrombosis, wound infections, anastomotic leaks. Long term complications include vitamin deficiencies in 20% to 50% of cases. Surgical complications (internal bowel herniation) related to the previous bariatric procedure may also impact 1% of pregnancies. Depending on the procedure, complications may lead to health problems: anemia (iron, B12 and folate deficiencies), osteopenia (loss in calcium and vitamin D), alopecia (zinc deficiency), muscle weakness (magnesium deficiency), visual disorders, neuropathy, and hemorrhagic disease (Vitamins A, C, E and K deficiencies).^,  For these reasons bariatric surgery requires long-term monitoring.