UvA-DARE (Digital Academic Repository) Fertility treatment in obese women Koning, A.M.H. Link to publication

Transcription

1 UvA-DARE (Digital Academic Repository) Fertility treatment in obese women Koning, A.M.H. Link to publication Citation for published version (APA): Koning, A. M. H. (2015). Fertility treatment in obese women General rights It is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s), other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons). Disclaimer/Complaints regulations If you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, stating your reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Ask the Library: or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam, The Netherlands. You will be contacted as soon as possible. UvA-DARE is a service provided by the library of the University of Amsterdam ( Download date: 12 May 2018

2 Fertility treatment in obese women Aafke Koning

3

4 Fertility treatment in obese women Aafke Marije Henriëtte Koning

5 ISBN: Cover design: Anne-Lot Hoek Layout: Ferdinand van Nispen, Citroenvlinder-dtp.nl, Bilthoven Print: GVO Drukkers & Vormgevers B.V. Ponsen & Looijen, Ede The author gratefully acknowledges financial support for printing this thesis by Division Gynecology and Obstetrics, Academic Medical Center, ABN AMRO, BMA Mosos, ChipSoft, Goodlife, Memidis Pharma, Origio Benelux BV, Smith&Nephew and ERBE

6 Fertility treatment in obese women ACADEMISCH PROEFSCHRIFT ter verkrijging van de graad van doctor aan de Universiteit van Amsterdam op gezag van de Rector Magnificus prof. dr. D.C. van den Boom ten overstaan van een door het College voor Promoties ingestelde commissie, in het openbaar te verdedigen in de Agnietenkapel op dinsdag 23 juni 2015, te 14:00 uur door Aafke Marije Henriëtte Koning geboren te Amsterdam

7 Promotiecommissie Promotor: Prof. dr. B.W.J. Mol Overige leden: Prof. dr. F. van der Veen Prof. dr. D.L. Willems Dr. M.J.M. Serlie Prof. dr. A. Franx Dr. A. Hoek Prof. dr. J.S.E. Laven Faculteit der Geneeskunde

8 Voor Cato, Boele en Faas

9

10 Table of contents Chapter 1 General introduction 9 Chapter 2 Chapter 3 Chapter 4 Chapter 5 Chapter 6 Chapter 7 Economic consequences of overweight and obesity in infertility: a framework for evaluating the costs and outcomes of fertility care Hum Reprod Update, 2010 May-Jun;16(3): Complications and outcome of assisted reproductive technology in overweight and obese women Hum Reprod, 2012 Feb;27(2): Effectiveness of a weight reduction program in obese, subfertile women Submitted Pregnancy complications after weight loss in obese women Submitted The long-term follow up of children born after a weight reduction program in obese, subfertile women Research letter Obesity: argument for withholding fertility treatment? Ned Tijdschr Geneeskd 2014;158:A Chapter 8 Summary and general discussion 101 References 110 Appendices Nederlandse samenvatting 123 List of abbreviations 128 List of co-authors 130 Portfolio 132 Dankwoord 134 Curriculum Vitae 138

11

12 Chapter 1 General introduction

13 Chapter 1 Impact of obesity The worldwide epidemic of obesity is overwhelming. The World Health Organisation (WHO) defined overweight as a body mass index (BMI) 25 kg/m² and obesity as BMI 30 kg/m². In women of childbearing age in the majority of developed countries the mean BMI is above 25 kg/m². In the Netherlands, in 2012 approximately 14% of women of childbearing age was obese. [1] Figure I depicts the dramatic increase in obesity in the Dutch population from 1981 until Furthermore, it shows that more women than men are obese. In addition to these figures the problem is also affecting the next generations leading to large numbers - such as 42 million infants and young children worldwide - being overweight or obese in 2013.[2] With the increasing size of the problem, morbidity and mortality related to obesity put their claim on health care. Two thirds of deaths worldwide are related to non-communicable diseases. Risk factors for these chronic diseases are, among others, poor diet and low physical activity both leading to obesity. [3] As health care provider it is important to know the extent of the problem and for future policies to have insight in the cost burden of this disease on society. Several attempts have been made to calculate direct and indirect health care costs of overweight and obesity. For example, Allender and Rayner proposed a model for calculating direct economic and health costs of overweight and obesity estimating GBP 3.2 billion health cost burden to the National Healthcare Services (NHS) in the UK. This equals 4.6% of total NHS costs in [4] Other authors have estimated total costs between 2 and 8% of national health care systems. [5] The WHO therefore proclaimed it of great importance to act on this growing problem. The Ministers and delegates attending the WHO European Ministerial Conference on Counteracting Obesity (November 2006) adopted the European Charter on Counteracting Obesity, in which they set several goals. One of these is making visible progress in decreasing obesity, especially related to children and adolescents, and aim to achieve this in most countries in the 4 5 years after 2006 and to reverse the trend by 2015 at the latest. Each country developed its own policy. In the Netherlands, several programs for reducing the overweight and obesity rate have been initiated since The different programs are preventive (universal or selective) and all with the emphasis on healthy food and exercise as stated in the guideline of the Health Council. Initiatives are very 10

14 General introduction percentage / / / / / / / /2012 total women men 1 Figure I Rijksinstituut voor Volksgezondheid en Milieu (RIVM) from beinvloedende-factoren/lichamelijke-eigenschappen/ontwikkeling-overgewicht-per-provincie/ fragmented and the effectiveness is most often not evaluated. Some results are mentioned in the yearly report of the Pact for Healthy Weight, a joint effort of 26 different parties from government, business and social organisations, all from local municipality initiatives. For example Zwolle, a town of 120,000 inhabitants, reported a decrease of overweight in children for the first time since joining the program Youth on healthy weight from 12% (2009) to 10% (2012). Other European countries like Finland and France have put taxes on unhealthy food in response to the European Charter. This policy has led to a decrease in the purchasing of this food.[6] Obesity and subfertility Besides the impact on general health, overweight and obesity have a negative impact on fertility in women. Subfertility, defined as not conceiving after 12 months of unprotected intercourse, has several different mechanisms. About one in every six couples meets this definition. If a couple seeks professional help for their wish for a child a diagnostic workup will identify roughly four known causes of subfertility. A tubal factor, due to for example infection or abdominal surgery, is present in about 15% of cases. A male factor is present in 40% of the cases, and can be identified with a wide variety in sperm count ranging from azoospermia to mild oligospermia. Ovulation disorders with oligoor amenorroea are present in 20% of the cases. Finally, disturbed interaction between cervical mucus and sperm, as identified by an abnormal postcoital test is present in 5% of the women, although the latter test and subsequent 11

15 Chapter 1 diagnosis are controversial and have been abandoned widely. Tubal factor, male factor, ovulation disorders and cervical abnormality all require targeted intervention with treatments such as In Vitro Fertilization (IVF) (tubal factor), Intrauterine Insemination (IUI) and Intracystoplasmic Sperm Injection (ICSI) (male factor), ovulation induction (anovulation) or IUI (cervical factor). Apart from these diagnoses, there is a substantial group of couples, 20 to 30%, where no clear cause of subfertility can be identified. Diminished ovarian reserve due to female age certainly plays a role in a considerable number of women. The prognostic index of these couples - calculated with the prognostic index developed by Hunault can be used to guide treatment. [7] These couples can be offered treatment if their prognosis of natural conception is lower than 30%, while couples with a better prognosis can be advised to delay treatment. Lifestyle is discussed if a subfertile couple consults a general practitioner or gynaecologist. Next to caffeine usage, smoking and alcohol, unhealthy weight is discussed. Obesity is known to reduce fertility due to different mechanisms. First, it can give ovulation disorders, with a substantial amount of women with polycystic ovarian syndrome (PCOS) being obese. [8] Moreover, obese women with a regular cycle also have a lower chance of natural conception than women with normal weight. [9] The mechanism behind this is not yet discovered. One hypothesis is that an adipokine, leptin, plays a role because of its obvious link to obesity and also its physiologic role in fertility. Leptin is a molecule secreted proportionally to the amount of adipose tissue in an organism. It is thought to inhibit food intake and lead to a feeling of satiety. The amount of adipose tissue in an organism is an energy storage reservoir for times of food shortage. Essential for reproduction is an energy equilibrium to ensure that energy needed for a pregnancy is available. [10] Obese people have elevated leptin levels which normally would lead to weight loss. Obese people fail to have this normal response, and therefore it is thought they are resistant to leptin. [11] Several findings have fuelled the idea that leptin also plays a role in reproduction. Knockout mice for the leptin gene are obese and infertile with atrophic gonads. [12] In literature two morbidly obese Indian cousins are described with a leptin deficiency. [13] One of the cousins treated with leptin lost weight and puberty set in. Also in normal female rodents when leptin is injected it accelerates the onset of puberty. [14] Another hypothesis is that abnormalities in concentrations of sex hormones, due to a decrease in SHBG and hyperinsulinemia, have an effect on 12

16 General introduction folliculogenesis and implantation. [15] Furthermore, several organs and tissues in the reproductive system, like the hypothalamus, ovary and endometrium, seem negatively influenced by obesity. [16] 1 Obesity and pregnancy If an obese woman becomes pregnant several other potential consequences of her body composition are at risk. Table I shows the pregnancy complications associated with overweight and obesity with their absolute risks and/or relative risks. Obese women have an increased miscarriage rate compared to women with normal weight. [17] Two systematic reviews report higher miscarriage rates in obese women, one after spontaneous conception and the other after spontaneous and assisted conception. [18;19] Both studies conclude however that the reliability of their results is limited by the heterogeneity of the included studies. Obesity also increases the risks of pregnancy complications such as preeclampsia and gestational diabetes (GDM ). [20-22] Because of the association between obesity and insulin resistance, it predisposes for type II diabetes and GDM. The exact mechanism by which obesity causes preeclampsia is not fully understood but both obesity and diabetes are associated with a higher risk of cardiovascular diseases, and hence preeclampsia. [23] Both preeclampsia and GDM also have negative consequences for the foetus with impact on growth and time of delivery. There is debate in literature about the relationship between obesity and preterm delivery. McDonald et al. in their meta-analysis found an association between overweight and obesity and induced preterm delivery and after correcting for publication bias also a significant overall higher risk of preterm delivery. [24] Preterm delivery is worldwide the leading cause of neonatal mortality and accounts for substantial morbidi ty. [25] Furthermore, obesity is associated with increased risks of infant mortality due to increased mortality risk in term bir ths. [26] Also, more caesarean sections and higher perioperative morbidity is seen in obese w omen. [27;28] And last, more stillbirths are seen in obese women. [29] Again the mechanism behind this association is not completely understood but other related conditions such as congenital anomalies, macrosomia and hypertensive disorders are associated with obesity. 13

17 Chapter 1 Table I Risk of maternal and fetal complications in overweight and obese women complication n study population study design outcome measure results author miscarriage 16,696 spontaneous, OI, oocyte donation, IVF/ICSI meta-analysis pregnancy loss < 20 wks 15.5% vs 12.5% (OR 1.67 (95% CI 1.25 to 2.25) for BMI 25) Metwally et al. 24,738 spontaneous conceptions systematic review different definitions miscarriage preeclampsia 12,330 primi, singleton retrospective hypertension with proteinuria gestational diabetes 1,699,073 different populations cohort (prospective and retrospective) 844,295 different populations cohort (prospective and retrospective) meta-analysis different definitions pre eclampsia, median incidence 3.9% meta-analysis different definitions GDM, prevalence % 11.8% vs 10.7% (OR 1.11 (95% CI 1.0 to 1.24) for BMI % vs 10.7% (OR 1.31 (95% CI 1.18 to 1.46) for BMI % vs 4% (OR 3.74 (95% CI 3.03 to 4.61) for BMI 30 vs BMI 18,5-30 LRs (95% CI) 1.7 (0.3 to 11.9) for BMI 25 and 0.73 (0.22 to 2.45) for BMI < 25 (OR 2.3) LRs 2.7 (1.0 to 7.3) for BMI 35 and 0.86 (0.68 to 1.07) for BMI < 35 (OR 3.7) OR 2.1 (95%CI 1.8 to 4.2) for overweight women OR 3.6 (95% CI 3.1 to 4.2) for obese women OR 8.6 (95% CI 5.1 to 16) for severely obese women Boots and Stephenson Hancke et al. Cnossen et al. Chu et al. preterm delivery caesarean delivery 1,095,834 different populations, cohort and case control 1,391,654 cohort (prospective and retrospective) meta-analysis birth < 37 weeks 9.9% vs 9.7% (RR 1.1 (95% CI 0.87 to 1.3) high BMI vs low BMI, after publication bias adjustment RR 1.2 (95%CI 1.1 to 1.4) 2.5% vs 1.9% (RR 1.2 (95% CI 1.2 to 1.4) for induced preterm delivery meta-analysis elective and emergency caesarean delivery, prevalence % OR 1.5 (95% CI 1.3 to 1.6) for overweight women OR 2.1 (95% CI 1.9 to 2.3) for obese women OR 2.9 (95% CI 2.3 to 3.8) for severely obese women McDonald et al. Chu et al. stillbirth 3,288,688 different populations, cohort and case control meta-analysis death of a fetus > 20wks 0.48% vs 0.4% OR 1.2 (95% CI 1.1 to 1.3) for BMI % vs 0.4% OR 1.5 (95% CI 1.4 to 1.6) for BMI 30 Aune et al. BMI = body mass index, OR = odds ratio, LR = likelihood ratio, OI = ovulation induction, IVF/ICSI = in vitro fertilisation/intracystoplasmic sperm injection, GDM = gestational diabetes 14

18 General introduction Obesity and fertility treatment Just as women with normal weight, obese women who have trouble getting pregnant often seek help. Obesity not only affects pregnancy chances, but it also can interfere with various aspects of fertility treatment. Obese women need a longer duration of ovarian stimulation and higher doses of gonadotrophins than normal weight women. [30] Also, oocyte retrieval with ultrasonography is technically more difficult in these women. [31] The consequences of obesity on assisted reproductive technology (ART) and pregnancy have led to different BMI limits for access to ART around the world, ranging from BMI kg/m². [32] In the Netherlands, different fertility treatment centres use a BMI limit to access ART. Several authors have argued against this policy for reasons of inequity and poor success of management of obesity. [33;34] The ESHRE (European Society of Human Reproduction and Endocrinology) Task Force has stipulated that there should be equal access to fertility treatment for all women. [35] In that statement, no specific comment was made on BMI. Irrespective of the position in this debate on the use of a BMI limit, all care providers agree that a first aim in any management strategy is to improve the lifestyle and aim for weight reduction. 1 Lifestyle intervention Lifestyle intervention aiming for weight loss in obese subfertile women prior to ART is advised in the National Institute for Health and Care Excellence (NICE) guideline. [36] Several observational studies have shown potential benefit of these interventions in subfertile women with overweight and obesity. [37-42] Different regimes with several combinations between diets, exercise, repetitive consultation and medication for ovulation induction or insulin sensitizers are used. In aforementioned articles different outcome measures were studied like weight loss, BMI difference, pregnancy rates or cycle restoration in anovulatory patients. All interventions seem to show some benefit. There are to our knowledge only two randomized controlled trials for lifestyle intervention in subfertile obese women, both with a small sample size (49 and 38 women). [43;44] These studies also both report increased fertility following the intervention. Currently a Cochrane meta-analysis is being conducted trying 15

19 Chapter 1 to answer the question what the effectiveness of lifestyle intervention in obese subfertile women is and will hopefully give guidance for clinical practice. Also, lifestyle intervention during pregnancy is thought to be able to change the amount of pregnancy complications in obese women. A meta-analysis concluded that dietary and lifestyle interventions in pregnancy can reduce maternal gestational weight gain and improve outcomes for both mother and baby. [45] Among these interventions, those based on diet were the most effective. The overall evidence rating of the studies was graded as low to very low however, because of heterogeneity observed in the effect size, deficiencies in the quality of the individual studies, and risk of publication and related biases. A recent large randomized study evaluating antenatal interventions in women who were overweight or obese, showed antenatal lifestyle advice not to reduce the risk delivering a baby weighing above the 90th centile nor improve maternal pregnancy and birth outcomes. [46] In summary, obesity is an important health problem, both in incidence as well as in its consequences for the individual. Among its many aspects, it has profound impact on reproduction, both on the capacity to conceive as well as on the course of pregnancy and neonatal outcome. Aims of thesis This thesis studies the impact of overweight and obesity on subfertility and fertility treatment. We will address the cost-effectiveness and safety of fertility treatment in overweight and obese women, we will address the effectiveness of lifestyle interventions in obese women who suffer subfertility and we will address ethical issues. We specifically address the following questions: 1) What is the impact of overweight and obesity on costs and effects of treatment of a population subfertile couples? 2) What is the safety of ART in overweight and obese women? 3) What is the effectiveness of a lifestyle intervention in obese subfertile women in terms of fertility, pregnancy outcome and child development? 4) Are there ethical arguments to withhold ART in obese subfertile women? 16

20 General introduction Chapter 2 answers the first question. We addressed the impact of overweight and obesity on costs and effects of fertility treatment of a population subfertile couples. Using epidemiological data and data on the effectiveness of treatments, we constructed a model to compare women with normal weight, overweight and obesity in two hypothetical cohorts receiving treatment for subfertility divided in an- and ovulatory women. We reported on costs and effectiveness of treatment. 1 Chapter 3 answers the second question. We studied the safety of ART in overweight and obese subfertile women. We performed a systematic review of studies that reported on complications of ART use in overweight and obese women and its effectiveness. Chapter 4, 5 and 6 answer the third question. We addressed the effectiveness of lifestyle intervention for obese women suffering from subfertility. In chapter 4 we assess in a retrospective cohort study the effectiveness of a weight reduction program (WRP) in terms of weight reduction and fertility. In chapter 5, we study the same cohort, as we try to answer whether a WRP has impact on the course and outcome of the pregnancies these women established. In chapter 6 we try to address the impact of a WRP on the long-term follow up of the children born after such a program. In chapter 7 we evaluate the different arguments used by advocates of a BMI limit for access to ART. We specifically address the justification for such policy. Finally in chapter 8 we summarize our findings and give recommendations for the treatment of obese subfertile patients. 17

21

22 Chapter 2 Economic consequences of overweight and obesity in infertility: A framework for evaluating the costs and outcomes of fertility care Koning AMH, Kuchenbecker WKH, Groen H, Hoek A, Land JA, Khan KS, Mol BWJ Hum Reprod Update, 2010 May-Jun;16(3):

23 Chapter 2 ABSTRACT Background Overweight and obesity are an epidemic in Western society, and have a strong impact on fertility. We studied the consequences of overweight and obesity with respect to fecundity, costs of fertility treatment and pregnancy outcome in subfertile women. Methods We searched the literature for systematic reviews and large studies reporting on the effect of weight on both fecundity and pregnancy outcome in subfertile women. We collected data on costs of treatment with ovulation induction, intrauterine insemination and in vitro fertilisation, as well as costs of pregnancy complications. We calculated, for ovulatory and anovulatory women separately, the number of expected pregnancies, complications and costs in a hypothetical cohort of 1,000 normal weight, overweight and obese women each. Results In our hypothetical cohort of 1,000 women compared to women with normal weight, live birth was decreased by 14 and 15% (from 806 live births to 692 and 687 live births) in overweight and obese anovulatory women respectively, for ovulatory women it was decreased by 22 and 24% (from 698 live births to 546 and 531 live births) respectively. These outcomes were associated with an increase in the number of complications and associated costs leading to cost per live birth in anovulatory overweight and obese women being 54 and 100% higher than their normal weight counterparts. For ovulatory women they were 44 and 70% higher respectively. Interpretation Overweight and obese subfertile women have a reduced probability of successful fertility treatment and their pregnancies are associated with more complications and higher costs. 20

24 Economic consequences Introduction The prevalence of overweight and obesity varies in populations and is estimated to range from 5% in some developing countries to 30% in developed countries. [47] The World Health Organization defines overweight as a body mass index (BMI) 25 kg/m², and obesity as a BMI 30 kg/m². [48] Considering the trends in childhood obesity, a significant increase in obesity related subfertility can be anticipated in the future. [49] Nowadays, the rate of obesity in women of child bearing age is 12% in Western Europe and 25% in North America. [50-53] 2 The strongest obesity related effect on fertility is anovulation. Polycystic ovarian syndrome (PCOS), the most noted cause of anovulation, is furthermore exacerbated by increased insulin resistance and hyperinsulinemia associated with overweight and obesity. [54] In 65% of patients with PCOS, obesity therefore contributes to anovulation. [55] On the other hand, even obese women with an ovulatory cycle have a lower chance of spontaneous conception. [9;56] In cases of chronic anovulation, ovulation induction (OI) with clomiphene citrate in overweight and obese women results in lower ovulation rates [57] and lower cumulative live birth rates for women with a BMI > 30 kg/m². [58] McClure et al. showed that in overweight women ovulation rates are lower due to higher cancellation rates, but if OI is successful no difference is found in pregnancy rates in different weight categories. [59] Mulders et al. also found obesity to be associated with higher cancellation rates and substantially higher miscarriage rates leading to a lower live birth rate per started cycle. [60] This decreased success rate is however not found in all studies. [61] The literature on the impact of body weight on the effectiveness of intrauterine insemination (IUI) is just as for OI, inconsistent. Koloszar et al. showed a negative impact of increasing body weight on the success rates of IUI, but Wang et al. could not confirm this finding. [62;63] Furthermore, several retrospective studies have shown a negative impact of overweight and obesity in women on the outcome of in vitro fertilisation (IVF). 21

25 Chapter 2 [64-66]The ongoing pregnancy rate and live birth rate is however consistently decreased especially due to an increased miscarriage rate in women with obesity.[30;67;68] Apart from these obesity related fertility problems, there is indisputable evidence that pregnancy in overweight and obese women is associated with an increased risk of complications, leading to higher maternal and neonatal morbidity and mortality and increased costs. [51;69;70] Pregnancy complications associated with obesity are hypertensive disorders, gestational diabetes, prolonged duration of labour, increased need of operative delivery, macrosomia, shoulder dystocia and increased blood loss. [27;71;72] Obesity is furthermore associated with an increased risk of adverse pregnancy outcomes such as unexplained still birth and neonatal admissions. [51;73-75] In view of the issues stated above, it is likely that overweight and obesity have a negative impact on the outcome as well as the costs of fertility treatment. The aim of this paper is to conceptualise the impact of overweight and obesity on fertility treatment and the resultant pregnancies, in terms of effectiveness, costs and cost-effectiveness. Methods We developed a framework within which the consequences of fertility treatment and outcomes of resultant pregnancy can be evaluated simultaneously for subfertile women in different body weight categories. We performed systematic reviews to obtain information on outcomes and costs to generate costeffectiveness estimates for inclusion in decision analytic models. To do so, we searched the literature for evidence on the effect of obesity on spontaneous pregnancy chances, success of assisted reproductive technology (ART), as well as pregnancy outcome. We used the following electronic databases: PubMed, Embase, DARE and the Cochrane Library to initially search for systematic reviews on each of the subjects. In absence of reviews, we identified large, reliable studies. 22

26 Economic consequences To identify studies that reported on the association between obesity and spontaneous pregnancy chances we combined the key words ( obesity, overweight, or body mass index ) and ( pregnancy or fertility ). By adding the key words ( assisted reproductive technology, intra uterine insemination ) and ( ovulation induction ) we looked for studies reporting on the effect of obesity on these treatments. To identify studies reporting on the association between obesity and pregnancy outcome, we used the key words: ( obesity, overweight or body mass index ) and ( pregnancy outcome ). 2 We included studies reporting on maternal morbidity as well as pregnancy outcome. The reported odds ratios (ORs) in the reviews were used, or if not available, calculated by using a 2 x 2 table cross classifying BMI and one of the aforementioned outcomes. These ORs were used as input for calculating the additional impact of overweight and obesity on both fecundity as well as pregnancy. The economic analysis was performed from a hospital perspective. Costs of fertility treatments were obtained from a series of Dutch studies, that reported on the costs of OI, IUI and costs of IVF. [76;77] Furthermore, we looked for studies reporting on costs of pregnancy in overweight women and costs of pregnancy complications in these women. To do so, we performed a search of several major journals in obstetrics and gynaecology for economic evaluations. We looked for studies that reported on the costs of each of the complications miscarriage, pre-eclampsia, gestational diabetes and caesarean delivery. We assumed no difference in multiple pregnancy rates between different weight categories. [78] Next, we assessed the impact of overweight and obesity on the costs and effects of fertility treatments. To achieve this, we distinguished between the case of ovulatory women and the case of anovulatory women. For each of these situations, we considered women with normal weight, overweight, and obese women. According to the WHO normal weight is defined as a BMI between 20 and 25 kg/m², overweight as a BMI between 25 and 30 kg/m² and obesity as a BMI over 30 kg/m². Because of differences in definitions of overweight and obesity in some studies we used in our review, we could not use the very strict BMI cut off points proposed by the WHO for our different weight groups. 23

27 Chapter 2 We then constructed a theoretical model, simulating the situation where women were treated for their subfertility. For each of the six categories, i.e. anovulatory women with normal weight, anovulatory overweight women and anovulatory obese women and ovulatory women with normal weight, ovulatory overweight women and ovulatory obese women, we calculated the expected pregnancy rates, the expected number of fertility treatments and the expected number of pregnancy complications for a hypothetical group of 1,000 women. We performed multiple sensitivity analyses on the following variables: success rate of IVF (range 40% to 60%), success rate of OI (range 70% to 90%) and IUI (range 30% to 50%). With these figures we calculated and then plotted in two figures different success rates of ART against the costs per live birth in anovulatory and ovulatory women in different weight categories. Results Literature identified The search for studies on the association between spontaneous pregnancy chances in overweight women revealed two reviews by Jensen and Gesink Law et al. as well as the study of Van der Steeg et al. [9;56;79] The results of these studies are shown in Table I. Both reviews as well as the study of Van der Steeg et al. showed that overweight women take longer to conceive than normal weight women. The reviews were retrospective studies in a cohort of women not seeking medical help for any subfertility, whereas Van der Steeg et al. studied women in fertility clinics. Based on these results, we assumed that among obese ovulatory women spontaneous pregnancy chances were 90% of those in normal weight or overweight women. Moreover, we assumed that spontaneous pregnancy chances prior to and in between ART cycles was 10% in all groups. From the literature no unequivocal conclusion could be drawn about the influence of obesity on IUI. Whereas Wang et al. reported an increased probability of success of IUI in women with a BMI > 30 kg/m 2, Koloszar et al. reported exactly the opposite, i.e. a decrease of success of IUI with increasing BMI. [62;63] In view of these conflicting results on IUI, for purpose of this review we considered no effect of BMI on IUI. 24

28 Economic consequences Table I Spontaneous fecundity and pregnancy chances after reproductive treatment in different weight categories author study period study population Spontaneous pregnancy Gesink Law first planned pregnancy Jensen first planned pregnancy data collection retrospective retrospective outcome measure chance conceiving per cycle chance conceiving per cycle results OR 0.92 (95% CI 0.84 to 1.01) for BMI OR 0.82 (95% CI 0.75 to 0.92) for BMI > 30 OR 0.66 (95% CI 0.49 to 0.89) for primipari with BMI > 30 OR 0.77 (95% CI 0.70 to 0.84) for BMI > 25 2 V/d Steeg subfertile ovulatory Intra Uterine Insemination Wang infertile couples undergoing IUI treatment Koloszar infertile ovulatory In Vitro Fertilisation Maheshwari infertile women undergoing IVF prospective retrospective time to pregnancy < 12 months chance conceiving per cycle HR 0.95 (95% CI 0.91 to 0.99) per extra kg/m2 from a BMI > 29 OR 1.5 (95% CI 1.1 to 1.9)* for BMI > 30 prospective pregnancy OR 0.66 (95% CI 0.49 to 0.88)* for BMI OR 0.42 (95% CI 0.25 to 0.73)* for BMI systematic review pregnancy OR 0.71 (95% CI 0.62 to 0.81) for BMI > 25 OR 0.68 (95% CI 0.55 to 0.83) for BMI > 30 IUI = intra uterine insemination, IVF = in vitro fertilisation, OR = odds ratio, HR = hazard ratio, * = calculated OR Maheshwari et al. published a systematic review of the literature from 1960 until 2006 on the outcome of IVF for overweight and obese women. [30] They reported an OR for pregnancy after IVF of 0.71 (CI 95% 0.62 to 0.81) for women with a BMI > 25 kg/m 2 compared to women with a BMI between 20 and 25 kg/m 2, and for women with a BMI > 30 kg/m 2 even 0.68 (CI 95% 0.55 to 0.83) (Table I). We applied these odds ratios in our model. Maheshwari et al. also found that overweight women require more total units of gonadotrophins during hyperstimulation for IVF, but these additional costs were not considered in the present analysis. 25

29 Chapter 2 We found one meta-analysis and three studies that reported on the impact of BMI on the effectiveness of OI in anovulatory women (Table II). [60;61;80;81] The study of Al-Azemi et al. showed a negative impact of obesity on live birth rate after OI with clomiphene citrate. The meta-analysis of Mulders et al. did not show a significant impact of BMI on the fecundity after OI with gonadotrophins. However, they found higher cancellation rates per cycle (OR 1.9) and higher miscarriage rates in the obese group (OR 3.1), thus leading to lower ongoing pregnancy rates per started cycle. Table II Odds ratio s of pregnancy chances of overweight and obese anovulatory women following ovulation induction author study period Azemi not reported infertile women undergoing CC-OI Mulders infertile women undergoing gonadotrophin-oi study population study design outcome measure retrospective cohort study live birth results OR 0.74* (95% CI 0.39 to 1.4) for BMI OR 0.15* (95% CI 0.07 to 0.30) for BMI >30 meta analysis pregnancy OR 1.22 (95% CI 0.77 to 1.93) obese vs lean women cancellation rate miscarriage rate OR 1.86 (95% CI 1.13 to 3.06) OR 3.05 (95% CI 1.45 to 6.44) Balen after 3 cycles CC failed OI --> gonadotrophins Imani infertile women undergoing CC-OI prospective cohort study prospective cohort study pregnancy OR 1.3* (95% CI 0.71 to 2.2) for BMI >25 OR 0.99* (95% CI 0.48 to 2.0) for BMI >30 ovulation OR 0.92 (95% CI 0.88 to 0.96) obese vs normal weight live birth OR 1.00 (95% CI 0.97 to 1.04) CC = clomiphene citrate, OI = ovulation induction, BMI = body mass index, * = calculated OR Balen et al. studied anovulatory women with a BMI up to 35 kg/m 2 and also did not find a significant difference in pregnancy rates after OI with gonadotrophins in overweight and obese women compared to women of normal weight. Imani et al. found among anovulatory women a hazard ratio of 0.92 for obese versus lean women for ovulation after OI with clomiphene citrate, but they also did not find a difference in live birth chances between the weight groups. We therefore assumed in our analysis that there is no influence of BMI on pregnancy rates after OI in anovulatory women. 26

30 Economic consequences Table III shows the additional risk of obstetric complications due to overweight and obesity. We applied meta-analyses conducted by Chu et al. and Cnossen et al.. [20;21;28;82] These studies report that there is an additional risk of stillbirth, caesarean delivery, preeclampsia and gestational diabetes with increasing BMI. Fedorcsak et al. studied the impact of overweight in women undergoing IVF treatment, and found an increased risk of miscarriage. [64] 2 Table III Odds ratios of maternal and fetal complications in overweight and obese women author study period study population study design outcome measure results Fedorcsak IVF/ICSI retrospective cohort Chu (82) birth registries, clinical medical records etc. Cnossen cohort (prospective and retrospective) Chu (28) cohort (prospective and retrospective) Chu (20) cohort (prospective and retrospective) meta analysis meta analysis meta analysis meta analysis abortion < 6wks OR 2.0* (95% CI 1.1 to 3.7) for BMI > 25 stillbirth OR 1.5 (95% CI 1.1 to 1.9) for overweight women OR 2.1 (95% CI 1.6 to 2.7) for obese women pre eclampsia LRs (95% CI) 1.7 (0.3 to 11.9) for BMI > or = 25 and 0.73 (0.22 to 2.45) for BMI < 25 (OR 2.3) LRs 2.7 (1.0 to 7.3) for BMI > or = 35 and 0.86 (0.68 to 1.07) for BMI < 35 (OR 3.7) cesarean delivery gestational diabetes BMI = body mass index, OR = odds ratio, LR = likelihood ratio, * = calculated OR OR 1.5 (95% CI 1.3 to 1.6) for overweight women OR 2.1 (95% CI 1.9 to 2.3) for obese women OR 2.9 (95% CI 2.3 to 3.8) for severely obese women OR 2.1 (95% CI 1.8 to 4.2) for overweight women OR 3.6 (95% CI 3.1 to 4.2) for obese women OR 8.6 (95% CI 5.1 to 16) for severely obese women Table IV presents expected costs of fertility treatment and pregnancy complications. We used several studies on costs of pregnancy complications and calculated the costs presented in euro s using the current exchange rates. [83-86] Furthermore, we used studies on costs in the Netherlands for OI, IUI and IVF treatment. [76;77;86] 27

31 Chapter 2 Table IV Costs of pregnancy complications and ART per pregnancy study (yr) complication/treatment costs per pregnancy ( ) Chen (2001) cesarean delivery 3,350* Barton (2006) hypertensive disorder 8,250* Moss (2007) gestational diabetes mellitus 345* Graziosi (2005) miscarriage 683 Goverde (2000) IVF 1,700* Eijkemans (2005) OI 250 Goverde (2000) IUI 450* * calculated from US and AUS dollars and Dutch guilders Expected outcome and costs Table V shows the result when the model was applied on a hypothetical cohort of 1,000 anovulatory women. Our model represents costs until birth, including the costs of delivery. In 1,000 normal weight anovulatory women, treatment with three cycles of OI and, if needed, followed by one or two cycles of IVF, would result in 900 pregnancies. Figure IA shows that costs per live birth are higher for overweight and obese anovulatory women with different success rates of OI and IVF and these differences in costs are roughly constant over a large range of success rates. Of these pregnancies 90 are expected to end in miscarriage and 810 women will have an ongoing pregnancy. The expected number of pregnancies complicated by preeclampsia, gestational diabetes and caesarean delivery, will be 81, 41 and 81, respectively, whereas 4 women will suffer stillbirth. Overall, 806 women are expected to deliver a child, for a total cost of 2,430 per woman, resulting in a cost of 3,016 per live birth. From Table V, it can also be seen that in overweight anovulatory women the effectiveness of treatment decreases, resulting in a decrease of the number of pregnancies and live births, an increase in costs and a relative increase of the number of complications. This results in a decrease in the number of live births of 114 (14%), and an expected increase in costs of almost 800 (32%) per patient. For obese anovulatory women, these figures are slightly worse, as the number of live births decreases to 119 (15%), and the expected increase in cost of approximately 1,700 (71%) per patient as compared to normal weight women. 28

32 Economic consequences Table V Hypothetical cohort of 1,000 anovulatory women in different weight categories normal weight overweight¹ obese² Cohort 1,000 1,000 1,000 Impact of overweight on OI (OR) Pregnant after 3 cycles OI (baseline rate 80%) Number of women undergoing IVF Impact of weight on effectiveness IVF (OR) Pregnant after 2 cycles IVF (baseline rate 50%) Expected number of pregnancies Impact of weight on miscarriage (OR) Expected number of miscarriages (baseline rate 10%) Number of women without ongoing pregnancy Number of women with ongoing pregnancy Impact of weight on pre-eclampsia (OR) % pregnancies complicated by pre-eclampsia Impact of weight on gestational diabetes (OR) % pregnancies complicated by gestational diabetes Impact of weight on caesarean deliveries (OR) % pregnancies caesarean delivery Impact of weight on stillbirth (OR) % pregnancies stillbirth Total women with live birth Total costs complications ( ) 1,000 1,788 2,724 Total expected costs (* 1,000) 2,430 3,218 4,154 Cost per live birth ( ) 3,016 4,653 6,045 Cost per pregnancy ( ) 3,001 4,618 5,982 ¹ applied BMI (kg/m²) threshold differed from study to study (range 25-27) ² applied BMI (kg/m²) threshold differed from study to study (range 29-35) 29

33 Chapter 2 Figure IA Sensitivity analyses showing the effect of variation of success rates of OI and IVF on costs per live birth in anovulatory women Costs ( ) /40 75/45 80/50 85/55 90/60 OI (%)/IVF (%) obese overweight normal weight Table VI shows the results for a theoretical cohort of 1,000 ovulatory women. In 1,000 normal weight ovulatory women, treatment consisted of three cycles of IUI and if this was unsuccessful one or two cycles of IVF, added with 10% spontaneous pregnancies that occur on waiting lists or in between cycles, would result in 780 pregnancies. Figure IB shows that over a large range of different success rates of IUI and IVF the costs per live birth are higher for overweight and obese ovulatory women. Of these pregnancies 78 are expected to suffer a miscarriage and 702 women will have an ongoing pregnancy. The expected number of pregnancies complicated by preeclampsia, gestational diabetes and caesarean delivery, will be 70, 35 and 70, respectively, whereas 4 women will suffer stillbirth. Overall, 698 women are expected to deliver a child, for a total cost of 4,258 per woman, resulting in a cost per live birth 6,096. Similarly as for anovulatory women, it can be shown that in overweight ovulatory women the effectiveness of treatment decreases, resulting in a decrease of the number of pregnancies and live births, an increase in costs and a relative increase of the number of complications. This results in a decrease of the number of live births of 153 (22%), with the expected increase in costs of 543 (13%), resulting in a cost per live birth of 8,800. For obese ovulatory women, these figures are worse with a decrease in live births of 167 (24%), with the expected increase in costs of almost 1,250 (29%), resulting in a cost per live birth of 10,

34 Economic consequences Table VI Hypothetical cohort of 1,000 ovulatory women in different weight categories normal weight overweight¹ obese² Cohort 1,000 1,000 1,000 Impact of overweight on spontaneous pregnancies (OR) Spontaneous pregnancies (baseline rate 10% ) Number of women undergoing IUI Treatment effect of IUI (OR) Pregnant after 3 cycles IUI (baseline rate 40% ) Number of women undergoing IVF Impact of overweight on effectiveness IVF Pregnant after 2 cycles IVF (baseline rate 50% ) Expected number of pregnancies Impact of weight on miscarriage (OR) Expected number of miscarriages (baseline rate 10%) Number of women without ongoing pregnancy Number of women with ongoing pregnancy Impact of weight on preeclampsia (OR) % pregnancies complicated by preeclampsia Impact of weight on gestational diabetes (OR) % pregnancies complicated by gestational diabetes Impact of weight on caesarean deliveries (OR) % pregnancies caesarean delivery Impact of weight on stillbirth (OR) % pregnancies stillbirth Total women with live birth Total costs complications ( ) 867 1,410 2,103 Total expected costs (* 1,000) 4,258 4,801 5,494 Cost per live birth ( ) 6,096 8,800 10,355 Cost per pregnancy ( ) 6,066 8,734 10,246 ¹ applied BMI (kg/m²) threshold differed from study to study (range 25-27) ² applied BMI (kg/m²) threshold differed from study to study (range 29-35) 31

35 Chapter 2 Figure IB Sensitivity analyses showing the effect of variations of success rates of IUI and IVF on costs per live birth in ovulatory women Costs ( ) obese overweight normal weight 0 30/40 35/45 40/50 45/55 50/60 IUI(%)/IVF(%) Discussion Overweight and obesity are an increasing problem in Western society. In this review, we collected data on the impact of overweight and obesity on fertility care. We found that both in ovulatory and in anovulatory subfertile women overweight and obesity resulted in a decreased fecundity and in an increase in the number of pregnancy complications and associated costs. However, there is no proven cause and effect between overweight and subfertility. It remains possible that excessive weight and subfertility are both symptoms of an unknown pathology. Our results roughly suggest that overweight leads to an additional cost of 1,500 per pregnancy and 100 fewer pregnancies per 1,000 anovulatory women undergoing fertility treatment, where this is 2,500 and 150 pregnancies respectively for ovulatory women. The validity of our findings depends on the robustness of our methodology. We put forward a framework that can be used to encourage development of more advanced models for generating cost-effectiveness information through robust economic evaluation. In our review of the literature, we found different and occasionally conflicting results on the impact of overweight and obesity on the effect of fertility treatment. When this was the case, we chose to consider no effect of overweight. Furthermore, the success rate of IVF decreases with 32

36 Economic consequences increasing BMI, thus overweight and obese women will have to undergo more cycles compared to normal weight women. [30;86] As a consequence, our findings may be an underestimation of the impact of overweight and obesity. Since the purpose of this review was not to give exact figures on costs but to show a trend in costs and cost-effectiveness, we feel this possible inaccuracy does not undermine the overall conclusion. 2 From our analysis several issues rise. First, as a higher BMI is associated with more pregnancy complications, there is the question as to whether women should lose weight before fertility treatment is started. A recent retrospective analysis by Maheshwari et al. concludes that cost of IVF is not different for several weight categories but because of obstetric complications associated with higher BMI women with overweight should be advised to lose weight prior to IVF. [86;87] Our analysis concurs with this conclusion and gives indicative results that merit consideration in counselling patients and guiding evidencebased discussions on current practice and policy. Weight loss may be achieved by lifestyle modification interventions, incorporating multiple approaches (diet, exercise, behaviour). Interventions of this kind are advised as a key component for the improvement of reproductive function in overweight women, specifically with PCOS, although the evidence of its effectiveness as demonstrated in clinical studies is limited. [37-39;42;61; 88-90] The cost-effectiveness of losing weight has never been assessed in large groups of subfertile women with respect to increasing treatment success for weight-related subfertility, prevention of complications during pregnancy and improvement of perinatal outcome. Until this has been demonstrated we do not think it should be obligatory for overweight subfertile women to undergo a lifestyle intervention program before starting fertility treatment. But in counselling patients, there should be attention for possible pregnancy complications with increasing BMI. It is clear that losing weight takes great effort and we feel that overweight should be considered a disease rather than an amenable condition. 33

37 Chapter 2 Second, apart from the unproven effectiveness of lifestyle interventions in overweight subfertile women, there is the question as to whether there should be upper limits for BMI above which couples should not be treated. Some authors have suggested limits for BMI for women undergoing fertility treatment, both with the arguments of patient safety concerns, as well as a lack of effectiveness of treatment of obese women. [30;91;92] However, we feel that from the perspective of effectiveness of treatment, our data show that there is no reason to withhold treatment. Although effectiveness rates decrease with increasing BMI, the same appears true for women undergoing assisted reproduction over the age of 40, which is a well accepted practice in many countries. Age is however a predictable and amenable factor considering the fact that many couples delay conception to for example pursue career opportunities. In our opinion, studies on weight loss interventions should show a clear increase of effectiveness of fertility treatment and a clear decrease in pregnancy related complications, before BMI limits can be implemented. In conclusion, in ovulatory and anovulatory subfertile women overweight and obesity is associated with a decrease in the number of pregnancies, a sharp increase in the number of complications with an additional rise of associated costs per pregnancy. There is not enough evidence however to prove that losing weight will improve the outcome of fertility treatment and decrease complications in pregnancies, and therefore strict BMI limits cannot be recommended yet. However, overweight and obese subfertile women should be counselled that overweight is a risk factor in pregnancy and is associated with several complications in both mothers and their children. 34

38

39

40 Chapter 3 Complications and outcome of assisted reproductive technology in overweight and obese women Koning AMH, Mutsaerts MAQ, Kuchenbecker WKH, Broekmans FJ, Land JA, Mol BW, Hoek A Hum Reprod, 2012 Feb;27(2):

41 Chapter 3 ABSTRACT Background Based on a presumed negative impact of overweight and obesity on reproductive capacity and pregnancy outcome, some national guidelines and clinicians have argued that there should be an upper limit for a woman s body mass index (BMI) to access assisted reproductive technology (ART). However, evidence on the risk of complications or expected success rate of ART in obese women is scarce. We therefore performed a systematic review on the subject. Methods We searched the literature for studies reporting on complications or success rates in overweight and obese women undergoing ART. Articles were scored on methodological quality. We calculated pooled odds ratios (ORs) to express the association between overweight and obesity on the one hand, and complications and success rates of ART on the other hand. We only pooled results if data were available per woman instead of per cycle or embryo transfer. Results We detected 14 studies that reported on the association between overweight and complications during or after ART, of which 6 reported on ovarian hyperstimulation syndrome (OHSS), 7 on multiple pregnancies and 6 on ectopic pregnancies. None of the individual studies found a positive association between overweight and ART complications. The pooled ORs for overweight versus normal weight for OHSS, multiple pregnancy and ectopic pregnancy were 1.0 [95% confidence interval (CI) 0.77 to 1.3], 0.97 (95% CI 0.91 to 1.04) and 0.96 (95% CI 0.54 to 1.7), respectively. In 27 studies that reported on BMI and the success of ART, the pooled ORs for overweight versus normal weight on live birth, ongoing and clinical pregnancy following ART were OR 0.90 (95% CI 0.82 to 1.0), 1.01 (95% CI 0.75 to 1.4) and OR 0.94 (95% CI 0.69 to 1.3), respectively. Conclusions Data on complications following ART are scarce and therefore a registration system should be implemented in order to gain more insight into this subject. In the available literature, there is no evidence of overweight or obesity increasing the risk of complications following ART. Furthermore, they only marginally reduce the success rates. Based on the currently available data, overweight and obesity in itself should not be a reason to withhold ART. 38

42 Complications and outcome Introduction Obesity is a growing problem throughout the Western world with more than half of the women in the UK and USA being overweight or obese, and up to 25% of women in their reproductive age being obese. [50;93] Several studies have shown that obesity is associated with a reduced fecundity. [9;56;94] In addition, obese women are more prone to anovulation and symptoms of polycystic ovarian syndrome (PCOS) are aggravated by obesity. [54] Pregnancy in overweight and obese women is associated with an increased risk of complications, leading to higher maternal and neonatal morbidity and mortality and increased costs. [27;51;69;70;75] 3 Due to the worldwide epidemic of obesity, an increasing proportion of women seeking medical help for subfertility will be overweight or obese. In their analysis, Vahratian and Smith found that, in comparison to normal weight women, obese women seek medical attention for their subfertility more often. [34] They receive, however, the least fertility-related services. Guidelines that regulate access to fertility care for overweight and obese women vary worldwide. In New Zealand, for example, women with a BMI of >32 kg/m 2 are excluded from any fertility treatment. Almost all clinics in the UK have an upper BMI limit for access to assisted reproductive technologies (ART), ranging from 25 to 40 kg/m 2. [91;92;95] The UK national guideline recommends that women with a BMI above 29 kg/m 2 should be informed about their lower pregnancy chances but does not explicitly prescribe a BMI cut-off point for treatment. [96] Recently, the ESHRE Task Force on Ethics and Law, reported on lifestyle-related factors, i.e. alcohol consumption, smoking and obesity, and access to medically assisted reproduction. [97] It was argued that if a high risk of serious harm for the future child is anticipated, fertility treatment should be denied. Although this risk is obvious for alcohol consumption and smoking, it is unclear whether this recommendation can be applied in daily clinical practice with respect to adipose subfertile women. Data regarding the impact of overweight and obesity on fertility treatment outcome are conflicting, with some studies showing a negative impact on pregnancy rates, whereas others report no impact. [61;64;98] In a systematic 39

43 Chapter 3 review of the literature on overweight and ART, Maheshwari et al. concluded that there is insufficient evidence of the effect of BMI on the outcome of ART to justify denial of treatment for overweight and obese women. [30] Apart from the impact of overweight and obesity on spontaneous pregnancy rates and the course of pregnancy, little is known about the impact of BMI on complications of ART. Some have suggested that women with a high BMI are more prone to complications during their fertility treatment and fertility treatment should therefore be denied. The aim of the present systematic review was to investigate the association between overweight or obesity and the occurrence of complications of ART, as well as the expected outcome in terms of clinical and ongoing pregnancy and live birth rates. To do so, we updated the review of Maheshwari et al., and added a new review with data on the impact of overweight and obesity on the risk of complications of ART. [30] Materials and Methods We searched the literature from January 1999 till July 2011 for studies that reported on the association between overweight and obesity on complications of ART, as well as for studies that reported on the association between overweight and obesity and the effectiveness of ART. We searched PubMed, Embase, DARE and the Cochrane Library for studies reporting on the association between BMI and complications of ART. Only studies in English were included. To do so, we combined the keywords ( obesity, overweight or body mass index ), ( complication, ovarian hyperstimulation syndrome, ectopic pregnancy, multiple pregnancy, infection, haemorrhage or injury ) and ( in vitro fertilisation ). In the search for studies on the association between overweight and obesity and the outcome of ART, we added the keywords ( outcome, pregnancy rate, live birth rate ). Additionally, we checked the references of the selected articles for relevant articles. Since overweight and obese women are more prone to PCOS, which has an increased risk of ovarian hyperstimulation syndrome (OHSS) after ART, we searched and described this group separately using keywords mentioned above in combination with ( polycystic ovarian syndrome ). [99] 40

44 Complications and outcome Study quality Two reviewers independently screened titles and abstracts of all retrieved studies (A.M.H.K. and M.A.Q.M.). We obtained full-text reports of studies that were likely to evaluate the association between BMI and complications or BMI and success rates of ART. If the title did not contain any of our keywords, it was excluded. Reviews were excluded. If not enough data were available for a 2 2 table, studies were excluded. Studies reporting on fertility treatment other than IVF/ICSI were excluded. Studies that did not report sufficient data for a 2 2 table, but for which data could possibly be obtained from the authors, were also evaluated. We extracted data on study characteristics, study quality and 2 2 tables of test accuracy using a predesigned data extraction form. In the case of multiple publications on the same data set, we used all publications to acquire complete data. The most recent and complete results were included in the analysis. If data on test accuracy or on other relevant characteristics were missing, we contacted the corresponding author. Disagreement on data was resolved by discussion and consensus. If consensus could not be reached, a third reviewer (B.W.M.) was consulted. 3 Methodological quality of selected papers was evaluated using QUADAS, a tool for quality assessment of studies of diagnostic accuracy. [100] We adjusted the original QUADAS list in order to evaluate items that we considered of specific importance for this systematic review. Included studies were evaluated on 15 items concerning patient selection, verification, description of the tests and of the study population. A comprehensive list of items on which the methodological study quality was assessed is available from the authors on request. We defined complications of ART as OHSS, multiple pregnancy, ectopic pregnancy, haemorrhage, infection or injury to pelvic structures. We defined success rates of ART as clinical pregnancy rate, ongoing pregnancy rate, live birth rate and delivery rate. If data on live birth rates were available, we used these data to calculate odds ratios (ORs) with 95% confidence intervals (CIs). If live birth rates were 41

45 Chapter 3 not available, we used pregnancy rates (clinical or ongoing pregnancy) for the calculations. Some studies report outcome per cycle, with several cycles per woman included. This would suggest better results in terms of pregnancy rates. Therefore, we report outcome (per first cycle) per woman where possible. In order to obtain clinically useful outcome, we chose to only pool data if results per woman were available. Data synthesis From each article, we calculated the OR from a 2 2 table cross-classifying BMI and one of the aforementioned outcomes. We used the WHO definitions for overweight (BMI > 25 kg/m 2 ) and obesity (BMI > 30 kg/m 2 ), where possible. Data of each outcome were pooled if the data were analysed per woman and if there were at least two studies with similar definition of the outcome and similar range of BMI for the comparison groups. Combined ORs and 95% CI were calculated using the Review Manager (RevMan 5.1) computer software of the Cochrane Collaboration. The heterogeneity was assessed with the I 2 -statistic for inconsistency. A value of >50% was considered as substantial heterogeneity. [101] In the case of statistical homogeneity, we used a fixed-effect model. In the case of substantial heterogeneity, we used the random-effect model instead of the fixed-effect model. Results Our literature search resulted in 91 studies on the association between BMI and complications of ART and 242 studies on the association between BMI and the effectiveness of ART, respectively. The separate search on the association between overweight women with PCOS and complications after ART or the effectiveness of ART yielded a total of 77 articles. Based on titles and abstracts derived from the search on studies between BMI and complications, 24 articles were identified as potentially eligible for the review (Fig. I). 42

46 Complications and outcome Figure I Flowchart in/exclusion studies for complications in ART included 67 excluded based on relevance/title/abstract 11 excluded only PCOS population (a) 8 excluded not enough data for 2 by 2 table (b) With crossreferencing another 9 studies included (c) 3 (a) [ ] (b) [ ] (c) [64;87;98; ] Eleven studies were excluded because they were solely concerned with women with PCOS and another eight because they lacked adequate data; however, another nine studies were identified by cross-referencing making 14 available for analysis. Table I shows the characteristics of the 14 studies and the results per woman if available, otherwise per cycle for each complication. [64;65;78;87;98; ] Only two out of the 14 studies were prospective. It was unclear whether a BMI limit was used for access to ART. None of the researchers reported blinding for BMI status in the assessment of end-points. Of the 14 studies, six reported on the relationship between OHSS and BMI, seven reported on the relationship between multiple pregnancies and BMI and six reported on the relationship between ectopic pregnancies and BMI. We found no studies that reported on the complications infection, haemorrhage or injury to pelvic structures in relation to BMI in women undergoing ART. Three out of the 14 studies used a different definition than the WHO definition for overweight (Table I legend). [65;78;127] 43

47 Chapter 3 Table I Percentages and odds ratios of complications following ART for overweight versus normal weight women Author Year ART N Per Data collection Complication (OR BMI > 25 vs BMI < 25) OHSS multiple pregnancy (p/p) ectopic pregnancy (p/p) Farhi 2011 IVF 233 woman retrospective 2.7% vs 3.8% (OR 0.7, 95% CI 0.1 to 3.7) Sathya 2010 IVF 308 woman retrospective 32.2% vs 26.3% (OR 1.3, 95% CI 0.6 to 3.1) 6.7% vs 2.6% (OR 2.7, 95% CI 0.3 to 22.7) Luke 2010 IVF 45,163 ET retrospective 31.9% vs 32.1% (OR 1.0, 95% CI 0.9 to 1.1) Zhang 2010 IVF/ICSI 2,628 woman retrospective 4.5% vs 3.8% (OR 1.2, 95% CI 0.5 to 2.9) Maheshwari 2009 IVF 1,756 woman retrospective 10.3% vs 10.2% (OR 1.0, 95% CI 0.7 to 1.4) 18.0% vs 20.7% (OR 0.9, 95% CI 0.6 to 1.3) Sneed 2008 IVF/ICSI 1,273 woman retrospective 1.7% vs 3.0% (OR 0.6, 95% CI 0.2 to 1.8) Matalliotakis 2008 IVF/ICSI 278 woman retrospective 27.6% vs 26.5% (OR 1.1, 95% CI 0.5 to 2.1)¹ Esinler 2008 ICSI 775 ET retrospective 0.8% vs 1.0% (OR 0.9, 95% CI 0.2 to 3.0) Dokras 2006 IVF/ICSI 1,291 woman retrospective 4.8% vs 4.8% (OR 1.0, 95% CI 0.6 to 1.6) 47.0% vs 52.0% (OR 0.8, 95% CI 0.6 to 1.2) 27.8% vs 31.3% (OR 0.9, 95% CI 0.6 to 1.2) 1.1% vs 1.2% (OR 1.0, 95% CI 0.06 to 15.5)¹ Van Swieten 2005 IVF/ICSI 162 woman prospective 4.9% vs 5.0% (OR 1.0, 95% CI 0.2 to 4.3) Spandorfer 2004 IVF/ICSI 828 cycle retrospective 35.7% vs 52.3% (OR 0.5, 95% CI 0.3 to 0.8)² Fedorcsak 2004 IVF/ICSI 2,660 woman retrospective 0.5% vs 1.7% (OR 0.3, 95% CI 0.03 to 2.0) Wittemer 2000 IVF/ICSI 325 cycle retrospective 15.0% vs 5.0% (OR 3.4, 95% CI 0.6 to 18.2) Lashen 1999 IVF 228 woman prospective 2.6% vs 5.3% (OR 0.5, 95% CI 0.1 to 2.4)³ OHSS ovarian hyperstimulation syndrome, IVF in vitro fertilisation, ICSI intracytoplasmic insemination, p/p per pregnancy, ¹BMI 24, ²BMI 27, ³BMI vs BMI>28 44

48 Complications and outcome None of the six studies that reported on OHSS found a significant difference between the risk of OHSS in normal and overweight women. The pooled OR for overweight woman was 1.0 (95% CI 0.77 to 1.3) (Fig. II). Figure II OHSS for overweight and normal weight women Study or Subgroup Dokras 2006 Farhi 2011 Maheshwari 2009 Van Swieten 2005 BMI > 25 BMI < 25 Odds Ratio Odds Ratio Events Total Events Total Weight 25.9% 3.2% 67.8% 3.1% M-H, Fixed, 95% CI 0.98 [0.59, 1.64] 0.72 [0.14, 3.67] 1.02 [0.74, 1.39] 0.99 [0.23, 4.31] M-H, Fixed, 95% CI Total (95% CI) Total events Heterogeneity: Chi² = 0.17, df = 3 (P = 0.98); I² = 0% Test for overall effect: Z = 0.01 (P = 0.99) 100.0% 1.00 [0.77, 1.29] BMI > 25 BMI < 25 3 Of the seven studies that reported on the relation between BMI and multiple pregnancies for women undergoing ART, the retrospective cohort study of Spandorfer et al. found a significantly higher risk of multiple pregnancies in women with normal weight following ART, whereas six other studies did not. [124] The pooled OR expressing the association between overweight and the risk of multiple pregnancies was 0.97 (95% CI 0.91 to 1.04) (Fig. III). Figure III Multiple pregnancies for overweight and normal weight women Study or Subgroup Dokras 2006 Esinler 2008 Luke 2010 Maheshwari 2009 Sathya 2010 BMI > 25 BMI < 25 Odds Ratio Odds Ratio Events Total Events Total Weight 3.8% 3.7% 89.2% 2.8% 0.5% M-H, Fixed, 95% CI 0.85 [0.60, 1.20] 0.82 [0.58, 1.16] 0.99 [0.92, 1.06] 0.85 [0.56, 1.27] 1.33 [0.57, 3.10] M-H, Fixed, 95% CI Total (95% CI) Total events Heterogeneity: Chi² = 2.75, df = 4 (P = 0.60); I² = 0% Test for overall effect: Z = 0.76 (P = 0.45) 100.0% 0.97 [0.91, 1.04] BMI > 25 BMI < 25 Six studies reported on the association between BMI and ectopic pregnancies. None of these studies revealed a significant difference between normal and overweight women. The pooled OR expressing the association between overweight and the risk of ectopic pregnancy was 0.96 (95% CI 0.54 to 1.7) (Fig. IV). 45

49 Chapter 3 Figure IV Ectopic pregnancies for overweight and normal weight women Study or Subgroup Fedorcsak 2004 Sathya 2010 Sneed 2008 Wittemer 2000 Zhang 2010 BMI > 25 BMI < 25 Odds Ratio Odds Ratio Events Total Events Total Weight 22.7% 5.4% 32.0% 5.3% 34.6% M-H, Fixed, 95% CI 0.26 [0.03, 2.03] 2.64 [0.31, 22.73] 0.55 [0.17, 1.76] 3.35 [0.62, 18.16] 1.17 [0.48, 2.88] M-H, Fixed, 95% CI Total (95% CI) Total events Heterogeneity: Chi² = 5.58, df = 4 (P = 0.23); I² = 28% Test for overall effect: Z = 0.14 (P = 0.89) 100.0% 0.96 [0.54, 1.70] BMI > 25 BMI < 25 After screening of titles and abstracts derived from the search on 242 studies reporting on the association between BMI and the effectiveness of ART, 40 articles were identified as potentially eligible for the review. After exclusion of papers that studied only patients with PCOS and those with insufficient data, 25 papers remained and a further two were identified by cross-referencing yielding 27 suitable for analysis (Fig. V). [64-67;78;87;98;115; ] Figure V Flowchart in/exclusion studies for outcome ART excluded based on relevance/title/abstract 11 excluded only PCOS population (a) 29 4 excluded not enough data for 2 by 2 table (b) With crossreferencing another 2 studies included (c) (a) [76;102;105;108;109; ] (b) [ ] (c) [123;131] 46

50 Complications and outcome Of the 27 included studies, four were prospective (Table II). Eleven studies reported on live birth rates, two on delivery rates, whereas 14 studies reported on pregnancy rates. The definition of delivery rate given in Dokras et al. is delivery after 20 weeks of gestation, whereas no definition is provided in Wittemer et al.. [98;129] Nine studies reported on clinical pregnancy rate, four on ongoing pregnancy rate and one reported on fecundity. Definitions of these outcome measures differed between studies. Six out of the 27 studies used a different definition for overweight and obesity than the WHO definition (Table II legend). [65;122;124;133;137;139] In four of the 27 included studies, a significant negative effect of an increased BMI on outcome was found, 22 studies did not find a significant association between BMI and ART outcome, and one study reported significantly more pregnancies in obese women. In 17 studies, results were presented per woman, and in 10 studies, results were presented per cycle or per embryo transfer (ET). Therefore, not all study results were comparable, and as a result, only seven studies could be pooled for the live birth rate per woman, our primary interest. 3 Overweight women undergoing ART had a significant lower live birth rate after ART than women with a normal weight [OR 0.90 (95% CI 0.82 to 1.0)] (Fig. VI), whereas obesity did not show a statistically significant difference in six studies including almost 10,000 women [OR 0.89 (95% CI 0.76 to 1.03)]. Clinical pregnancy rates were not different for overweight and normal weight women [OR 0.94 (95% CI 0.69 to 1.3)] or for obese women when compared with normal weight women [OR 0.97 (95% CI 0.59 to 1.6)]. For ongoing pregnancy rates, the pooled ORs were 1.01 (95% CI 0.75 to 1.4) and 0.96 (95% CI 0.64 to 1.4) for overweight and obesity, respectively (data not shown). 47

51 Chapter 3 Table II Percentages and odds ratios of different outcome measures of ART in overweight and obese versus normal weight women Author Year Design ART Data collection N Outcome Per Percentage and OR (95% CI) BMI > 25 vs BMI < 25 Farhi 2011 cohort IVF retrospective 233 live birth woman 41.1% vs 38.1% OR 1.1 (0.6 to 2.0) NA Percentage and OR (95% CI) BMI > 30 vs BMI < 30 Hill 2011 cohort IVF prospective 117 live birth woman 44.1% vs 32.8% OR 1.6 (0.8 to 3.4) 33.3% vs 39.6% OR 0.8 (0.3 to 2.1) Sathya 2010 cohort IVF retrospective 308 clinical pregnancy woman 40.9% vs 43.2% OR 0.9 (0.6 to 1.5) 32.9% vs 44.3% OR 0.6 (0.4 to 1.1) Luke 2010 cohort IVF retrospective 45,163 live birth ET 34.1% vs 37.6% OR 0.9 (0.8 to 0.9) 32.2% vs 35.9% OR 0.9 (0.8 to 0.9) Kilic 2010 C-S IVF retrospective 1,970 clinical pregnancy woman 47.0% vs 45.1% OR 1.1 (0.9 to 1.3) 49.9% vs 43.9% OR 1.3 (1.1 to 1.5) Zhang 2010 cohort IVF/ICSI retrospective 2,628 live birth woman 23.9% vs 26.2% OR 0.9 (0.7 to 1.1) 26.0% vs 25.8% OR 1.0 (0.4 to 2.4) Bellver 2010 cohort IVF/ICSI retrospective 6,500 live birth cycle 27.6% vs 30.9% OR 0.9 (0.8 to 1.0) 23.6% vs 30.5% OR 0.7 (0.6 to 0.9) Maheshwari 2009 C-S IVF retrospective 1,756 live birth woman 22.9% vs 25.9% OR 0.9 (0.7 to 1.1) 22.2% vs 25.0% OR 0.9 (0.6 to 1.2) Orvieto 2009 cohort IVF/ICSI retrospective 516 clinical pregnancy cycle NA 19.6% vs 28.8% OR 0.6 (0.3 to 1.2) Matalliotakis 2008 cohort IVF/ICSI retrospective 278 live birth woman 47.1% vs 46.4% OR 1.0 (0.6 to 1.7)¹ NA Sneed 2008 cohort IVF/ICSI retrospective 1,273 live birth woman 21.2% vs 24.4% OR 0.8 (0.6 to 1.1) 21.5% vs 23.3% OR 0.9 (0.7 to 1.2) Esinler 2008 cohort ICSI retrospective 775 clinical pregnancy ET 46.3% vs 44.6% OR 1.1 (0.8 to 1.4) 48.3% vs 44.9% OR 1.2 (0.8 to 1.6) Martinuzzi 2008 cohort IVF/ICSI retrospective 417 ongoing woman 45.0% vs 46.9% OR 0.9 (0.6 to 1.4) 42.3% vs 46.8% OR 0.8 (0.5 to 1.5) pregnancy Thum 2007 cohort IVF/ICSI retrospective 8,145 live birth woman 26.3% vs 29.3% OR 0.9 (0.8 to 1.0)² 24.1% vs 28.7% OR 0.8 (0.6 to 1.0)³ Dokras 2006 cohort IVF/ICSI retrospective 1,291 delivery rate woman 41.7% vs 41.6% OR 1.0 (0.8 to 1.3) 40.6% vs 42.0% OR 0.9 (0.7 to 1.2) 48

52 Complications and outcome Table II continued Percentage and OR (95% CI) BMI > 30 vs BMI < 30 N Outcome Per Percentage and OR (95% CI) BMI > 25 vs BMI < 25 Author Year Design ART Data collection woman 10.5% vs 25.9% OR 0.3 (0.1 to 1.0)¹ NA Ku 2006 cohort ICSI retrospective 223 clinical pregnancy Dechaud 2006 cohort IVF/ICSI retrospective 789 ongoing cycle 13.0% vs 16.6% OR 0.8 (0.4 to 1.3) 12.5% vs 16.2% OR 0.7 (0.3 to 1.8) pregnancy Styne-Gross 2005 cohort IVF retrospective 536 ongoing woman 74.2% vs 72.2% OR 1.1 (0.7 to 1.7) 74.0% vs 72.5% OR 1.1 (0.6 to 1.9) donor pregnancy oocyte Lintsen 2005 cohort IVF retrospective 8,105 live birth cycle 14.3% vs 15.5% OR 0.9 (0.8 to 1.1) NA woman 32.8% vs 45.5% OR 0.6 (0.3 to 1.1) 41.4% vs 40.6% OR 1.0 (0.5 to 2.3) Van Swieten 2005 cohort IVF/ICSI prospective 162 clinical pregnancy cycle 56.3% vs 56.4% OR 1.0 (0.7 to 1.5)⁴ NA Spandorfer 2004 cohort IVF/ICSI retrospective 828 clinical pregnancy Fedorcsak 2004 cohort IVF/ICSI retrospective 2,660 live birth woman 19.2% vs 21.7% OR 0.9 (0.7 to 1.1) 17.8% vs 21.3% OR 0.8 (0.6 to 1.1) cycle 24.7% vs 50.6% OR 0.3 (0.2 to 0.6) NA Loveland 2001 cohort IVF retrospective 139 ongoing pregnancy woman 26.0% vs 38.0% OR 0.6 (0.2 to 1.3)² NA Salha 2001 CC IVF prospective 100 clinical pregnancy Wang 2000 cohort IVF/ICSI retrospective 3,586 fecundity cycle 40.4% vs 47.4% OR 0.8 (0.7 to 0.9) 37.3% vs 46.0% OR 0.7 (0.6 to 0.9) Wittemer 2000 cohort IVF/ICSI retrospective 325 delivery rate cycle 14.3% vs 16.9% OR 0.8 (0.4 to 1.7) NA woman 23.7% vs 20.4% OR 1.2 (0.6 to 2.3)⁵ NA Lashen 1999 CC IVF prospective 228 clinical pregnancy C-S cross-sectional, CC case control, IVF in vitro fertilisation, ICSI intracytoplasmic insemination, ET embryo transfer, NA not available ¹ BMI 24, ² BMI 26, ³ BMI 31, ⁴ BMI 27, ⁵ BMI vs BMI >

53 Chapter 3 Figure VI Live birth rate for overweight and normal weight women Study or Subgroup Dokras 2006 Farhi 2011 Fedorcsak 2004 Hill 2011 Maheshwari 2009 Sneed 2008 Zhang 2010 BMI > 25 BMI < 25 Odds Ratio Odds Ratio Events Total Events Total Weight 19.3% 2.8% 23.3% 1.3% 21.1% 15.2% 17.0% M-H, Fixed, 95% CI 1.00 [0.80, 1.25] 1.13 [0.64, 1.99] 0.86 [0.69, 1.06] 1.62 [0.76, 3.43] 0.85 [0.68, 1.06] 0.83 [0.64, 1.08] 0.88 [0.69, 1.13] M-H, Fixed, 95% CI Total (95% CI) Total events Heterogeneity: Chi² = 4.74, df = 6 (P = 0.58); I² = 0% Test for overall effect: Z = 2.02 (P = 0.04) 100.0% 0.90 [0.82, 1.00] BMI > 25 BMI < 25 Our search retrieved 77 studies on the association between overweight women with PCOS and the effectiveness of ART. Two studies report on complications in PCOS women with respect to their BMI. Ozgun et al. report on multiple pregnancy rate, showing no difference between BMI < 30 versus >30 kg/m 2 (P = 1). [144] Liberty et al. describe women with haemorrhage after oocyte retrieval, in which they conclude that lean women with PCOS (BMI kg/m 2 ) are at more risk of this complication (4.5 versus 0%) than obese women with PCOS. [109] Four studies reported on BMI and outcome in PCOS women (Table III). [141;143;144;150] All studies showed better outcome for normal weight women, with an extreme OR 0.2 (95% CI 0.01 to 1.9) for live birth rates for overweight woman in McCormick et al.. [141] Discussion In this systematic review, we assessed the association between overweight and obesity on the one hand and complications and success rates of ART on the other hand. We summarized the current evidence on complications following ART with respect to BMI. We found no impact of overweight or obesity on OHSS, multiple pregnancies or ectopic pregnancies. A slight negative effect on live birth rate is found in overweight and obese women. Literature specifically reporting on the effect of BMI on the aforementioned complications in women with PCOS was lacking. 50

54 Complications and outcome Table III Percentages and odds ratios of outcome measures following ART for overweight and normal weight women with PCOS Percentage and OR (95% CI) BMI > 30 vs BMI < 30 Outcome Percentage and OR (95% CI) BMI > 25 vs BMI < 25 Definition PCOS Author Year ART N Per Data collection live birth NA 22.2% vs 46.2% OR 0.3 (0.1 to 1.3) Ozgun 2011 IVF/ICSI 44 woman prospective Rotterdam criteria Orvieto 2009 IVF/ICSI 100 cycle retrospective Rotterdam clinical 15.5% vs 28.6% OR 0.5 (0.2 to 1.2) NA criteria pregnancy McCormick 2007 IVF/ICSI 17 cycle retrospective Rotterdam live birth NA 45.5% vs 83.3% OR 0.2 (0.01 to 1.9)¹ criteria Kjotrod 2004 IVF/ICSI 73 woman prospective Rotterdam live birth NA 37.5% vs 39.4% OR 0.9 (0.4 to 2.4)² criteria IVF in vitro fertilisation, ICSI intracytoplasmic insemination, NA not available, ¹BMI vs BMI > 30, ²BMI

55 Chapter 3 One of the drawbacks of our review is the lack of publications on complications following ART in general. The articles we found did not study complications of ART in relation to BMI as their primary outcome; as a consequence, not one was powered to show a significant difference. Also, with exception of two studies, all were retrospective with the possible selection bias and confounding effect. As our research has been based on this limited and of moderate quality data, the outcome of our study should be put in this perspective. Serious complications like infection, haemorrhage, thromboembolism or even mortality are not documented in relation to BMI. The Safety Interest Group (SIG) Safety and Quality of the ESHRE recommended registering mortality following ART worldwide because of this lack of information. Following this review of the literature, we underscore the importance of achieving better data collection of morbidity following ART. Furthermore, we were not able to differentiate further between obesity classes because of lacking data. As mentioned earlier, several clinics throughout the world use upper limits for BMI for ART with scarce evidence on morbid obese women as a consequence. One might expect more complications or lower outcome results in obesity class III than in class I (WHO definition obesity class I BMI kg/m 2 and class III BMI > 40 kg/m 2 ). [121] However, in view of the lack of evidence on the outcomes of ART in severe obesity and in view of the little impact that we found from mild obesity, it remains difficult to define a limit. We found an OR of 0.90 for the association between overweight and live birth, indicating a 10% reduction in the success rates of IVF in overweight women. The absolute effect of such an association obviously depends on the absolute live birth rates after IVF. When, for example, the probability of live birth is 30% in a normal weight woman, then the success rates of an overweight woman drop to 27.8%. In other words, when we treat 46 normal weight women with IVF, there is one additional pregnancy when compared with when we treat 46 overweight women with the same treatment. Another drawback is that most articles are retrospective observational cohort studies with the disadvantage of observation bias and confounding. This bias could be overcome with prospective studies where covariate data can be collected extensively and in sufficient detail to more accurately evaluate 52

56 Complications and outcome confounding and effect modification. Moreover, the studies applied different cut-off points to define overweight and obesity, as well as different definitions of outcome and the population of women differed between studies. Because of these differences, we were not able to pool all studies in our analysis, e.g. half of the studies on live birth rate with 60,000 women could not be used in our analysis hypothetically changing our conclusion. [67;121;122;129;130;139] For argument sake, we therefore performed the analysis including the results of Luke et al., comprising ET, to establish the possible impact on our pooled OR. The OR for live birth rate remained within the same margins with OR 0.86 (95% CI 0.83 to 0.9) for overweight and OR 0.85 (95% CI 0.81 to 0.9) for obese women, now also showing a significantly lower live birth rate for obese women. 3 We expected to find a difference in complication rates in controlled ovarian hyperstimulation due to difficulty monitoring follicles with ultrasound in overweight or obese women. Therefore, we searched literature specifically for this association but we did not identify any study reporting on this subject. As OHSS is found more often in women with PCOS, we presumed that there might be a difference between normal and overweight women. [99] However, no literature was found on the subject. Our results on the effectiveness of ART concur with those of Maheshwari et al.. [30] Since publication of that review, 14 relevant additional studies were published on the subject. We found a moderate negative effect of obesity on ART outcome. Women with a BMI of 25 kg/m 2 or higher achieve reasonable, but slightly lower live birth rates when compared with normal weight women. The fact that the pooled OR for overweight women is significant, but the pooled OR of obese women lacks statistical significance is probably due to the smaller group of obese women when compared with overweight women, as point estimates for both groups were from the same magnitude. A large study on subfertile women undergoing ART with autologous and donor oocytes subclassifying between several obesity classes did show significantly lower live birth rates in increasing obesity classes with autologous oocytes. [121] However, they compared obesity classes to normal weight women, whereas we tried to find a cut-off value and compare, for example, BMI > 30 kg/m 2 with BMI < 30 kg/m 2. 53

57 Chapter 3 In the debate whether overweight or obese women should be denied access to ART, it is of interest to compare the effect of BMI with that of female age on ART outcome, as this is a well-accepted criterion to deny or allow ART. [151] In view of the decreasing success rates of ART in older women, most countries have an upper limit for age with respect to access to ART programmes. In the Netherlands, national guidelines recommend not offering ART to women older than 41 years because of poor pregnancy chances; but in clinical practice, this threshold has moved to an age of 43. Figure VII shows live birth rates following IVF in relation to female age for women with a BMI of <25 and >25 kg/m 2, which decrease from 26% for younger women to 10% for women aged 40. [152] When we consider an effect of overweight and obesity as pooled OR 0.9, the profound effect of age is much stronger when compared with the moderate effect of overweight on the live birth rate following IVF. This combined OR 0.9 should be viewed in the light of the earlier mentioned remarks on the level of evidence of the included studies. Figure VII Live birth rates following IVF for overweight and normal weight women in different age groups 54

58 Complications and outcome Although high BMI is associated with obstetric and perinatal complications, the BMI limits now applied in some countries are not justified by empirical evidence. Since weight loss programmes have shown limited success, thus indicating how difficult it is to lose weight, certainly in older patients, the gain in outcome variables must be balanced against the detrimental effect of older age. [128] Indeed, to delay attempts to conceive to the age of 40 is probably more detrimental and perhaps more changeable than the decision to gain weight. In the Netherlands, at present, a multicentre randomized controlled trial is in process which compares lifestyle intervention prior to conventional fertility care (including ART) with direct conventional fertility care in women with a BMI of 29 kg/m 2. [153] If this study shows better results in the treatment group, this would provide scientific evidence for recommending lifestyle intervention before fertility treatment is started. 3 Apart from the slightly negative effect of BMI 25 kg/m 2 on ART outcome, there is a substantial risk of pregnancy complications in overweight and obese women, such as hypertension, gestational diabetes, macrosomia and an increased risk of Caesarean section and perinatal death. [27;71;72] Moreover, the general health risk in terms of increased prevalence of, for example, cardiovascular disease and type 2 diabetes of overweight is well described. [154] In our opinion, overweight women should be informed about these overall health risks, obstetric risks and slightly lower success rates of ART when seeking fertility care. In conclusion, there is a lack of studies reporting on complications following ART. In the published literature, we did not find more complications in subfertile overweight and obese women when compared with subfertile normal weight women. We found slightly lower success rates in these overweight women following ART. Therefore, we feel there is no evidence for excluding women above a certain BMI limit from ART but counselling on the increased obstetrical risks is mandatory. Furthermore, following the recommendations of the SIG Safety and Quality of the ESHRE, an international complication registration system for ART should be implemented to increase safety of women undergoing ART. 55

59

60 Chapter 4 Effectiveness of a weight reduction program in obese, subfertile women Koning AMH, Zafarmand MH, Van den Dool G, De Lepper A, Van Veen L, Mol BWJ Submitted for publication

61 Chapter 4 ABSTRACT Purpose To evaluate, in subfertile obese women, the effect of a weight reduction program (WRP) in achieving weight loss and improving fertility. Methods We performed a retrospective cohort study to compare the effectiveness of a WRP versus no intervention in subfertile obese (body mass index (BMI) 32 kg/m 2 ) women. The intervention was a WRP from May 2006 through December The control cohort comprised obese women who visited the fertility clinic between January 2000 through December 2005, before the WRP had been implemented. The WRP consisted of counseling sessions with a personal coach. We compared the mean weight loss in both cohorts and the amount of women losing 5% of body weight. Hazard rate ratios for ongoing pregnancy rates and live birth rates were calculated and adjusted for possible confounders. Results We included 102 women in the intervention group and 100 in the control group. The mean weight reduction was 4.3 kg in the intervention group while in the control group there was a mean weight gain of 0.5 kg compared to baseline (p-value <0.001). There were 33 (32%) women in the intervention group versus 25 (25%) women in the control group who lost more than 5% of body weight compared to their baseline weight (crude odds ratio (OR) 1.40 (95% CI 0.73 to 2.7). The ongoing pregnancy rates were 52 (51%) and 41 (41%) (adjusted hazard rate ratio (HRR) 1.9 (95% CI 1.0 to 3.4)), while live birth rates were 48 (47%) and 37 (37%) (adjusted HRR 2.0 (95% CI 1.1 to 3.8) for the intervention group and control group, respectively. Conclusions These data indicate that in obese subfertile women a WRP is potentially effective. 58

62 Effectiveness weight reduction program Introduction The prevalence of overweight (Body Mass Index [BMI] 25 kg/m 2 ) and obesity (BMI 30 kg/m 2 ) has rapidly increased in the past decades. [155;156] In the Netherlands, 44% of women is overweight and 14% is obese. [1] In 2000, the World Health Organization (WHO) described obesity as one of the greatest public health challenges of the 21 st century. [155] The Dutch government declared the prevention and treatment of overweight and obesity a crucial element in national health policies.[157] In 2008, the Dutch Institute for Healthcare Improvement developed a guideline on the diagnostics and management of obesity. [158] In the Netherlands in 2010 this led to the completion of a standard care document with the goal of increasing the quality of care for obese people leading to better health, greater quality of life and greater participation in society. [159] This document was drawn up by a coalition of 16 organizations consisting of health care providers, insurance companies and patient associations. Health care providers can use this document to assess the scope of the problem in an individual and get advice on prevention and/or treatment of obesity or comorbidities. 4 In reproductive medicine, a steady rise has been seen in the number of obese patients presenting with subfertility. As early as the 1990 s, overweight has been associated with a reduced fecundity and obesity has shown to lead to female subfertility. [ ] For obese women, a small weight reduction could increase the chance of conception as well as decrease the risk of miscarriage. [161;163;166;167] Although the association between obesity and fertility is clear, it is unknown whether interventions aimed to reduce weight in obese subfertile women are truly effective. The relation between overweight and obesity and reproductive outcome made us initiate a weight reduction program (WRP) to support and stimulate obese subfertile women to lose weight. Here, we evaluate the effectiveness of this intervention, and the 4-year-followup results in achieving weight reduction, pregnancy rates and the use of assisted reproductive technology (ART). 59

63 Chapter 4 Methods Population We performed a retrospective cohort study among obese subfertile women who came for fertility treatment between January 2000 and December The study was performed in the fertility clinic of the Albert Schweitzer Hospital in the Netherlands. WRP intervention The intervention group consisted of obese subfertile women participating in a lifestyle program. From May 2006 onwards, all subfertile women with a BMI 32 kg/m 2 were offered participation in such a program. The aim of the program was to obtain weight reduction for ovulatory and anovulatory women in the first six months of participation before initiating ART. The WRP aimed to alter participants lifestyle by introducing nutritional changes and physical exercise in their daily activities. One of the IVF-specialized nurses acted as a personal coach (PC) for WRP participants. She attended courses in motivational interviewing and cognitive therapy. Every two to three weeks, participants had an in-hospital counseling session. The participant s partner was also requested to attend these sessions. In the first one to three PC counseling sessions, the PC showed the participant how to regain control over her own body and lifestyle. The sessions focused on daytime activities, nutritional habits and environmental factors. Hereafter, a focus was applied on implementing routine healthy nutrition and physical exercise. However, as counseling alone often proved to be insufficient, clinical psychologists, specialists in internal medicine and a fertility gynecologist were regularly consulted. We also commonly referred participants to the gym located next to the hospital. No dietary restrictions were subscribed. During every session the body weight was assessed by the PC. Control group The control group consisted of obese subfertile women seen in our clinic from January 2000 through December These were women with a BMI 32 kg/m 2 who visited our fertility clinic and attended an intake counseling session, received an information booklet and were urgently advised to lose weight to enhance their chances of conception. 60

64 Effectiveness weight reduction program During the whole study period between 2000 and 2010, all women with a BMI 40 kg/m 2 were referred to the department of internal medicine for a health screening, as they have highly increased risks for comorbidities such as cardiovascular disease or diabetes mellitus. All women received routine preconception counseling, including counseling on pregnancy complications related to obesity. Endpoints The primary endpoint was weight reduction after six months, defined as at least 5% loss of the initial body weight. For participants in the intervention group, body weight was assessed by the PC, whereas body length was selfreported. In the control group, length and body weight were self-reported by the women. Secondary endpoints were conception, ongoing pregnancy, live birth, multiple pregnancy and ART measured over a period of four years starting from inclusion. If a woman had more than one pregnancy during this period, only the first was used for analysis. If the first pregnancy was a miscarriage, we continued to follow the woman until live birth occurred. 4 Pregnancy and neonatal outcome data were obtained from the pregnancy charts from the hospital. If the woman was referred back to a midwife after conception, we tried to obtain the pregnancy data from the midwife. Statistical analysis Baseline characteristics from the two groups were tabularized and compared for statistical differences. Dichotomous variables were compared using the Chi-Square test, while continuous variables were compared using the Student s t-test. Descriptive results are presented as frequencies with row numbers and percentages or medians with interquartile ranges (IQRs), while the average reduction in weight or BMI and the time period between intake to ART or conception are presented in mean and standard errors. To prevent bias associated with missing data, we used multiple imputations for covariates with missing values on the basis of the correlation of missing variables with other participant characteristics. Although five imputed datasets have been suggested to be sufficient theoretically to reduce sampling variability from the imputation process [169], we imputed 20 datasets as has 61

65 Chapter 4 been recommended recently. [170] In addition, we added variables related to covariates as predictors to the imputation model to increase the plausibility of the missing-at-random assumption. The amount of missing values ranged from 1% to 46% (for semen analysis). We report the pooled results of the analyses performed in each of the 20 imputed datasets. We compared the WRP for any weight loss and a weight loss 5% using the logistic regression. The results as the odds ratios (ORs) with 95% confidence intervals (CIs) are presented after adjustment for age, alcohol consumption, BMI, smoking, ART, type of subfertility and duration of subfertility. Then we compared the main outcomes between the two groups using the Cox regression models. Results from the Cox regression analyses are presented as crude and adjusted hazard rate ratios (HRRs) with 95% CIs. Analyses were adjusted for age, alcohol consumption, BMI, smoking, ART, type of subfertility and duration of subfertility in a multivariable analysis. P values were 2-sided. Time to live birth in the intervention and the control group was compared by constructing a Kaplan-Meier curve. The curves were compared using the logrank test. Different curves for cumulative ongoing pregnancies and amount of ongoing pregnancies per three months for both groups are presented with special attention for the mode of conception. We performed separate analyses on the main outcomes for normal ovulatory women and women with an ovulation problem using the Cox regression models. Finally, we assessed the relation between weight loss and ongoing pregnancy. To do so, we divided the intervention group in four categories based on quartiles of weight loss, i.e. the quartile with the weight loss >75 th %, with a weight loss between 75 th and 50 th %, between 50 th and 25 th %, and less than 25 th %. All statistical analyses were performed using SPSS version 21.0 (IBM Corp., Armonk, NY, USA). Ethics This retrospective study followed the Declaration of Helsinki (2008) on medical protocol and ethics concerning the data collection and analysis and publication of the results. 62

66 Effectiveness weight reduction program Results Figure I shows the study population and exclusions. In the intervention period, 113 women were potentially eligible for the study. There were three women in whom the initial weight was unknown, while eight women conceived naturally before the start of the program. Thus, 102 women underwent the WRP and were analyzed. In the same period, 13 women with a BMI < 32 kg/m² also participated in the WRP, but these women were not analyzed. In the control period, 101 women with a BMI 32 kg/m² were seen. In one of these women, the original weight was unknown, leaving 100 women for analysis. Figure I Flowchart study population Study population 4 Intervention group N=113 Control group N=101 Excluded N=3 no initial weight measurement N=8 pregnant before starting Excluded N=1 no initial weight measurement For further analysis N=102 For further analysis N=100 Population characteristics Table I shows the baseline characteristics of the two groups. In the intervention group, women were younger (mean age 29 versus 32 years, p-value 0.015) and had a higher BMI (mean 37 versus 35 kg/m², p-value 0.001) compared to the control group. Also, in the intervention group, less women smoked (13% versus 28%, p-value 0.039), more women had an ovulation disorder (55% versus 37%, p-value 0.003) and more women declared consuming alcohol (30% versus 19%, p-value 0.001). Duration and type of subfertility, sperm count and parity were comparable between the groups. A median of seven counseling sessions were needed to motivate approximately two out of three participants to lose weight. Twenty-two participants (22%) dropped out of the program after one or two sessions. Three of them did lose 63

67 Chapter 4 weight. Fourteen women (14%) in the control group only attended an intake session, and it is unknown why they did not pursue treatment. Weight reduction Table II shows the weight reduction for both cohorts. The intervention group lost a mean of 4.3 kg compared to a mean weight gain of 0.5 kg in the control group (mean difference 4.8 kg, p-value <0.001). There were 33 (32%) women in the intervention group versus 25 (25%) women in the control group who lost more than 5% of body weight compared to their baseline weight (crude OR 1.40 (95% CI 0.73 to 2.7), p-value 0.312; adjusted OR 1.05 (95% CI 0.43 to 2.6), p-value 0.919). In the intervention group 72 women (71%) achieved any weight reduction (defined as 0.1kg) compared to 41 women (41%) in the control group (crude OR 3.1 (95% CI 1.7 to 5.7), p<0.001). This relation was still significant after further adjustment (adjusted OR 2.6 (95% CI 1.2 to 5.9), p-value 0.022). Table I Baseline characteristics of the study population Maternal Characteristics Intervention Group ( n=102) % missing Control Group ( n=100) % missing P** Maternal age (y), median (IQR*) 29.0 ( ) ( ) Weight at intake (kg), median ( ) ( ) (IQR*) Body mass index at intake (kg/ 37.3 ( ) ( ) m²), median (IQR*) Duration of subfertility at intake 25.0 ( ) ( ) (months), median (IQR*) Type of subfertility, n (%) Primary 64 (62.7) 56 (56.0) Secondary 36 (35.3) 44 (44.0) Type of cycle, n (%) Amenorrhea 10 (9.8) 3.0 (3.0) Oligomenorrhea 46 (45.1) 34 (34.0) Normal cycle 36 (35.3) 60 (60.0) Sperm count ( 10 6 /ml), median 33.4 ( ) ( ) (IQR*) <1 ( 10 6 /ml), n (%) 8 (7.8) 5 (5.0) 1-10 ( 10 6 /ml), n (%) 10 (9.8) 10 (10.0) >10 ( 10 6 /ml), n (%) 44 (43.1) 32 (32.0) Nulliparous, n (%) 56 (54.9) (59.0) Maternal smoking, n (%) 13 (12.7) (28.0) Maternal alcohol consumption, n (%) 31 (30.4) (19.0) * IQR: interquartile range, ** P-value based on χ 2 test for categorical or t-test for continuous variables 64

68 Effectiveness weight reduction program Table II Characteristics of weight reduction Intervention Group (n=102) Control Group (n=100) Sessions (n), median (IQR*) 7 (3-10) - - Participants with any weight reduction, n (%) 72 (70.6) 41 (41) <0.001 Average weight change (Kg), (Std. Error) -4.3 (0.8) +0.5 (1.0) <0.001 Average body mass index change (Kg/m²), (Std. Error) -1.4 (0.3) +0.2 (0.4) Participants with >5% weight reduction, n (%) 33 (32.4) 25 (25.0) * IQR: interquartile range P Pregnancies Table III shows the fertility outcomes. There were 55 (54%) women who conceived in the intervention group versus 46 (46%) in the control group. The number of ongoing pregnancies was 52 (51%) in the intervention group versus 41 (41%) in the control group, respectively (crude hazard rate ratio (HRR) 1.3 (95% CI 0.85 to 1.9), adjusted HRR 1.9 (95% CI 1.0 to 3.4)) (table IV). Live birth rates were 48 (47%) in the intervention group and 37 (37%) in the control group (crude HRR 1.3 (95%CI 0.86 to 2.0), adjusted HRR 2.0 (95% CI 1.1 to 3.8) (table IV)). 4 We divided the intervention and control group in ovulatory and anovulatory women and again calculated the HRR s for ongoing pregnancies and live birth rate (table IV). We found no statistical difference between these groups. 65

69 Chapter 4 Table III Comparisons of the outcomes of the groups Intervention Group Control Group ( n=102) ( n=100) P ART received, n (%) 48 (47.1) 79 (79) <0.001 Type of ART received <0.001 IVF/ICSI 23 (47.9) 52 (65.8) OI 12 (25.0) 5 (6.3) IUI 9 (18.8) 21 (26.6) Multiple 4 (8.3) 1 (1.3) Conceptions, n (%) 55 (53.9) 46 (46) <0.001 Spontaneous, n (%) 27 (49.1) 7 (15.2) ART assisted, n (%) 28 (50.9) 39 (84.8) EUG, n (%) 1 (1.0) 3 (3.0) - Miscarriage, n (%) 2 (2.0) 2 (2.0) - Termination of pregnancy, n (%) 1 (1.0) 1 (1.0) - Ongoing Pregnancies, n (%) 52 (51.0) 41(41.0) Singleton pregnancies, n (%) Multiple pregnancies, n (%) 51 (98.1) 35 (85.4) 1 (1.9) 6 (14.6) Live birth, n (%) 48 (47.1) 37 (37.0) Time period between intake to ART (months), mean(std. Error) Time period between intake to conception (months), mean(std. Error) 8.8 (1.4) 4.7 (0.8) (2.2) 13.2 (1.8)

70 Effectiveness weight reduction program Table IV Crude and adjusted hazard rate ratios (HRR) and 95% Confidence Intervals comparing outcomes of intervention group to control group Outcomes Crude model (95% CI) p Adjusted model (95% CI) Ongoing pregnancy 1.28 ( ) ( )* Live birth 1.32 ( ) ( )* p Ovulatory women Ongoing pregnancy 1.72 ( ) ( ) Live birth 1.80 ( ) ( ) Anovulatory women Ongoing pregnancy 0.89 ( ) ( ) Live birth 0.91 ( ) ( ) *Adjusted for duration of subfertility, type of subfertility, ART received, smoking, alcohol consumption, female age at intake, initial BMI at intake, type of cycle and sperm count (cat) Adjusted for all above-mentioned variables except for the type of cycle 4 In the intervention group one triplet pregnancy occurred after intrauterine insemination with mild ovarian hyperstimulation. In the control group six twin pregnancies occurred, all after IVF/ICSI treatment. The mean time to conception (spontaneous and assisted) did not differ between groups, mean 14.1 months (standard error (se) 2.2) in the intervention group versus 13.2 months (se 1.8) in the control group (p-value 0.77 (Table III)). Figure II shows the Kaplan Meier curves for live birth. Figures IIIa and IIIb show the cumulative pregnancies for both groups according to the mode of conception. Figure IVa and IVb show the amount of ongoing pregnancies per three months with or without ART. In the group of women that received ART one spontaneous pregnancy occurred in the intervention group versus five in the control group. Eight of the 55 pregnancies in the intervention group occurred in the 22 women that dropped out of the program. Of these three occurred spontaneously within six months, one after intra uterine insemination, and four more spontaneously in the four years follow up. Two women received ICSI treatment with a severe male factor causing the subfertility. 67

71 Chapter 4 Figure II Kaplan-Meier curves for live birth. Live Birth (Percentage) 1,0 0,8 0,6 0,4 Control Group Intervention Group Control Group-censored Intervention Groupcensored 0,2 0,0,00 10,00 20,00 30,00 40,00 50,00 Time to Conception (Months) Figures IIIa and b Cumulative ongoing pregnancy curve for the study population with special attention for the contribution per mode of conception. Note: these curves are an exact representation of the cumulative ongoing pregnancies, not Kaplan Meier curves. In the subgroup analysis in the intervention group no weight loss, weight loss equal to 2.9 kg, and weight loss equal to 7.5 kg were correspondent to the 25 th, 50 th, and 75 th percentiles, respectively. In these quartiles 11, 14, 13 and 14 ongoing pregnancies occurred, respectively. The difference was not statistically significant (p-value for trend=0.55). In the group of women with more than 7.5 kg weight loss, the ongoing pregnancy rate was not different from the other quartiles (p-value 0.862). 68

72 Effectiveness weight reduction program Figures IVa and b Amount of ongoing pregnancies per 3 months divided in ART and no ART for the intervention and control group ART treatment The number of couples using ART was 48 (47%) versus 79 (79%) in the intervention and control group (p-value<0.001) (table III). Of 48 treatments in the intervention group 23 (48%) were IVF/ICSI treatment, compared to 52 (66%) in the control group (p-value<0.001). The mean time to commencement of ART treatment for the intervention group was 8.8 months (se 1.4) versus 4.7 months (se 0.8) in the control group (p-value 0.014). 4 Discussion In this retrospective study we showed that with a WRP for subfertile obese women (BMI 32 kg/m²) an increased proportion of women lost weight as compared to the control group. There was no significant difference in the amount of women losing 5% of their initial weight. There were significantly more ongoing pregnancies and a higher live birth rate in the intervention group compared to the control group. We found no relationship between weight loss and pregnancy. Strengths and weaknesses The main limitation of this study is its non-randomised and retrospective character. Moreover, the comparison with the control group for weight reduction was difficult because most women in this group (65%) only had one weight reported prior to ART or pregnancy. To assess if there was weight gain or loss in the control group we retrieved second weights of these women from hospital files. The median time between the weights is however 25.6 months. This could lead to over- or underestimation of the actual weight before starting 69

73 Chapter 4 ART or pregnancy for the women with substantial weight gain or loss. The mean time up to fertility treatment was approximately five months in the control group hereby making large changes in weight not very likely. We realize that the methodology of this retrospective evaluation limits the scientific value of the results, as there will be information and selection bias. Baseline characteristics differ between groups with the control group being more than 2 years older with potentially a slightly higher risk of miscarriage and lower pregnancy chance. Also, in the control group more women smoked. BMI at inclusion, however, was lower in the control group with the body weight being self-reported by women. Self-reported weight is significantly less accurate than measured weight, with a lower weight reported, especially in obese individuals. [171] In our study this could implicate that at baseline the difference in BMI is less or even nil thereby making the two groups better comparable. Because of our retrospective design there were missing data. To account for these missings we used multiple imputation. This is an accepted method of reducing bias and increasing precision in analysis. [170] The treatment of obesity is challenging for both health care provider and patient. Most research on the effectiveness of treatment of obesity is performed on diets. When compared to conventional food all diets lead to weight loss on the short term. Most effective diet is a low calorie diet with approximately kg weight loss in one year. If cognitive behavior therapy is combined with dietary intervention, better results are achieved than information on lifestyle intervention alone. [172] One of the difficulties in trying to achieve weight loss is dropout during intervention. A median dropout rate of 24% was found in a systematic review on subfertile overweight and obese women trying to get pregnant, with lower pregnancy rates and less weight loss in the dropout group. [173] In our study the dropout rate was 22% which is comparable. Another problem of treating obesity is consolidation of the weight loss that has been achieved. If weight loss is >10% of initial weight most people (59%) can limit relapse to 2 kg in long-term research. [174] Patients that include physical exercise with diet therapies are more effective on long-term. [175] If cognitive 70

74 Effectiveness weight reduction program behavioral treatment is combined with dietary interventions better results on the short and long-term are achieved. [176] We chose the personal approach with a PC who was trained in motivational interviewing and behavioral treatment. The intervention had a tremendous effect on weight reduction, as it led 71% of the participants to lose weight. There was however no significant difference in the amount of women losing 5% of body weight. We chose this cutoff point because of the positive effect on fertility described in anovulatory subfertile women and also for the positive effect on general health, in addition to the feasibility of achieving this weight loss in 6 months. [167;177;178] The higher number of ongoing pregnancies at 12 weeks amenorrhea and higher live birth rate in the intervention group is possibly due to the weight reduction. Several other potential factors like waist hip ratio, exercise or diet were not recorded. Exercise for example led to higher live birth rates after IVF in a cohort study of obese subfertile women independent of body weight reduction. [179] We found no association between weight loss and pregnancy. This might be due to early conception in the study period, when women will not have lost weight. Indeed, most spontaneous pregnancies in the intervention group occur in the beginning of the study period (figure IVa). 4 We should keep in mind that the increased pregnancy rate in the intervention group might be partly driven by natural conceptions. As figure IVa demonstrates, the natural conception rate in couples in the intervention group is 25%, thus contributing half of all the conceptions occurring in that group. Although the impact of weight loss might play a role here, one should also consider the fact that ART was postponed in this group, thereby allowing natural conception to occur. Thus, the effect of a WRP or other lifestyle programs could partly be explained by the fact that natural conception plays a role here. Indeed, previous studies with a comparable design in a non-selected population showed a similar phenomenon. [ ] In conclusion, our results add to the scientific evidence asserting the potential benefit of a WRP in terms of weight reduction and (spontaneous) pregnancy chances in obese women with subfertility. This does not only increase pregnancy chances, but is also likely to improve general health. 71

75

76 Chapter 5 Pregnancy complications after weight loss in obese women Koning AMH, Zafarmand MH, Van den Dool G, de Lepper AM, Van Veen L, Mol BWJ Submitted for publication

77 Chapter 5 ABSTRACT Purpose To evaluate the effect of a weight reduction program (WRP) for subfertile obese women on the occurrence of pregnancy complications. Methods We performed a retrospective cohort study of pregnant women (BMI 32 kg/m 2 ) who conceived after a WRP or without such intervention. The intervention group comprised obese women participating in a WRP who conceived and had an ongoing pregnancy. The control group comprised obese women who conceived and had an ongoing pregnancy before implementation of the WRP. We compared the pregnancy complications hypertension, preeclampsia, gestational diabetes, premature birth, stillbirth and intra uterine fetal death between the two groups and assessed, in the intervention group, the relationship between weight loss and pregnancy complications. The analysis was done for singleton pregnancies only, as well as for the ongoing pregnancies including the multiple pregnancies. Results Among 102 women in the intervention group, and 100 women in the control group, ongoing pregnancy rates were 52 (51%) and 41 (41%), respectively, with singleton pregnancy rates of 51 (50%) and 35 (35%). Among women with a singleton pregnancy, the rate of pregnancy complications was 19 (37%) in the intervention group and 12 (34%) in the control group (relative risk (RR) 1.1 (95% CI 0.61 to 1.9), p-value 0.34). When multiple gestations were also considered, these rates were 20 (38%) and 18 (44%) respectively (RR 0.87 (95% CI 0.54 to 1.4, p-value 0.60)). The amount of weight loss in different quartiles had no impact on pregnancy complications (p-value for trend 0.74). Conclusions In subfertile obese women, a WRP does not reduce pregnancy complications. 74

78 Pregnancy complications Introduction Obesity in pregnancy is associated with an increased risk of pregnancy complications, for example gestational diabetes and hypertensive disorder [21;28], as well as compromised neonatal outcome, such as macrosomia and congenital malformations. [ ] In view of this, several authors support weight loss programs before starting assisted reproductive technology (ART). Some authors propose even to withhold ART to women who are severely obese. [186;187] Small observational studies have shown the restoration of ovulation and the occurrence of natural pregnancies in subfertile obese women who managed to lose weight. [161;165;166;188] In pregnant women who are overweight or obese, a recent review of 44 randomised controlled trials reporting on 7,278 women, showed that dietary and lifestyle interventions in pregnancy can reduce maternal gestational weight gain and improve outcomes for both mother and baby. [45] Among the interventions, those based on diet are the most effective and are associated with reductions in maternal gestational weight gain and improved obstetric outcomes. However, a recent large randomized study evaluating antenatal lifestyle advice in women who were overweight or obese, showed neither a reduction in delivering a baby weighing above the 90th centile nor improve maternal pregnancy and birth outcomes. [46] 5 With respect to studies in subfertile women, we were not aware of any results on pregnancy outcome in relation to lifestyle interventions, specifically on weight loss. There are however several studies pending on the subject. [153;189;190] In view of the growing evidence of the negative impact of obesity on fertility, we started a weight reduction program (WRP) in our fertility clinic. We previously reported an increased pregnancy rate and reduced use of ART after introduction of our WRP (chapter 4). Here, we want to evaluate if the WRP also led to a decrease in pregnancy complications. 75

79 Chapter 5 Methods Population We performed a retrospective cohort study among obese subfertile women who came for fertility treatment between January 2000 and December The study was performed in the fertility clinic of the Albert Schweitzer Hospital in the Netherlands. Details of the study design have been described in chapter 4 of this thesis. Subfertility was defined as unprotected intercourse for 12 months without conception. Obesity was defined as BMI 32 kg/m². We compared the impact of a WRP intervention versus usual care on pregnancy complications in first ongoing singleton pregnancies and for all first ongoing pregnancies including the multiple gestations. Ethics This retrospective study followed the Declaration of Helsinki (2008) on medical protocol and ethics concerning the data collection and analysis and publication of the results. WRP intervention From May 2006 onwards, obese women were proposed to participate in a lifestyle intervention. The aim of the WRP was to achieve weight loss in six months prior to commencement of ART. The intervention group was formed by women who conceived of an ongoing pregnancy that occurred after having had the WRP. The WRP intervention has been described earlier (chapter 4). In short, it consisted of dietary advice and face to face counselling sessions with a nurse trained in motivational interviewing. Control group We compared the course of pregnancy of these women with that of obese women who had an ongoing pregnancy after visiting the fertility clinic from January 2000 up to December 2005, before implementation of the WRP (control group). In the control group, couples were counselled about the impact of obesity on fertility and pregnancy and advised to lose weight. 76

80 Pregnancy complications For participants in the intervention group, body weight was assessed by the personal coach, whereas body length was self-reported. In the control group, length and body weight were self-reported by the patients. For both groups, BMI was subsequently calculated by the authors. Outcome measurements The following pregnancy complications were studied. Gestational diabetes (GDM) was defined as an abnormal result of a glucose day curve (fasting glucose 6.0 or 2 hours postprandial 7.0 mmol/l). Pregnancy induced hypertension (PIH) was defined as at least two measurements of a systolic blood pressure 140 mmhg or diastolic blood pressure 90 mmhg. Preeclampsia (PE) was defined as PIH with proteinuria ( 0.3 gr/24 hours). Premature birth (spontaneous or iatrogenic) was defined as delivery before 37 weeks of gestation. Stillbirth was defined as death at delivery or within 24 hours after 24 weeks of gestation. Intra uterine fetal death was defined as fetal demise after 16 weeks gestation. We studied these endpoints as a composite for any pregnancy complication, as well as individual endpoints. Other endpoints were birth weight, being large for gestational age (LGA) defined as birth weight > 90 th percentile, and being small for gestational age (SGA) defined as birth weight < 10 th percentile. Furthermore the following delivery complications were compared: hemorrhage postpartum defined as 1000 ml blood loss in the first 24 hours after childbirth, shoulder dystocia, and mode of delivery (spontaneous, instrumental vaginal delivery, cesarean sections). We also reported on 1 and 5 minutes Apgar scores and congenital anomalies. 5 If a patient had more than one pregnancy during the study period, only the first was used for analysis. The data were collected from the pregnancy charts of patients on pregnancy and childbirth. Finally, we assessed the relation between weight loss and pregnancy complications. To do so, we divided the intervention group in four categories based on percentiles weight loss, i.e. the quartile with the weight loss >75 th %, with a weight loss between 50 th and 75 th %, between 25 th and 50 th %, and less than 25 th %. 77

81 Chapter 5 Statistical analysis Baseline characteristics from the two groups were tabularized and compared for statistical differences. Dichotomous variables were compared using the Chi-Square test or the Fisher s exact test (when the number of observations were less than 5 in the 2 by 2 tables), while continuous variables were compared using the Student s t-test. Descriptive results are presented as frequencies with row numbers and percentages or medians with interquartile ranges (IQRs). To prevent bias associated with missing data, we used multiple imputations for covariates with missing values on the basis of the correlation of missing variables with other participant characteristics. Although five imputed datasets have been suggested to be sufficient theoretically to reduce sampling variability from the imputation process, [169] we imputed 20 datasets as has been recommended recently.[170] In addition, we added variables related to covariates as predictors to the imputation model to increase the plausibility of the missing-at-random assumption. The amount of missing values ranged from 1% to 31% (for maternal alcohol consumption). We report the pooled results of the analyses performed in each of the 20 imputed datasets. We compared the composite pregnancy complication outcome as well as the individual pregnancy complications (as dichotomous variables) between the intervention and the control group using the Chi-Square test or Fisher s exact test. We also compared the secondary pregnancy outcomes between the two groups. Finally, we assessed the relation between weight loss and ongoing pregnancy. To do so, we divided the intervention group in four categories based on quartiles of weight loss, i.e. the quartile with the weight loss >75 th %, with a weight loss between 75 th and 50 th %, between 50 th and 25 th %, and less than 25 th %. All analyses have been done in all women with ongoing pregnancy and those with a singleton pregnancy separately. P values were 2-sided, and a p-value < 0.05 was supposed to indicate statistical significance. All statistical analyses were performed using SPSS version 21.0 (IBM Corp., Armonk, NY, USA). 78

82 Pregnancy complications Results Population characteristics Out of 102 women in the WRP group, and 100 women in the control group, there were 52 women who had an ongoing pregnancy in the intervention group (51%), and 41 women in the control group (41%). Table I shows the baseline characteristics of the women who had an ongoing pregnancy. In the intervention group less women smoked (6% versus 32%), more women consumed alcohol (27% versus 20%), less women received ART (52% versus 95%) and less multiple gestations were present (2% versus 15%) with respect to the women in the control group. Age, BMI and parity were comparable between groups. Table I Baseline characteristics of the study population Maternal Characteristics Intervention Group (n=52) % missing Control Group (n=41) % missing P* Maternal age (y), median (IQR) 28.1 ( ) ( ) Weight at intake (kg), median ( ) ( ) (IQR) Body mass index at intake (kg/ 35.8 ( ) ( ) m2), median (IQR) Nulliparous, n (%) 33 (63.5) (68.3) Maternal smoking, n (%) 3 (5.8) (31.7) Maternal alcohol consumption, 14 (26.9) (19.5) n (%) ART received, n (%) 27 (51.9) 0 39 (95.1) 0 <0.001 Type of ART received 0 0 <0.001 IVF/ICSI 11 (21.2) 24 (58.6) OI 9 (17.3) 4 (9.8) IUI 5 (9.6) 11 (26.8) Multiple 2 (3.8) 0 (0) Multiple pregnancies, n (%) 1 (1.9) 0 6 (14.6) ^ IQR: interquartile range, * P-value based on χ2 test for categorical or t-test for continuous variables, ^ P-value based on the Fisher s exact test 5 Outcome in singleton pregnancies Table II shows the number of pregnancy complications for the singleton pregnancies in each group. Overall, there were 19/51 (37%) women with a complication in the intervention group versus 12/35 (34%) in the control group (RR 1.1 (95% CI 0.61 to 1.9)). A trend towards less preeclampsia was seen in the 79

83 Chapter 5 intervention group (8% versus 23%, RR 0.34 (95% CI 0.11 to 1.1)). The other pregnancy complications were comparable between groups. One intra uterine fetal death occurred in the intervention group at 20 weeks gestation. Table II Primary outcomes Intervention Group Singleton pregnancies Control Group (n=51) (n=35) P* Weight at intake pregnancy (kg), median ( ) ( ) (IQR) BMI at intake pregnancy (kg/m2), median 36.1 ( ) 36.6 ( ) (IQR) Gestational weight gain (kg), median 9.5 ( ) 9.9 ( ) (IQR) Live birth, n (%) 47 (92.2) 32 (91.4) Gestational age at birth (weeks), median 39.4 ( ) 39.1 ( ) (IQR) Complications of pregnancy, n (%) 19 (37.3) 12 (34.3) Pregnancy induced hypertension, 8 (15.7) 3 (8.6) 0.513ˆ n (%) Preeclampsia, n (%) 4 (7.8) 8 (22.9) 0.062ˆ Gestational Diabetes Mellitus, 10 (19.6) 6 (17.1) n (%) Premature Birth, n (%) 3 (5.9) 4 (11.4) 0.435ˆ Stillbirth 0 (0.0) 0 (0.0) - Intrauterine fetal death 1 (2.0) 0 (0.0) 1.0ˆ IQR: interquartile range, * P-value based on χ2 test for categorical or t-test for continuous variables, ^ P-value based on the Fisher s exact test Table III shows the delivery complications and neonatal outcomes in the women with live birth of a singleton. Mean birth weight in the intervention group was higher than in the control group (3601 versus 3338 gram, p-value 0.03). In the intervention group, there were no SGA children versus five (16%) in the control group (p-value 0.009). We found no differences in LGA children, mode of delivery, hemorrhage postpartum, shoulder dystocia, Apgar scores or congenital anomalies. 80

84 Pregnancy complications Table III Delivery complications and neonatal outcomes Intervention Group (n=47) Singleton pregnancies Control Group (n=32) p* Birthweight (g), mean (Std. Error) 3,601 (74.0) 3,338 (97.7) SGA, n (%) 0 (0.0) 5 (15.6) 0.009ˆ LGA, n (%) 14 (29.8) 8 (25.0) Apgar scores, median (IQR) 1 minute 9 (8-9) 9 (8-9) minutes 10 (9-10) 10 (9-10) Congenital anomalies, n (%) 0 (0.0) 1 (3.1) 0.353ˆ Mode of delivery Spontaneous, n (%) 30 (63.8) 16 (50.0) Instrumental vaginal, n (%) 7 (14.9) 5 (15.6) Cesarian section, n (%) 10 (21.3) 11 (34.4) Hemorrhage postpartum, n (%) 4 (8.5) 3 (9.4) 1.0 ˆ Shoulder dystocia, n (%) 1 (2.1) 0 (0.0) 1.0 ˆ IQR: interquartile range, * P-value based on χ2 test for categorical or t-test for continuous variables, ^ P-value based on the Fisher s exact test 5 Among singleton pregnancies, in the intervention group the 25th, 50th, and 75th percentiles corresponded with no weight loss, weight loss equal to 3.2 kg, and weight loss equal to 7.8 kg, respectively. In these quartiles, any pregnancy complication occurred in 5, 5, 5 and 4 women, respectively (p-value for trend=0.74). In women with more than 7.8 kg weight loss, the ongoing pregnancy rate was not different from the other quartiles (p-value=0.58). Outcome in pregnancies including multiple gestations When we considered also the women with a multiple gestation the following results occurred. The composite outcome for any pregnancy complication occurred in 20/52 women (39%) in the intervention group versus 18/41 women (44%) in the control group (RR 0.87 (95% CI 0.54 to 1.4, p- value 0.60). Significantly less preeclampsia (RR 0.34 (95% CI 0.13 to 0.87), p-value 0.017) was found in the intervention group. The other pregnancy complications were not different between groups. Considering the secondary outcomes, the mean 81

85 Chapter 5 birthweight was 3561 grams in the intervention group versus 3190 grams in the control group (p-value 0.008). There were no SGA children in the intervention group versus six in the control group (p-value 0.005). Otherwise we observed no differences between the groups for LGA, Apgar scores, congenital anomalies, mode of delivery, hemorrhage postpartum or shoulder dystocia. Among all pregnancies including the multiple gestations, in the intervention group the 25th, 50th, and 75th percentiles were correspondent to no weight loss, weight loss equal to 3.3 kg, and weight loss equal to 7.7 kg, respectively. In these quartiles 5, 5, 6 and 4 composite pregnancy complications occurred, respectively. The difference was not statistically significant (p-value for trend=0.80). In the group of women with more than 7.7 kg weight loss, the ongoing pregnancy rate was not different from the other quartiles (p-value=0.51). Discussion In this retrospective study we found no differences in pregnancy complications between obese subfertile women with an ongoing singleton pregnancy after a WRP versus obese women without such intervention. In the WRP group, the number of SGA infants was limited to zero, while in the control group this happened in 16%. Also, a trend towards more preeclampsia was found in the control group. There are several limitations to our research. First, our study was retrospective, thus being potentially affected by selection and information bias. Furthermore, the small number of women limits the power of our study. We aimed to evaluate if a WRP for obese subfertile women was effective, first for improving fertility outcome, but also to influence the course and outcome of pregnancy. In the intervention group no SGA infants were born. This could be a positive effect of the WRP, although several other factors associated with SGA were different between groups. First, there was a trend towards less preeclampsia in the intervention group compared to the control group (8% versus 23%, RR

86 Pregnancy complications (95% CI 0.11 to 1.1), p-value 0.06). This could potentially be a positive effect of the WRP. Also, less women smoked in the intervention group, which obviously affected their chance of SGA. [191] Finally, IVF/ICSI might be associated with SGA. [192] In the control group, more women conceived after an IVF/ICSI procedure compared to the intervention group. The multiple gestations were not equally divided between the two groups with one triplet pregnancy in the intervention group versus six twin pregnancies in the control group (p-value 0.045). This is due to a higher rate of ART in the control group. Because of the pregnancy complications that are associated with multiple gestation such as hypertensive disorders, premature labor as well as cesarean delivery and SGA, [193;194] we analyzed the group for singletons only and including the multiple pregnancies. The composite outcome for pregnancy complications remained the same with inclusion of the multiple gestations, but more preeclampsia was seen in the control group as expected. Women pregnant after ART have a higher chance of obstetric complications and associated worse neonatal outcome than women with spontaneously conceived pregnancies. [192;195;196] The women in our intervention group participated six months in a WRP prior to starting ART. This provides a double advantage: firstly two out of three women will change their lifestyle with weight reduction as a result and secondly they will not receive ART in this timespan and thus - as we have shown in chapter 4- have a higher spontaneous conception rate and lower need for ART. 5 We hypothesized that the weight loss achieved by our WRP could lead to a decrease in weight related pregnancy complications. In women following bariatric surgery, with a substantial amount of weight loss, this effect has been demonstrated. [197] Although, women in the intervention group lost a substantial amount of weight, their BMI at pregnancy intake was comparable to that of women in the control group (table II), while at the start of intervention, their BMI had been higher than the BMI of women in the control group (35.8 kg/m² vs 34.5 kg/m²; p-value 0.051). They possibly regained weight between the intervention and pregnancy, or regained weight in the first trimester of pregnancy. Despite weight loss, their mean BMI 36 kg/m² at pregnancy intake was far above 30 kg/m² which is associated with an increased risk of pregnancy complications. 83

87 Chapter 5 Consolidation of achieved weight loss is one of the challenges in treating obesity. Face-to-face follow up in the period after weight loss, is superior in maintaining weight loss than internet based information or newsletters. [198] Therefore, although in terms of subfertility and general health there is substantial benefit with the weight loss after WRP, the prevention of pregnancy complications would require either a stronger decrease of the BMI, or a continuation of the WRP beyond six months. The question is how feasible this is. Also a prolonged program would require higher costs, thus jeopardizing its cost effectiveness. There is an abundance of literature supporting an association between excessive gestational weight gain and pregnancy complications such as gestational diabetes and hypertensive disorders. Rasmussen et al. state however that the evidence is inconclusive because of inconsistent results and methodological flaws. [199] In our study, there was no difference in gestational weight gain with a mean of 9.5 kg weight gain in the intervention group versus 9.9 kg in the control group (p-value 0.57). A review of literature showed that in obese subfertile women, there was a high dropout rate in women receiving lifestyle intervention of 24% and within the dropout group there was substantially less weight loss. [173] In our intervention group the drop out rate was 22% within three sessions with the personal coach. In conclusion, we found no association between participation in a WRP and pregnancy complications in a subsequent pregnancy. Our results therefore do not support the idea that a WRP leads to less pregnancy complications. Because of the retrospective nature of our study and the small number of women more prospective studies are needed to confirm this finding. 84

88

89

90 Chapter 6 The long-term follow up of children born after a weight reduction program in obese, subfertile women Koning AMH, Van den Dool G, Barendregt S, Van Marrewijk I, Mol BWJ Research letter

91 Chapter 6 Children born from obese mothers have a higher risk of being macrosomic than children from women with normal weight. [200;201] Macrosomia at birth increases the risk of becoming overweight or obese later in life as well as developing the metabolic syndrome with increased cardiovascular risk and diabetes mellitus. [200] This association is partly associated with the prenatal environment in utero. [202] High maternal prepregnancy BMI is a probable early marker of adult obesity. [203] The number of overweight and obese adults globally is 1.5 billion, with 170 million children being overweight or obese. [156;204] Hypothetically preconceptional weight loss in obese women could lead to a decrease in childhood obesity. Here we want to evaluate the long-term outcome in terms of overweight and obesity for the children born after our lifestyle intervention in a group of subfertile obese women. We performed a retrospective cohort study determining the effect of a weight reduction program (WRP) in a general hospital in the Netherlands among obese subfertile women who came for fertility treatment between January 2000 and December 2010 (chapter 4). The intervention group (2005 to 2010, n=102) participated in a WRP for six months prior to assisted reproductive technology (ART), the control group (2000 to 2005, n=100) was advised to lose weight and got ART without delay. The amount of women who lost more than 5% of body weight compared to their baseline weight was 33 (32%) in the intervention group versus 25 (25%) in the control group (crude odds ratio (OR) 1.4 (95% CI 0.73 to 2.7). The live birth rates were significantly higher in the intervention group (n=48 (47%) versus n=37 (37%)) with an adjusted HRR 2.0 (95% CI 1.1 to 3.8). Four years after the end of the study period, we assessed the growth of the singleton children in these two groups (n=79). We tried to contact the women by telephone and asked the data on body weight and height that were taken at the Youth Health Care (YHC) center at 3, 6, 14, 24, 36 and 45 months of age of their child. The body mass index (BMI) was calculated by the authors for 24, 36 and 45 months. Data was obtained for 42 children. We were unable to reach 17 women, 11 women denied participation, one women had died of an ischemic stroke and eight women had lost the booklets of the YHC. We compared the mean BMIs of the children (table I). In the intervention group 88

92 Follow up children 5 (21%) large for gestational age (LGA) were born versus 3 (17%) in the control group (p-value 0.56). Overall, in the intervention group 5 (21%) children were overweight or obese according to the age and gender specific international standards versus 4 (22%) in the control group (p-value 0.60). [205] Figure I graphically shows the mean weights until the age of four years. In this follow up study of a retrospective cohort of subfertile obese women we did not find a difference in mean weights or BMI in early childhood between the two groups. Weight loss prior to pregnancy is widely propagated for obese women, based on the argument that it achieves healthier pregnancies and children. Our data do not support the idea that the positive impact of (mild) weight loss on fertility in obese women also leads to better outcome for their offspring on the long-term. Table I Intervention Group Control Group (n=24) (n=18) P Birthweight (grams), mean (SD) 3597 (427) 3365 (555) 0.14 LGA, n (%) 5 (21%) 3 (17%) 0.56 Overweight or obese, n (%) 5 (21%) 4 (22%) 0.60 BMI at 24 mnths (kg/m²), mean (SD) 16.7 (1.1) 16.5 (1.2) 0.54 BMI at 36 mnths, (kg/m²), mean (SD) 16.3 (1.6) 16.5 (1.3) 0.77 BMI at 45 mnths, (kg/m²), mean (SD) 16.8 (1.3) 16.4 (1.5) SD = standard deviation, comparison using the Fisher s exact test 89

93 Chapter 6 Figure I Mean weights up to 45 months Weight at 3 months (g) Weight at 6 months (g) Weight at 14 months (g) Weight at 24 months (g) Weight at 36 months (g) Weight at 45 months (g) Weight at 3 months (g) Weight at 6 months (g) Weight at 14 months (g) Weight at 24 months (g) Weight at 36 months (g) Weight at 45 months (g) Mean intervention group intervention control group Error Bars: 95% CI 90

94

95

96 Chapter 7 Obesity: argument for withholding fertility treatment? Koning AMH, Mol BWJ, Dondorp WJ [translated from Dutch] Ned Tijdschr Geneeskd 2014;158:A7258

97 Chapter 7 ABSTRACT Obesity can lead to anovulation and subfertility. Around the world fertility treatment is withheld from women above a certain BMI, ranging from 25 to 40 kg/m². The proponents of this policy use three different arguments to justify their restrictions: risks for the woman, health and wellbeing of the future child, and importance for society. In this article we critically appraise these arguments. In conclusion, we think obese women should be informed about the consequences of their weight on fertility and pregnancy complications and encouraged to lose weight. If, however, a woman is unable to lose weight despite effort, we feel there is no argument to withhold treatment from her. This would be unjustified with respect to the treatment of other women with a high risk of complications. 94

98 Withholding treatment Numerous people have the wish to start a family. This will be no different for women with obesity (BMI 30 kg/m²) than for women with a normal weight. Obese women have a higher chance of subfertility than women with a healthy weight. Furthermore obese women are anovulatory more often and subfertile obese women with a regular cycle have a lower chance of a spontaneous pregnancy.[9] For these reasons obese women will more often need assistance to conceive. In the Netherlands and abroad there are different fertility clinics which have a BMI-limit for withholding treatment. In the Netherlands this limit usually lies around BMI 35 kg/m², and varies worldwide from kg/m². The exclusion of a specific group of subfertile women of fertility treatment that is available for others, asks for explicit justification on the grounds of valid arguments. In this article we will check if the arguments used for a BMI-limit are tenable. Looking at the considerations of the advocates of a BMI-limit in further detail, there are 3 kinds: risks for the woman, health and wellbeing of the future child and the consequences for society. Hereunder we discuss these arguments. Risks for the woman When using fertility treatment a woman is exposed to the accompanying risks. In obese women the risks of treatment are not per se increased, [206] but the risks of pregnancy complications are. The discussion will then be about these risks. According to the numerous articles on pregnancy complications, i.e. hypertensive disorders, gestational diabetes and cesarean section, the risk of such complications is indeed increased in obese women and increases with every BMI-class. [27] For example a woman with overweight has an almost 2 times higher risk of preeclampsia than a woman with a normal weight, [21] which means based on highest prevalence one in every five overweight women gets preeclampsia during pregnancy. For women with a BMI > 35 kg/m² this is one in every four women. Question is however what this means. First of all, a higher risk than the mean IVF population doesn t mean that it is irresponsible to take that risk. It is a question of proportionality: a higher risk can still be acceptable in light of the gain a woman can expect from treatment. Through the same reasoning IVF is thought acceptable with other women who 7 95

99 Chapter 7 are at increased risk of pregnancy complications because of medical conditions. Women with diabetes mellitus have an increased risk of hypertensive disorders and congenital abnormalities, macrosomia, stillbirth and premature labour. [207] Diabetes mellitus is however not an exclusion criterion for fertility treatment. Second of all it should be open for discussion who is to decide about this. It is without question that a physician cannot put her patient at risks that are disproportional, even not at her request. But in cases where there is at least discussion conceivable, we can defend that a competent and well informed woman in principle has the right to her own deliberation when considering risks taken for herself in realising her child wish. Not allowing them this would be unjustified paternalism. Risks for the child Children of obese mothers are at increased risk of labour complications and perinatal mortality. Maternal obesity is associated with a significant higher risk on several congenital malformations (oddsratio (OR): ; absolute risk: circa 1.2%), including neural tube defects, cardiovascular risks and stillbirth (OR: 2.1, 95% CI: 1.2 to 3.6).[207] Furthermore, pregnancies complicated by preeclampsia result more often in premature birth with related morbidity. In addition there is a positive relation found between higher maternal BMI and the chance of a child getting overweight themselves. Also children from obese mothers have a higher risk on other illnesses and disorders associated with the metabolic syndrome (hypertension, dyslipidemia and glucose-intolerance). There is general agreement that caregivers in fertility, because of their causal and intentional involvement in realising the child wish of a woman or couple, should in their consideration of treatment keep in mind not only the interest of the help seeker(s) but also the wellbeing of the future child. In the Netherlands a point of view has been formulated by the occupational group. [208] Although there are different visions in daily practice about when this is at hand, it is emphasized that rejection of treatment on grounds of the welfare of the child can only be considered in exceptional circumstances, actually when there is great risk of serious harm. [209] Despite aforementioned high relative risks it cannot be maintained that based on this criterion fertility treatment must be withheld from women with obesity. 96

100 Withholding treatment Consequences for society Society has an interest in restraining the costs involved in compensation of fertility treatments. Could this possibly be a reason to at least restrict the reimbursement of fertility treatment to a BMI value? In our simulation study of subfertile ovulatory and anovulatory women the costs per live birth were respectively 77 and 100% higher for obese women compared to normal weight women. [210] The largest proportion of extra costs were derived from pregnancy complications which as mentioned before occur more in women with a BMI > 30 kg/m². Besides this there is a slightly lower success rate with obese women (OR: ). Because of the higher risk of pregnancy complications and the higher costs per live birth, society could save money with restricting reimbursement of the costs of fertility treatment to a certain BMI. There are other priorities in health care besides fertility treatment and therefore it is inevitable that society draws a line somewhere. In the Netherlands only 3 IVF treatments are reimbursed for example up until the age of 43. The first restriction affects all IVF patients roughly in the same manner, and this age restriction is justified in light of the deteriorating success levels beyond the 40th life year. A BMI limit however is of another category because then a potentially well treatable patient group is excluded from reimbursement. An argument used for this policy is limited resources should be used to maximum effectiveness. [211] Question is however why women with an additional health problem do not have the same right to aid to conceive as women that are just subfertile and who s treatment will be less expensive because of that. Further so, the ones who advocate excluding treatment above a certain BMI level seem to use this instrument rather selectively. Excluding women with other comorbidities is not called for. Overweight and lifestyle We conclude that none of three arguments that are used in favour of a BMIlimit are convincing. Against all three the objection is that it excludes a specific patient category on grounds that are not alleged against treatment of others with comparable risks. Or did we overlook a difference which suggests that no equal cases are at hand here? You could introduce the argument that to a certain point obesity is a changeable condition, while for example diabetes mellitus or age are not. If potential risk for mother and future child together with the costs for society can be prevented by first aiming at weight loss, then it is evident 7 97

101 Chapter 7 that healthcare providers should advise this. There are some observational studies that suggest a positive relationship between weight loss and fertility. In the Netherlands a randomised controlled trial has been conducted past years looking at the effect of lifestyle intervention on the chances of live birth in obese subfertile women. If the results confirm it is useful and cost-effective to start with lifestyle intervention in obese women, than this should be policy from now on. This advice should not be without obligations: one should expect the women to make a serious attempt to lose weight. [97] You cannot assume however that it is achievable to lose weight for everybody. Dropout is a considerable problem with lifestyle intervention, with less or no weight loss as a consequence.[173] As noted by the Task Force Ethics and Law of the European Society of Human Reproduction and Embryology (ESHRE) in 2010, women that have attempted weight loss without success should not be excluded from fertility treatment. [97] As we have shown there is no justification for this policy. Conclusion Obesity is a cause of subfertility and pregnancy complications. Possibly lifestyle intervention can change this. A consultation at the fertility clinic should be considered an opportunity to inform patients about this and offer lifestyle intervention. But if weight loss is not achieved, this should not automatically shut the door to treatment for women with a weight above a certain BMI. Looking at the risks and costs that are considered acceptable for other fertility patients this would be unjustified. 98

102

103

104 Chapter 8 Summary and General discussion

105 Chapter 8 Overweight and obesity are increasing worldwide. This has major adverse consequences for health in general and fertility in women in particular. With the increasing number of women in reproductive age being obese, there is also an increasing need for fertility treatment. And with more pregnant women being obese, the amount of pregnancy complications will also continue to increase. In the USA, women with a body mass index (BMI) 35 kg/m² most frequently sought medical attention to become pregnant, but received surgical fertilityrelated services least frequent. [34] Assisted reproductive technology (ART) is restricted worldwide beyond a certain BMI, ranging from 25 to 40 kg/m². [32] In the Netherlands several fertility centres have a BMI limit, mostly around kg/m². [ ] This restriction is based on different arguments from safety for the woman to risk for the future child. This thesis studies several aspects of reproductive problems in overweight and obese women. We address the (cost) effectiveness and safety of fertility treatment in overweight and obese women, the effectiveness of lifestyle intervention in obese women who suffer from subfertility, and we discuss ethical issues regarding fertility treatment in obese women. In chapter 2 we used decision analysis and economic modelling to assess the costs of a live birth after ART in normal weight, overweight and obese women. We found that in our hypothetical cohort of 1,000 anovulatory subfertile obese women live birth was decreased by 15% (from 806 to 687 live births) compared to women with normal weight. For ovulatory women, live birth rate was decreased by 24% (from 698 to 531 live births). In parallel the number of complications increased. Associated costs were also higher: the cost per live birth for anovulatory obese women was 100% higher than for their normal weight counterparts. For ovulatory women, these costs were 70% higher. ART in obese women is less effective and their pregnancies are more costly than in women with normal weight. In chapter 3 we asked ourselves the question if fertility treatment in overweight and obese women has different complication rates than in women with normal weight and what the effectiveness of fertility treatment is. We performed a systematic review on complications of ART in relation to body weight. We identified 14 studies that reported on the relationship between overweight 102

106 General discussion and complications during or after ART. None of the individual studies found a positive association between overweight and ART complications. The pooled odds ratios (OR) for overweight versus normal weight for ovarian hyperstimulation syndrome, multiple pregnancy and ectopic pregnancy were 1.0 (95% confidence interval (CI) 0.77 to 1.3), 0.97 (95% CI 0.91 to 1.04) and 0.96 (95% CI 0.54 to 1.7), respectively. In 27 studies that reported on BMI and the success of ART, the pooled ORs for overweight versus normal weight on live birth, ongoing and clinical pregnancy following ART were OR 0.90 (95% CI 0.82 to 1.0), 1.01 (95% CI 0.75 to 1.4) and OR 0.94 (95% CI 0.69 to 1.3), respectively. These data show that both in terms of safety and in terms of effectiveness, the impact of overweight and obesity is very limited. In chapter 4 we evaluated the effectiveness of a six months weight reduction program (WRP) for obese subfertile women in a retrospective cohort study. The mean weight reduction was 4.3 kg in the intervention group and a mean weight gain of 0.5 kg in the control group compared to the baseline values at the start of treatment (p-value <0.001). There were 33 (32%) women in the intervention group versus 25 (25%) women in the control group who lost more than 5% of body weight compared to baseline (crude odds ratio (OR) 1.40 (95% CI 0.73 to 2.7). The ongoing pregnancy rates were 52 (51%) and 41 (41%) (adjusted hazard rate ratio (HRR) 1.9 (95% CI 1.0 to 3.4)) while live birth rates were 48 (47%) and 37 (37%) (adjusted HRR 2.0 (95% CI 1.1 to 3.8) for the intervention group and control group, respectively. In our study there is potential benefit from a WRP in obese subfertile women. In chapter 5 we compared pregnancy outcome in these women. The composite outcome of pregnancy complications was not different between groups: 19 (37%) women in the intervention group and 12 (34%) women in the control group had at least one pregnancy complication (p-value 0.34). We found no relationship between weight loss and pregnancy complications. Our data do not support a positive effect of lifestyle intervention on the occurrence of pregnancy complications. In chapter 6 we reported on the long term follow up of the singleton children from these women. The mean weights and BMI s in early childhood are not different between the groups. 8 Chapter 7 discusses the three different arguments used for a BMI limit for access to ART. First, the argument of safety for the woman. Increased risks of pregnancy complications in obese women are evident but these are not 103

107 Chapter 8 different than in other women with associated disease that are treated with ART and also not disproportional. Second, risks for the future child. There are increased risks of congenital anomalies and premature birth. But if treatment is rejected on grounds of the welfare of the child this can only be considered in exceptional circumstances, such as when there is great risk of serious harm. Third, the cost burden of treatment. It appears that this argument is used selectively while other groups with an additional health problem do get reimbursed for their fertility treatment. With the potential gain of lifestyle intervention in obese subfertile women this should be offered. If weight loss is not achieved however, this should not automatically shut the door to treatment for women with weight above a certain BMI. Looking at the risks and costs that are considered acceptable for other fertility patients this would be unjustified. Obesity: result of modern society Obesity is considered a chronic disease by the World Health Organisation (WHO), this implies that it is incurable. For a long time it s development was considered as nature and not nurture. The Barker thrifty hypothesis changed this view as it set out a framework for the importance of the early foetal life environment and the programming of our bodies in terms of metabolism and growth. [215] More recently Armitage et al. proposed the developmental overnutrition hypothesis. [202] This states that high maternal glucose, free fatty acid and amino acid concentrations result in permanent changes in appetite control, neuroendocrine functioning and/or energy metabolism in the developing foetus, thus leading to risk of adiposity. Obviously, today s society with access to abundant food for the majority of people plays a substantial role in the arising of this disease. With the knowledge that treating obesity is difficult, prevention should be the cornerstone, preferably starting at childhood. Several programs in the Netherlands have been initiated to inform people and give advice on a healthy lifestyle. With commercials on national television and the use of social media several national campaigns have been launched. For example drink water to promote drinking of water, healthy sports club to advise clubs on healthy alternatives in the cafetaria and healthy school cafetaria with advice on recommended healthy food and drinks. Besides these informative measures other directive methods are used by governments to fight obesity. 104

108 General discussion It has been well established that drinking sugar sweetened beverages leads to weight gain and obesity. [216] To fight obesity consumption of these drinks can be effectively reduced by raising their prices via taxation. This was for instance shown in a recent British modelling study. [217] Also, in the Netherlands it was demonstrated in an experimental environment that sugar sweetened drink purchases were reduced substantially after raising value added tax from 6 to 19%. [218] In the same manner, reduction of alcohol and tobacco consumption by taxation has been very effective. [219;220] Hence, there are excellent reasons to prevent obesity in a similar way. Another method which is advised by different organisations is the labelling of food with a traffic light method. This is a very simple and clear way to see at a glance if the product contains too much sugar, salt or fat. The food industry is trying to prevent the introduction of these forms of legislation and regulation. Their main goal will mostly remain maximising profit. The lobby they use is by some compared to the tobacco industry lobby because it misleads and hides negative data. [221]They pronounce, for example, that there is no unhealthy food but only an unhealthy diet, or claim that obesity is not because of excessive food but because of a lack of physical activity. [222]Children are an important part of their marketing, they are using packaging, television commercials, websites and apps to promote their product, sometimes aggressively. Children are cognitively incapable of appreciating the commercial purpose of television advertising and are particularly vulnerable to its persuasive techniques. [223] The food industry has resisted legislation and adopted self-regulation of children s marketing. In a systematic review on the effects of self-regulation a heterogeneous set of results is found. Surveys reported in papers in peer-reviewed journals provide ample evidence of continuing high levels of promotion of less healthy food products and high levels of exposure of children to this type of promotion. Significant reductions were absent in many locations except in response to statutory regulation. Industry-sponsored reports however, indicate a remarkable reduction in the promotion of unhealthy products and children s exposure. [224] Reports from a variety of other authoritative sources show weak or absent reductions, or insufficient evidence of change as a result of self-regulation, but some reduction following statutory regulation. 8 Treatment of obesity has been proven difficult, with dieting being the most effective intervention in terms of weight loss. There is, however, substantial 105

109 Chapter 8 dropout in these programs with figures ranging from 10 to 40% in systematic reviews. [225;226] Furthermore, sustaining the lost weight is a challenge. These figures support the idea that a person with obesity is set differently in utero than a person with normal weight. Also children born macrosomic are at risk of becoming obese and having metabolic syndrome (dyslipidemia, hypertension, obesity and insulin resistance). [227] A retrospective study showed that children born from obese mothers are twice as likely to be obese at 2 years of age when compared to mothers with normal weight. [228] In our current western society characterised by an abundance of (unhealthy) food and intensive marketing of a giant food industry, next to a more sedentary life, obesity seems the logic result. One of its negative consequences is subfertility in obese women. Society is responding with preventive measures, the results of these measures seem inadequate thus far in turning the table. Obesity: judged by appearance Today it is not general policy to treat obese women for subfertility. Besides the increased risks in pregnancy perhaps other factors play a role in denying fertility treatment. The majority of fertility physicians in the USA feels there should be a BMI limit for access to fertility treatment, and that this limit should be lower than the limit they currently apply in their facility. [229] These feelings are perhaps based on, or a result of judgment of obese women. It has been demonstrated extensively that obese people are judged and discriminated based on their appearance. [230] This even seems to be the case in health care providers for the obese. An interview based research study of participants at an obesity conference showed that they attribute negative findings significantly more to fat than to thin people. [231] To our knowledge no research has specifically been performed for fertility care providers but it is conceivable that, in line with the above conference study, they also think negatively of obese women in terms of their individual responsibility for not becoming pregnant or responding to treatment. This will probably have a negative impact on whether or how these women are treated. Prejudice concerning weight leads to substandard care in health care. [232] 106

110 General discussion In other groups of women judgement has played a role in access to fertility treatment, for example in lesbian couples and single women. Opponents of treating these women used different arguments like only treat people who are infertile or that children need a father and normal upbringing. Several studies have shown that these assumptions and prejudices are totally false. [233] The Ethics Committee in the USA proclaimed it therefore unjustified to withhold these women treatment. [234] The Committee s arguments were however foremost arguments in light of the welfare of the future child. The comparison with obesity is therefore weak because the risk for the woman is the argument used most frequently by proponents of a BMI limit. Another reason to deny fertility treatment is age. Seeking treatment for subfertility at an older age is also judged by society and health care providers. Worldwide, different age limits (39-46 years) are used for access to fertility care. Some argue that women should not be treated beyond a certain age [235] while others warn against discriminating women based on their age. [236] The proponents of not treating older women suggest that it is their own fault they don t conceive and therefore should not be treated. In conclusion, health care professionals should be aware of their prejudices and be able to care for their obese patients without being influenced by implicit negative feelings. Implications of findings in this thesis Based on current scientific knowledge, obese subfertile women should be offered lifestyle intervention prior to ART with the objective of losing weight and achieving a spontaneous pregnancy. They should be properly informed about their increased risks of pregnancy complications and that this may have distinct consequences for themselves as well as for their offspring. Clinical implications Chapter 3 clearly shows overweight and obesity have a negative impact on fertility. This does, however, not result in unsafe situations, be it for the subfertile woman during fertility treatment, or for the woman and her offspring during and after pregnancy. We conclude that when a woman with excess weight suffers from infertility it is of key importance to help her improve her lifestyle and lose weight. There should be attention for the reason the woman 8 107

111 Chapter 8 has excess weight and give guidance in improving possible bad habits. Most people with obesity have tried to lose weight but have regained it. Therefore it is important to explain what different approaches are available and offer an extensive follow up plan. This can be done by a dietician or trained nurse. Lifestyle intervention should be available in fertility clinics and at least six months treatment should be advised. A combination of diet, exercise and cognitive behavior therapy has the best chance of success. About 25% of women will conceive spontaneously during the intervention. After the lifestyle intervention ART should be offered, even when a woman did not lose weight despite effort. When time is an issue in women above 35 years or in case of low ovarian reserve lifestyle intervention and ART can be offered simultaneously. In order to have a good chance of success with the lifestyle intervention we advise not to judge the patient and consider the excess weight as a health issue instead of a choice. If a woman gets pregnant, in our opinion second line pregnancy care or shared care should be offered depending on her BMI. Fertility treatment of obese women is lengthy, more costly and less successful than that of a woman with normal weight. However, we show in chapter 4 that the success of ART is still good with an only 10% lower live birth rate compared to women with normal weight and therefore certainly worth the effort. Guidelines There is to our knowledge no clinical guideline specifically for ART in obese women. The National Institute for Health and Care Excellence(NICE) has a guideline for weight management before, during and after pregnancy. [237] They recommend all health care professionals to discuss the risks of obesity with these women and help them achieving a more healthy weight with lifestyle intervention. Based on the Dutch guideline for diagnostics and treatment of obesity we recommend the following [158]: from a BMI above 30 kg/m² combined lifestyle intervention should be available. In women with a BMI kg/m² and a waist circumference above 88 cm the same risk of comorbidities are at hand as for a woman with BMI 30 kg/m² and they should therefore also be offered combined lifestyle intervention. Women with a BMI 40kg/m² have such a significant increased risk of comorbidities such as type 2 diabetes and cardiovascular disease that they should first be screened for these disorders by 108

112 General discussion a specialist for internal diseases and treated if necessary. A weight reduction of 5% will lead to a decrease in comorbidity risk and is achievable for a subgroup in six months. Also a substantial amount of women will get pregnant during the six months of lifestyle intervention. If they are not pregnant after this time and were not able to lose weight despite their effort, they should nevertheless be treated for their subfertility. Future research Although many studies have shown potential benefit of lifestyle intervention in obese subfertile women, the exact mechanism behind this effect remains to be discovered. In order to determine the positive effects of lifestyle intervention prior to ART, several parameters should be studied in more detail besides weight loss: exercise, waist circumference, insulin resistance, diet, effect of cognitive behavior treatment. Also patient satisfaction and preference should be monitored. There is substantial dropout in lifestyle intervention programs with less weight loss as a result. In order to maximize the positive effects of lifestyle intervention we recommend to search for factors associated with dropout. A decrease in pregnancy complications is found in women after bariatric surgery with substantial weight loss. The effect of lifestyle intervention prior to pregnancy on pregnancy complications is still unclear. We did not find a difference between the two groups in our retrospective cohort study but this should be studied prospectively. Long term follow up of children born after lifestyle intervention is lacking. With the potential gain of preventing childhood obesity this could help achieving financial support from the government for implementation of lifestyle intervention

113 Chapter 8 References 1. RIVM. Nederland de Maat Genomen. Maat_Genomen WHO. WHO fact sheet Bauer UE, Briss PA, Goodman RA, Bowman BA. Prevention of chronic disease in the 21st century: elimination of the leading preventable causes of premature death and disability in the USA. Lancet 2014; 384: Allender S, Rayner M. The burden of overweight and obesity-related ill health in the UK. Obes.Rev. 2007; 8: Seidell JC. Societal and personal costs of obesity. Exp.Clin.Endocrinol.Diabetes 1998; 106 Suppl 2: Health Council. Voor dik en dun. Preventie van overgewicht en obesitas en het risico op eetstoornissen Den Haag: Gezondheidsraad, 2010; publicatienr. 2010/ Hunault CC, Habbema JD, Eijkemans MJ, Collins JA, Evers JL, te Velde ER. Two new prediction rules for spontaneous pregnancy leading to live birth among subfertile couples, based on the synthesis of three previous models. Hum Reprod. 2004; 19: Pasquali R, Gambineri A, Pagotto U. The impact of obesity on reproduction in women with polycystic ovary syndrome. BJOG. 2006; 113: van der Steeg JW, Steures P, Eijkemans MJ et al. Obesity affects spontaneous pregnancy chances in subfertile, ovulatory women. Hum.Reprod. 2008; 23: Chehab FF. 20 YEARS OF LEPTIN: Leptin and reproduction: past milestones, present undertakings, and future endeavors. J.Endocrinol. 2014; 223: T37-T Considine RV, Caro JF. Leptin and the regulation of body weight. Int.J.Biochem.Cell Biol. 1997; 29: Chehab FF, Lim ME, Lu R. Correction of the sterility defect in homozygous obese female mice by treatment with the human recombinant leptin. Nat.Genet. 1996; 12: Clement K, Vaisse C, Lahlou N et al. A mutation in the human leptin receptor gene causes obesity and pituitary dysfunction. Nature 1998; 392: Chehab FF, Mounzih K, Lu R, Lim ME. Early onset of reproductive function in normal female mice treated with leptin. Science 1997; 275: Azrad M, Gower BA, Hunter GR, Nagy TR. Intra-abdominal adipose tissue is independently associated with sex-hormone binding globulin in premenopausal women. Obesity.(Silver.Spring) 2012; 20: Jungheim ES, Travieso JL, Hopeman MM. Weighing the impact of obesity on female reproductive function and fertility. Nutr.Rev. 2013; 71 Suppl 1: S3-S Feodor NS, Andersen PK, Strandberg-Larsen K, Nybo Andersen AM. Risk factors for miscarriage from a prevention perspective: a nationwide follow-up study. BJOG. 2014; 121: Boots C, Stephenson MD. Does obesity increase the risk of miscarriage in spontaneous conception: a systematic review. Semin.Reprod.Med. 2011; 29: Metwally M, Ong KJ, Ledger WL, Li TC. Does high body mass index increase the risk of miscarriage after spontaneous and assisted conception? A meta-analysis of the evidence. Fertil.Steril. 2008; 90: Chu SY, Callaghan WM, Kim SY et al. Maternal obesity and risk of gestational diabetes mellitus. Diabetes Care 2007; 30: Cnossen JS, Leeflang MM, de Haan EE et al. Accuracy of body mass index in predicting pre-eclampsia: bivariate meta-analysis. BJOG. 2007; 114: Hancke K, Gundelach T, Hay B, Sander S, Reister F, Weiss JM. Pre-pregnancy obesity compromises obstetric and neonatal outcomes. J.Perinat.Med Sibai B, Dekker G, Kupferminc M. Pre-eclampsia. Lancet 2005; 365: McDonald SD, Han Z, Mulla S, Beyene J. Overweight and obesity in mothers and risk of preterm birth and low birth weight infants: systematic review and meta-analyses. BMJ 2010; 341: c

114 References 25. Liu L, Johnson HL, Cousens S et al. Global, regional, and national causes of child mortality: an updated systematic analysis for 2010 with time trends since Lancet 2012; 379: Johansson S, Villamor E, Altman M, Bonamy AK, Granath F, Cnattingius S. Maternal overweight and obesity in early pregnancy and risk of infant mortality: a population based cohort study in Sweden. BMJ 2014; 349: g Weiss JL, Malone FD, Emig D et al. Obesity, obstetric complications and cesarean delivery rate--a population-based screening study. Am.J.Obstet.Gynecol. 2004; 190: Chu SY, Kim SY, Schmid CH et al. Maternal obesity and risk of cesarean delivery: a meta-analysis. Obes.Rev. 2007; 8: Aune D, Saugstad OD, Henriksen T, Tonstad S. Maternal body mass index and the risk of fetal death, stillbirth, and infant death: a systematic review and meta-analysis. JAMA 2014; 311: Maheshwari A, Stofberg L, Bhattacharya S. Effect of overweight and obesity on assisted reproductive technology--a systematic review. Hum Reprod.Update. 2007; 13: Tamer EC, Senturk LM. The impact of body mass index on assisted reproduction. Curr.Opin.Obstet. Gynecol. 2009; 21: Zachariah M, Fleming R, Acharya U. Management of obese women in assisted conception units: a UK survey. Hum Fertil.(Camb.) 2006; 9: Pandey S, Maheshwari A, Bhattacharya S. Should access to fertility treatment be determined by female body mass index? Hum Reprod. 2010; 25: Vahratian A, Smith YR. Should access to fertility-related services be conditional on body mass index? Hum Reprod. 2009; 24: Pennings G, De WG, Shenfield F, Cohen J, Tarlatzis B, Devroey P. ESHRE Task Force on Ethics and Law 14: equity of access to assisted reproductive technology. Hum Reprod. 2008; 23: NICE. Fertility: assessment and treatment for people with fertility problems. London: National Institute for Clinical Excellence Clark AM, Ledger W, Galletly C et al. Weight loss results in significant improvement in pregnancy and ovulation rates in anovulatory obese women. Hum Reprod. 1995; 10: Hoeger KM, Kochman L, Wixom N, Craig K, Miller RK, Guzick DS. A randomized, 48-week, placebo-controlled trial of intensive lifestyle modification and/or metformin therapy in overweight women with polycystic ovary syndrome: a pilot study. Fertil.Steril. 2004; 82: Kiddy DS, Hamilton-Fairley D, Bush A et al. Improvement in endocrine and ovarian function during dietary treatment of obese women with polycystic ovary syndrome. Clin.Endocrinol.(Oxf) 1992; 36: Legro RS, Barnhart HX, Schlaff WD et al. Clomiphene, metformin, or both for infertility in the polycystic ovary syndrome. N.Engl.J.Med. 2007; 356: Norman RJ, Noakes M, Wu R, Davies MJ, Moran L, Wang JX. Improving reproductive performance in overweight/obese women with effective weight management. Hum Reprod.Update. 2004; 10: Tang T, Glanville J, Orsi N, Barth JH, Balen AH. The use of metformin for women with PCOS undergoing IVF treatment. Hum Reprod. 2006; 21: Moran L, Tsagareli V, Norman R, Noakes M. Diet and IVF pilot study: short-term weight loss improves pregnancy rates in overweight/obese women undertaking IVF. Aust.N.Z.J.Obstet.Gynaecol. 2011; 51: Sim K.A., Dezarnaulds G.M., Denyer G.S., Skilton M.R., Caterson I.D. Weight loss improves reproductive outcomes in obese women undergoing fertility treatment: a randomized controlled trial. Clin.Obes. 2014; 4: Thangaratinam S, Rogozinska E, Jolly K et al. Effects of interventions in pregnancy on maternal weight and obstetric outcomes: meta-analysis of randomised evidence. BMJ 2012; 344: e Dodd JM, Turnbull D, McPhee AJ et al. Antenatal lifestyle advice for women who are overweight or obese: LIMIT randomised trial. BMJ 2014; 348: g

115 Chapter James WP, Jackson-Leach R, Ni Mhurchu C et al. Overweight and Obesity (High Body Mass Index). Comparative quantification of health risks: global and regional burden of disease attributable to selected major risk factors. WHO 2004; World Health Organisation. Obesity: preventing and managing the global epidemic. Report of a WHO consultation. World Health Organ.Tech.Rep.Ser. 2000; Schokker DF, Visscher TL, Nooyens AC, van Baak MA, Seidell JC. Prevalence of overweight and obesity in the Netherlands. Obes.Rev. 2007; 8: Haslam DW, James WP. Obesity. Lancet 2005; 366: Linne Y. Effects of obesity on women s reproduction and complications during pregnancy. Obes. Rev. 2004; 5: Watson R. EU aims to tackle growing problem of obesity. BMJ 2005; 331: Butler D. The fertility riddle. Nature 2004; 432: Pasquali R, Patton L, Gambineri A. Obesity and infertility. Curr.Opin.Endocrinol.Diabetes Obes. 2007; 14: Pasquali R, Pelusi C, Genghini S, Cacciari M, Gambineri A. Obesity and reproductive disorders in women. Hum Reprod.Update. 2003; 9: Jensen TK, Scheike T, Keiding N, Schaumburg I, Grandjean P. Fecundability in relation to body mass and menstrual cycle patterns. Epidemiology 1999; 10: Imani B, Eijkemans MJ, te Velde ER, Habbema JD, Fauser BC. Predictors of patients remaining anovulatory during clomiphene citrate induction of ovulation in normogonadotropic oligoamenorrheic infertility. J.Clin.Endocrinol.Metab 1998; 83: Legro RS, Barnhart HX, Schlaff WD et al. Clomiphene, metformin, or both for infertility in the polycystic ovary syndrome. N.Engl.J.Med. 2007; 356: McClure N, McQuinn B, McDonald J, Kovacs GT, Healy DL, Burger HG. Body weight, body mass index, and age: predictors of menotropin dose and cycle outcome in polycystic ovarian syndrome? Fertil.Steril. 1992; 58: Mulders AG, Laven JS, Eijkemans MJ, Hughes EG, Fauser BC. Patient predictors for outcome of gonadotrophin ovulation induction in women with normogonadotrophic anovulatory infertility: a meta-analysis. Hum Reprod.Update. 2003; 9: Balen AH, Platteau P, Andersen AN et al. The influence of body weight on response to ovulation induction with gonadotrophins in 335 women with World Health Organization group II anovulatory infertility. BJOG. 2006; 113: Koloszar S, Daru J, Kereszturi A, Zavaczki Z, Szollosi J, Pal A. Effect of female body weight on efficiency of donor AI. Arch.Androl 2002; 48: Wang JX, Warnes GW, Davies MJ, Norman RJ. Overweight infertile patients have a higher fecundity than normal-weight women undergoing controlled ovarian hyperstimulation with intrauterine insemination. Fertil.Steril. 2004; 81: Fedorcsak P, Dale PO, Storeng R et al. Impact of overweight and underweight on assisted reproduction treatment. Hum Reprod. 2004; 19: Lashen H, Ledger W, Bernal AL, Barlow D. Extremes of body mass do not adversely affect the outcome of superovulation and in-vitro fertilization. Hum Reprod. 1999; 14: Wang JX, Davies M, Norman RJ. Body mass and probability of pregnancy during assisted reproduction treatment: retrospective study. BMJ 2000; 321: Lintsen AM, Pasker-de Jong PC, de Boer EJ et al. Effects of subfertility cause, smoking and body weight on the success rate of IVF. Hum Reprod. 2005; 20: Wang JX, Davies MJ, Norman RJ. Obesity increases the risk of spontaneous abortion during infertility treatment. Obes.Res. 2002; 10: Cedergren MI. Maternal morbid obesity and the risk of adverse pregnancy outcome. Obstet.Gynecol. 2004; 103: Sebire NJ, Jolly M, Harris JP et al. Maternal obesity and pregnancy outcome: a study of 287,213 pregnancies in London. Int.J.Obes.Relat Metab Disord. 2001; 25: Edwards LE, Hellerstedt WL, Alton IR, Story M, Himes JH. Pregnancy complications and birth outcomes in obese and normal-weight women: effects of gestational weight change. Obstet.Gynecol. 1996; 87:

116 References 72. Garbaciak JA, Jr., Richter M, Miller S, Barton JJ. Maternal weight and pregnancy complications. Am.J.Obstet.Gynecol. 1985; 152: Cnattingius S, Bergstrom R, Lipworth L, Kramer MS. Prepregnancy weight and the risk of adverse pregnancy outcomes. N.Engl.J.Med. 1998; 338: Kristensen J, Vestergaard M, Wisborg K, Kesmodel U, Secher NJ. Pre-pregnancy weight and the risk of stillbirth and neonatal death. BJOG. 2005; 112: Usha Kiran TS, Hemmadi S, Bethel J, Evans J. Outcome of pregnancy in a woman with an increased body mass index. BJOG. 2005; 112: Eijkemans MJ, Polinder S, Mulders AG, Laven JS, Habbema JD, Fauser BC. Individualized cost-effective conventional ovulation induction treatment in normogonadotrophic anovulatory infertility (WHO group 2). Hum Reprod. 2005; 20: Goverde AJ, McDonnell J, Vermeiden JP, Schats R, Rutten FF, Schoemaker J. Intrauterine insemination or in-vitro fertilisation in idiopathic subfertility and male subfertility: a randomised trial and cost-effectiveness analysis. Lancet 2000; 355: Esinler I, Bozdag G, Yarali H. Impact of isolated obesity on ICSI outcome. Reprod.Biomed.Online. 2008; 17: Gesink Law DC, Maclehose RF, Longnecker MP. Obesity and time to pregnancy. Hum Reprod. 2007; 22: Al-Azemi M, Omu FE, Omu AE. The effect of obesity on the outcome of infertility management in women with polycystic ovary syndrome. Arch.Gynecol.Obstet. 2004; 270: Imani B, Eijkemans MJ, te Velde ER, Habbema JD, Fauser BC. A nomogram to predict the probability of live birth after clomiphene citrate induction of ovulation in normogonadotropic oligoamenorrheic infertility. Fertil.Steril. 2002; 77: Chu SY, Kim SY, Lau J et al. Maternal obesity and risk of stillbirth: a metaanalysis. Am.J.Obstet.Gynecol. 2007; 197: Chen KT, Sell RL, Tuomala RE. Cost-effectiveness of elective cesarean delivery in human immunodeficiency virus-infected women(1). Obstet.Gynecol. 2001; 97: Graziosi GC, van der Steeg JW, Reuwer PH, Drogtrop AP, Bruinse HW, Mol BW. Economic evaluation of misoprostol in the treatment of early pregnancy failure compared to curettage after an expectant management. Hum Reprod. 2005; 20: Moss JR, Crowther CA, Hiller JE, Willson KJ, Robinson JS. Costs and consequences of treatment for mild gestational diabetes mellitus - evaluation from the ACHOIS randomised trial. BMC.Pregnancy.Childbirth. 2007; 7: Barton JR, Istwan NB, Rhea D, Collins A, Stanziano GJ. Cost-savings analysis of an outpatient management program for women with pregnancy-related hypertensive conditions. Dis.Manag. 2006; 9: Maheshwari A, Scotland G, Bell J, McTavish A, Hamilton M, Bhattacharya S. The direct health services costs of providing assisted reproduction services in overweight or obese women: a retrospective cross-sectional analysis. Hum Reprod. 2009; 24: Hoeger KM. Role of lifestyle modification in the management of polycystic ovary syndrome. Best. Pract.Res.Clin.Endocrinol.Metab 2006; 20: Huber-Buchholz MM, Carey DG, Norman RJ. Restoration of reproductive potential by lifestyle modification in obese polycystic ovary syndrome: role of insulin sensitivity and luteinizing hormone. J.Clin.Endocrinol.Metab 1999; 84: Norman RJ, Noakes M, Wu R, Davies MJ, Moran L, Wang JX. Improving reproductive performance in overweight/obese women with effective weight management. Hum Reprod.Update. 2004; 10: Gillett WR, Putt T, Farquhar CM. Prioritising for fertility treatments--the effect of excluding women with a high body mass index. BJOG. 2006; 113: Zachariah M, Fleming R, Acharya U. Management of obese women in assisted conception units: a UK survey. Hum Fertil.(Camb.) 2006; 9: Balen AH, Anderson RA. Impact of obesity on female reproductive health: British Fertility Society, Policy and Practice Guidelines. Hum Fertil.(Camb.) 2007; 10:

117 Chapter Ramlau-Hansen CH, Thulstrup AM, Nohr EA, Bonde JP, Sorensen TI, Olsen J. Subfecundity in overweight and obese couples. Hum Reprod. 2007; 22: Awartani KA, Nahas S, Al Hassan SH, Al Deery MA, Coskun S. Infertility treatment outcome in sub groups of obese population. Reprod.Biol.Endocrinol. 2009; 7: NICE. Fertility: assessment and treatment for people with fertility problems. London: National Institute for Clinical Excellence Ref Type: Report 97. Dondorp W, de WG, Pennings G et al. Lifestyle-related factors and access to medically assisted reproduction. Hum Reprod. 2010; 25: Dokras A, Baredziak L, Blaine J, Syrop C, VanVoorhis BJ, Sparks A. Obstetric outcomes after in vitro fertilization in obese and morbidly obese women. Obstet.Gynecol. 2006; 108: Tummon I, Gavrilova-Jordan L, Allemand MC, Session D. Polycystic ovaries and ovarian hyperstimulation syndrome: a systematic review*. Acta Obstet.Gynecol.Scand. 2005; 84: Whiting P, Rutjes AW, Reitsma JB, Bossuyt PM, Kleijnen J. The development of QUADAS: a tool for the quality assessment of studies of diagnostic accuracy included in systematic reviews. BMC.Med. Res.Methodol. 2003; 3: Higgins JP, Thompson SG, Deeks JJ, Altman DG. Measuring inconsistency in meta-analyses. BMJ 2003; 327: Beydoun HA, Stadtmauer L, Beydoun MA, Russell H, Zhao Y, Oehninger S. Polycystic ovary syndrome, body mass index and outcomes of assisted reproductive technologies. Reprod.Biomed. Online. 2009; 18: Esinler I, Bayar U, Bozdag G, Yarali H. Outcome of intracytoplasmic sperm injection in patients with polycystic ovary syndrome or isolated polycystic ovaries. Fertil.Steril. 2005; 84: Fedorcsak P, Dale PO, Storeng R, Tanbo T, Abyholm T. The impact of obesity and insulin resistance on the outcome of IVF or ICSI in women with polycystic ovarian syndrome. Hum Reprod. 2001; 16: Kolibianakis E, Zikopoulos K, Albano C et al. Reproductive outcome of polycystic ovarian syndrome patients treated with GnRH antagonists and recombinant FSH for IVF/ICSI. Reprod.Biomed. Online. 2003; 7: Kumbak B, Kahraman S. Efficacy of metformin supplementation during ovarian stimulation of lean PCOS patients undergoing in vitro fertilization. Acta Obstet.Gynecol.Scand. 2009; 88: Kurzawa R, Ciepiela P, Baczkowski T, Safranow K, Brelik P. Comparison of embryological and clinical outcome in GnRH antagonist vs. GnRH agonist protocols for in vitro fertilization in PCOS nonobese patients. A prospective randomized study. J.Assist.Reprod.Genet. 2008; 25: Li HW, Van EM, De VJ, Duncan WC, Anderson RA. Gonadotrophin ovulation induction is a successful treatment for World Health Organisation Group II anovulatory subfertility in women aged up to 40 and with body mass index up to 34. Hum Fertil.(Camb.) 2010; 13: Liberty G, Hyman JH, Eldar-Geva T, Latinsky B, Gal M, Margalioth EJ. Ovarian hemorrhage after transvaginal ultrasonographically guided oocyte aspiration: a potentially catastrophic and not so rare complication among lean patients with polycystic ovary syndrome. Fertil.Steril. 2010; 93: Lu XE, Huang HF, Li MG, Zhu YM, Qiang YL, Dong MY. Resistin levels of serum and follicular fluid in non-obese patients with polycystic ovary syndrome during IVF cycles. J.Zhejiang.Univ Sci.B 2005; 6: Salobir B, Prezelj J, Kocijancic A, Meden-Vrtovec H. Basal glucose level as an independent predictor of fertilization rate in patients without polycystic ovary syndrome included in an in-vitro fertilization program. Wien.Klin.Wochenschr. 2002; 114: Yarali H, Zeyneloglu HB. Gonadotrophin treatment in patients with polycystic ovary syndrome. Reprod.Biomed.Online. 2004; 8: Aramwit P, Pruksananonda K, Kasettratat N, Jammeechai K. Risk factors for ovarian hyperstimulation syndrome in Thai patients using gonadotropins for in vitro fertilization. Am.J.Health Syst. Pharm. 2008; 65:

118 References 114. Brannian JD, Schmidt SM, Kreger DO, Hansen KA. Baseline non-fasting serum leptin concentration to body mass index ratio is predictive of IVF outcomes. Hum Reprod. 2001; 16: Dechaud H, Anahory T, Reyftmann L, Loup V, Hamamah S, Hedon B. Obesity does not adversely affect results in patients who are undergoing in vitro fertilization and embryo transfer. Eur.J.Obstet. Gynecol.Reprod.Biol. 2006; 127: Jacobsen AF, Skjeldestad FE, Sandset PM. Ante- and postnatal risk factors of venous thrombosis: a hospital-based case-control study. J.Thromb.Haemost. 2008; 6: Jayaprakasan K, Jayaprakasan R, Al-Hasie HA et al. Can quantitative three-dimensional power Doppler angiography be used to predict ovarian hyperstimulation syndrome? Ultrasound Obstet. Gynecol. 2009; 33: la Cour FN, Gerds TA, Forman JL, Silver JD, Nyboe AA, Popovic-Todorovic B. Risk charts to identify low and excessive responders among first-cycle IVF/ICSI standard patients. Reprod.Biomed.Online. 2011; 22: Lee TH, Liu CH, Huang CC et al. Serum anti-mullerian hormone and estradiol levels as predictors of ovarian hyperstimulation syndrome in assisted reproduction technology cycles. Hum Reprod. 2008; 23: Tur R, Barri PN, Coroleu B, Buxaderas R, Martinez F, Balasch J. Risk factors for high-order multiple implantation after ovarian stimulation with gonadotrophins: evidence from a large series of 1878 consecutive pregnancies in a single centre. Hum Reprod. 2001; 16: Luke B, Brown MB, Stern JE, Missmer SA, Fujimoto VY, Leach R. Female obesity adversely affects assisted reproductive technology (ART) pregnancy and live birth rates. Hum Reprod. 2011; 26: Matalliotakis I, Cakmak H, Sakkas D, Mahutte N, Koumantakis G, Arici A. Impact of body mass index on IVF and ICSI outcome: a retrospective study. Reprod.Biomed.Online. 2008; 16: Sathya A, Balasubramanyam S, Gupta S, Verma T. Effect of body mass index on in vitro fertilization outcomes in women. J.Hum Reprod.Sci. 2010; 3: Spandorfer SD, Kump L, Goldschlag D, Brodkin T, Davis OK, Rosenwaks Z. Obesity and in vitro fertilization: negative influences on outcome. J.Reprod.Med. 2004; 49: van Swieten EC, Leeuw-Harmsen L, Badings EA, van der Linden PJ. Obesity and Clomiphene Challenge Test as predictors of outcome of in vitro fertilization and intracytoplasmic sperm injection. Gynecol.Obstet.Invest 2005; 59: Zhang D, Zhu Y, Gao H et al. Overweight and obesity negatively affect the outcomes of ovarian stimulation and in vitro fertilisation: a cohort study of 2628 Chinese women. Gynecol.Endocrinol. 2010; 26: Farhi J, Ben-Haroush A, Sapir O, Fisch B, Ashkenazi J. High-quality embryos retain their implantation capability in overweight women. Reprod.Biomed.Online. 2010; 21: Sneed ML, Uhler ML, Grotjan HE, Rapisarda JJ, Lederer KJ, Beltsos AN. Body mass index: impact on IVF success appears age-related. Hum Reprod. 2008; 23: Wittemer C, Ohl J, Bailly M, Bettahar-Lebugle K, Nisand I. Does body mass index of infertile women have an impact on IVF procedure and outcome? J.Assist.Reprod.Genet. 2000; 17: Bellver J, Ayllon Y, Ferrando M et al. Female obesity impairs in vitro fertilization outcome without affecting embryo quality. Fertil.Steril. 2010; 93: Hill MJ, Hong S, Frattarelli JL. Body mass index impacts in vitro fertilization stimulation. ISRN.Obstet.Gynecol. 2011; 2011: Kilic S, Yilmaz N, Zulfikaroglu E et al. Obesity alters retrieved oocyte count and clinical pregnancy rates in high and poor responder women after in vitro fertilization. Arch.Gynecol.Obstet. 2010; 282: Ku SY, Kim SD, Jee BC et al. Clinical efficacy of body mass index as predictor of in vitro fertilization and embryo transfer outcomes. J.Korean Med.Sci. 2006; 21: Loveland JB, McClamrock HD, Malinow AM, Sharara FI. Increased body mass index has a deleterious effect on in vitro fertilization outcome. J.Assist.Reprod.Genet. 2001; 18:

119 Chapter Martinuzzi K, Ryan S, Luna M, Copperman AB. Elevated body mass index (BMI) does not adversely affect in vitro fertilization outcome in young women. J.Assist.Reprod.Genet. 2008; 25: Orvieto R, Meltcer S, Nahum R, Rabinson J, Anteby EY, Ashkenazi J. The influence of body mass index on in vitro fertilization outcome. Int.J.Gynaecol.Obstet. 2009; 104: Salha O, Dada T, Sharma V. Influence of body mass index and self-administration of hcg on the outcome of IVF cycles: a prospective cohort study. Hum Fertil.(Camb.) 2001; 4: Styne-Gross A, Elkind-Hirsch K, Scott RT, Jr. Obesity does not impact implantation rates or pregnancy outcome in women attempting conception through oocyte donation. Fertil.Steril. 2005; 83: Thum MY, El-Sheikhah A, Faris R et al. The influence of body mass index to in-vitro fertilisation treatment outcome, risk of miscarriage and pregnancy outcome. J.Obstet.Gynaecol. 2007; 27: Jungheim ES, Lanzendorf SE, Odem RR, Moley KH, Chang AS, Ratts VS. Morbid obesity is associated with lower clinical pregnancy rates after in vitro fertilization in women with polycystic ovary syndrome. Fertil.Steril. 2009; 92: McCormick B, Thomas M, Maxwell R, Williams D, Aubuchon M. Effects of polycystic ovarian syndrome on in vitro fertilization-embryo transfer outcomes are influenced by body mass index. Fertil.Steril. 2008; 90: Mulders AG, Laven JS, Eijkemans MJ, Hughes EG, Fauser BC. Patient predictors for outcome of gonadotrophin ovulation induction in women with normogonadotrophic anovulatory infertility: a meta-analysis. Hum Reprod.Update. 2003; 9: Orvieto R, Nahum R, Meltcer S et al. Ovarian stimulation in polycystic ovary syndrome patients: the role of body mass index. Reprod.Biomed.Online. 2009; 18: Ozgun MT, Uludag S, Oner G, Batukan C, Aygen EM, Sahin Y. The influence of obesity on ICSI outcomes in women with polycystic ovary syndrome. J.Obstet.Gynaecol. 2011; 31: Pagidas K, Carson SA, McGovern PG et al. Intercourse compliance, ovulation, and treatment success in the National Institute of Child Health and Human Development-Reproductive Medicine Network s Pregnancy in Polycystic Ovary Syndrome (PPCOS) Trial. Fertil.Steril. 2010; 94: Fadini R, Dal Canto MB, Renzini MM et al. Predictive factors in in-vitro maturation in unstimulated women with normal ovaries. Reprod.Biomed.Online. 2009; 18: Kilic S, Moraloglu O, Yuksel B, Ozaksit G, Tasdemir N, Dogan M. IL-18 and IL-18 binding protein concentrations in serum and ovarian follicular fluid do not affect in vitro fertilization outcomes in obese, overweight, and non-obese patients. J.Reprod.Immunol. 2009; 79: Kumbak B, Akbas H, Sahin L, Karlikaya G, Karagozoglu H, Kahraman S. Ovarian stimulation in women with high and low body mass index: GnRH agonist versus GnRH antagonist. Reprod.Biomed. Online. 2010; 20: Kupka MS, Gnoth C, Buehler K, Dahncke W, Kruessel JS. Impact of female and male obesity on IVF/ ICSI: results of 700,000 ART-cycles in Germany. Gynecol.Endocrinol. 2011; 27: Kjotrod SB, von D, V, Carlsen SM. Metformin treatment before IVF/ICSI in women with polycystic ovary syndrome; a prospective, randomized, double blind study. Hum Reprod. 2004; 19: Templeton A, Morris JK, Parslow W. Factors that affect outcome of in-vitro fertilisation treatment. Lancet 1996; 348: Maheshwari A, Scotland G, Bell J, McTavish A, Hamilton M, Bhattacharya S. Direct health services costs of providing assisted reproduction services in older women. Fertil.Steril. 2010; 93: Mutsaerts MA, Groen H, ter Bogt NC et al. The LIFESTYLE study: costs and effects of a structured lifestyle program in overweight and obese subfertile women to reduce the need for fertility treatment and improve reproductive outcome. A randomised controlled trial. BMC.Womens Health 2010; 10: Visscher TL, Seidell JC. The public health impact of obesity. Annu.Rev.Public Health 2001; 22: WHO. Obesity: preventing and managing the global epidemic. Report of a WHO consultation. World Health Organ Tech.Rep.Ser. 2000; 894: i Finucane MM, Stevens GA, Cowan MJ et al. National, regional, and global trends in body-mass index since 1980: systematic analysis of health examination surveys and epidemiological studies with 960 country-years and 9.1 million participants. Lancet 2011; 377:

120 Referencesn 157. VWS. Kiezen voor gezond leven. kamerstukken/2006/10/06/bijlage-bij-preventienota-kiezen-voor-gezond-leven.html CBO. Richtlijn Diagnostiek en behandeling van obesitas bij volwassenen en kinderen. diliguide.nl/document/ Partnerschap overgewicht. Zorgstandaard obesitas. site_files/uploads/pon_zorgstandaard_obesitas_2011_a4_v1%2004.pdf Rich-Edwards JW, Goldman MB, Willett WC et al. Adolescent body mass index and infertility caused by ovulatory disorder. Am.J.Obstet.Gynecol. 1994; 171: Hollmann M, Runnebaum B, Gerhard I. Effects of weight loss on the hormonal profile in obese, infertile women. Hum.Reprod. 1996; 11: Jensen TK, Scheike T, Keiding N, Schaumburg I, Grandjean P. Fecundability in relation to body mass and menstrual cycle patterns. Epidemiology 1999; 10: Galletly C, Clark A, Tomlinson L, Blaney F. Improved pregnancy rates for obese, infertile women following a group treatment program. An open pilot study. Gen.Hosp.Psychiatry 1996; 18: Laven JS, Imani B, Eijkemans MJ, Fauser BC. New approach to polycystic ovary syndrome and other forms of anovulatory infertility. Obstet.Gynecol.Surv. 2002; 57: Huber-Buchholz MM, Carey DG, Norman RJ. Restoration of reproductive potential by lifestyle modification in obese polycystic ovary syndrome: role of insulin sensitivity and luteinizing hormone. J.Clin.Endocrinol.Metab 1999; 84: Clark AM, Thornley B, Tomlinson L, Galletley C, Norman RJ. Weight loss in obese infertile women results in improvement in reproductive outcome for all forms of fertility treatment. Hum.Reprod. 1998; 13: Clark AM, Ledger W, Galletly C et al. Weight loss results in significant improvement in pregnancy and ovulation rates in anovulatory obese women. Hum.Reprod. 1995; 10: Kuchenbecker WK, Ruifrok AE, Bolster JH, Heineman MJ, Hoek A. [Subfertility in overweight women]. Ned.Tijdschr.Geneeskd. 2006; 150: Carpenter JR, Kenward MG. Missing Data in Randomised Controlled Trials a Practical Guide. National Institute for Health Research 2008; 18: Sterne JA, White IR, Carlin JB et al. Multiple imputation for missing data in epidemiological and clinical research: potential and pitfalls. BMJ 2009; 338: b May AM, Barnes DR, Forouhi NG et al. Prediction of measured weight from self-reported weight was not improved after stratification by body mass index. Obesity.(Silver.Spring) 2013; 21: E137-E Messier SP, Loeser RF, Miller GD et al. Exercise and dietary weight loss in overweight and obese older adults with knee osteoarthritis: the Arthritis, Diet, and Activity Promotion Trial. Arthritis Rheum. 2004; 50: Mutsaerts MA, Kuchenbecker WK, Mol BW, Land JA, Hoek A. Dropout is a problem in lifestyle intervention programs for overweight and obese infertile women: a systematic review. Hum.Reprod. 2013; 28: Phelan S, Hill JO, Lang W, Dibello JR, Wing RR. Recovery from relapse among successful weight maintainers. Am.J.Clin.Nutr. 2003; 78: Anderson JW, Konz EC, Frederich RC, Wood CL. Long-term weight-loss maintenance: a meta-analysis of US studies. Am.J.Clin.Nutr. 2001; 74: Stahre L, Hallstrom T. A short-term cognitive group treatment program gives substantial weight reduction up to 18 months from the end of treatment. A randomized controlled trial. Eat.Weight. Disord. 2005; 10: Kiddy DS, Hamilton-Fairley D, Bush A et al. Improvement in endocrine and ovarian function during dietary treatment of obese women with polycystic ovary syndrome. Clin.Endocrinol.(Oxf) 1992; 36: NICE guideline. Obesity: identification, assessment and management of overweight and obesity in children, young people and adults Palomba S, Falbo A, Valli B et al. Physical activity before IVF and ICSI cycles in infertile obese women: an observational cohort study. Reprod.Biomed.Online. 2014; 29: Brandes M, Hamilton CJ, de Bruin JP, Nelen WL, Kremer JA. The relative contribution of IVF to the total ongoing pregnancy rate in a subfertile cohort. Hum Reprod. 2010; 25:

121 Chapter Eijkemans MJ, Lintsen AM, Hunault CC et al. Pregnancy chances on an IVF/ICSI waiting list: a national prospective cohort study. Hum.Reprod. 2008; 23: Horne G, Farrell C, Pease EH, Brison DR, Falconer DA, Lieberman BA. Waiting for in vitro fertilization treatment: spontaneous and ART live births. Hum.Fertil.(Camb.) 2003; 6: Ehrenberg HM, Mercer BM, Catalano PM. The influence of obesity and diabetes on the prevalence of macrosomia. Am.J.Obstet.Gynecol. 2004; 191: Stephansson O, Dickman PW, Johansson A, Cnattingius S. Maternal weight, pregnancy weight gain, and the risk of antepartum stillbirth. Am.J.Obstet.Gynecol. 2001; 184: Waller DK, Shaw GM, Rasmussen SA et al. Prepregnancy obesity as a risk factor for structural birth defects. Arch.Pediatr.Adolesc.Med. 2007; 161: Farquhar CM, Gillett WR. Prioritising for fertility treatments--should a high BMI exclude treatment? BJOG. 2006; 113: Balen AH, Dresner M, Scott EM, Drife JO. Should obese women with polycystic ovary syndrome receive treatment for infertility? BMJ 2006; 332: Jensen TK, Scheike T, Keiding N, Schaumburg I, Grandjean P. Fecundability in relation to body mass and menstrual cycle patterns. Epidemiology 1999; 10: Baillargeon. Evaluation of Clinical Outcomes and Costs of a Lifestyle Intervention in Obese Infertile Women. NCT Legro RS, Brzyski RG, Diamond MP et al. The Pregnancy in Polycystic Ovary Syndrome II study: baseline characteristics and effects of obesity from a multicenter randomized clinical trial. Fertil. Steril. 2014; 101: Aagaard-Tillery KM, Porter TF, Lane RH, Varner MW, Lacoursiere DY. In utero tobacco exposure is associated with modified effects of maternal factors on fetal growth. Am.J.Obstet.Gynecol. 2008; 198: Pandey S, Shetty A, Hamilton M, Bhattacharya S, Maheshwari A. Obstetric and perinatal outcomes in singleton pregnancies resulting from IVF/ICSI: a systematic review and meta-analysis. Hum Reprod.Update. 2012; 18: Chauhan SP, Scardo JA, Hayes E, Abuhamad AZ, Berghella V. Twins: prevalence, problems, and preterm births. Am.J.Obstet.Gynecol. 2010; 203: Conde-Agudelo A, Belizan JM, Lindmark G. Maternal morbidity and mortality associated with multiple gestations. Obstet.Gynecol. 2000; 95: Kallen B, Finnstrom O, Nygren KG, Otterblad OP, Wennerholm UB. In vitro fertilisation in Sweden: obstetric characteristics, maternal morbidity and mortality. BJOG. 2005; 112: Pinborg A, Wennerholm UB, Romundstad LB et al. Why do singletons conceived after assisted reproduction technology have adverse perinatal outcome? Systematic review and meta-analysis. Hum Reprod.Update. 2013; 19: Galazis N, Docheva N, Simillis C, Nicolaides KH. Maternal and neonatal outcomes in women undergoing bariatric surgery: a systematic review and meta-analysis. Eur.J.Obstet.Gynecol.Reprod. Biol. 2014; 181: Wing RR, Tate DF, Gorin AA, Raynor HA, Fava JL. A self-regulation program for maintenance of weight loss. N.Engl.J.Med. 2006; 355: Rasmussen KM, Catalano PM, Yaktine AL. New guidelines for weight gain during pregnancy: what obstetrician/gynecologists should know. Curr.Opin.Obstet.Gynecol. 2009; 21: Boney CM, Verma A, Tucker R, Vohr BR. Metabolic syndrome in childhood: association with birth weight, maternal obesity, and gestational diabetes mellitus. Pediatrics 2005; 115: e290-e Gaudet L, Ferraro ZM, Wen SW, Walker M. Maternal obesity and occurrence of fetal macrosomia: a systematic review and meta-analysis. Biomed.Res.Int. 2014; 2014: Armitage JA, Poston L, Taylor PD. Developmental origins of obesity and the metabolic syndrome: the role of maternal obesity. Front Horm.Res. 2008; 36: Brisbois TD, Farmer AP, McCargar LJ. Early markers of adult obesity: a review. Obes.Rev. 2012; 13: Lobstein T, Baur L, Uauy R. Obesity in children and young people: a crisis in public health. Obes.Rev. 2004; 5 Suppl 1:

122 References 205. Cole TJ, Lobstein T. Extended international (IOTF) body mass index cut-offs for thinness, overweight and obesity. Pediatr.Obes. 2012; 7: Koning AM, Mutsaerts MA, Kuchenbecker WK et al. Complications and outcome of assisted reproduction technologies in overweight and obese women. Hum.Reprod. 2012; 27: Spong CY, Mercer BM, D alton M, Kilpatrick S, Blackwell S, Saade G. Timing of indicated latepreterm and early-term birth. Obstet.Gynecol. 2011; 118: NVOG. Mogelijke morele contra-indicaties bij vruchtbaarheidsbehandelingen,versie nvog-documenten.nl/index.php?pagina=/richtlijn/pagina.php&fselecttg_62=75&fselected- Sub=62&fSelectedParent= Pennings G. Measuring the welfare of the child: in search of the appropriate evaluation principle. Hum.Reprod. 1999; 14: Koning AM, Kuchenbecker WK, Groen H et al. Economic consequences of overweight and obesity in infertility: a framework for evaluating the costs and outcomes of fertility care. Hum.Reprod. Update. 2010; 16: Hamilton M. This house believes that obese women should not be treated until they have lost weight. Hum Repr. 2011; 26, suppl 1: i Hales CN, Barker DJ. The thrifty phenotype hypothesis. Br.Med.Bull. 2001; 60: Malik VS, Pan A, Willett WC, Hu FB. Sugar-sweetened beverages and weight gain in children and adults: a systematic review and meta-analysis. Am.J.Clin.Nutr. 2013; 98: Briggs AD, Mytton OT, Kehlbacher A, Tiffin R, Rayner M, Scarborough P. Overall and income specific effect on prevalence of overweight and obesity of 20% sugar sweetened drink tax in UK: econometric and comparative risk assessment modelling study. BMJ 2013; 347: f Waterlander WE, Ni MC, Steenhuis IH. Effects of a price increase on purchases of sugar sweetened beverages. Results from a randomized controlled trial. Appetite 2014; 78: Jha P, Peto R. Global effects of smoking, of quitting, and of taxing tobacco. N.Engl.J.Med. 2014; 370: Wagenaar AC, Salois MJ, Komro KA. Effects of beverage alcohol price and tax levels on drinking: a meta-analysis of 1003 estimates from 112 studies. Addiction 2009; 104: Chopra M, Darnton-Hill I. Tobacco and obesity epidemics: not so different after all? BMJ 2004; 328: FNLI. Stille kracht: route voorwaarts voor de nederlandse levensmiddelenindustrie. fnli.nl/wp-content/uploads/2015/01/de-stille-kracht-route-voorwaarts-voor-de-nederlandse-levensmiddelenindustrie.pdf Carter OB, Patterson LJ, Donovan RJ, Ewing MT, Roberts CM. Children s understanding of the selling versus persuasive intent of junk food advertising: implications for regulation. Soc.Sci.Med. 2011; 72: Galbraith-Emami S, Lobstein T. The impact of initiatives to limit the advertising of food and beverage products to children: a systematic review. Obes.Rev. 2013; 14: Avenell A, Brown TJ, McGee MA et al. What are the long-term benefits of weight reducing diets in adults? A systematic review of randomized controlled trials. J.Hum Nutr.Diet. 2004; 17: Pirozzo S, Summerbell C, Cameron C, Glasziou P. Advice on low-fat diets for obesity. Cochrane. Database.Syst.Rev. 2002; CD Catalano PM, Ehrenberg HM. The short- and long-term implications of maternal obesity on the mother and her offspring. BJOG. 2006; 113: Whitaker RC. Predicting preschooler obesity at birth: the role of maternal obesity in early pregnancy. Pediatrics 2004; 114: e29-e

123 Chapter Harris ID, Python J, Roth L, Alvero R, Murray S, Schlaff WD. Physicians perspectives and practices regarding the fertility management of obese patients. Fertil.Steril. 2011; 96: Puhl RM, Heuer CA. The stigma of obesity: a review and update. Obesity.(Silver.Spring) 2009; 17: Tomiyama AJ, Finch LE, Incollingo Belsky AC et al. Weight bias in 2001 versus 2013: Contradictory attitudes among obesity researchers and health professionals. Obesity.(Silver.Spring) Hebl MR, Xu J. Weighing the care: physicians reactions to the size of a patient. Int.J.Obes.Relat Metab Disord. 2001; 25: Murray C, Golombok S. Solo mothers and their donor insemination infants: follow-up at age 2 years. Hum Reprod. 2005; 20: The Ethics Committee of the American Society for Reproductive Medicine. Access to fertility treatment by gays, lesbians, and unmarried persons: a committee opinion. Fertil.Steril. 2013; 100: Caplan AL, Patrizio P. Are you ever too old to have a baby? The ethical challenges of older women using infertility services. Semin.Reprod.Med. 2010; 28: Smajdor A. The ethics of IVF over 40. Maturitas 2011; 69: NICE. Weight management before, during and after pregnancy. ph

124

125

126 Nederlandse samenvatting