Women, their Offspring and iMproving lifestyle for Better cardiovascular health of both (WOMB project): a protocol of the follow-up of a multicentre randomised controlled trial.

The overall aim of the Women, their Offspring and iMproving lifestyle for Better cardiovascular health of both (WOMB) project is to examine the effects of a preconception lifestyle intervention on health in obese women (WOMB women) and their children (WOMB kids). It concerns a follow-up of the LIFEstyle study, a multicentre randomised controlled trial (RCT) (Netherlands Trial Registry number 1530), that was originally set up to investigate the effects and costs of a structured lifestyle programme in overweight and obese infertile couples to prevent unnecessary infertility treatment and improve reproductive outcome. 1 Compared with prompt infertility treatment, the LIFEstyle intervention did not increase the healthy singleton live birth rate, although it did raise the chance of spontaneous conception. 2 Furthermore, live birth rate was higher in women who succeeded in losing weight. 3 Currently, 4–7 years after the LIFEstyle study, we will investigate the effects of the intervention on women’s and offspring’s lifestyle and health, with specific focus on cardiovascular health. Overweight and obese women have a higher risk of developing cardiovascular diseases (CVD) compared with normal weight women. 4 They also have a higher chance of medical and obstetric complications during pregnancy, such as gestational hypertension, pre-eclampsia and gestational diabetes. 5 The cardiovascular and metabolic health of women before and during pregnancy may influence the development of cardiovascular structure and function, and metabolic balance in the offspring, either mediated by increased risk of obstetric complications or by affecting placental and foetal growth and physiology. 6 Through these pathways, obese women may transfer the obesity risk to their children by non-Mendelian, for example, epigenetic, mechanisms. 7 This intergenerational cycle of obesity and consequent susceptibility for non-communicable diseases (NCDs), including CVD, may be broken by improving the health of obese women before and during pregnancy, making a preconceptional lifestyle intervention the ideal window to improve the health of the current and subsequent generations. 8 In general, women are especially receptive to advice about lifestyle before and during pregnancy 9 10 and optimising their lifestyle for the benefit of their offspring’s health will be a powerful motivator. Therefore, lifestyle interventions before and during pregnancy may be more effective than lifestyle interventions at any other time during the lifespan. Optimising lifestyle of women before and during pregnancy may be an innovative way of improving cardiovascular health and preventing CVD both in women and in their offspring. The design of the LIFEstyle study creates a unique opportunity to investigate the long-term effects of preconception lifestyle advice to overweight and obese women. The LIFEstyle study was performed within the Dutch Consortium for Healthcare Evaluation and Research in Obstetrics and Gynaecology. The WOMB project will be carried out in a collaboration between the Academic Medical Centre (coordinating centre), the VU University Medical Centre, the University Medical Centre Groningen and Wageningen University and Research.

In the following section, we will adhere to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines to the extent to which it is possible in a cohort profile. More details on handling of (quantitative) variables and statistical analyses will be provided in the papers describing the findings; the STROBE guidelines will be followed in papers resulting from this study. 11 The WOMB cohort consists of women who participated in the LIFEstyle study (WOMB women) and their offspring (WOMB kids). Between June 2009 and June 2012, women were recruited for the LIFEstyle study at fertility clinics of six university medical centres and 17 general hospitals, spread over The Netherlands. Women, aged between 18 and 39 years, who presented with infertility and a body mass index (BMI) ≥29 kg/m² could be included in the study. Women were diagnosed with infertility because of chronic anovulation, oligomenorrhoea or amenorrhoea or, in case of a functioning ovulatory cycle, unsuccessful conception for at least 12 months. Exclusion criteria were severe endometriosis, premature ovarian insufficiency, endocrinopathy (eg, diabetes type I and Cushing’s syndrome), the use of donor semen because of azoospermia and untreated preconceptional hypertension or hypertension-related complications in a previous pregnancy. Of the 822 eligible women, 577 agreed to participate and were randomly assigned to the intervention or the control group after providing written informed consent. Three women withdrew informed consent later on and 10 women were lost to follow-up, leaving 280 women in the control group and 284 women in the intervention group ( figure 1 ). See table 1 for baseline characteristics of both groups. The LIFEstyle flow diagram: recruitment and study design. Baseline characteristics of eligible participants and participants lost to attrition for WOMB women, according to trial group* Differences between the eligible intervention group and the eligible control group were compared with the use of Student’s t-test for means, Mann-Whitney U test for medians and χ 2 test or Fisher’s exact test for proportions. There were no significant differences between the groups. *Baseline is at randomisation for the LIFEstyle study. †Number (%) unless otherwise specified. ‡Ethnic background was self-reported. §Couples could have more than one diagnosis. ¶The denominator is the number of women with anovulatory infertility. BMI, body mass index; PCOS, polycystic ovary syndrome; WOMB, Women, their Offspring and iMproving lifestyle for Better cardiovascular health of both. The 284 women allocated to the intervention group received a 6 months structured lifestyle programme aiming at 5%–10% loss of their original body weight. During this weight loss period, they did not receive infertility treatment. The programme was developed according to National Institutes of Health 12 recommendations during a single centre pilot study. 13 14 The participants were guided by trained coaches with a degree in nursing or dietetics. Women were stimulated to reduce energy intake by 600 kcal/day (with a minimum total intake of 1200 kcal/day), supported by an online diet diary. 15 They were also encouraged to increase physical activity, aiming at 10 000 steps per day, monitored by a step counter, and at least 30 min of exercise of moderate intensity two or three times a week. Furthermore, the coaches provided individualised motivational counselling, directed at awareness of healthy lifestyle and formulating goals. The intervention was terminated earlier if women became pregnant. In case of a miscarriage, women could resume the intervention. To enhance compliance with the intervention, the infertility treatment was offered as soon as women had reached minimal 5% wt loss or a BMI <29 kg/m². The 280 women in the control group received infertility treatment according to the Dutch infertility guidelines, irrespective of their BMI. In total, 564 women will be eligible for the follow-up measurements of WOMB women. The study population comprises all children of the 564 women eligible for follow-up who were conceived within 24 months after inclusion in the study. In total, 341 children were born of whom 7 children died antepartum, during or short after labour. After exclusion of the children from multiple pregnancies (n=29), 305 singletons will be eligible for WOMB kids ( figure 2 ). Flow diagram enrolment WOMB women and WOMB kids. WOMB, Women, their Offspring and iMproving lifestyle for Better cardiovascular health of both. In the LIFEstyle study, information was obtained about demographics, medical, gynaecological and obstetric history, anthropometry, blood pressure, preconceptional lifestyle, eating behaviour, infertility treatment, pregnancy outcome, quality of life and medical costs of the intervention. Blood samples were taken and weight was measured at 0, 3 and 6 months after entering the study. In the intervention group, step counters and dietary assessments were used during the 6 months of the intervention. The majority of these data was collected during the 6 months after inclusion in the study. Key measures are summarised in table 2 . Measurements of LIFEstyle study *Number of weeks after inclusion, 0=baseline. †Till 24 weeks in intervention group only. PCOS, polycystic ovary syndrome. In the WOMB project, data will be collected about present lifestyle, and mental and physical health, with a specific focus on cardiovascular health. An overview of all key measures is given in table 3 (WOMB women) and table 4 (WOMB kids). Planned measurements of WOMB women PCOS, polycystic ovary syndrome; WOMB, Women, their Offspring and iMproving lifestyle for Better cardiovascular health of both. Planned measurements of WOMB kids WOMB, Women, their Offspring and iMproving lifestyle for Better cardiovascular health of both. The follow-up assessment consists of three stages (flowchart in figure 3 ) and comprises of (online) questionnaires (focusing on physical and mental health, diet, sleep of woman and child and growth and development of the child), accelerometry, physical measurements and collection of biospecimens (blood, faeces and buccal swab samples) as well as child behavioural observations. A subsample of 100 women and 100 children are planned for in depth cardiovascular measurements, including intima–media thickness and ultrasound assessment of cardiac function. Flowchart follow-up assessment of WOMB project. WOMB, Women, their Offspring and iMproving lifestyle for Better cardiovascular health of both. The physical measurements and observations are done in a mobile research vehicle in the vicinity of the participants’ homes to optimise participation and collect the data in a standardised situation. Following the RCT design, we will examine the effect of the preconception lifestyle intervention on later lifestyle and health of the women and their offspring by means of intention-to-treat analyses. In addition, we will perform per-protocol analyses in which women who did not complete the intervention will be excluded. Finally, the registered variations in adherence to the prescribed lifestyle intervention will allow us to also perform exploratory dose–response analyses in which associations between measures of variation in diet (eg, fruit/vegetable intake, soft drink intake and snack intake) or physical activity (eg, weekly moderate to vigorous physical activity) will be linked to outcomes (blood pressure, glucose/lipid levels, weight, child development and health) in both short and long term for women and their children. Furthermore, we will investigate whether physical, psychological, socioeconomic or genetic characteristics contribute to the effectiveness of the intervention, giving insight into potential gene–environment interactions, personality factors or susceptible socioeconomic groups and ultimately allowing us to refine and personalise future lifestyle intervention programmes to maximally target specific groups. The main strength of the WOMB project is the fact that it concerns the follow-up of a RCT in which a structured lifestyle intervention was given preconceptionally. We will capitalise on the randomised design in which interference by other factors, like genetic variability and environmental characteristics, will be balanced between the intervention and the control arm. Inevitably, there will be loss to follow-up, since it is 4–7 years ago that the women were included in the LIFEstyle study. In general, attrition will lead to a reduction in power, and there may be selective attrition reducing the representativeness of the study sample. 16 To control for selection bias, we will always report on the baseline characteristics of those who did and did not participate in follow-up. The original study was not powered on long-term outcome of the women in advance, let alone the follow-up of their offspring, neither was it powered on cardiovascular outcomes. Therefore, we conducted power analyses with respect to the main outcome BMI, which can be considered a risk factor for later CVD. 17 18 With respect to BMI, we need 90 women in each arm (31.9% of the total population) to detect a 0.5 kg/m 2 difference (27.5 kg/m 2 vs 28 kg/m 2 ) with a power of 80% (alpha level of 5% and an SD of 1.2). For the children, the participation rate needs to be higher: 70.1% (214 children), to detect a comparable difference (15.3 kg/m 2 vs 15.8 kg/m 2 ) with the same power (alpha level of 5% and an SD of 1.3). All in all, power to detect meaningful differences between the groups, especially with respect to secondary outcomes, will depend on the attrition levels. To increase power and generalisability, we are planning to pool data with other studies that use a comparable conceptual framework and an equivalent design and implementation 19 (of which the Finnish RADIEL study is one 20 ). The study will be conducted according to the principles of the Declaration of Helsinki (revised version of October 2013) and in accordance with the Dutch Medical Research Involving Human Subjects Act. For WOMB women and WOMB kids separate written informed consents will be obtained, as well as for each follow-up stage. The participants and/or parents will be asked to sign a separate consent form to store biological material in the biobank for up to 50 years. The data will be handled confidentially and analysed coded, in compliance with the Dutch Personal Data Protection Act. The study imposes no risks on the participants. Taking a venous blood sample (women 40 mL and children 15 mL) can be a small discomfort for the participants, but is considered safe. In the informed consent, the participants/parents have the ability to choose to participate in the study without blood sampling. The study findings of the WOMB project will be widely announced and reported to the scientific community, healthcare professionals, policy makers, future parents and general public. Dissemination is crucial in actually achieving the ultimate goal of WOMB project to prevent CVD in two generations by optimising lifestyle of women before and during pregnancy. The project has a Dutch website mainly focused on participants and a scientifically orientated international website ( http://womb-project.eu/ ). In 2015, we started a social media community on Facebook ( http://www.facebook/hetwombproject ), primarily directed to Dutch women in the reproductive age group, with the goal to maximally reach our target population at the time that results of the study will be known (2018/2019). This community is regularly fed by interesting news facts, short films about our project, blogs of our researchers and guest blogs of other stakeholders. We will also specifically focus on the dissemination of the study results among nutritionists, general practitioners, midwives and gynaecologists, who are in the position to advice (future) pregnant women about (pre)conceptional lifestyle.

The results of the present ongoing WOMB project, the follow-up of the LIFEstyle study, will show us if a preconception lifestyle intervention can have long-term health benefits for obese women and their children. The project will also provide more detailed information about the relationship between preconceptional diet and physical activity and later health: for the women, but also with respect to the prenatal programming effect on the conceived children. In general, this project will bring forth new knowledge on the prevention of obesity and our opportunities of breaking the intergenerational cycle of obesity and consequent susceptibility for NCDs, including CVD.