LAUSR

STUDY QUESTION What is the impact of BMI on uncomplicated pregnancies and healthy births in women who did or did not have medically assisted reproduction (MAR, i.e. ART or hormonal stimulation without manipulation of eggs or embryos) in the Flanders region (Belgium)? SUMMARY ANSWER Women with a higher BMI who use MAR are at the highest risk of pregnancy and birth complications. WHAT WE KNOW ALREADY Medically assisted reproduction (MAR) is used increasingly worldwide and is associated with increased risk of adverse perinatal outcomes. Obesity is also increasing globally and obese women are more likely to seek MAR since obesity is associated with infertility. When obese women undergo MAR, the risk of adverse outcomes may be enhanced but it is not clear to what extent. STUDY DESIGN, SIZE, DURATION We conducted a registry-based study using the data from the Study Centre for Perinatal epidemiology database for years 2009-2015, region of Flanders, Belgium. This included 428 336 women. PARTICIPANTS/MATERIALS, SETTING, METHODS The average age was 30.0 years (SD 4.78), 194 061 (45.31%) were nulliparous, and 6.3% (n = 26 971) conceived with MAR. We examined the association of BMI and MAR with the following composite primary outcomes: 'uncomplicated pregnancy and birth' and 'healthy baby'. We conducted Poisson regression and adjusted for maternal age, parity, gestational weight gain, smoking and previous caesarean section. MAIN RESULTS AND THE ROLE OF CHANCE In our study, 36.80% (n = 157 623) of women had an uncomplicated pregnancy and birth according to the definition used. The predicted probability of having an uncomplicated pregnancy and birth for women with a BMI of 25 kg/m2 who conceived spontaneously was 0.33 (0.32 to 0.35), while it was 0.28 (0.24 to 0.32) for women who used hormonal stimulation and 0.26 (0.22 to 0.29) for women who used IVF/ICSI. This probability reduced with increasing BMI category for both MAR and non-MAR users. For women with a BMI of 30 kg/m2, the predicted probability of having an uncomplicated pregnancy and birth was 0.28 (0.26 to 0.30) for women who conceived spontaneously, and 0.22 (0.16 to 0.29) and 0.20 (0.14 to 0.26) for women who used hormonal stimulation only or IVF/ICSI, respectively. The predicted probability of having a healthy baby for women with a BMI of 25 kg/m2 who conceived spontaneously was 0.92 (0.91 to 0.93), 0.89 (0.87 to 0.92) for women who used hormonal stimulation only and 0.85 (0.84 to 0.87) for women who used IVF/ICSI. LIMITATIONS, REASONS FOR CAUTION The database did not include data on socio-economic status, pre-pregnancy morbidities and paternal BMI. Subsequently, we could not adjust for these factors in the analysis. WIDER IMPLICATIONS OF THE FINDINGS Obese women who use MAR are at the highest risk of pregnancy and birth complications. This increase in interventions also has cost and resource implications which is relevant for funding policies. Weight loss interventions prior to MAR seem plausible but their (cost-) effectiveness needs urgent investigation. STUDY FUNDING/COMPETING INTEREST(S) F.W. received an Erasmus Plus training grant to visit A.B., L.A. and R.D. and conducted this study during this visit. The authors have no competing interest to declare. TRIAL REGISTRATION NUMBER N/A.