Subjects
Pregnant NW (BMI 18.5–24.9 kg/m2) and OB (BMI ≥30 kg/m2), aged 20–45 years, were recruited from six antenatal health units in the Gothenburg area as part of the Pregnancy Obesity Nutrition and Child Health (PONCH) study, as described22. Exclusion criteria were non-European descent, diabetes mellitus (type 1, type 2, or gestational) or other chronic diseases, pregnancy-related complications, use of tobacco or neuroleptic drugs, and vegetarianism or veganism. The women came to the Sahlgrenska University Hospital for study visits once each trimester (weeks 8–12, 24–26, and 35–37; T1, T2, and T3, respectively) during pregnancy and at 6, 12, and 18 months (± 10 days) after pregnancy. Gestational age at study visits were on average 81 ± 8 days at T1, 175 ± 8 days at T2 and 253 ± 7 days at T3, and did not differ between NW and OB (p >0.4).
GDF15 was measured in a subset of the total 132 NW and 50 OB that enlisted in PONCH. Subjects for the present study were selected from those who had complete serum samples and body composition measurements from T1 or T2 until 18 months after pregnancy. Of 71 NW who had complete visits, 38 were randomly selected. Selected NW did not differ from non-selected NW in BMI, body composition, or homeostatic model assessment of insulin resistance (HOMA-IR). All OB who had the maximal number of visits were selected. The number of subjects for NW after pregnancy was matched with OB (n=30 after pregnancy). A few subjects were lacking in T2 samples. This resulted in differences in the numbers for visits at T1 (38 NW and 28 OB), T2 (30 NW and 29 OB), and T3 (38 NW and 35 OB) and at 6 months (30 NW and 30 OB), 12 months (30 NW and 24 OB), and 18 months (30 NW and 11 OB) postpartum. The reasons given for postpartum drop-outs in the OB group were lack of time, moving to a different part of Sweden, or a new pregnancy.