Background
Maternal obesity is becoming more prevalent in obstetrics and has been linked with pregnancy complications and perinatal outcomes. The gradient of association of increasing maternal obesity and pregnancy outcome is less well studied.

Aims
To determine the influence of an increasing gradient of obesity, categorised by the body mass index (BMI), on pregnancy outcomes and to determine the BMI thresholds at which pregnancy complications occur.

Materials and Methods
Secondary analysis of an observational study on pregnancy and obesity. The BMI at the first prenatal visit was grouped into BMI categories (<18.5, 18.5–24.9, 25–29.9, 30–34.9, 35–39.9, 40–44.9, and ≥45) and compared with the normal category (BMI 18.5–25) for pregnancy outcomes and adjusted for known cofounders. Results A total of 4,490 women were stratified into the pre‐pregnancy BMI categories: <18.5 (n = 276), 18.5–24.9 (n = 1965), 25–29.9 (n = 1072), 30–34.9 (n = 551), 35–39.9 (n = 317), 40–44.9 (n = 167), and ≥45 (n = 142). The maternal demographics were significantly different between BMI groups (P < 0.001). Compared to women with a normal BMI, different BMI thresholds convey an increased risk for specific pregnancy complications: BMI≥25 for gestational diabetes (P < 0.001), induction of labour (P < 0.001), caesarean delivery (P < 0.001) and large for gestational age neonate (P < 0.001); BMI≥30 for pre‐eclampsia (P < 0.001), wound infection (P = 0.001), shoulder dystocia (P < 0.001) and meconium (P = 0.006); BMI≥35 for urinary tract infection (P < 0.001) and postpartum haemorrhage (P < 0.001); BMI≥40 for endometritis (P < 0.001). Conclusions Body mass index thresholds exist at which pregnancy complications significantly increase and they vary depending on outcome ranging from BMI ≥25 to a BMI ≥40.