Abstract
BACKGROUND: Twin pregnancies carry perinatal risks. Gestational weight gain (GWG) is modifiable, but Institute of Medicine (IOM) twin recommendations may not suit Chinese women. METHOD: A retrospective cohort study included 3109 twin pregnancies delivered at Northwest Women's and Children's Hospital in Xi'an, China, during 2018-2022. Pre-pregnancy BMI was classified according to WHO criteria, with Chinese criteria used in sensitivity analyses. Optimal BMI-specific GWG ranges were defined as the interquartile range among low-risk pregnancies. GWG adequacy (below/within/above) was evaluated against study-derived and IOM-recommended ranges. Associations were estimated using multivariable generalized estimating equations, with threshold-effect and smooth-curve analyses to assess nonlinearity. RESULTS: Optimal GWG ranges were: underweight, 15.91-23.68 kg; normal, 14.80-21.46 kg; overweight, 12.95-19.98 kg; obese, 7.03-17.76 kg. GWG below range increased risks of moderate preterm birth (MPTB) (OR 1.62, 95% CI 1.29-2.04), very preterm birth (VPTB) (OR 1.90, 95% CI 1.34-2.70), low birth weight (LBW) (OR 1.87, 95% CI 1.61-2.17), and small for gestational age (SGA) (OR 1.31, 95% CI 1.08, 1.59). Excessive GWG raised large for gestational age (LGA) risk (OR 1.98, 95% CI 1.59, 2.46) but lowered LBW (OR 0.69, 95% CI 0.59-0.81) and SGA (OR 0.75, 95% CI 0.60, 0.95). Nonlinear analyses showed U-shaped risks for MPTB, VPTB, and Apgar ≤ 7; inverted U-shaped for gestational age and birth weight; J-shaped for LGA. CONCLUSIONS: BMI-specific GWG recommendations derived from this large Chinese twin cohort are lower than IOM values. Both inadequate and excessive GWG increase adverse perinatal risks.