Preconception TSH and Adverse Pregnancy Outcomes in China: A Nationwide Prospective Cohort Study

Background The relationship between maternal thyroid function and pregnancy outcomes remains controversial and the safe range of TSH is still unclear in women planning pregnancy. Methods This population-based cohort study enrolled Chinese women who became pregnant in 30 provinces from 2010 to 2012 from the National Free Preconception Checkups Project. The maternal TSH level within 6 months before pregnancy and different pregnancy outcomes were collected and analyzed using restricted cubic spline regression model for dose-response relationship and potential optimal cutoff values. Logistic regression was used to reveal the relationship between different TSH groups and the risk of adverse outcomes. Results Among 175 112 women, a J-shaped association was revealed between TSH and large for gestational age (LGA; P < 0.001). When TSH was lower than 1.27 or 0.91 mIU/L, lower TSH was associated with higher odds ratio of low birth weight (LBW; P = 0.003) or preterm delivery (P < 0.001). There was no significant association of preconception TSH with SGA, macrosomia, fetal anomalies, stillbirth, natural or induced abortion, and cesarean delivery. The range of TSH for odds ratio lower than 1.0 was within 0.91 to 1.82 mIU/L in dose-response association. Compared with TSH 0.91 to 1.82 mIU/L, TSH low (< 0.40 mIU/L and 0.40-0.90 mIU/L) and high (1.83-2.49 mIU/L, 2.50-3.99 mIU/L, and >4.00 mIU/L) were associated with higher risk of preterm delivery and LGA. There was no significant association between TSH groups and the risk of LBW except for TSH < 0.40 mIU/L. Conclusion Preconception TSH was associated with preterm delivery, LGA, and LBW. Preconception TSH had a bidirectional effect on LGA, indicating a potential mechanism regarding influence of TSH on birth weight. TSH within 0.91 to 1.82 mIU/L was the potential safe range for preconception women.