3.1 Baseline characteristics of pregnant women and their neonates
Table1 shows the baseline characteristics of pregnant women in Shunde. The mean ± SD for gestational week, preconception body mass index (BMI), current BMI were 21.6 ± 9.5wks, 21.2 ± 3.2 kg/m2, 23.4 ± 3.8 kg/m2, respectively. The median UIC of the pregnant women was 116 (64, 190) μg/L, indicating iodine deficiency in pregnant women in Shunde according to WHO-recommended criteria. A weak but no significant declined trend was found in UIC across different trimesters (P = 0.055). 63.9% of pregnant women had UIC < 150μg/L, 18.0% had UIC < 50μg/L, 24.8% were 50-99μg/L, 21.1% were 100-149μg/L, 21.0% had 150-249μg/L, and 15.1% were UIC ≥ 250μg/L (Figure 2).
The median maternal TSH of pregnant women was 1.09 (0.61, 1.72) mIU/L. An increasing tendency was found in the median maternal TSH across pregnancy (P < 0.001) (Table1).
The mean ± SD for delivery week was 38.6 ± 1.3wks (Table 2). Higher birth weight, birth length, birth head circumference was found in boys (all P < 0.001). The length-for-age (LFA) Z-scores was higher in girls (P < 0.001). The median neonatal TSH was 4.44 (3.02, 6.22) mIU/L, 40.4% of neonatal TSH ≥ 5.0mIU/L. No difference was found in neonatal TSH and other Z-scores between boys and girls (all P > 0.05).
3.2 Factors influencing physical growth of neonates.
As presented in Table 3, higher birth head circumference and HFA (head circumference-for-age) Z-scores were found in neonates with maternal age < 35yrs and given birth after 37wks (all P < 0.001). Compared with neonates with maternal UIC ranged 150-249μg/L, birth weight and WFA (weight-for-age) Z-scores were lower in those with maternal UIC within 100-149μg/L (both P<0.05). No significance was found between maternal TSH during pregnancy and physical growth in neonates (all P > 0.05). Birth length, weight, head circumference and corresponding Z-scores in term neonates were higher than preterm infants (P < 0.001). Birth length, birth weight and corresponding Z-scores were higher in neonates of multiparous women (parity ≥ 1). Both birth weight and WFA Z-scores were higher in neonates born via caesarean section (CS) (P=0.004, P=0.006). No significances were found between other factors with neonatal physical growth (all P>0.05).
Maternal age (< 35yrs or ≥ 35yrs), parity (0 or ≥ 1), current BMI, UIC, maternal TSH, neonatal gender (boys or girls), delivery week (≥ 37wks or < 37wks), delivery way (NVD, natural vaginal delivery or CS) were adjusted in multiple linear models to analyze the association with neonatal birth weight and WAF Z-scores (Table 4). No association of birth weight and WFA Z-scores of neonates were found in maternal age, TSH and UIC during pregnancy (all P > 0.05). Birth weight and WFA Z-scores were 18.9g and 0.040-points higher in neonates with current BMI of pregnant women increased per unit, respectively (P < 0.001). Compared with neonates of nulliparous women (parity = 0), birth weight in neonates of multiparous women increased 64.7g (P < 0.001) and WFA Z-scores showed a 0.136-point improvement (P < 0.001). And birth weight was 95.4g higher in boys (P < 0.001). Compared with term neonates, birth weight and WFA Z-scores of preterm infants decreased 732.5g (P < 0.001) and 1.545-point (P < 0.001), respectively. Birth weight in neonates born via CS was 41.1g higher than those via NVD (P = 0.008), and WFA Z-scores increased 0.086-point.
3.3 Factors influencing neonatal TSH
Higher neonatal TSH was found in neonates with maternal age < 35yrs, nulliparous women, term neonates or born via NVD (all P < 0.05). Compared with neonates with maternal UIC ranged 150-249μg/L, no changes was found in neonatal TSH among different UIC groups (P > 0.05) (Table 5). Neonatal TSH was positively and significant correlated with maternal TSH during pregnancy (r = 0.066, P = 0.001, data not shown). Neonates with TSH ≥ 5.0mIU/L were associated with high maternal TSH during pregnancy (1.14 vs 1.06 mIU/L, P=0.010). Parity=0, term delivery and given birth by natural vaginal were related to high rate of neonatal TSH ≥ 5.0mIU/L (all P < 0.05). No differences were found between neonatal TSH and current BMI during pregnancy and neonatal gender (all P > 0.05).
Maternal age (< 35yrs or ≥ 35yrs), parity (0 or ≥ 1), current BMI, UIC, maternal TSH, neonatal gender (boys or girls), delivery week (≥ 37wks or < 37wks), delivery way (NVD or CS) were adjusted in multiple linear models and logistics regression to analyze the association with neonatal TSH (Table 6). Maternal age, current BMI and neonatal gender were found no relationship both with neonatal TSH and neonatal TSH ≥ 5.0mIU/L (all P > 0.05). Neonatal TSH increased by 0.28mIU/L per unit in neonates with nulliparous mother (P = 0.005), 1.12mIU/L higher in term neonates (P < 0.001) and 0.44mIU/L higher in neonates born via NVD (P < 0.001), respectively. The risk of neonatal TSH ≥ 5.0mIU/L was 1.178 times higher in neonates of nulliparous mother than multiparous (95%CI: 1.001, 1.387, P = 0.049), 2.485 times higher in term neonates than those in preterm infants (CI: 1.582, 3.904, P < 0.001) and 1.444 times higher in neonates born via NVD than CS (95% CI: 1.218, 1.710, P < 0.001), respectively.
No correlation was found between maternal UIC during pregnancy and neonatal TSH and neonatal TSH ≥ 5.0mIU/L (all P > 0.05). Neonatal TSH increased by 0.18mIU/L with increasing maternal TSH per unit (P = 0.001). And the risk of neonatal TSH ≥ 5.0mIU/L was 1.013 folds positively associated with maternal TSH (95% CI: 1.012, 1.201, P = 0.026).