[A birth cohort study on the effect of chronotype during pregnancy on small for gestational age and the mediating effect of glucose-lipid metabolism].

Objective: To explore the relationship between maternal sleep time and the risk of small for gestational age (SGA), and to evaluate the role of glucose-lipid metabolism in the association. Methods: A total of 6 821 women who was second pregnancy were recruited from pregnancies consulted at Hefei First People's Hospital, Anhui Province Maternity & Child Health Hospital and the First Affiliated Hospital of Anhui Medical University from March 2015 to April 2019, and a face-to-face questionnaire survey was conducted to collect general demographic characteristics, dietary habits and routine lifestyles. Sleep information including bedtime, getup and sleep duration were reported by pregnant woman herself, and this survey as well as the third trimester of gestation. Pregnancy and birth outcomes were collected at delivery. A total of 5 488 mother-pairs with complete data were obtained in the final data. The non-linear relationship between chronotype and SGA risk was explored by restricted cubic spline regression model, and the role of glucose-lipid metabolism in the association between sleep midpoint and SGA was explored by using the mediating model based on bootstrap method. Results: The incidence of SGA was 8.4% (459/5 488) in eligible pregnant women. Compared with the pregnant women who went to bed before 21∶00, the risk of SGA of women who went to bed after 23∶00 am (OR=1.54, 95%CI: 1.01-2.34) was significantly higher in the multivariate logistic regression model. Additionally, the risk of SGA in pregnant women who got up after 8∶00 am was significantly higher than those women who got up before 8 o'clock (OR=1.31, 95%CI:1.05-1.62). However, the significant association between sleep duration and SGA was not found. In the restricted cubic spline regression, the risk of SGA was significantly increased from the specific midpoint of 02∶45 am (P<0.05). Moreover, mediation model showed that the negative effect of late sleep in the second trimester on SGA may be partially explained through glucose-lipid metabolism(all P<0.05). Conclusion: Maternal sleep status at the second trimester of gestation may be more susceptible to SGA. Lately sleep midpoint may be a potential independent risk factor for increased risk of SGA, and furtherly affect the occurrence of SGA by changing the level of glucose and lipid metabolism.