Strengths and limitations
This study was a prospective cohort study with a large sample size. Sleep information was collected before birth outcomes, so the results were credible because of small recall bias. In this study, the PSQI with high reliability and validity was used to assess the effects of sleep at different stages of pregnancy and sleep changes on birth outcomes, adding to the relevant research field. The study has some limitations. First, although this is a prospective cohort study, the sleep information of pregnant women over the past week was obtained from subjective reports, which may have had recall bias. Some studies have shown that pregnant women actually sleep about 30 minutes less than they subjectively report, so sleep duration during pregnancy may be overestimated. However, it was found that subjectively reported sleep data had a more significant effect on adverse birth outcomes than was assessed by objective methods41. When conditions permit, future studies can combine subjective reports with sleep assessed by more reliable devices such as polysomnography and wrist motion detectors. Secondly, although we adjusted socio-demographic characteristics, living habits, and health status, other residual confounding may still exist, such as fatigue, restless leg syndrome, sleep apnea, etc., which can be considered for inclusion in future studies. Moreover, the subjects included in this study only included pregnant women who went to Shuangliu Maternal and Child Health Hospital in Chengdu, China, and could not be extended to people in other areas, which could be further discussed by conducting multi-center studies in the future. This study did not collect sleep information during the third trimester of pregnancy, so it could not assess the relationship between sleep during the whole pregnancy and birth outcome. Future studies can add the third trimester to make the study results more comprehensive.
Interpretation, the association between sleep during pregnancy and adverse birth outcomes
This study found that short sleep duration during the second trimester was associated with PTB. Consistent with recent meta-study39. Some studies have found the same results in different pregnancies. Micheli et al. assessed sleep in the third trimester (28-32 weeks) of 1091 singleton pregnancies and found that women who slept less than 5 hours had an increased risk of preterm delivery39. Similarly, Li et al. assessed sleep duration in 1082 healthy women with single fetal pregnancies at 8-12, 24-28, and 32-36 weeks of gestation and found that participants with short sleep duration (≤7 h) at 32-36 weeks were more likely to report PTB23. However, other studies have reported different results. Previous case-control studies by Guendelman et al. also found no link between short sleep duration and PTB40. Two other large-sample prospective cohort studies assessing the relationship between sleep duration and poor birth outcomes in late pregnancy in Chinese women and throughout pregnancy in Japanese women also found no association between short sleep duration and the risk of PTB20,24. Different definitions of sleep duration, gestational age of concern, corrected covariates, and sample size may explain this controversial result. The mechanisms underlying the current association between short sleep duration and PTB are not clear, and some mechanisms may explain the association between lack of sleep and PTB. One possibility is the effect of excessive inflammatory reaction. Sleep deprivation will promote the increase of inflammatory cytokines such as interleukin-6 (IL-6) and IL-8, thereby stimulating the production of prostaglandins in pregnancy tissue, leading to cervical maturation and uterine contraction23,41.
Studies have reported that longer sleep duration (>9 hours or >10 hours) is significantly associated with impaired glucose tolerance, coronary heart disease, cardiovascular events, stroke, and mortality30,42. A study by Yang et al. reported an increased incidence of PTB in pregnant women who slept longer43. But, Kajeepeta et al. showed that women who reported long sleep duration and fatigue in the first 6 months of pregnancy had an increased risk of PTB, while women who reported long sleep duration (≥ 9 hours) and no fatigue had no statistically significant risk of PTB, and fatigue may be a new risk factor for PTB43. Notably, this study found that longer sleep duration during the second trimester was associated with a lower risk of PTB, and differences in study design and definition of long sleep duration may lead to conflicting findings. We did not find a mechanism to explain the protective effect of longer sleep duration on pregnant women. It may be that longer sleep duration counteracts the effects of fatigue. In conclusion, the results of this study need further verification. In addition, this study did not find an association between sleep duration in early pregnancy and PTB, which is consistent with the results of Li and Nakahara et al20,23. Data analysis in this study showed that, compared with the second trimester, women in the first trimester subjectively reported longer sleep duration. The difference in sleep duration between the first trimester and the second trimester may explain the relationship between sleep duration in different stages of pregnancy and PTB, and the lack of sleep information in some subjects may also be one of the reasons.
Analysis of the data in this study found that sleep quality during pregnancy was not associated with PTB. A study by Du et al. in China also reported consistent results20,23. Other findings suggest that poor sleep quality during pregnancy may be a risk factor for PTB. A small cohort study in the United States found that PSQI> 5 in early pregnancy was associated with PTB, with a 25% increase in the chance of PTB for every percentage point increase (OR: 1.25, 95% CI: 1.04, 1.50)44. However, this study only corrected for obstetric risk, income, and stress. A Chinese cohort with a sample size of 688 Li et al., after adjusting for prepregnancy weight and birth weight, found that women with poor sleep quality in the second and third trimesters had a four-fold (OR: 5.35, 95%CI: 2.10, 13.63) and two-fold (OR: 3.01, 95%CI: 1.26, 7.19) increased risk of PTB, respectively45. The results were inconsistent, possibly due to sample size, pregnancy, and adjusted confounding factors.
The study also did not find an association between sleep during pregnancy and LBW and SGA. Consistent with some research findings14,18,18,26. However, two large-sample cohort studies in China during the third trimester found that poor sleep quality (PSQI>5) was associated with LBW (OR: 1.50, 95%CI: 1.08, 2.08) 26 and sleep duration ≤ 7h was a risk factor for SGA (OR: 2.67, 95%CI: 1.18, 6.54) 2317. Another prospective study also reported that women with poor sleep quality or sleep deprivation (<7 hours vs. >9 hours) at 30 weeks gestation had lower baby weight47. We speculate that the controversial findings may be related to the study environment, sleep classification, and the focus on differences in sleep gestational age. The relationship between sleep during pregnancy and birth weight of newborn remains to be further verified.
To the best of our knowledge, there have been relatively few studies on nap duration and poor birth outcomes, focusing on birth weight. A Chinese birth cohort study recruiting 10,111 women found that women who reported napping >1 hour had a lower risk of LBW delivery compared to women who reported no napping (OR: 0.61, 95%CI: 0.44, 0.83)27. In contrast, the present study found that women who napped for more than 90 minutes had an increased risk of giving birth small for gestational age, but that association disappeared after controlling for potential risk factors. Differences in gestational age and number of outcomes may explain the difference in results. A similar study in Brazil also examined the relationship and found no correlation between nap duration during pregnancy and birth weight, although the sample size was only 176 and the outcome was birth weight z-score18. This study found no association between lunchtime sleep during pregnancy and other birth outcomes, and more evidence on the effect of napping on birth outcomes is needed.
Overall, the available research results are inconsistent, but some of the findings suggest that poor sleep leads to the possibility of adverse birth outcome cannot be ignored. Adverse birth outcomes are not only bad for the short-term physical health of newborns harmful effects and increased susceptibility to disease in adulthood. Identification of possible risk factors is helpful for pregnancy preparation, prevention, screening and early intervention during pregnancy, and will have a positive impact on the reduction of the incidence of adverse birth outcomes and good birth and good upbringing.