Evaluating causal associations of chronotype with pregnancy and perinatal outcomes and its interactions with insomnia and sleep duration: a mendelian randomization study

Abstract

Background: Observational studies suggested chronotype was associated with pregnancy and perinatal outcomes. Whether these associations are causal is unclear. Our aims are to use Mendelian randomization (MR) to explore (1) associations of evening preference with stillbirth, miscarriage, gestational diabetes, hypertensive disorders of pregnancy, perinatal depression, preterm birth and offspring birthweight; and (2) differences in associations of insomnia and sleep duration with those outcomes between chronotype preferences.

Methods: We conducted two-sample MR using 105 genetic variants reported in a genome-wide association study (N = 248,100) to instrument for lifelong predisposition to evening- versus morning-preference. We generated variant-outcome associations in European ancestry women from UK Biobank (UKB, N = 176,897), Avon Longitudinal Study of Parents and Children (ALSPAC, N = 6826), Born in Bradford (BiB, N = 2940) and the Norwegian Mother, Father and Child Cohort Study (MoBa, N = 57,430), and extracted equivalent associations from FinnGen (N = 190,879). We used inverse variance weighted (IVW) as main analysis, with weighted median and MR-Egger as sensitivity analyses. Relying on the individual participant data from UKB, ALSPAC, BiB and MoBa, we also conducted IVW analyses of insomnia and sleep duration on the pregnancy and perinatal outcomes, stratified by genetically predicted chronotypes.

Results: In IVW and sensitivity analyses, we did not find robust evidence of associations of chronotype with the outcomes. Insomnia was associated with a higher risk of preterm birth among evening preference women (odds ratio 1.61, 95% confidence interval: 1.17, 2.21), but not among morning preference women (odds ratio 0.87, 95% confidence interval: 0.64, 1.18), with an interaction P-value = 0.01. There was no evidence of interactions between insomnia and chronotype on other outcomes, or between sleep duration and chronotype on any outcomes.

Conclusions: This study raises the possibility of a higher risk of preterm birth among women with insomnia who also have an evening preference. Our findings warrant replications due to imprecise estimates.

Keywords: Chronotype; Mendelian randomization; Pregnancy and perinatal outcomes; Sleep.