Authors
Qian Yang, M. Carolina Borges, Eleanor Sanderson, Kate Tilling, Deborah A Lawlor on behalf of the MR-PREG consortia

Institution(s)
MRC Integrative Epidemiology Unit, University of Bristol
Population Health Sciences, Bristol Medical School, University of Bristol
NIHR Bristol Biomedical Research Centre, University Hospitals Bristol NHS Foundation and University of Bristol

Introduction
Insomnia is common during pregnancy and has been linked with higher risks of adverse pregnancy and perinatal outcomes in previous observational studies using multivariable regressions. However, it remains unclear whether those associations are causal. We aim to test whether insomnia is causally associated with pregnancy loss, gestational diabetes, hypertensive disorders in pregnancy, perinatal depression, preterm birth and low/high offspring birthweight using two-sample Mendelian randomization (MR).

Methods
We randomly split individual-participant data from UK Biobank (N=208171), and conducted a two-sample MR study with 81 genetic variants for a lifelong difference in having insomnia (usually versus sometimes to never) to investigate its effects on outcomes. We compared MR results to those from multivariable regressions of insomnia in pregnancy in Avon Longitudinal Study of Parents and Children (N=11748). Funding sources include MRC, Wellcome Trust, NIH, ERC and BHF. Ethical approvals were received from NREC for both cohorts.

Results
In the main (inverse variance weighted) two-sample MR, insomnia related to higher risks of any pregnancy loss (odds ratio [OR] 1.44 95% CI: 1.07-1.93), miscarriage (OR 1.43, 95% CI: 1.01-2.02), perinatal depression (OR 3.33, 95% CI: 1.25-8.85) and low offspring birthweight (OR 2.96, 95% CI: 1.55-5.68). There was no evidence of effects with other outcomes. Sensitivity analyses suggested that results for perinatal depression and low offspring birthweight might have been biased by unbalanced horizontal pleiotropy. In multivariable regressions, insomnia at 18 weeks of gestation was associated with higher risks of pregnancy loss (OR 1.27, 95% CI 1.10-1.45), stillbirth (OR 2.19, 95% CI 1.18-4.07), miscarriage (OR 1.32, 95% CI 1.13-1.55) and perinatal depression (OR 3.17, 95% CI 1.13-1.55) after adjusting for potential confounders. Estimates were similar for insomnia at 32 weeks of gestation.

Conclusions
Results from MR and multivariable regression were in agreement, despite different sources of biases. Our findings suggest insomnia may increase the risks of pregnancy loss, perinatal depression and low offspring birthweight. Further replication of these findings in a large Norwegian birth cohort is planned.

References
1.Warland J, et al. Sleep Med Rev. 2018; 41:197-219. 2.Lawlor DA. Int J Epidemiol. 2016; 45: 908-15.
This research was conducted using the UK Biobank resource under application number 23938.