Emma Bradley 
University of Cambridge, Cambridge University School of Clinical Medicine, Cambridge, UK 
Correspondence to: Emma Bradley, eb928@cam.ac.uk
Additional information
- Ethical approval: N/a
- Consent: N/a
- Funding: No industry funding
- Conflicts of interest: N/a
- Author contribution: Emma Bradley – Conceptualization, Writing – original draft, review and editing
- Guarantor: Emma Bradley
- Provenance and peer-review: Unsolicited and externally peer-reviewed
- Data availability statement: N/a
Keywords: Prenatal illicit drug exposure, lamotrigine pharmacokinetics in pregnancy, endometrial transcriptomics after miscarriage, recurrent miscarriage inflammatory dysregulation, maternal mental health comorbidity.
Peer Review
Received: 24 March 2026
Last revised: 01 April 2026
Accepted: 03 April 2026
Version accepted: 2
Published: 11 April 2026
Abstract
Evaluating the impacts of maternal illicit substance use during pregnancy on outcomes
Ellie Buttery1 and Miss Michelle Kemp2
University of Cambridge, Cambridge, United Kingdom
Rosie Hospital, CUH
Ethical approval: N/a
Consent: N/a
Funding: No industry funding
Conflicts of interest: N/a
Author contribution: as per abstract – Conceptualization, Writing – original draft, review and editing
Guarantor: Emma Bradley
Provenance and peer-review: Unsolicited and externally peer-reviewed
Data availability statement: N/a
Background: Illicit substance use during pregnancy poses serious risks for both mother and child, linking to adverse pre- and peri-natal outcomes and potential long-term cognitive, behavioural, and psychiatric effects in exposed children. This study examines the maternal and neonatal impacts of illicit substance use on pregnancy outcomes.
Methods: Patient records from a tertiary hospital (2023-2024) were identified using ICD-10 codes for -pregnancy complications, delivery outcomes, and substance use (including cannabis). Eligible records were manually reviewed, and data were extracted on demographics, clinical variables, substance use patterns, mental health, smoking, domestic violence, specialist referrals, pregnancy and delivery outcomes, and social care involvement.
Results: Thirty-nine women aged 18–44 years (mean 28.9, SD 6.7) were included. Cannabis was the most common substance (n = 31), with smaller numbers using cocaine, opioids, alcohol, or benzodiazepines. Mental health comorbidity was frequent – particularly anxiety/depression (n = 21), PTSD (n = 10), and personality disorders (n = 8). Most engaged with specialist services (n = 18) and maternity liaison (n = 26). Smoking occurred in 27 pregnancies. Complications included preterm birth (n = 4), placental disorders (n = 5), intrauterine growth restriction (n = 5), and antepartum foetal demise (n = 2). Deliveries were mainly spontaneous vaginal (n = 47), with 18 caesareans and 8 instrumental births. Social care removal occurred in 8 cases.
Discussion: Complication rates were high: 5% of pregnancies ended in stillbirth and 13% in other adverse outcomes, reflecting risks from drug use and associated lifestyles. Common mental health comorbidity highlights a key intervention point. Strengthening mental health and addiction services may reduce harm to mothers and neonates and improve outcomes.
Keywords: Obstetrics & Gynaecology, Maternal substance use, Pregnancy outcomes, Neonatal outcomes.
Miscarriage Hijacks Transcriptional Dynamics in the Endometrium
S. S. Mousavi1, Mireia Taus Nebot2 and Jan Brosens2
1. Department of Engineering, Robinson College, University of Cambridge
2. Division of Biomedicine, Warwick Medical School, University of Warwick
Ethical Approval: This study used previously collected, anonymised data obtained during a prior study conducted in accordance with ethical standards.
Consent: Informed consent was obtained in the original study or was waived where appropriate.
Data Availability: Data are not publicly available due to ethical restrictions.
Funding: N/a
Conflicts of Interest: The authors declare no conflicts of interest.
Author Contribution: S. S. Mousavi planned, formulated, and conducted majority of the data analyses, and presentation. M. Taus Nebot (Middle coauthor) collected and preprocessed the original data provided technical directions on further preprocessing of the data and clarified the problem at hand suggesting potential solutions. J. Brosens (Senior coauthor) collected the original biopsies and provided clinical background, with regular insight discussion and guidance on the interpretation of the results.
Miscarriage is a common mechanism of pregnancy loss during the first 20 weeks of gestation. Epidemiological studies highlighted a significant reduction in the probability of live birth per menstrual cycle following a single miscarriage, regardless of birth, prompting investigation into the mechanism and effects of miscarriage in the endometrium (Kolte et al., 2021). This study investigates the effect of miscarriage on transcriptional dynamics within and between cycles. Paired biopsy transcriptomic data from 161 patients for two menstrual cycles were normalised for tissue heterogeneity and post-ovulation LH+ time variation. Exploratory analysis of intracyclical and intercyclical dynamics were conducted using dimension reduction, regression, and network-based analysis.
It is found that five or more miscarriages can distort the individual uniqueness of the transcription profile. No significant trends in intracyclical gene expression with miscarriage was seen. Intercyclical change in expression of transcripts involved in inflammatory pathways had greater variance with increased miscarriage whilst the -opposite was seen for decidualisation. Consistent changes in general structure of gene networks were observed with increasing miscarriage. Pathway-level analysis indicated inflammatory genes as the only pathway enriched at high module switch rates. Differential coexpression analysis highlighted correlation of TRAJ and U6snRNA, SPRR2D and AHI1 becoming significantly weaker with increasing miscarriage.
Miscarriage establishes genetic memory which potentially leads to reduction in control of inflammatory pathways while in turn making decidualisation more rigid, it less responsive to the dynamic endometrium. Methods were developed here for pinpointing specific transcripts responsible for the effect of miscarriages which should be applied to larger datasets.
Acknowledgements
Prof. Jan Brosens and Mireia Taus Nebot as supervisors, providing the data and direction; Prof. Sascha Ott and Dr Andrew Sharkey for their advice on the project.
Reference
Kolte, A.M., Westergaard, D., Lidegaard, Ø., Brunak, S. & Nielsen, H.S. (2021). Chance of live birth: a nationwide, registry-based cohort study. Human Reproduction. 36 (4), 1065–1073. https://doi.org/10.1093/humrep/deaa326.
