Rationale This study assessed the associations between pregnancy loss (miscarriage, stillbirth and abortion) and future maternal cardiovascular diseases (CVD, myocardial infarction (MI), stroke, ischaemic stroke, intracerebral haemorrhage and subarachnoid haemorrhage) in women in the UK Biobank from 2006–2016.

Cardiovascular diseases (CVDs) are a high priority public health issue, presenting a high disease burden. Pregnancy has been described as a stress test for women’s bodies, where complications with pregnancy, including pregnancy loss, may be an indicator of increased risk of future CVDs. However, findings regarding associations between pregnancy loss and future CVDs are mixed, with a scarcity of UK studies. Furthering understanding of female specific risk factors has the potential to increase the clinical utility of risk prediction and public health prevention.

Methods A prospective cohort study was conducted, including women in the UK Biobank cohort with self-reported exposure status recorded at baseline (n=246,124), with a median follow up time of 7.03 years (IQR 6.36–7.75) and a total of 1,762,729 person-years. Cardiovascular outcomes were ascertained through linkage of participant healthcare records and UK mortality databases. Cox proportional hazards models were used to ascertain Hazard ratios (HR) and 95% confidence intervals (CI). Patient and participant involvement (PPI) was included to inform understanding of public health impacts of study findings.

Results Within the study cohort, 79,482 (32.29%) women had experienced one or more pregnancy losses. During follow up 2,567 (1.04%) women developed CVD. A history of one or more stillbirths was associated with an increased risk of stroke (Adjusted HR 1.55, 95%CI 1.06–2.27), in particular subarachnoid haemorrhage (Adjusted HR: 1.52, 95%CI 0.62–3.74). Miscarriage was not consistently associated with an increased risk of any cardiovascular outcome and no association between a history of therapeutic abortion and increased risk of CVD was demonstrated.

Conclusion Analyses of the relationship between pregnancy loss and subtypes of CVD indicated that the direction and magnitude of associations were not universal in women in the UK Biobank or across populations. There is evidence of association between some types of pregnancy loss are associated with an increased risk of CVDs, though further research into associations with haemorrhagic stroke, and the associations between abortion and CVD, are needed. History of miscarriage and/or stillbirth should be considered for inclusion in cardiovascular risk assessment tools, potential pooled with other pregnancy complications, to identify and target support for women at increased risk.