Adverse pregnancy outcomes and long-term stroke risks

This commentary refers to ‘Adverse pregnancy outcomes and long-term risk of stroke: a Swedish nationwide co-sibling study’, by C. Crump et al., https://doi.org/10.1093/eurheartj/ehaf366 and the discussion piece ‘Interpreting stroke risk after adverse pregnancy outcomes: Methodological considerations for sibling analyses and long-term follow-up’, by M. Hasebe and C.-Y. Su, https://doi.org/10.1093/eurheartj/ehaf625. We thank Dr Masashi Hasebe and Chen-Yang Su for their interest in our study on long-term stroke risks in women with adverse pregnancy outcomes. Our findings in a large national cohort show that women with any of the five major adverse pregnancy outcomes (preterm delivery, small for gestational age, preeclampsia, other hypertensive disorders, or gestational diabetes) had significantly increased risks of stroke that remained elevated up to 46 years after delivery.1 This study builds on our prior research that showed long-term increased risks of stroke after preterm birth in both the mothers2 and offspring.3 Our analyses were adjusted for multiple covariates that are potential confounders of associations between adverse pregnancy outcomes and stroke risk, including maternal age, socio-economic factors, biometric factors, prior cardiometabolic and sleep disorders, and atrial fibrillation. In addition, we conducted co-sibling analyses to control for unmeasured shared genetic and early-life environmental factors. For the five adverse pregnancy outcomes examined, the number of exposure-discordant sets of siblings ranged from 8965 to 99 094. The results showed moderate (up to ∼50%) attenuation of certain risk estimates and no attenuation of others. Co-sibling analyses do not exclude residual confounding by genetic or environmental factors. Nonetheless, they are a valuable approach for controlling for unmeasured shared familial factors beyond traditional covariate adjustment.