Associations between parental type 2 diabetes risk and offspring birthweight and placental weight: a survival analysis using the Walker cohort

Hypothesis: Low birth weight (BW) is associated with the development of type 2 diabetes. Genome wide analyses have identified a strong genetic component to this association, with many BW-associated loci also involved in glucose metabolism. We hypothesized that offspring BW and placental weight (PW) are correlated with parental type 2 diabetes risk, reflecting the inheritance of diabetes-risk alleles which also influence foetal growth.
Methods: The Walker cohort, a collection of birth records from Dundee, Scotland, from the 1950s-60s was used to test this hypothesis, linking BW and PW measurements to parental health outcomes. Using data from SCI-Diabetes and the National Death Registry, we obtained health records for over 20,000 Walker parents. We performed Fine-Gray survival analyses of parental type 2 diabetes risk with competing risk of death, and Cox regression analyses of risk of death, independently in the maternal and paternal datasets, modelled by their offspring´s BW and PW.
Results: We found significant associations between paternal type 2 diabetes risk and offspring BW (subdistribution hazard ratio (SHR): 0.92, CI: 0.87 - 0.97) and PW (SHR: 0.87, CI: 0.81 - 0.94). The association of maternal type 2 diabetes with offspring BW or PW was not significant. Offspring BW was also associated with risk of death in both mothers (HR: 0.91, CI: 0.89 – 0.93) and fathers (HR: 0.95, CI: 0.92 – 0.98), and offspring PW was associated with maternal mortality (HR: 1.08, CI: 1.04 – 1.13) when adjusted for BW.
Conclusions: We identified associations between offspring BW and paternal type 2 diabetes risk, most likely resulting from the independent effects of common type 2 diabetes-susceptibility alleles on foetal growth, as described by the foetal insulin hypothesis. Moreover, we identified novel associations between offspring PW and reduced paternal type 2 diabetes risk, a relationship which might also be caused by the inheritance of diabetes-predisposition variants, limiting foetal growth. We found differing associations between offspring BW and PW and parental risk of death. These results provide novel epidemiological support for the use of offspring BW and PW as predictors for future risk of type 2 diabetes and death in mothers and fathers.