Background The aim of the current study was to investigate whether exposure to prenatal maternal anxiety and depression influenced later offspring glucose, lipid and inflammatory markers via intrauterine mechanisms.

Methods Data from a prospective birth cohort based in the South West of England were used. Our analysis included 2839 mother-children duos and 2361 father-child duos for outcomes assessed at mean age 9.9 years (non-fasting cholesterol, triglycerides, low density and high density lipoprotein cholesterol (LDLc and HDLc), C-reactive protein (CRP) and interleukin 6 (IL–6) and 2011 and 1726 parent-child duos for outcomes at mean age 15.4 years (fasting glucose, insulin, lipids and CRP). We compared associations of maternal exposures with offspring outcomes to those of the same paternal exposures with offspring outcomes. The rationale for this comparison was that if maternal depression/anxiety influenced offspring outcomes via intrauterine mechanisms we would expect stronger maternal compared with paternal associations. We also examined whether any association of exposures during pregnancy reflected a postnatal effect, with persistence of depression/anxiety into the postnatal period.

Results Maternal anxiety at 18 and 32 weeks gestation, and maternal depression at 32 weeks gestation were associated with increased CRP in children at 9.9 years (mean difference (95% CI): 0.031 (0.005 to 0.057), 0.030 (0.004 to 0.056), and 0.021 (0.003 to 0.040) respectively), but not at 15.4 years. These associations remained when adjusting for potential confounders (maternal age, ethnicity, pre-pregnancy BMI, parity, social class, smoking and alcohol consumption). Paternal anxiety and depression (measured at 18 weeks gestation)were also associated with increased CRP in children at 9.9 years (mean difference (95% CI): 0.039 (0.003 to 0.076) and 0.026 (0–0.052) respectively), but not at 15.4 years. The magnitudes of the paternal associations were similar to those seen in mothers. Maternal and paternal postnatal depression/anxiety symptoms were also associated with offspring CRP at age 9.9 and appeared to explain much of the antenatal association.

There were no consistent associations between maternal or paternal anxiety or depression during the antenatal or postnatal periods and any of offspring glucose, insulin, IL-6 or lipids at either age.

Conclusion We have found evidence of a relationship between maternal and paternal anxiety and depression during pregnancy and CRP levels in childhood, which does not persist to adolescence. Our results suggest that these associations are unlikely to be explained by intrauterine mechanisms and may be explained by shared familial confounding or postnatal effects.