Population-based neuroimaging reveals traces of childbirth in the maternal brain

Abstract

1Pregnancy and childbirth involve maternal brain adaptations that promote attachment to and protection 2 of the newborn. Using brain imaging and machine learning, we provide evidence for a positive relation-3 ship between number of childbirths and a 'younger-looking' brain in 12,021 women, which could not be 4 explained by common genetic variation. The findings demonstrate that parity can be linked to brain 5 health later in life. 6 Page 2 of 23Population-based neuroimaging reveals traces of childbirth in the maternal brain -de Lange et. al., 2019 During pregnancy and postpartum, fundamental biological processes are instigated to support maternal 7 adaptation and ensure protection of the offspring [1]. In rodents, brain adaptations across pregnancy 8 and postpartum include altered neurogenesis in the dentate gyrus [2], and changes in volume, dendritic 9 morphology, and cell proliferation in the hippocampus [1,3]. In humans, reduction in total brain vol-10 ume has been observed during pregnancy, with reversion occurring within six months of parturition [4]. 11Regional changes in brain structure are evident during the postpartum period, with effects depending 12 on region and time since delivery [5][6][7][8]. While some maternal brain changes revert postpartum, others 13 extend well beyond this phase [1,[7][8][9] and may influence the course of neurobiological aging later in 14 life. Some regional grey matter changes have been found to endure for at least 2 years post-pregnancy 15 in humans [1], and aged parous rats have increased hippocampal long-term potentiation and show fewer 16 signs of brain aging [1,10]. In addition to the direct and indirect bodily and environmental adaptations 17 in response to pregnancy and child-rearing, such long-lasting effects on brain health in humans could 18 also reflect genetic pleiotropy, as reproductive behaviors are complex, heritable traits with a polygenic 19 architecture that partly overlaps with a range of other traits that influence brain-health trajectories [11]. 20Based on the evidence of long-lasting effects of parity on the maternal brain, we investigated struc-21 tural brain characteristics in 12,021 women from the UK Biobank, hypothesizing that women who had 22 given (live) birth (n = 9568) would show less evidence of brain aging compared to their nulliparous peers 23 (n = 2453). We used machine learning and brain age prediction to test I) if a classifier could identify 24 women as parous or nulliparous based on morphometric brain characteristics, and II) whether brain age 25 gap (estimated brain age − chronological age) differed between parous and nulliparous women. Mean 26 age (± SD) was 54.72 (7.29) years for the full sample; 55.23 (7.22) years for parous and 52.79 (7.23) 27 years for nulliparous women. To investigate the impact of number of childbirths, we tested for associa-28 tions between number of births and the probabilistic scores from the group classification and brain age 29 gap, respectively, in addition to comparing women who had given 1-2 births, 3-...

Women’s brain aging: effects of sex-hormone exposure, pregnancies, and genetic risk for Alzheimer’s disease

“…Higher estrogen levels during menopause were as-sociated with higher brain age gap in carriers and lower brain age gap in non-carriers. Two main conclusions can be drawn from the results: I) In light of our recent findings showing a link between parity and lower brain age relative to chronological age (49), estrogen exposure and parity may differentially relate to brain aging; II) The genotype-specific associations between age at onset and dosage of HRT and brain aging suggest that genetic factors are important to consider in clinical settings.…”

“…Parous women may also have shorter menstrual cycles and lower levels of estradiol than nulliparous women (44,45), indicating that age at first birth also influences lifetime estrogen exposure. Studies have demonstrated a protective effect of parity on brain aging (31,(46)(47)(48), and we recently showed lower brain age in parous compared to nulliparous women in the UK Biobank cohort (49). In the present paper, we investigate the association between estrogen exposure and brain aging beyond the effects of parity.…”

“…We analysed the association between endogenous as well as exogenous estrogen exposure and predicted brain age relative to chronological age (brain age gap) in 16,854 women with a mean age of 54.70 years ± 7.29 (standard deviation). Brain age prediction estimates an individual's apparent brain aging based on MRI-derived brain characteristics, and is becoming a widely applied marker of the neuroanatomical aging process (50)(51)(52)(53). We estimated brain age using the XG-Boost Python Package (54).…”

“…We estimated brain age using the XG-Boost Python Package (54). To test for associations between estimates of estrogen exposure and structures and modalities (55-57), we included separate brain age models based on T1-weighted MRI measures (full T1w model, cortical thickness, cortical and subcortical volume) (51,53,58,59), and diffusion-weighted MRI in a sub-sample of 9,829 women. Cumulative estrogen exposure was estimated by an index of cumulative estrogen exposure (ICEE) including age at menarche and menopause, time since menopause, body mass index (BMI), and duration of HRT use (7).…”

“…To examine the effect of APOE e4 genotype on the association between estrogen exposure and brain age gap, we performed follow-up analyses including interaction terms between APOE e4 genotype and estimates of estrogen exposure for each of the measures (see Materials and Methods for details). In order to investigate the link between estrogen exposure and brain aging beyond the effects of parity (49), all analyses were corrected for number of childbirths.…”