Long-term effects of prenatal infection on the human brain: a prospective multimodal neuroimaging study

Abstract

There is convincing evidence from rodent studies suggesting that prenatal infections affect the offspring's brain, but evidence in humans is limited. Here, we assessed the occurrence of common infections during each trimester of pregnancy and examined associations with brain outcomes in adolescent offspring. Our study was embedded in the Generation R Study, a large-scale sociodemographically diverse prospective birth cohort. We included 1094 mother-child dyads and investigated brain morphology (structural MRI), white matter microstructure (DTI), and functional connectivity (functional MRI), as outcomes at the age of 14. We focused on both global and focal regions. To define prenatal infections, we composed a score based on the number and type of infections during each trimester of pregnancy. Models were adjusted for several confounders. We found that prenatal infection was negatively associated with cerebral white matter volume (B = -0.069, 95% CI -0.123 to -0.015, p = 0.011), and we found an association between higher prenatal infection scores and smaller volumes of several frontotemporal regions of the brain. After multiple testing correction, we only observed an association between prenatal infections and the caudal anterior cingulate volume (B = -0.104, 95% CI -0.164 to -0.045, p < 0.001). We did not observe effects of prenatal infection on other measures of adolescent brain morphology, white matter microstructure, or functional connectivity, which is reassuring. Our results show potential regions of interest in the brain for future studies; data on the effect of severe prenatal infections on the offspring's brain in humans are needed.

Fig. 1. Flowchart study population with global and focal brain outcomes for all three modalities (sMRI, DTI, and fMRI).

The exclusion criteria are indicated with red. The global measures are indicated with orange and the focal measures are indicated with green. Global sMRI measures include volumes of total brain, lateral ventricles, cerebellum, cortical gray matter, and cerebral white matter. Focal sMRI measures include all the brain regions from the Desikan-Killany atlas, and subcortical regions such as the brain stem, caudate, hippocampus, amygdala, putamen, amygdala, thalamus, accumbens, and corpus callosum (posterior/ mid posterior/ central/ mid anterior/ anterior). Global DTI measures include global mean diffusivity and global fractional anisotropy. Focal DTI measures include the fractional anisotropy (FA) and mean diffusivity (MD) of seven previously described white matter tracts. Global fMRI measures include global graph theory measures, such as global efficiency, modularity, and characteristic path length. Focal fMRI measures include within and between functional connectivity matrices for thirteen previously described networks.

Fig. 2. Prenatal infection and significant (p_uncorrected < 0.05) adolescent brain morphology regions.

A–H shows the regression plots for all significant global and focal regions including the effect estimates. To distinguish between global and focal regions, the effects in global regions are indicated with a red line and the effects in focal regions are indicated with a blue line. Below each graph the effect size, confidence interval and p-value are noted.

Fig. 3. Results of all cortical brain regions.

The starred regions entail the significant cortical regions, namely the volumes of entorhinal, caudal anterior cingulate, lateral occipital, parahippocampal, pars orbitalis, rostral middle frontal, and transverse temporal.