Severe nausea and vomiting in pregnancy: psychiatric and cognitive problems and brain structure in children

Abstract

Background: Two studies have suggested that severe prolonged nausea and vomiting during pregnancy is associated with emotional and behavioral problems in offspring, with smaller sample size and short-term follow-up. Moreover, little information is available on the role of the brain structure in the associations.

Methods: In a US-based cohort, the association was investigated between severe prolonged nausea and vomiting in pregnancy (extending after the second trimester and termed SNVP), psychiatric and cognitive problems, and brain morphology, from the Adolescent Brain Cognitive Development (ABCD) study, from 10,710 children aged 9-11 years. We validated the emotional including psychiatric findings using the Danish National Cohort Study with 2,092,897 participants.

Results: SNVP was significantly associated with emotional and psychiatric problems (t = 8.89, Cohen's d = 0.172, p = 6.9 × 10^-19) and reduced global cognitive performance (t = - 4.34, d = - 0.085, p = 1.4 × 10^-5) in children. SNVP was associated with low cortical area and volume, especially in the cingulate cortex, precuneus, and superior medial prefrontal cortex. These lower cortical areas and volumes significantly mediated the relation between SNVP and the psychiatric and cognitive problems in children. In the Danish National Cohort, severe nausea and vomiting in pregnancy were significantly associated with increased risks of behavioral and emotional disorders in children (hazard ratio, 1.24; 95% confidence interval, 1.16-1.33).

Conclusions: SNVP is strongly associated with psychiatric and cognitive problems in children, with mediation by brain structure. These associations highlight the clinical importance and potential benefits of the treatment of SNVP, which could reduce the risk of psychiatric disorder in the next generation.

Keywords: Cingulate cortex; Cognitive performance; Cortical structure; Nausea and vomiting; Precuneus; Psychiatric problems; Superior medial prefrontal cortex.

Fig. 1

a A histogram showing the differences in cognitive measurements of 10,710 children between the children of mothers with prolonged nausea and vomiting in pregnancy (SNVP) and controls. The Y axis is the cognitive score (from the abcd_tbss) and the error bar is the SEM. All cognitive measurements except for flanker inhibitory control and attention (nihtbx_flanker), executive function (nihtbx_cardsort), language (nihtbx_picture), and pattern comparison processing speed (nihtbx_pattern) were significantly lower in the children with mothers with NVP during pregnancy (FDR corrected, p < 0.05). b, c Histograms showing the differences in psychiatric and emotional measures in the children of mothers with SNVP and controls. The Y axis is the psychiatric score. All psychiatric measurements were significantly lower in the group with exposure to SNVP during pregnancy (FDR corrected, p < 0.05). *FDR corrected p < 0.05. Note: nihtbx_picvocab: NIH Toolbox Picture Vocabulary Test Age 3+ v2.0 Uncorrected Standard Score; nihtbx_flanker: NIH Toolbox Flanker Inhibitory Control and Attention Test Ages 8-11 v2.0 Uncorrected Standard Score; nihtbx_list: NIH Toolbox List Sorting Working Memory Test Age 7+ v2.0 Uncorrected Standard Score; nihtbx_cardsort: NIH Toolbox Dimensional Change Card Sort Test Ages 8-11 v2.0 Uncorrected Standard Score; nihtbx_pattern: NIH Toolbox Pattern Comparison Processing Speed Test Age 7+ v2.0 Uncorrected Standard Score; nihtbx_picture: NIH Toolbox Picture Sequence Memory Test Age 8+ Form A v2.0 Uncorrected Standard Score; nihtbx_reading: NIH Toolbox Oral Reading Recognition Test Age 3+ v2.0 Uncorrected Standard Score; nihtbx_fluidcomp: Cognition Fluid Composite Uncorrected Standard Score; nihtbx_cryst: Crystallized Composite Uncorrected Standard Score; nihtbx_totalcomp: Cognition Total Composite Score Uncorrected Standard Score; cbcl_scr_syn_anxdep: Anxious/Depressed CBCL Syndrome Scale; cbcl_scr_syn_withdep: Withdrawn/Depressed CBCL Syndrome Scale; cbcl_scr_syn_somatic: Somatic Complaints CBCL Syndrome Scale; cbcl_scr_syn_social: Social Problems CBCL Syndrome Scale; cbcl_scr_syn_attention: Attention Problems CBCL Syndrome Scale; cbcl_scr_syn_rulebreak: Rule-Breaking Behavior CBCL Syndrome Scale; cbcl_scr_syn_aggressive: Aggressive Behavior CBCL Syndrome Scale; cbcl_scr_syn_internal: Internalizing Problems CBCL Syndrome Scale; cbcl_scr_syn_external: Externalizing Problems CBCL Syndrome Scale; cbcl_scr_dsm5_depress: Depressive Problems CBCL DSM5 Scale; cbcl_scr_dsm5_anxdisord: Anxiety Problems CBCL DSM5 Scale; cbcl_scr_dsm5_somaticpr: Somatic Problems CBCL DSM5 Scale; cbcl_scr_dsm5_adhd: ADHD CBCL DSM5 Scale; cbcl_scr_dsm5_opposit: Oppositional Defiant Problems CBCL DSM5 Scale; cbcl_scr_dsm5_conduct: Conduct Problems CBCL DSM5 Scale; cbcl_scr_07_sct: Sluggish Cognitive Tempo (SCT) CBCL Scale2007 Scale; cbcl_scr_07_ocd: Obsessive-Compulsive Problems (OCD) CBCL Scale2007 Scale; cbcl_scr_07_stress: Stress Problems CBCL Scale2007 Scale

Fig. 2

a Brain regions with their cortical volume significantly lower in children whose mothers had SNVP during pregnancy (FDR corrected, p < 0.05). b Brain regions with their cortical area significantly lower in the children whose mothers had SNVP during pregnancy (FDR corrected, p < 0.05)

Fig. 3

Mediation analysis of the effects of SNVP in the mothers on cognition and psychiatric problems that are mediated by brain structure. a The mediation by cortical volume from the exposure to SNVP on psychiatric problems was significant (β = 0.176, p = 2.0 × 10⁻⁵). Path A: the effect of the exposure to SNVP on the mediator (cortical volume). Path B: the effect of the mediator (cortical volume) on the outcome (psychiatric problems). Path C shows that the regression coefficient (beta value) of the exposure to SNVP on the cognition was high when the cortical volume was not taken into account. The beta values show the regression coefficient of the effect of the independent variable (the exposure to SNVP) on the outcome (psychiatric problems). Path C′ indicates the direct effect of the exposure to SNVP on the outcome (psychiatric problems) controlling for the mediator (cortical volume). Path C′ shows some reduction in the regression coefficient when the effect of the cortical volume was taken into account. Path AB indicates the extent to which taking the cortical volume into account can explain the 3.7% effect of the cortical area, which is significant as noted above at p = 2.0 × 10⁻⁵. b The mediation by cortical area from the exposure to SNVP on psychiatric problems was significant (β = 0.172, p = 3.1 × 10⁻⁵). c The mediation by cortical volume from the exposure to SNVP on cognition was significant (β = − 0.122, p = 8.5 × 10⁻⁷). d The mediation by cortical area from the exposure to SNVP on cognition was significant (β = − 0.136, p = 8.0 × 10⁻⁷)