Association of Prenatal Maternal Anxiety With Fetal Regional Brain Connectivity | Anxiety Disorders | JAMA Network Open | JAMA Network
[Skip to Navigation]
Sign In
Figure.  Associations Between Resting State Functional Connectivity and Behavior in the Fetal Brain
Associations Between Resting State Functional Connectivity and Behavior in the Fetal Brain

Fetal blood oxygen level dependent signals were measured from 100 regions of interest (ROIs); estimated locations on the fetal brain surface are shown in the eFigure in the Supplement. A, ROIs were grouped into 14 modules. Blue spheres in brains are the locations of participating ROIs per module. Red lines indicate negative ROI-ROI correlations; green lines, positive correlations; and blue lines, 50% positive. Thicker lines indicate more ROI-ROI connections per module pair. Asterisks indicate network interactions that overlap for trait and state anxiety. B, Significant individual connections between modules pairs are shown as well as whether associations with behavior were positive (+) or negative (-). B indicates bilateral; L, left; and R, right.

Table 1.  Maternal and Fetal Clinical Characteristics
Maternal and Fetal Clinical Characteristics
Table 2.  Multivariate Distance Matrix Regression of Whole-Brain Connections
Multivariate Distance Matrix Regression of Whole-Brain Connections
Table 3.  Strongest Positive and Negative Correlations Between Resting State Functional Connectivity and STAI Score
Strongest Positive and Negative Correlations Between Resting State Functional Connectivity and STAI Score
Table 4.  Strongest Positive and Negative Correlations Between Resting State Functional Connectivity and SSAI Score
Strongest Positive and Negative Correlations Between Resting State Functional Connectivity and SSAI Score
1.
Fairbrother  N, Young  AH, Janssen  P, Antony  MM, Tucker  E.  Depression and anxiety during the perinatal period.   BMC Psychiatry. 2015;15:206. doi:10.1186/s12888-015-0526-6 PubMedGoogle ScholarCrossref
2.
Rees  S, Channon  S, Waters  CS.  The impact of maternal prenatal and postnatal anxiety on children’s emotional problems: a systematic review.   Eur Child Adolesc Psychiatry. 2019;28(2):257-280. doi:10.1007/s00787-018-1173-5 PubMedGoogle ScholarCrossref
3.
Gavin  NI, Gaynes  BN, Lohr  KN, Meltzer-Brody  S, Gartlehner  G, Swinson  T.  Perinatal depression: a systematic review of prevalence and incidence.   Obstet Gynecol. 2005;106(5 Pt 1):1071-1083. doi:10.1097/01.AOG.0000183597.31630.db PubMedGoogle ScholarCrossref
4.
Kim  DR, Bale  TL, Epperson  CN.  Prenatal programming of mental illness: current understanding of relationship and mechanisms.   Curr Psychiatry Rep. 2015;17(2):5. doi:10.1007/s11920-014-0546-9 PubMedGoogle ScholarCrossref
5.
O’Connor  TG, Heron  J, Glover  V; Alspac Study Team.  Antenatal anxiety predicts child behavioral/emotional problems independently of postnatal depression.   J Am Acad Child Adolesc Psychiatry. 2002;41(12):1470-1477. doi:10.1097/00004583-200212000-00019 PubMedGoogle ScholarCrossref
6.
Beversdorf  DQ, Manning  SE, Hillier  A,  et al.  Timing of prenatal stressors and autism.   J Autism Dev Disord. 2005;35(4):471-478. doi:10.1007/s10803-005-5037-8 PubMedGoogle ScholarCrossref
7.
Walder  DJ, Laplante  DP, Sousa-Pires  A, Veru  F, Brunet  A, King  S.  Prenatal maternal stress predicts autism traits in 6½ year-old children: Project Ice Storm.   Psychiatry Res. 2014;219(2):353-360. doi:10.1016/j.psychres.2014.04.034 PubMedGoogle ScholarCrossref
8.
Weinstock  M.  Alterations induced by gestational stress in brain morphology and behaviour of the offspring.   Prog Neurobiol. 2001;65(5):427-451. doi:10.1016/S0301-0082(01)00018-1 PubMedGoogle ScholarCrossref
9.
Huizink  AC, Mulder  EJH, Buitelaar  JK.  Prenatal stress and risk for psychopathology: specific effects or induction of general susceptibility?   Psychol Bull. 2004;130(1):115-142. doi:10.1037/0033-2909.130.1.115 PubMedGoogle ScholarCrossref
10.
Davis  EP, Snidman  N, Wadhwa  PD, Glynn  LM, Schetter  CD, Sandman  CA.  Prenatal maternal anxiety and depression predict negative behavioral reactivity in infancy.   Infancy. 2004;6(3):319-331. doi:10.1207/s15327078in0603_1Google ScholarCrossref
11.
Brouwers  EPM, van Baar  AL, Pop  VJM.  Maternal anxiety during pregnancy and subsequent infant development.   Infant Behav Dev. 2001;24(1):95-106. doi:10.1016/S0163-6383(01)00062-5Google ScholarCrossref
12.
Laplante  DP, Barr  RG, Brunet  A,  et al.  Stress during pregnancy affects general intellectual and language functioning in human toddlers.   Pediatr Res. 2004;56(3):400-410. doi:10.1203/01.PDR.0000136281.34035.44 PubMedGoogle ScholarCrossref
13.
Sandman  CA, Buss  C, Head  K, Davis  EP.  Fetal exposure to maternal depressive symptoms is associated with cortical thickness in late childhood.   Biol Psychiatry. 2015;77(4):324-334. doi:10.1016/j.biopsych.2014.06.025 PubMedGoogle ScholarCrossref
14.
Lebel  C, Walton  M, Letourneau  N, Giesbrecht  GF, Kaplan  BJ, Dewey  D.  Prepartum and postpartum maternal depressive symptoms are related to children’s brain structure in preschool.   Biol Psychiatry. 2016;80(11):859-868. doi:10.1016/j.biopsych.2015.12.004 PubMedGoogle ScholarCrossref
15.
Qiu  A, Rifkin-Graboi  A, Chen  H,  et al.  Maternal anxiety and infants’ hippocampal development: timing matters.   Transl Psychiatry. 2013;3:e306. doi:10.1038/tp.2013.79 PubMedGoogle Scholar
16.
Jones  SL, Dufoix  R, Laplante  DP,  et al.  Larger amygdala volume mediates the association between prenatal maternal stress and higher levels of externalizing behaviors: sex specific effects in Project Ice Storm.   Front Hum Neurosci. 2019;13:144. doi:10.3389/fnhum.2019.00144 PubMedGoogle ScholarCrossref
17.
Acosta  H, Tuulari  JJ, Scheinin  NM,  et al.  Maternal pregnancy-related anxiety is associated with sexually dimorphic alterations in amygdala volume in 4-year-old children.   Front Behav Neurosci. 2019;13:175. doi:10.3389/fnbeh.2019.00175 PubMedGoogle ScholarCrossref
18.
Wen  DJ, Soe  NN, Sim  LW,  et al.  Infant frontal EEG asymmetry in relation with postnatal maternal depression and parenting behavior.   Transl Psychiatry. 2017;7(3):e1057. doi:10.1038/tp.2017.28 PubMedGoogle Scholar
19.
Field  T, Diego  M.  Maternal depression effects on infant frontal EEG asymmetry.   Int J Neurosci. 2008;118(8):1081-1108. doi:10.1080/00207450701769067 PubMedGoogle ScholarCrossref
20.
Sarkar  S, Craig  MC, Dell’Acqua  F,  et al.  Prenatal stress and limbic-prefrontal white matter microstructure in children aged 6-9 years: a preliminary diffusion tensor imaging study.   World J Biol Psychiatry. 2014;15(4):346-352. doi:10.3109/15622975.2014.903336 PubMedGoogle ScholarCrossref
21.
Rifkin-Graboi  A, Bai  J, Chen  H,  et al.  Prenatal maternal depression associates with microstructure of right amygdala in neonates at birth.   Biol Psychiatry. 2013;74(11):837-844. doi:10.1016/j.biopsych.2013.06.019 PubMedGoogle ScholarCrossref
22.
Scheinost  D, Kwon  SH, Lacadie  C,  et al.  Prenatal stress alters amygdala functional connectivity in preterm neonates.   Neuroimage Clin. 2016;12:381-388. doi:10.1016/j.nicl.2016.08.010 PubMedGoogle ScholarCrossref
23.
Scheinost  D, Sinha  R, Cross  SN,  et al.  Does prenatal stress alter the developing connectome?   Pediatr Res. 2017;81(1-2):214-226. doi:10.1038/pr.2016.197 PubMedGoogle ScholarCrossref
24.
Wu  Y, Lu  Y-C, Jacobs  M,  et al.  Association of prenatal maternal psychological distress with fetal brain growth, metabolism, and cortical maturation.   JAMA Netw Open. 2020;3(1):e1919940. doi:10.1001/jamanetworkopen.2019.19940 PubMedGoogle Scholar
25.
Middleton  H, Shaw  I.  Distinguishing mental illness in primary care. We need to separate proper syndromes from generalised distress.   BMJ. 2000;320(7247):1420-1421. doi:10.1136/bmj.320.7247.1420 PubMedGoogle ScholarCrossref
26.
Cohen  S, Kamarck  T, Mermelstein  R.  A global measure of perceived stress.   J Health Soc Behav. 1983;24(4):385-396. doi:10.2307/2136404 PubMedGoogle ScholarCrossref
27.
Cox  JL, Holden  JM, Sagovsky  R.  Detection of postnatal depression: development of the 10-item Edinburgh Postnatal Depression Scale.   Br J Psychiatry. 1987;150:782-786. doi:10.1192/bjp.150.6.782 PubMedGoogle ScholarCrossref
28.
Spielberger  CD, Sydeman  SJ. State-trait anxiety inventory and state-trait anger expression inventory. In: Maruish ME, ed. The Use of Psychological Testing for Treatment Planning and Outcome Assessment. Lawrence Erlbaum Associates; 1994:292-321.
29.
Cox  RW.  AFNI: software for analysis and visualization of functional magnetic resonance neuroimages.   Comput Biomed Res. 1996;29(3):162-173. doi:10.1006/cbmr.1996.0014 PubMedGoogle ScholarCrossref
30.
Gholipour  A, Limperopoulos  C, Clancy  S,  et al. Construction of a Deformable Spatiotemporal MRI Atlas of the Fetal Brain: Evaluation of Similarity Metrics and Deformation Models. In: Medical Image Computing and Computer-Assisted Intervention—MICCAI 2014. Lecture Notes in Computer Science. Springer International Publishing; 2014:292-299. doi:10.1007/978-3-319-10470-6_37
31.
Tustison  NJ, Avants  BB, Cook  PA,  et al.  N4ITK: improved N3 bias correction.   IEEE Trans Med Imaging. 2010;29(6):1310-1320. doi:10.1109/TMI.2010.2046908 PubMedGoogle ScholarCrossref
32.
Joshi  A, Scheinost  D, Okuda  H,  et al.  Unified framework for development, deployment and robust testing of neuroimaging algorithms.   Neuroinformatics. 2011;9(1):69-84. doi:10.1007/s12021-010-9092-8 PubMedGoogle ScholarCrossref
33.
Scheinost  D, Onofrey  JA, Kwon  SH,  et al. A fetal fMRI specific motion correction algorithm using 2nd order edge features. Paper presented at: the 2018 IEEE 15th International Symposium on Biomedical Imaging. 2018:1288-1292. Accessed March 8, 2019. http://ieeexplore.ieee.org
34.
Ojemann  JG, Akbudak  E, Snyder  AZ, McKinstry  RC, Raichle  ME, Conturo  TE.  Anatomic localization and quantitative analysis of gradient refocused echo-planar fMRI susceptibility artifacts.   Neuroimage. 1997;6(3):156-167. doi:10.1006/nimg.1997.0289 PubMedGoogle ScholarCrossref
35.
Behzadi  Y, Restom  K, Liau  J, Liu  TT.  A component based noise correction method (CompCor) for BOLD and perfusion based fMRI.   Neuroimage. 2007;37(1):90-101. doi:10.1016/j.neuroimage.2007.04.042 PubMedGoogle ScholarCrossref
36.
Hallquist  MN, Hwang  K, Luna  B.  The nuisance of nuisance regression: spectral misspecification in a common approach to resting-state fMRI preprocessing reintroduces noise and obscures functional connectivity.   Neuroimage. 2013;82:208-225. doi:10.1016/j.neuroimage.2013.05.116 PubMedGoogle ScholarCrossref
37.
Wheelock  MD, Hect  JL, Hernandez-Andrade  E,  et al.  Sex differences in functional connectivity during fetal brain development.   Dev Cogn Neurosci. 2019;36:100632. doi:10.1016/j.dcn.2019.100632 PubMedGoogle Scholar
38.
Jo  HJ, Saad  ZS, Simmons  WK, Milbury  LA, Cox  RW.  Mapping sources of correlation in resting state FMRI, with artifact detection and removal.   Neuroimage. 2010;52(2):571-582. doi:10.1016/j.neuroimage.2010.04.246 PubMedGoogle ScholarCrossref
39.
Muschelli  J, Nebel  MB, Caffo  BS, Barber  AD, Pekar  JJ, Mostofsky  SH.  Reduction of motion-related artifacts in resting state fMRI using aCompCor.   Neuroimage. 2014;96:22-35. doi:10.1016/j.neuroimage.2014.03.028 PubMedGoogle ScholarCrossref
40.
Zhao  L, Feng  X, Meyer  C, Wu  Y, du Plessis  AJ, Limperopoulos  C. Fetal brain automatic segmentation using 3D deep convolutional neural network. Paper presented at: ISMRM 27th Annual Meeting; Montreal, Quebec, Canada; May 11-16, 2019.
41.
Worsley  KJ, Marrett  S, Neelin  P, Vandal  AC, Friston  KJ, Evans  AC.  A unified statistical approach for determining significant signals in images of cerebral activation.   Hum Brain Mapp. 1996;4(1):58-73. doi:10.1002/(SICI)1097-0193(1996)4:1<58::AID-HBM4>3.0.CO;2-O PubMedGoogle ScholarCrossref
42.
Friston  KJ, Williams  S, Howard  R, Frackowiak  RSJ, Turner  R.  Movement-related effects in fMRI time-series.   Magn Reson Med. 1996;35(3):346-355. doi:10.1002/mrm.1910350312 PubMedGoogle ScholarCrossref
43.
Li  X, Hect  J, Thomason  M, Zhu  D. Interpreting age effects of human fetal brain from spontaneous fMRI using deep 3D convolutional neural networks. arXiv. Preprint posted on June 9, 2019. https://arxiv.org/abs/1906.03691
44.
Thomason  ME, Scheinost  D, Manning  JH,  et al.  Weak functional connectivity in the human fetal brain prior to preterm birth.   Sci Rep. 2017;7:39286. doi:10.1038/srep39286 PubMedGoogle ScholarCrossref
45.
Thomason  ME, Hect  JL, Rauh  VA,  et al.  Prenatal lead exposure impacts cross-hemispheric and long-range connectivity in the human fetal brain.   Neuroimage. 2019;191:186-192. doi:10.1016/j.neuroimage.2019.02.017 PubMedGoogle ScholarCrossref
46.
van den Heuvel  MI, Turk  E, Manning  JH,  et al.  Hubs in the human fetal brain network.   Dev Cogn Neurosci. 2018;30:108-115. doi:10.1016/j.dcn.2018.02.001 PubMedGoogle ScholarCrossref
47.
Craddock  RC, James  GA, Holtzheimer  PE  III, Hu  XP, Mayberg  HS.  A whole brain fMRI atlas generated via spatially constrained spectral clustering.   Hum Brain Mapp. 2012;33(8):1914-1928. doi:10.1002/hbm.21333 PubMedGoogle ScholarCrossref
48.
Makropoulos  A, Gousias  IS, Ledig  C,  et al.  Automatic whole brain MRI segmentation of the developing neonatal brain.   IEEE Trans Med Imaging. 2014;33(9):1818-1831. doi:10.1109/TMI.2014.2322280 PubMedGoogle ScholarCrossref
49.
Peer  M, Abboud  S, Hertz  U, Amedi  A, Arzy  S.  Intensity-based masking: a tool to improve functional connectivity results of resting-state fMRI.   Hum Brain Mapp. 2016;37(7):2407-2418. doi:10.1002/hbm.23182 PubMedGoogle ScholarCrossref
50.
Shehzad  Z, Kelly  C, Reiss  PT,  et al.  A multivariate distance-based analytic framework for connectome-wide association studies.   Neuroimage. 2014;93(Pt 1):74-94. doi:10.1016/j.neuroimage.2014.02.024 PubMedGoogle ScholarCrossref
51.
McArtor  DB, Lubke  GH, Bergeman  CS.  Extending multivariate distance matrix regression with an effect size measure and the asymptotic null distribution of the test statistic.   Psychometrika. 2017;82(4):1052-1077. doi:10.1007/s11336-016-9527-8 PubMedGoogle ScholarCrossref
52.
Aggarwal  CC, Hinneburg  A, Keim  DA. On the surprising behavior of distance metrics in high dimensional space. In:  Database Theory—ICDT 2001. Springer Berlin Heidelberg; 2001:420-434. doi:10.1007/3-540-44503-X_27
53.
Gower  JC.  Some distance properties of latent root and vector methods used in multivariate analysis.   Biometrika. 1966;53(3-4):325-338. doi:10.1093/biomet/53.3-4.325Google ScholarCrossref
54.
Zapala  MA, Schork  NJ.  Statistical properties of multivariate distance matrix regression for high-dimensional data analysis.   Front Genet. 2012;3:190. doi:10.3389/fgene.2012.00190 PubMedGoogle ScholarCrossref
55.
Rivals  I, Personnaz  L, Taing  L, Potier  M-C.  Enrichment or depletion of a GO category within a class of genes: which test?   Bioinformatics. 2007;23(4):401-407. doi:10.1093/bioinformatics/btl633 PubMedGoogle ScholarCrossref
56.
Eggebrecht  AT, Elison  JT, Feczko  E,  et al; IBIS Network.  Joint attention and brain functional connectivity in infants and toddlers.   Cereb Cortex. 2017;27(3):1709-1720. doi:10.1093/cercor/bhw403 PubMedGoogle ScholarCrossref
57.
Rubinov  M, Sporns  O.  Weight-conserving characterization of complex functional brain networks.   Neuroimage. 2011;56(4):2068-2079. doi:10.1016/j.neuroimage.2011.03.069 PubMedGoogle ScholarCrossref
58.
Xia  M, Wang  J, He  Y.  BrainNet Viewer: a network visualization tool for human brain connectomics.   PLoS One. 2013;8(7):e68910. doi:10.1371/journal.pone.0068910 PubMedGoogle Scholar
59.
Van den Bergh  BRH, Mulder  EJH, Mennes  M, Glover  V.  Antenatal maternal anxiety and stress and the neurobehavioural development of the fetus and child: links and possible mechanisms. A review.   Neurosci Biobehav Rev. 2005;29(2):237-258. doi:10.1016/j.neubiorev.2004.10.007 PubMedGoogle ScholarCrossref
60.
Van den Bergh  BR, Mulder  EJ, Visser  GH, Poelmann-Weesjes  G, Bekedam  DJ, Prechtl  HF.  The effect of (induced) maternal emotions on fetal behaviour: a controlled study.   Early Hum Dev. 1989;19(1):9-19. doi:10.1016/0378-3782(89)90100-X PubMedGoogle ScholarCrossref
61.
Matthews  SG.  Early programming of the hypothalamo-pituitary-adrenal axis.   Trends Endocrinol Metab. 2002;13(9):373-380. doi:10.1016/S1043-2760(02)00690-2 PubMedGoogle ScholarCrossref
62.
Hobson  JA, Pace-Schott  EF.  The cognitive neuroscience of sleep: neuronal systems, consciousness and learning.   Nat Rev Neurosci. 2002;3(9):679-693. doi:10.1038/nrn915 PubMedGoogle ScholarCrossref
63.
O’Connor  TG, Heron  J, Golding  J, Beveridge  M, Glover  V.  Maternal antenatal anxiety and children’s behavioural/emotional problems at 4 years: report from the Avon Longitudinal Study of Parents and Children.   Br J Psychiatry. 2002;180:502-508. doi:10.1192/bjp.180.6.502 PubMedGoogle ScholarCrossref
64.
Van den Bergh BRH.  The influence of maternal emotions during pregnancy on fetal and neonatal behavior.   J Prenat Perinat Psychol Health. 1990;5(2):119.Google Scholar
65.
Gitau  R, Cameron  A, Fisk  NM, Glover  V.  Fetal exposure to maternal cortisol.   Lancet. 1998;352(9129):707-708. doi:10.1016/S0140-6736(05)60824-0 PubMedGoogle ScholarCrossref
66.
Dedovic  K, Duchesne  A, Andrews  J, Engert  V, Pruessner  JC.  The brain and the stress axis: the neural correlates of cortisol regulation in response to stress.   Neuroimage. 2009;47(3):864-871. doi:10.1016/j.neuroimage.2009.05.074 PubMedGoogle ScholarCrossref
67.
Mansell  T, Novakovic  B, Meyer  B,  et al; BIS Investigator Team.  The effects of maternal anxiety during pregnancy on IGF2/H19 methylation in cord blood.   Transl Psychiatry. 2016;6:e765. doi:10.1038/tp.2016.32 PubMedGoogle Scholar
68.
Conradt  E, Lester  BM, Appleton  AA, Armstrong  DA, Marsit  CJ.  The roles of DNA methylation of NR3C1 and 11β-HSD2 and exposure to maternal mood disorder in utero on newborn neurobehavior.   Epigenetics. 2013;8(12):1321-1329. doi:10.4161/epi.26634 PubMedGoogle ScholarCrossref
69.
Jobe  EM, Zhao  X.  DNA methylation and adult neurogenesis.   Brain Plast. 2017;3(1):5-26. doi:10.3233/BPL-160034 PubMedGoogle ScholarCrossref
70.
Wu  L, Wang  L, Shangguan  S,  et al.  Altered methylation of IGF2 DMR0 is associated with neural tube defects.   Mol Cell Biochem. 2013;380(1-2):33-42. doi:10.1007/s11010-013-1655-1 PubMedGoogle ScholarCrossref
71.
Peterson  A, Thome  J, Frewen  P, Lanius  RA.  Resting-state neuroimaging studies: a new way of identifying differences and similarities among the anxiety disorders?   Can J Psychiatry. 2014;59(6):294-300. doi:10.1177/070674371405900602 PubMedGoogle ScholarCrossref
72.
Mochcovitch  MD, da Rocha Freire  RC, Garcia  RF, Nardi  AE.  A systematic review of fMRI studies in generalized anxiety disorder: evaluating its neural and cognitive basis.   J Affect Disord. 2014;167:336-342. doi:10.1016/j.jad.2014.06.041 PubMedGoogle ScholarCrossref
73.
Menon  V, Uddin  LQ.  Saliency, switching, attention and control: a network model of insula function.   Brain Struct Funct. 2010;214(5-6):655-667. doi:10.1007/s00429-010-0262-0 PubMedGoogle ScholarCrossref
74.
Seeley  WW, Menon  V, Schatzberg  AF,  et al.  Dissociable intrinsic connectivity networks for salience processing and executive control.   J Neurosci. 2007;27(9):2349-2356. doi:10.1523/JNEUROSCI.5587-06.2007 PubMedGoogle ScholarCrossref
75.
Greicius  MD, Supekar  K, Menon  V, Dougherty  RF.  Resting-state functional connectivity reflects structural connectivity in the default mode network.   Cereb Cortex. 2009;19(1):72-78. doi:10.1093/cercor/bhn059 PubMedGoogle ScholarCrossref
76.
Raichle  ME.  The brain’s default mode network.   Annu Rev Neurosci. 2015;38:433-447. doi:10.1146/annurev-neuro-071013-014030 PubMedGoogle ScholarCrossref
77.
Dosenbach  NUF, Fair  DA, Miezin  FM,  et al.  Distinct brain networks for adaptive and stable task control in humans.   Proc Natl Acad Sci U S A. 2007;104(26):11073-11078. doi:10.1073/pnas.0704320104 PubMedGoogle ScholarCrossref
78.
Sripada  RK, King  AP, Welsh  RC,  et al.  Neural dysregulation in posttraumatic stress disorder: evidence for disrupted equilibrium between salience and default mode brain networks.   Psychosom Med. 2012;74(9):904-911. doi:10.1097/PSY.0b013e318273bf33 PubMedGoogle ScholarCrossref
79.
Zhang  Y, Liu  F, Chen  H,  et al.  Intranetwork and internetwork functional connectivity alterations in post-traumatic stress disorder.   J Affect Disord. 2015;187:114-121. doi:10.1016/j.jad.2015.08.043 PubMedGoogle ScholarCrossref
80.
Chen  AC, Etkin  A.  Hippocampal network connectivity and activation differentiates post-traumatic stress disorder from generalized anxiety disorder.   Neuropsychopharmacology. 2013;38(10):1889-1898. doi:10.1038/npp.2013.122 PubMedGoogle ScholarCrossref
81.
Poppenk  J, Evensmoen  HR, Moscovitch  M, Nadel  L.  Long-axis specialization of the human hippocampus.   Trends Cogn Sci. 2013;17(5):230-240. doi:10.1016/j.tics.2013.03.005 PubMedGoogle ScholarCrossref
82.
Satpute  AB, Mumford  JA, Naliboff  BD, Poldrack  RA.  Human anterior and posterior hippocampus respond distinctly to state and trait anxiety.   Emotion. 2012;12(1):58-68. doi:10.1037/a0026517 PubMedGoogle ScholarCrossref
83.
Schneier  FR, Pomplun  M, Sy  M, Hirsch  J.  Neural response to eye contact and paroxetine treatment in generalized social anxiety disorder.   Psychiatry Res. 2011;194(3):271-278. doi:10.1016/j.pscychresns.2011.08.006 PubMedGoogle ScholarCrossref
84.
Adamson  B, Letourneau  N, Lebel  C.  Prenatal maternal anxiety and children’s brain structure and function: a systematic review of neuroimaging studies.   J Affect Disord. 2018;241:117-126. doi:10.1016/j.jad.2018.08.029 PubMedGoogle ScholarCrossref
85.
Mennes  M, Stiers  P, Lagae  L, Van den Bergh  BRH.  Antenatal maternal anxiety modulates the BOLD response in 20-year-old men during endogenous cognitive control.   Brain Imaging and Behavior. 2019. doi:10.1007/s11682-018-0027-6 PubMedGoogle Scholar
86.
Burghy  CA, Stodola  DE, Ruttle  PL,  et al.  Developmental pathways to amygdala-prefrontal function and internalizing symptoms in adolescence.   Nat Neurosci. 2012;15(12):1736-1741. doi:10.1038/nn.3257 PubMedGoogle ScholarCrossref
87.
Adamson  B, Letourneau  N, Lebel  C.  Corrigendum to prenatal maternal anxiety and children’s brain structure and function: a systematic review of neuroimaging studies.   J Affect Disord. 2019;253:S0165-0327(18)33091-X. doi:10.1016/j.jad.2018.12.002 PubMedGoogle Scholar
88.
Rifkin-Graboi  A, Meaney  MJ, Chen  H,  et al.  Antenatal maternal anxiety predicts variations in neural structures implicated in anxiety disorders in newborns.   J Am Acad Child Adolesc Psychiatry. 2015;54(4):313-21.e2. doi:10.1016/j.jaac.2015.01.013 PubMedGoogle ScholarCrossref
89.
Smyser  CD, Inder  TE, Shimony  JS,  et al.  Longitudinal analysis of neural network development in preterm infants.   Cereb Cortex. 2010;20(12):2852-2862. doi:10.1093/cercor/bhq035 PubMedGoogle ScholarCrossref
90.
Doria  V, Beckmann  CF, Arichi  T,  et al.  Emergence of resting state networks in the preterm human brain.   Proc Natl Acad Sci U S A. 2010;107(46):20015-20020. doi:10.1073/pnas.1007921107 PubMedGoogle ScholarCrossref
91.
Thomason  ME, Dassanayake  MT, Shen  S,  et al.  Cross-hemispheric functional connectivity in the human fetal brain.   Sci Transl Med. 2013;5(173):173ra24. doi:10.1126/scitranslmed.3004978 PubMedGoogle Scholar
92.
Golfenshtein  N, Hanlon  AL, Deatrick  JA, Medoff-Cooper  B.  Parenting stress in parents of infants with congenital heart disease and parents of healthy infants: the first year of life.   Compr Child Adolesc Nurs. 2017;40(4):294-314. doi:10.1080/24694193.2017.1372532 PubMedGoogle ScholarCrossref
Limit 200 characters
Limit 25 characters
Conflicts of Interest Disclosure

Identify all potential conflicts of interest that might be relevant to your comment.

Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.

Err on the side of full disclosure.

If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.

Not all submitted comments are published. Please see our commenting policy for details.

Limit 140 characters
Limit 3600 characters or approximately 600 words
    Original Investigation
    Pediatrics
    December 7, 2020

    Association of Prenatal Maternal Anxiety With Fetal Regional Brain Connectivity

    Author Affiliations
    • 1Division of Diagnostic Imaging and Radiology, Children’s National, Washington DC
    • 2Division of Neonatology, Children’s National, Washington DC
    • 3Department of Pediatrics, The George Washington University School of Medicine, Washington DC
    JAMA Netw Open. 2020;3(12):e2022349. doi:10.1001/jamanetworkopen.2020.22349
    Key Points

    Question  What is the association between maternal psychological distress and fetal brain functional connectivity?

    Findings  In this cohort study of 50 women with 59 fetal scans, prenatal maternal trait and state anxiety were associated with increased resting state functional connectivity between sensorimotor and brainstem areas and reduced connectivity between temporoparietal cortices and basal ganglia.

    Meaning  Our findings suggest an association between prenatal exposure to anxiety and disrupted connectivity of neural networks.

    Abstract

    Importance  Maternal psychological distress during pregnancy is associated with adverse obstetric outcomes and neuropsychiatric deficits in children. Currently unavailable in vivo interrogation of fetal brain function could provide critical insights into the onset and timing of altered neurodevelopmental trajectories.

    Objective  To investigate the association between prenatal maternal stress, anxiety, and depression and in vivo fetal brain resting state functional connectivity.

    Design, Setting, and Participants  This cohort study included pregnant women scanned between January 2016 and April 2019. A total of 50 pregnant women with healthy pregnancies were prospectively recruited from low-risk obstetric clinics in the Washington DC area and were scanned at Children’s National in Washington DC.

    Exposures  Maternal stress, anxiety, and depression.

    Main Outcomes and Measures  The association of prenatal maternal stress, anxiety, and depression with whole-brain connectivity was analyzed using multivariate distance matrix regression. Prenatal maternal stress, anxiety, and depression were assessed using the Perceived Stress Scale, Spielberger State Anxiety Inventory and Spielberger Trait Anxiety Inventory, and the Edinburgh Postnatal Depression Scale, respectively. Whole-brain connectivity was measured from 100 functionally defined regions of interest.

    Results  This study analyzed 59 resting-state functional connectivity magnetic resonance image data sets from the fetuses (mean [SD] gestational age, 33.52 [4 weeks]) of 50 healthy pregnant women (mean [SD] age, 33.77 [5.51]). Mean (SD) scores for the questionnaires were as follows: Spielberger State Anxiety Inventory, 26.66 (6.72) (range, 20-48); Spielberger Trait Anxiety Inventory, 28.09 (6.62) (range, 20-50); Perceived Stress Scale, 9.27 (5.13) (range, 1-25); and Edinburgh Postnatal Depression Scale 3.24 (2.84) (range, 0-14). Prenatal maternal anxiety scores measured using the Spielberger Trait and State Anxiety Inventories were associated with differences in fetal connectivity (Spielberger State Anxiety Inventory: pseudo-R2 = 0.019, P = .04; Spielberger Trait Anxiety Inventory: pseudo-R2 = 0.021, P = .007). Interhemispheric connections, such as those involving the parietofrontal and occipital association cortices, were associated with reduced maternal prenatal anxiety, and those between the brainstem and sensorimotor areas were associated with higher anxiety scores.

    Conclusions and Relevance  In this cohort study, an association was found between prenatal maternal anxiety and disturbances in fetal brain functional connectivity, suggesting altered fetal programming. Early onset of functional deviations suggests the need for more widespread screening of pregnant women for symptoms of anxiety.

    Introduction

    Up to 50% of women report symptoms of stress, depression, or anxiety during pregnancy based on systematic reviews.1-3 Maternal mental health disorders are associated with adverse pregnancy outcomes and an increased risk for neuropsychiatric disorders, such as autism and attention-deficit/hyperactivity disorder.4-7 The high prevalence of prenatal psychological distress and its association with poor obstetric outcomes as well as motor deficits, sociocognitive, and socioaffective impairments in exposed children underscores the need for identifying the earliest effects of in utero exposure to the developing brain.

    Available clinical and imaging evidence supports the negative association of maternal stress with postnatal growth and brain development. Previous studies have reported low birth weights8,9 in neonates exposed to high antenatal stress or anxiety. Clinical studies have reported neurobehavioral deficits beginning in infancy and early childhood, including higher reactivity,10 impaired motor coordination,11 and language delays.12 Postnatal brain imaging findings have provided insights into potential neural substrates for these deficits. These include reduced cortical thickness,13,14 amygdala and hippocampal volume changes,15-17 asymmetric electroencephalographic patterns in the frontal lobes,18,19 white matter microstructural changes,20,21 and impaired connectivity.22,23 More recently, a study by Wu and colleagues24 provided, to our knowledge, the first report of impaired brain metabolism, reduced hippocampal growth, and accelerated cortical folding in fetuses of women experiencing psychological distress. However, the effect of prenatal maternal stress on the developing neural circuitry during this critical period of brain development has not been investigated.

    In this study, we investigated the association between maternal psychological distress and in vivo resting state brain functional connectivity in late second- to third-trimester fetuses. We hypothesized that elevated maternal psychological distress would be associated with disturbances in functional connectivity in neural circuitry related to stress, anxiety, or depression. Herein, psychological distress refers to symptoms of prenatal maternal depression, stress, or anxiety that have not been clinically diagnosed as a mental health illness or disorder.24,25 Using multivariate distance matrix regression (MDMR), we examined the association between maternal psychological distress using well-validated self-report questionnaires and the developing connections in the human fetal brain. Our goal was to assess the association between in utero exposure to elevated levels of stress, depression, and anxiety and the fetal connectome that may serve as an early biomarker of altered brain development and later neurodevelopmental disabilities.

    Methods
    Participants

    Healthy pregnant women with normal ultrasonography and fetal biometry findings were prospectively recruited as part of a cohort study to investigate fetal brain development in complex congenital heart disease. Fetuses with known or suspected genetic or chromosomal abnormalities and fetuses of pregnant women with known psychiatric, metabolic, or genetic disorders; complicated pregnancies (ie, preeclampsia and gestational diabetes); multiple pregnancies; alcohol and tobacco use; maternal medications; and contraindications to magnetic resonance imaging (MRI) were excluded. Of the women who were scanned and who answered the questionnaires, only those with resting state data that met the criteria described below were included in the analysis. The institutional review board of Children’s National in Washington DC, approved this study. Written informed consent was obtained from each study participant. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.

    Maternal Stress, Anxiety, and Depression Scores

    Widely used and validated stress, anxiety, and depression questionnaires were administered to participants on the same day they had a fetal MRI. These were the Perceived Stress Scale [PSS],26 Edinburgh Postnatal Depression Scale [EDPS],27 Spielberger State Anxiety Inventory [SSAI], and Spielberger Trait Anxiety Inventory [STAI].28 The PSS is a 10-item questionnaire that evaluates an individual’s perceived stress level over the past month. The EPDS has a similar number of items and assesses depression over the past week. Both the STAI and SSAI are composed of 20 items and evaluate trait (ie, general feeling) and state (ie, feeling for the day) anxiety, respectively. Scores of 15 or higher, 10 or higher, and 40 or higher in the PSS, EPDS, and SSAI and STAI indicate that the individual has symptoms of stress, depression, and anxiety, respectively. Scoring above these thresholds does not mean a clinical diagnosis of stress or anxiety; instead, these cutoffs are used to identify individuals who may need additional intervention.

    MRI Acquisition

    A 1.5 Tesla MRI scanner (GE Healthcare) with an 8-channel receiver coil was used to acquire images of the fetal brain. Anatomical T2- weighted images (ie, sagittal, axial, and coronal sections) were collected using a single-shot fast spin-echo sequence with the following settings: TR, 1100 ms; TE, 160 ms; flip angle, 90°; and section thickness, 2 mm. Resting state echo planar images (EPI) were collected using the following parameters: TR, 3000 ms; TE, 60 ms; voxel size, 2.578 × 2.578 × 3 mm; flip angle, 90°; field of view, 33 cm; matrix size, 128 × 128; and scan duration, 7 minutes.

    Processing of Resting State Data

    Fetal resting state data were preprocessed using tools from the Analysis of Functional NeuroImages unless otherwise indicated.29 Fetal EPI images based on blood oxygen level dependent (BOLD) contrast were slice-time corrected, followed by exclusion of the first 4 volumes of the time series, and then manually oriented to radiologic convention using an age-matched gestational age template.30 Resting state data were then despiked, bias-field corrected (using the N4BiasFieldCorrection tool31), and corrected for motion.32,33 After motion correction, EPI images were manually aligned to the T2-weighted brain images to ensure overlap between EPI and anatomic images; this improved later automatic affine coregistration. The EPI images were then intensity scaled to a global mode of 100034 and smoothed using an isotropic 5-mm full width at half maximum gaussian kernel. After smoothing, bandpass filtering (0.009-0.08 Hz), nuisance regression with volume censoring, and normalization to a 32-week gestational age template were performed.35-37 Residual BOLD signals were analyzed.

    Regressors included tissue- and motion-based signals.35,38,39 Specifically, tissue signals were derived from white matter and cerebrospinal fluid defined using an in-house deep learning–based segmentation algorithm40 and registered onto EPI images. Motion regressors included linearly detrended rigid motion parameters, their temporal derivatives, and quadratic terms.41,42 To further minimize the effects of motion on functional connectivity, high motion volumes, defined as those with frame-to-frame translational motion greater than 1 mm and rotational motion less than 1.5°, were censored from the hemodynamic time series.37,43-46 The preceding frame was also removed. Volumes in which more than 10% of voxels had signals deviating from the voxel time series’ median absolute deviation were excluded. Only fetuses with 4 or more minutes of available data after processing were included in the analyses.

    Statistical Analysis

    The BOLD signals were measured from 100 regions of interest (ROIs) defined using a spectral clustering algorithm applied on functional data47 and refined using intensity-based masking.48,49 The location of the 100 ROIs are shown in the eFigure in the Supplement. Functional ROIs were named based on their overlap with the newborn automated anatomical labeling template (eTable 3 in the Supplement lists all ROI labels); as such, labels are estimated locations in the brain and do not refer to precise anatomic locations. Whole-brain differences in connectivity across all 100 ROIs associated with maternal depression, anxiety, and stress scores were then evaluated using MDMR.50,51 Functional connectivity across all ROI pairs (4950 connections) was computed at the subject level using pairwise Pearson correlations. These scores were z transformed to facilitate statistical analysis. Differences in connectivity profiles for each ROI among all participants were then evaluated using a Manhattan distance, a metric commonly used for high-dimensional data.52 These steps yielded an n × n distance matrix, where n is the number of participants, for each ROI; this distance matrix represents the dissimilarity among individual fetal connectivity networks. An MDMR was then performed51 to test how maternal neurobehavioral test scores accounted for differences in connectivity profiles.53,54 The significance of the MDMR pseudo–F statistic was assessed using permutation testing (permutations, 100 000). From the factors that we initially evaluated (gestational age at the time of the scan, head motion [ie, mean framewise displacement], SSAI score, STAI score, PSS, and EPDS score), we only included those that helped explain variability in connectivity profiles. These regressors included remaining volumes after preprocessing, SSAI scores, and STAI scores. The omnibus MDMR model and main effects were considered significant at a 2-tailed P < .05.

    We then used enrichment analysis to characterize the significant main associations in the MDMR model. Enrichment analysis is widely used in large-scale genomic studies.55 Recently, it has been used for analyzing associations between behavior and functional connectivity.37,56 We first subdivided the 100 ROIs into groups using a community detection algorithm that was set up to detect smaller module sizes.57 Modules are nonoverlapping ROI clusters that tend to connect densely with other members of its subgroup and sparsely to the rest of the network. We then used enrichment analysis to identify within- and between-module connections in which significant associations between maternal neurobehavioral scores and functional connectivity (or, resting state functional connectivity [RSFC]) were clustered. We computed the Pearson correlation between maternal scores and all 4950 ROI-ROI pairwise connections. We then used a 1-df χ2 test to assess whether the actual number of strong brain-behavior associations (P < .05) within a functional network pair was more than what would be expected if all strong RSFC-behavior associations were equally distributed across all possible network pairs (ie, enrichment). The significance of the χ2 test was assessed by comparing the statistic with values generated using permutation testing, by which connectivity was correlated with 1000 permuted values of the maternal questionnaire score. Interactions between network modules were visualized using the BrainNet Viewer.58

    Results

    This analysis included 59 resting-state MRI scans performed on 50 fetuses between 24.71 and 39.43 gestational weeks (mean [SD] gestational age, 33.52 [4.00] weeks). Of these fetuses, 26 (44.1%) were female and 24 (40.7%) were male. All fetuses had structurally normal brains on conventional MRIs that were evaluated by an experienced pediatric neuroradiologist (G.V.). The median Apgar score at 5 minutes for the fetuses was 9 (range, 6-10); the mean (SD) birth weight was 3308.51 (511.37) g. The mean (SD) age of the pregnant women scanned was 33.77 (5.51) years. Most of the pregnant women in the study were college graduates (43 [86%]) and reported professional employment (41 [82%]). Table 1 gives a summary of the cohort.

    Quality Assurance of Resting State Data

    The mean (SD) scan duration for fetuses was 5.4 (0.87) minutes (range, 4-7 minutes; equivalent to a mean [SD] of 108.7 [17.3] remaining brain volumes) (eTable 1 in the Supplement), and the mean (SD) maximum framewise displacement was 1.34 (0.18) mm. eTable 2 in the Supplement gives additional details on head motion.

    Maternal Stress, Depression, and Anxiety Scores

    Of the 50 pregnant women, 4 (8%) were positive for state anxiety, 6 (12%) for trait anxiety, 7 (14%) for stress, and 1 (2%) for depression. Mean (SD) scores on the questionnaires were as follows: SSAI, 26.66 (6.72) (range, 20-48); STAI, 28.09 (6.62) (range, 20-50); PSS, 9.27 (5.13) (range, 1-25); and EPDS, 3.24 (2.84) (range, 0-14).

    Modules in the Fetal Brain

    The fetal functional network was decomposed into 14 modules. The majority of modules were confined to 1 hemisphere except for a few, such as module 2, which included bilateral superior frontal and anterior cingulate gyri. The regions that comprised each module are shown in the Figure, A.

    Association Between Psychological Distress Scores and Connectivity

    In the MDMR model, the association between psychological distress and connectivity was significant (pseudo-R2 = 0.056, P = .03). The association between STAI score and connectivity was significant (pseudo-R2 = 0.021, P = .007), as was SSAI (pseudo-R2 = 0.019, P = .04) (Table 2). For the STAI, there were 285 connections with significant connectivity associated with maternal trait anxiety scores. Of these 4950 connections, 140 (2.8%) connections were positively correlated with trait anxiety and 145 (2.9%) were negatively correlated with STAI scores. For the SSAI, there were 235 strong association between behavior and RSFC. Of these, 96 were positively associated and 139 negatively correlated with maternal SSAI scores. eTables 4 and 5 in the Supplement list the ROI-ROI pairs that were associated with trait and state anxiety, respectively.

    Networks Associated With Prenatal Maternal Anxiety

    Enrichment analysis showed that significant associations between STAI and brain connectivity tended to cluster across 12 functional network pairs involving 8 of the 14 modules (Figure, B). Three of these were within-module (modules 2, 7, and 13) interactions; the rest were between modules. Of the 285 strong associations between RSFC and STAI, 100 (35%) connected these functional networks; 73 (73%) of these connections showed a positive association with trait anxiety.

    Figure, B shows the 13 significantly enriched functional network pairs for SSAI; the majority of these were between-network links. Of the 235 significant brain-behavior interactions, 88 were clustered across 13 network pairs formed by 11 modules. Connectivity between most (58%) of the ROI-ROI pairs were inversely associated with maternal state anxiety scores.

    Three network-pairs were involved in both state and trait anxiety. These pairs were modules 1 and 10, modules 7 and 14, and modules 3 and 11. Most of the ROI-ROI pairs in the first 2 network pairs were negatively associated with STAI and SSAI. Module 10 included areas of the left inferior parietal lobule and middle frontal gyrus, and module 1 comprised occipital regions. The deep gray matter and midbrain were components of module 14, and the areas surrounding the supramarginal gyrus were part of module 7. Modules 3 and 11 comprised bilateral mid-superior frontal gyrus and sensorimotor regions. Table 3 and Table 4 show the top 10 ROI-ROI connections that were associated with maternal trait and state anxiety.

    Discussion

    We report for the first time, to our knowledge, alterations in fetal functional connectivity associated with maternal anxiety. Connectivity strength between some regions correlated positively with maternal anxiety as measured using the STAI and SSAI and negatively in others. These associations between anxiety and RSFC were observed in multiple functional networks. Involved networks for state and trait anxiety overlapped; for instance, connectivity between the inferior parietal lobule and contralateral occipital regions was negatively associated with both trait and state anxiety. Likewise, links to the superior dorsal-frontal areas of the brain, mainly the somatosensory areas, were positively associated with both types. In some cases, the associations between state and trait anxiety and RSFC implicated distinct functional networks; for instance, brainstem and fusiform face area associations were positively associated with symptoms of trait anxiety only.

    Our study revealed large-scale brain networks that were affected by increasing levels of anxiety in pregnant women. Although the neurobiological mechanisms underlying the effects of maternal anxiety on fetal brain development have yet to be fully explored, neurobehavioral and brain imaging studies59-64 have reported an association of maternal anxiety with atypical brain development. Fetal ultrasonography studies during the third trimester, for instance, have shown that fetuses of anxious women demonstrated increased wakefulness and heart rate variability,59-61 a pattern that may be related to the disorganized sleep-wake cycles observed in various neuropsychopathologies in children and adults.62 Numerous postnatal neurobehavioral studies63,64 have also shown an association between prenatal maternal anxiety and cognitive and emotional development of children.

    The current literature59,65 suggests that cortisol mediates some of these outcomes. Fetal cortisol levels have been found to linearly correlate with maternal anxiety levels,65 and excess amounts in the fetus may disrupt the development of the hypothalamic-pituitary-axis, limbic system, and prefrontal areas.59 Mineralocorticoid and glucocorticoid receptors are ubiquitous in the brain,66 and this may explain the wide array of adverse neurodevelopmental outcomes reported in children exposed to prenatal anxiety. Rather than localized effects, anxiety likely impacts multiple neural systems. The involvement of different networks in our study is consistent with this. Epigenetic processes also appear to be involved. Maternal anxiety has been associated with both decreased DNA methylation at cytosine-phosphate-guanine (CpG) sites in the brain67 and increased DNA methylation in the placenta.68 Similar to cortisol receptors, there are numerous CpG islands (ie, areas of the genome with a high frequency of a consecutive cytosine and guanine nucleotides) in the brain, and their methylation has been implicated in embryonic and adult neurogenesis.69 Abnormal methylation has also been linked to neurologic deficits (ie, neural tube defects).70 We speculate that the prevalence of CpG sites in the brain,67 as in the case of cortisol receptors, may account for some of the distributed associations of maternal anxiety with fetal functional connectivity.

    Mood disorders have been shown to disrupt large scale network organization.71,72 Altered interactions between the salience network (ie, brainstem, fronto-insular and dorsal anterior cingulate cortices),73,74 default mode network (ie, posterior cingulate, precuneus and medial prefrontal cortices),75,76 and executive control networks (ie, dorsolateral prefrontal and lateral and posterior parietal cortices)77 have been demonstrated in anxiety.78,79 Of interest, our findings showed that some of the more consistently affected regions and connections in the fetal connectome belonged to these networks. For example, we showed strengthened associations between areas of the brain involved with arousal and salience, such as connections between the brainstem and sensorimotor and dorsal frontal regions and anterior cingulate with the hippocampus, with increasing levels of maternal trait anxiety. Also, various brain nuclei related to anxiety reside in the brainstem, possibly helping mediate the observed increase in connectivity. Connections between other areas associated with stress and anxiety, such as the hippocampus and the insula, also showed increased connectivity associated with increased symptoms of maternal anxiety.

    Although hippocampal connectivity was positively associated with trait and state anxiety scores, the hippocampal circuits activated varied. In trait anxiety, the right hippocampus interacted with the medial and dorsal superior frontal gyrus; in state anxiety, the connections were mostly to the precuneus and middle cingulate region. These connectivity profiles have previously been described in high anxiety states in adults,80 but the specificity of each network to either state or trait anxiety is unclear. Previous studies81,82 have also suggested that anterior and posterior hippocampal connectivity differs, with the former engaging with the ventromedial prefrontal cortex and likely responsible for moment-to-moment or state changes. Although our findings suggest changes specific to type of anxiety, namely state and trait, additional studies that focus on fetal hippocampal connectivity with a larger sample of pregnant women exhibiting symptoms of either trait or state anxiety will help to further elucidate selective vulnerability of specific brain regions. Longitudinal studies that allow assessment of the severity and chronicity of anxiety would be especially insightful. Studies like these are important given the altered hippocampal growth trajectories and altered connectivity in infants15 and children83 exposed to prenatal anxiety.

    Our data showed an association between reduced connectivity in regions that are part of the executive control network and increasing maternal anxiety. The frontoparietal cortices were some of the more commonly affected areas with reduced connectivity. For example, the strength of connectivity between the inferior parietal lobule and superior to middle frontal cortex decreased in association with increasing levels of maternal anxiety. The association between inferior parietal lobule-occipital and fusiform connectivity and increased maternal state and trait anxiety was also observed; the same neurocircuitry has been implicated in social anxiety in adults.84 Default mode network–related regions, such as the medial frontal cortex and angular gyrus, also showed an association of reduced connectivity with increased maternal anxiety.

    Our study is, to our knowledge, the first to report an association between maternal trait and state anxiety and altered fetal functional brain connectivity, supporting a fetal programming hypothesis. This builds on findings from previous studies in newborns, infants, children, and adults exposed to anxiety in utero that have shown network dysfunction, including reduced amygdala-thalamus connectivity,22 lack of inferior frontal cortex modulation in an endogenous cognitive control task,85 and aberrant amygdala and prefrontal cortex circuitry.86 Our resting state findings are also consistent with volumetric and diffusion tensor imaging studies that showed structural abnormalities in the limbic, temporal, and frontal regions in fetuses and infants exposed to prenatal distress.87 Hippocampal volumes have been shown to be reduced in fetuses24 of women with elevated stress levels. Similarly, slower hippocampal growth has been reported in infants of women with anxiety symptoms, suggesting an association of prenatal maternal anxiety with regional brain growth.15 In newborns and infants exposed to anxiety in utero, fractional anisotropy, a diffusion tensor imaging metric that reflects neuronal integrity, has been shown to be decreased in regions critical to emotional and cognitive development (the insula and dorsolateral frontal regions), visual processing (middle occipital cortex), and social functioning (angular region and posterior cingulate),88 areas also affected in the current study. Taken together, these findings suggest that prenatal maternal anxiety may affect the development of a diverse set of brain regions and networks, including temporal and frontal and prefrontal areas, and that this may, in turn, impact long-term neurodevelopmental outcomes.

    In our study, we showed that functional connections between areas that developmentally associate earlier (ie, brainstem and sensorimotor areas and local short-range connections) were stronger in high maternal trait anxiety states. In contrast, those that emerged later in development (ie, more distant anteroposterior, interhemispheric connections) were weakened by higher levels of anxiety.89-91 Strengthening of earlier emerging connections suggests a preference for networks that support more fundamental processes (ie, sensory and motor processing, arousal) as opposed to connections between association regions that will eventually subserve higher-order cognitive functions such as executive control. This may reflect an adaptive response of the fetal brain to in utero exposure to anxiety, the mechanisms of which require further investigation that is outside the scope of the current study.

    Limitations

    This study has limitations. We examined a moderately large sample of fetuses, but younger fetuses (ie, at lower gestational ages) were not equally represented in the sample owing in part to the technical difficulties associated with the acquisition and processing of these images. As a result, investigating the onset and timing of connectivity changes remained challenging. We anticipate that improvements in acquisition and processing techniques will help alleviate some of these issues. A larger sample that also includes clinically diagnosed pregnant women, may also give better power to detect associations between other types of maternal psychological distress (ie, stress and depression) and functional brain connectivity. Second, only 12% of the women in the study were positive for trait anxiety using the STAI, and only 8% presented with symptoms of state anxiety using the SSAI. A complete understanding of the functional connectome in fetuses exposed to prenatal anxiety entails an assessment of women with higher anxiety scores and clinically diagnosed and managed anxiety disorders. However, our study emphasizes the need for mental health surveillance in pregnant women as functional brain changes appears at subclinical levels. Third, maternal anxiety was measured using self-report. Although biases may be inherent to this method, well-validated and tested questionnaires have been shown to reliably quantify subjective perceptions of psychologic distress.92 To complement surveys, studies are under way that use objective measures of stress and anxiety. Fourth, to maximize the number of data sets analyzed, few fetuses had 2 scans included in the sample. Although this number accounts for a small subset of the total cohort, futures studies that account for repeated scans within the context of the analytic technique used (ie, MDMR) would be ideal. Fifth, whether the functional connectivity findings observed during the fetal period predict infant and childhood neurobehavioral outcomes needs to be validated. Longitudinal follow-up studies that evaluate postnatal neurodevelopmental outcomes in these fetuses are currently under way.

    Conclusions

    In this cohort study, alterations in late second- to third-trimester fetal brain functional connectivity were associated with maternal anxiety. Maternal anxiety and fetal connectivity correlations either decreased or increased depending on the networks involved. Interhemispheric connections, such as those involving the medial frontal regions and basal ganglia, were found to be weakened. In contrast, connections such as those between the brainstem and sensorimotor areas, were strengthened in association with higher trait anxiety scores. Some networks were associated with both trait and state anxiety and overlapped, whereas some were distinct to 1 type. Areas associated with salience network, DMN, and central executive network were commonly implicated. These findings suggest an association between altered fetal programming in fetuses and maternal anxiety and the need for mental health surveillance and interventions for pregnant women.

    Back to top
    Article Information

    Accepted for Publication: August 16, 2020.

    Published: December 7, 2020. doi:10.1001/jamanetworkopen.2020.22349

    Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2020 De Asis-Cruz J et al. JAMA Network Open.

    Corresponding Author: Catherine Limperopoulos, PhD, Division of Neonatology, Children’s National, 111 Michigan Ave NW, Washington, DC 20010 (climpero@childrensnational.org).

    Author Contributions: Drs De Asis-Cruz and Limperopoulos had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

    Concept and design: De Asis-Cruz, Limperopoulos.

    Acquisition, analysis, or interpretation of data: All authors.

    Drafting of the manuscript: De Asis-Cruz, Zhao, Kapse.

    Critical revision of the manuscript for important intellectual content: All authors.

    Statistical analysis: De Asis-Cruz, Krishnamurthy, Kapse.

    Obtained funding: Limperopoulos.

    Administrative, technical, or material support: Zhao, Andescavage, Quistorff, Lopez, Limperopoulos.

    Supervision: De Asis-Cruz, Andescavage, Quistorff, Limperopoulos.

    Conflict of Interest Disclosures: Dr Krishnamurthy reported receiving grants from Children's National Health System during the conduct of the study. No other disclosures were reported.

    Funding/Support: This study was funded by grant R01 HL116585-01 from the National Heart, Lung, and Blood Institute, National Institutes of Health (Dr De Asis-Cruz) and grant U54HD090257 from the Intellectual and Developmental Disabilities Research Center (Dr De Asis-Cruz).

    Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

    Additional Contributions: We thank the participants of this study.

    References
    1.
    Fairbrother  N, Young  AH, Janssen  P, Antony  MM, Tucker  E.  Depression and anxiety during the perinatal period.   BMC Psychiatry. 2015;15:206. doi:10.1186/s12888-015-0526-6 PubMedGoogle ScholarCrossref
    2.
    Rees  S, Channon  S, Waters  CS.  The impact of maternal prenatal and postnatal anxiety on children’s emotional problems: a systematic review.   Eur Child Adolesc Psychiatry. 2019;28(2):257-280. doi:10.1007/s00787-018-1173-5 PubMedGoogle ScholarCrossref
    3.
    Gavin  NI, Gaynes  BN, Lohr  KN, Meltzer-Brody  S, Gartlehner  G, Swinson  T.  Perinatal depression: a systematic review of prevalence and incidence.   Obstet Gynecol. 2005;106(5 Pt 1):1071-1083. doi:10.1097/01.AOG.0000183597.31630.db PubMedGoogle ScholarCrossref
    4.
    Kim  DR, Bale  TL, Epperson  CN.  Prenatal programming of mental illness: current understanding of relationship and mechanisms.   Curr Psychiatry Rep. 2015;17(2):5. doi:10.1007/s11920-014-0546-9 PubMedGoogle ScholarCrossref
    5.
    O’Connor  TG, Heron  J, Glover  V; Alspac Study Team.  Antenatal anxiety predicts child behavioral/emotional problems independently of postnatal depression.   J Am Acad Child Adolesc Psychiatry. 2002;41(12):1470-1477. doi:10.1097/00004583-200212000-00019 PubMedGoogle ScholarCrossref
    6.
    Beversdorf  DQ, Manning  SE, Hillier  A,  et al.  Timing of prenatal stressors and autism.   J Autism Dev Disord. 2005;35(4):471-478. doi:10.1007/s10803-005-5037-8 PubMedGoogle ScholarCrossref
    7.
    Walder  DJ, Laplante  DP, Sousa-Pires  A, Veru  F, Brunet  A, King  S.  Prenatal maternal stress predicts autism traits in 6½ year-old children: Project Ice Storm.   Psychiatry Res. 2014;219(2):353-360. doi:10.1016/j.psychres.2014.04.034 PubMedGoogle ScholarCrossref
    8.
    Weinstock  M.  Alterations induced by gestational stress in brain morphology and behaviour of the offspring.   Prog Neurobiol. 2001;65(5):427-451. doi:10.1016/S0301-0082(01)00018-1 PubMedGoogle ScholarCrossref
    9.
    Huizink  AC, Mulder  EJH, Buitelaar  JK.  Prenatal stress and risk for psychopathology: specific effects or induction of general susceptibility?   Psychol Bull. 2004;130(1):115-142. doi:10.1037/0033-2909.130.1.115 PubMedGoogle ScholarCrossref
    10.
    Davis  EP, Snidman  N, Wadhwa  PD, Glynn  LM, Schetter  CD, Sandman  CA.  Prenatal maternal anxiety and depression predict negative behavioral reactivity in infancy.   Infancy. 2004;6(3):319-331. doi:10.1207/s15327078in0603_1Google ScholarCrossref
    11.
    Brouwers  EPM, van Baar  AL, Pop  VJM.  Maternal anxiety during pregnancy and subsequent infant development.   Infant Behav Dev. 2001;24(1):95-106. doi:10.1016/S0163-6383(01)00062-5Google ScholarCrossref
    12.
    Laplante  DP, Barr  RG, Brunet  A,  et al.  Stress during pregnancy affects general intellectual and language functioning in human toddlers.   Pediatr Res. 2004;56(3):400-410. doi:10.1203/01.PDR.0000136281.34035.44 PubMedGoogle ScholarCrossref
    13.
    Sandman  CA, Buss  C, Head  K, Davis  EP.  Fetal exposure to maternal depressive symptoms is associated with cortical thickness in late childhood.   Biol Psychiatry. 2015;77(4):324-334. doi:10.1016/j.biopsych.2014.06.025 PubMedGoogle ScholarCrossref
    14.
    Lebel  C, Walton  M, Letourneau  N, Giesbrecht  GF, Kaplan  BJ, Dewey  D.  Prepartum and postpartum maternal depressive symptoms are related to children’s brain structure in preschool.   Biol Psychiatry. 2016;80(11):859-868. doi:10.1016/j.biopsych.2015.12.004 PubMedGoogle ScholarCrossref
    15.
    Qiu  A, Rifkin-Graboi  A, Chen  H,  et al.  Maternal anxiety and infants’ hippocampal development: timing matters.   Transl Psychiatry. 2013;3:e306. doi:10.1038/tp.2013.79 PubMedGoogle Scholar
    16.
    Jones  SL, Dufoix  R, Laplante  DP,  et al.  Larger amygdala volume mediates the association between prenatal maternal stress and higher levels of externalizing behaviors: sex specific effects in Project Ice Storm.   Front Hum Neurosci. 2019;13:144. doi:10.3389/fnhum.2019.00144 PubMedGoogle ScholarCrossref
    17.
    Acosta  H, Tuulari  JJ, Scheinin  NM,  et al.  Maternal pregnancy-related anxiety is associated with sexually dimorphic alterations in amygdala volume in 4-year-old children.   Front Behav Neurosci. 2019;13:175. doi:10.3389/fnbeh.2019.00175 PubMedGoogle ScholarCrossref
    18.
    Wen  DJ, Soe  NN, Sim  LW,  et al.  Infant frontal EEG asymmetry in relation with postnatal maternal depression and parenting behavior.   Transl Psychiatry. 2017;7(3):e1057. doi:10.1038/tp.2017.28 PubMedGoogle Scholar
    19.
    Field  T, Diego  M.  Maternal depression effects on infant frontal EEG asymmetry.   Int J Neurosci. 2008;118(8):1081-1108. doi:10.1080/00207450701769067 PubMedGoogle ScholarCrossref
    20.
    Sarkar  S, Craig  MC, Dell’Acqua  F,  et al.  Prenatal stress and limbic-prefrontal white matter microstructure in children aged 6-9 years: a preliminary diffusion tensor imaging study.   World J Biol Psychiatry. 2014;15(4):346-352. doi:10.3109/15622975.2014.903336 PubMedGoogle ScholarCrossref
    21.
    Rifkin-Graboi  A, Bai  J, Chen  H,  et al.  Prenatal maternal depression associates with microstructure of right amygdala in neonates at birth.   Biol Psychiatry. 2013;74(11):837-844. doi:10.1016/j.biopsych.2013.06.019 PubMedGoogle ScholarCrossref
    22.
    Scheinost  D, Kwon  SH, Lacadie  C,  et al.  Prenatal stress alters amygdala functional connectivity in preterm neonates.   Neuroimage Clin. 2016;12:381-388. doi:10.1016/j.nicl.2016.08.010 PubMedGoogle ScholarCrossref
    23.
    Scheinost  D, Sinha  R, Cross  SN,  et al.  Does prenatal stress alter the developing connectome?   Pediatr Res. 2017;81(1-2):214-226. doi:10.1038/pr.2016.197 PubMedGoogle ScholarCrossref
    24.
    Wu  Y, Lu  Y-C, Jacobs  M,  et al.  Association of prenatal maternal psychological distress with fetal brain growth, metabolism, and cortical maturation.   JAMA Netw Open. 2020;3(1):e1919940. doi:10.1001/jamanetworkopen.2019.19940 PubMedGoogle Scholar
    25.
    Middleton  H, Shaw  I.  Distinguishing mental illness in primary care. We need to separate proper syndromes from generalised distress.   BMJ. 2000;320(7247):1420-1421. doi:10.1136/bmj.320.7247.1420 PubMedGoogle ScholarCrossref
    26.
    Cohen  S, Kamarck  T, Mermelstein  R.  A global measure of perceived stress.   J Health Soc Behav. 1983;24(4):385-396. doi:10.2307/2136404 PubMedGoogle ScholarCrossref
    27.
    Cox  JL, Holden  JM, Sagovsky  R.  Detection of postnatal depression: development of the 10-item Edinburgh Postnatal Depression Scale.   Br J Psychiatry. 1987;150:782-786. doi:10.1192/bjp.150.6.782 PubMedGoogle ScholarCrossref
    28.
    Spielberger  CD, Sydeman  SJ. State-trait anxiety inventory and state-trait anger expression inventory. In: Maruish ME, ed. The Use of Psychological Testing for Treatment Planning and Outcome Assessment. Lawrence Erlbaum Associates; 1994:292-321.
    29.
    Cox  RW.  AFNI: software for analysis and visualization of functional magnetic resonance neuroimages.   Comput Biomed Res. 1996;29(3):162-173. doi:10.1006/cbmr.1996.0014 PubMedGoogle ScholarCrossref
    30.
    Gholipour  A, Limperopoulos  C, Clancy  S,  et al. Construction of a Deformable Spatiotemporal MRI Atlas of the Fetal Brain: Evaluation of Similarity Metrics and Deformation Models. In: Medical Image Computing and Computer-Assisted Intervention—MICCAI 2014. Lecture Notes in Computer Science. Springer International Publishing; 2014:292-299. doi:10.1007/978-3-319-10470-6_37
    31.
    Tustison  NJ, Avants  BB, Cook  PA,  et al.  N4ITK: improved N3 bias correction.   IEEE Trans Med Imaging. 2010;29(6):1310-1320. doi:10.1109/TMI.2010.2046908 PubMedGoogle ScholarCrossref
    32.
    Joshi  A, Scheinost  D, Okuda  H,  et al.  Unified framework for development, deployment and robust testing of neuroimaging algorithms.   Neuroinformatics. 2011;9(1):69-84. doi:10.1007/s12021-010-9092-8 PubMedGoogle ScholarCrossref
    33.
    Scheinost  D, Onofrey  JA, Kwon  SH,  et al. A fetal fMRI specific motion correction algorithm using 2nd order edge features. Paper presented at: the 2018 IEEE 15th International Symposium on Biomedical Imaging. 2018:1288-1292. Accessed March 8, 2019. http://ieeexplore.ieee.org
    34.
    Ojemann  JG, Akbudak  E, Snyder  AZ, McKinstry  RC, Raichle  ME, Conturo  TE.  Anatomic localization and quantitative analysis of gradient refocused echo-planar fMRI susceptibility artifacts.   Neuroimage. 1997;6(3):156-167. doi:10.1006/nimg.1997.0289 PubMedGoogle ScholarCrossref
    35.
    Behzadi  Y, Restom  K, Liau  J, Liu  TT.  A component based noise correction method (CompCor) for BOLD and perfusion based fMRI.   Neuroimage. 2007;37(1):90-101. doi:10.1016/j.neuroimage.2007.04.042 PubMedGoogle ScholarCrossref
    36.
    Hallquist  MN, Hwang  K, Luna  B.  The nuisance of nuisance regression: spectral misspecification in a common approach to resting-state fMRI preprocessing reintroduces noise and obscures functional connectivity.   Neuroimage. 2013;82:208-225. doi:10.1016/j.neuroimage.2013.05.116 PubMedGoogle ScholarCrossref
    37.
    Wheelock  MD, Hect  JL, Hernandez-Andrade  E,  et al.  Sex differences in functional connectivity during fetal brain development.   Dev Cogn Neurosci. 2019;36:100632. doi:10.1016/j.dcn.2019.100632 PubMedGoogle Scholar
    38.
    Jo  HJ, Saad  ZS, Simmons  WK, Milbury  LA, Cox  RW.  Mapping sources of correlation in resting state FMRI, with artifact detection and removal.   Neuroimage. 2010;52(2):571-582. doi:10.1016/j.neuroimage.2010.04.246 PubMedGoogle ScholarCrossref
    39.
    Muschelli  J, Nebel  MB, Caffo  BS, Barber  AD, Pekar  JJ, Mostofsky  SH.  Reduction of motion-related artifacts in resting state fMRI using aCompCor.   Neuroimage. 2014;96:22-35. doi:10.1016/j.neuroimage.2014.03.028 PubMedGoogle ScholarCrossref
    40.
    Zhao  L, Feng  X, Meyer  C, Wu  Y, du Plessis  AJ, Limperopoulos  C. Fetal brain automatic segmentation using 3D deep convolutional neural network. Paper presented at: ISMRM 27th Annual Meeting; Montreal, Quebec, Canada; May 11-16, 2019.
    41.
    Worsley  KJ, Marrett  S, Neelin  P, Vandal  AC, Friston  KJ, Evans  AC.  A unified statistical approach for determining significant signals in images of cerebral activation.   Hum Brain Mapp. 1996;4(1):58-73. doi:10.1002/(SICI)1097-0193(1996)4:1<58::AID-HBM4>3.0.CO;2-O PubMedGoogle ScholarCrossref
    42.
    Friston  KJ, Williams  S, Howard  R, Frackowiak  RSJ, Turner  R.  Movement-related effects in fMRI time-series.   Magn Reson Med. 1996;35(3):346-355. doi:10.1002/mrm.1910350312 PubMedGoogle ScholarCrossref
    43.
    Li  X, Hect  J, Thomason  M, Zhu  D. Interpreting age effects of human fetal brain from spontaneous fMRI using deep 3D convolutional neural networks. arXiv. Preprint posted on June 9, 2019. https://arxiv.org/abs/1906.03691
    44.
    Thomason  ME, Scheinost  D, Manning  JH,  et al.  Weak functional connectivity in the human fetal brain prior to preterm birth.   Sci Rep. 2017;7:39286. doi:10.1038/srep39286 PubMedGoogle ScholarCrossref
    45.
    Thomason  ME, Hect  JL, Rauh  VA,  et al.  Prenatal lead exposure impacts cross-hemispheric and long-range connectivity in the human fetal brain.   Neuroimage. 2019;191:186-192. doi:10.1016/j.neuroimage.2019.02.017 PubMedGoogle ScholarCrossref
    46.
    van den Heuvel  MI, Turk  E, Manning  JH,  et al.  Hubs in the human fetal brain network.   Dev Cogn Neurosci. 2018;30:108-115. doi:10.1016/j.dcn.2018.02.001 PubMedGoogle ScholarCrossref
    47.
    Craddock  RC, James  GA, Holtzheimer  PE  III, Hu  XP, Mayberg  HS.  A whole brain fMRI atlas generated via spatially constrained spectral clustering.   Hum Brain Mapp. 2012;33(8):1914-1928. doi:10.1002/hbm.21333 PubMedGoogle ScholarCrossref
    48.
    Makropoulos  A, Gousias  IS, Ledig  C,  et al.  Automatic whole brain MRI segmentation of the developing neonatal brain.   IEEE Trans Med Imaging. 2014;33(9):1818-1831. doi:10.1109/TMI.2014.2322280 PubMedGoogle ScholarCrossref
    49.
    Peer  M, Abboud  S, Hertz  U, Amedi  A, Arzy  S.  Intensity-based masking: a tool to improve functional connectivity results of resting-state fMRI.   Hum Brain Mapp. 2016;37(7):2407-2418. doi:10.1002/hbm.23182 PubMedGoogle ScholarCrossref
    50.
    Shehzad  Z, Kelly  C, Reiss  PT,  et al.  A multivariate distance-based analytic framework for connectome-wide association studies.   Neuroimage. 2014;93(Pt 1):74-94. doi:10.1016/j.neuroimage.2014.02.024 PubMedGoogle ScholarCrossref
    51.
    McArtor  DB, Lubke  GH, Bergeman  CS.  Extending multivariate distance matrix regression with an effect size measure and the asymptotic null distribution of the test statistic.   Psychometrika. 2017;82(4):1052-1077. doi:10.1007/s11336-016-9527-8 PubMedGoogle ScholarCrossref
    52.
    Aggarwal  CC, Hinneburg  A, Keim  DA. On the surprising behavior of distance metrics in high dimensional space. In:  Database Theory—ICDT 2001. Springer Berlin Heidelberg; 2001:420-434. doi:10.1007/3-540-44503-X_27
    53.
    Gower  JC.  Some distance properties of latent root and vector methods used in multivariate analysis.   Biometrika. 1966;53(3-4):325-338. doi:10.1093/biomet/53.3-4.325Google ScholarCrossref
    54.
    Zapala  MA, Schork  NJ.  Statistical properties of multivariate distance matrix regression for high-dimensional data analysis.   Front Genet. 2012;3:190. doi:10.3389/fgene.2012.00190 PubMedGoogle ScholarCrossref
    55.
    Rivals  I, Personnaz  L, Taing  L, Potier  M-C.  Enrichment or depletion of a GO category within a class of genes: which test?   Bioinformatics. 2007;23(4):401-407. doi:10.1093/bioinformatics/btl633 PubMedGoogle ScholarCrossref
    56.
    Eggebrecht  AT, Elison  JT, Feczko  E,  et al; IBIS Network.  Joint attention and brain functional connectivity in infants and toddlers.   Cereb Cortex. 2017;27(3):1709-1720. doi:10.1093/cercor/bhw403 PubMedGoogle ScholarCrossref
    57.
    Rubinov  M, Sporns  O.  Weight-conserving characterization of complex functional brain networks.   Neuroimage. 2011;56(4):2068-2079. doi:10.1016/j.neuroimage.2011.03.069 PubMedGoogle ScholarCrossref
    58.
    Xia  M, Wang  J, He  Y.  BrainNet Viewer: a network visualization tool for human brain connectomics.   PLoS One. 2013;8(7):e68910. doi:10.1371/journal.pone.0068910 PubMedGoogle Scholar
    59.
    Van den Bergh  BRH, Mulder  EJH, Mennes  M, Glover  V.  Antenatal maternal anxiety and stress and the neurobehavioural development of the fetus and child: links and possible mechanisms. A review.   Neurosci Biobehav Rev. 2005;29(2):237-258. doi:10.1016/j.neubiorev.2004.10.007 PubMedGoogle ScholarCrossref
    60.
    Van den Bergh  BR, Mulder  EJ, Visser  GH, Poelmann-Weesjes  G, Bekedam  DJ, Prechtl  HF.  The effect of (induced) maternal emotions on fetal behaviour: a controlled study.   Early Hum Dev. 1989;19(1):9-19. doi:10.1016/0378-3782(89)90100-X PubMedGoogle ScholarCrossref
    61.
    Matthews  SG.  Early programming of the hypothalamo-pituitary-adrenal axis.   Trends Endocrinol Metab. 2002;13(9):373-380. doi:10.1016/S1043-2760(02)00690-2 PubMedGoogle ScholarCrossref
    62.
    Hobson  JA, Pace-Schott  EF.  The cognitive neuroscience of sleep: neuronal systems, consciousness and learning.   Nat Rev Neurosci. 2002;3(9):679-693. doi:10.1038/nrn915 PubMedGoogle ScholarCrossref
    63.
    O’Connor  TG, Heron  J, Golding  J, Beveridge  M, Glover  V.  Maternal antenatal anxiety and children’s behavioural/emotional problems at 4 years: report from the Avon Longitudinal Study of Parents and Children.   Br J Psychiatry. 2002;180:502-508. doi:10.1192/bjp.180.6.502 PubMedGoogle ScholarCrossref
    64.
    Van den Bergh BRH.  The influence of maternal emotions during pregnancy on fetal and neonatal behavior.   J Prenat Perinat Psychol Health. 1990;5(2):119.Google Scholar
    65.
    Gitau  R, Cameron  A, Fisk  NM, Glover  V.  Fetal exposure to maternal cortisol.   Lancet. 1998;352(9129):707-708. doi:10.1016/S0140-6736(05)60824-0 PubMedGoogle ScholarCrossref
    66.
    Dedovic  K, Duchesne  A, Andrews  J, Engert  V, Pruessner  JC.  The brain and the stress axis: the neural correlates of cortisol regulation in response to stress.   Neuroimage. 2009;47(3):864-871. doi:10.1016/j.neuroimage.2009.05.074 PubMedGoogle ScholarCrossref
    67.
    Mansell  T, Novakovic  B, Meyer  B,  et al; BIS Investigator Team.  The effects of maternal anxiety during pregnancy on IGF2/H19 methylation in cord blood.   Transl Psychiatry. 2016;6:e765. doi:10.1038/tp.2016.32 PubMedGoogle Scholar
    68.
    Conradt  E, Lester  BM, Appleton  AA, Armstrong  DA, Marsit  CJ.  The roles of DNA methylation of NR3C1 and 11β-HSD2 and exposure to maternal mood disorder in utero on newborn neurobehavior.   Epigenetics. 2013;8(12):1321-1329. doi:10.4161/epi.26634 PubMedGoogle ScholarCrossref
    69.
    Jobe  EM, Zhao  X.  DNA methylation and adult neurogenesis.   Brain Plast. 2017;3(1):5-26. doi:10.3233/BPL-160034 PubMedGoogle ScholarCrossref
    70.
    Wu  L, Wang  L, Shangguan  S,  et al.  Altered methylation of IGF2 DMR0 is associated with neural tube defects.   Mol Cell Biochem. 2013;380(1-2):33-42. doi:10.1007/s11010-013-1655-1 PubMedGoogle ScholarCrossref
    71.
    Peterson  A, Thome  J, Frewen  P, Lanius  RA.  Resting-state neuroimaging studies: a new way of identifying differences and similarities among the anxiety disorders?   Can J Psychiatry. 2014;59(6):294-300. doi:10.1177/070674371405900602 PubMedGoogle ScholarCrossref
    72.
    Mochcovitch  MD, da Rocha Freire  RC, Garcia  RF, Nardi  AE.  A systematic review of fMRI studies in generalized anxiety disorder: evaluating its neural and cognitive basis.   J Affect Disord. 2014;167:336-342. doi:10.1016/j.jad.2014.06.041 PubMedGoogle ScholarCrossref
    73.
    Menon  V, Uddin  LQ.  Saliency, switching, attention and control: a network model of insula function.   Brain Struct Funct. 2010;214(5-6):655-667. doi:10.1007/s00429-010-0262-0 PubMedGoogle ScholarCrossref
    74.
    Seeley  WW, Menon  V, Schatzberg  AF,  et al.  Dissociable intrinsic connectivity networks for salience processing and executive control.   J Neurosci. 2007;27(9):2349-2356. doi:10.1523/JNEUROSCI.5587-06.2007 PubMedGoogle ScholarCrossref
    75.
    Greicius  MD, Supekar  K, Menon  V, Dougherty  RF.  Resting-state functional connectivity reflects structural connectivity in the default mode network.   Cereb Cortex. 2009;19(1):72-78. doi:10.1093/cercor/bhn059 PubMedGoogle ScholarCrossref
    76.
    Raichle  ME.  The brain’s default mode network.   Annu Rev Neurosci. 2015;38:433-447. doi:10.1146/annurev-neuro-071013-014030 PubMedGoogle ScholarCrossref
    77.
    Dosenbach  NUF, Fair  DA, Miezin  FM,  et al.  Distinct brain networks for adaptive and stable task control in humans.   Proc Natl Acad Sci U S A. 2007;104(26):11073-11078. doi:10.1073/pnas.0704320104 PubMedGoogle ScholarCrossref
    78.
    Sripada  RK, King  AP, Welsh  RC,  et al.  Neural dysregulation in posttraumatic stress disorder: evidence for disrupted equilibrium between salience and default mode brain networks.   Psychosom Med. 2012;74(9):904-911. doi:10.1097/PSY.0b013e318273bf33 PubMedGoogle ScholarCrossref
    79.
    Zhang  Y, Liu  F, Chen  H,  et al.  Intranetwork and internetwork functional connectivity alterations in post-traumatic stress disorder.   J Affect Disord. 2015;187:114-121. doi:10.1016/j.jad.2015.08.043 PubMedGoogle ScholarCrossref
    80.
    Chen  AC, Etkin  A.  Hippocampal network connectivity and activation differentiates post-traumatic stress disorder from generalized anxiety disorder.   Neuropsychopharmacology. 2013;38(10):1889-1898. doi:10.1038/npp.2013.122 PubMedGoogle ScholarCrossref
    81.
    Poppenk  J, Evensmoen  HR, Moscovitch  M, Nadel  L.  Long-axis specialization of the human hippocampus.   Trends Cogn Sci. 2013;17(5):230-240. doi:10.1016/j.tics.2013.03.005 PubMedGoogle ScholarCrossref
    82.
    Satpute  AB, Mumford  JA, Naliboff  BD, Poldrack  RA.  Human anterior and posterior hippocampus respond distinctly to state and trait anxiety.   Emotion. 2012;12(1):58-68. doi:10.1037/a0026517 PubMedGoogle ScholarCrossref
    83.
    Schneier  FR, Pomplun  M, Sy  M, Hirsch  J.  Neural response to eye contact and paroxetine treatment in generalized social anxiety disorder.   Psychiatry Res. 2011;194(3):271-278. doi:10.1016/j.pscychresns.2011.08.006 PubMedGoogle ScholarCrossref
    84.
    Adamson  B, Letourneau  N, Lebel  C.  Prenatal maternal anxiety and children’s brain structure and function: a systematic review of neuroimaging studies.   J Affect Disord. 2018;241:117-126. doi:10.1016/j.jad.2018.08.029 PubMedGoogle ScholarCrossref
    85.
    Mennes  M, Stiers  P, Lagae  L, Van den Bergh  BRH.  Antenatal maternal anxiety modulates the BOLD response in 20-year-old men during endogenous cognitive control.   Brain Imaging and Behavior. 2019. doi:10.1007/s11682-018-0027-6 PubMedGoogle Scholar
    86.
    Burghy  CA, Stodola  DE, Ruttle  PL,  et al.  Developmental pathways to amygdala-prefrontal function and internalizing symptoms in adolescence.   Nat Neurosci. 2012;15(12):1736-1741. doi:10.1038/nn.3257 PubMedGoogle ScholarCrossref
    87.
    Adamson  B, Letourneau  N, Lebel  C.  Corrigendum to prenatal maternal anxiety and children’s brain structure and function: a systematic review of neuroimaging studies.   J Affect Disord. 2019;253:S0165-0327(18)33091-X. doi:10.1016/j.jad.2018.12.002 PubMedGoogle Scholar
    88.
    Rifkin-Graboi  A, Meaney  MJ, Chen  H,  et al.  Antenatal maternal anxiety predicts variations in neural structures implicated in anxiety disorders in newborns.   J Am Acad Child Adolesc Psychiatry. 2015;54(4):313-21.e2. doi:10.1016/j.jaac.2015.01.013 PubMedGoogle ScholarCrossref
    89.
    Smyser  CD, Inder  TE, Shimony  JS,  et al.  Longitudinal analysis of neural network development in preterm infants.   Cereb Cortex. 2010;20(12):2852-2862. doi:10.1093/cercor/bhq035 PubMedGoogle ScholarCrossref
    90.
    Doria  V, Beckmann  CF, Arichi  T,  et al.  Emergence of resting state networks in the preterm human brain.   Proc Natl Acad Sci U S A. 2010;107(46):20015-20020. doi:10.1073/pnas.1007921107 PubMedGoogle ScholarCrossref
    91.
    Thomason  ME, Dassanayake  MT, Shen  S,  et al.  Cross-hemispheric functional connectivity in the human fetal brain.   Sci Transl Med. 2013;5(173):173ra24. doi:10.1126/scitranslmed.3004978 PubMedGoogle Scholar
    92.
    Golfenshtein  N, Hanlon  AL, Deatrick  JA, Medoff-Cooper  B.  Parenting stress in parents of infants with congenital heart disease and parents of healthy infants: the first year of life.   Compr Child Adolesc Nurs. 2017;40(4):294-314. doi:10.1080/24694193.2017.1372532 PubMedGoogle ScholarCrossref
    ×