Following the arrows, reconstruction of motion-corrupted stacks of 2-dimensional (2-D) slices to a high-resolution 3-D isotropic image; fetal brain reorientation; extraction of fetal brain from surrounding tissue; brain structure segmentation (red indicates cerebrum; cyan, cerebellum; brown, brainstem; green, left hippocampus; yellow, right hippocampus); and volumetric and shape analysis.
Regions with negative t values (blue color) signify where the group positive for maternal psychological distress is significantly smaller than the group negative for maternal psychological distress. Regions with no color overlay indicate no significant difference between the 2 groups.
eFigure 1. T2-Weighted Brain MR Image Segmentation
eFigure 2. Flow Diagram Summarizing Our Subject Recruitment in This Study
eFigure 3. Associations Between Maternal Trait Anxiety (A) or Stress (B) and Fetal Cerebellar Volume in CHD and Controls
eTable 1. Maternal Psychological Distress Scales in Controls (92 Subjects; 149 Visits) and CHD (48 Subjects; 74 Visits) as Well as in Women With 2 Visits (57 Controls, 26 CHD Subjects)
eTable 2. Congenital Heart Disease Diagnoses
eTable 3. Brain Volumes in SV vs 2V CHD, and CHD Fetuses With vs Without Aortic Obstruction
eTable 4. Brain Volumes in CHD Fetuses With One and More Maternal Stressors (Stress, Anxiety, and/or Depression)
Customize your JAMA Network experience by selecting one or more topics from the list below.
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.
Wu Y, Kapse K, Jacobs M, et al. Association of Maternal Psychological Distress With In Utero Brain Development in Fetuses With Congenital Heart Disease. JAMA Pediatr. 2020;174(3):e195316. doi:10.1001/jamapediatrics.2019.5316
Is there an association of prenatal maternal stress, anxiety, and depression with brain growth in fetuses with congenital heart disease?
This longitudinal case-control study of 140 fetuses, including 48 fetuses with congenital heart disease having 74 magnetic resonance imaging scans and 92 healthy fetuses with 149 magnetic resonance imaging scans, showed that psychological distress in women with fetal congenital heart disease appears to be prevalent and associated with impaired fetal cerebellar and hippocampal development during the second half of gestation.
This study’s findings suggest that universal screening for prenatal psychological distress and integrated cognitive-behavioral interventions are needed to better support pregnant women and optimize neurodevelopment in fetuses with congenital heart disease.
Prenatal maternal psychological distress can result in detrimental mother and child outcomes. Maternal stress increases with receipt of a prenatal diagnosis of fetal congenital heart disease (CHD); however, the association between maternal stress and the developing brain in fetuses with CHD is unknown.
To determine the association of maternal psychological distress with brain development in fetuses with CHD.
Design, Setting, and Participants
This longitudinal, prospective, case-control study consecutively recruited 48 pregnant women carrying fetuses with CHD and 92 healthy volunteers with low-risk pregnancies from the Children’s National Health System between January 2016 and September 2018. Data were analyzed between January 2016 and June 2019.
Fetal CHD and maternal stress, anxiety, and depression.
Main Outcomes and Measures
Maternal stress, anxiety, and depression were measured using the Perceived Stress Scale, Spielberger State-Trait Anxiety Inventory, and Edinburgh Postnatal Depression Scale, respectively. Volumes of fetal total brain, cerebrum, left and right hippocampus, cerebellum, and brainstem were determined from 3-dimensionally reconstructed T2-weighted magnetic resonance imaging (MRI) scans.
This study included 223 MRI scans from 140 fetuses (74 MRIs from 48 fetuses with CHD and 149 MRIs from 92 healthy fetuses) between 21 and 40 weeks’ gestation. Among 48 women carrying fetuses with CHD, 31 (65%) tested positive for stress, 21 (44%) for anxiety, and 14 (29%) for depression. Among 92 pregnant women carrying healthy fetuses, 25 (27%) tested positive for stress, 24 (26%) for anxiety, and 8 (9%) for depression. Depression scores were higher among 17 women carrying fetuses with single-ventricle CHD vs 31 women carrying fetuses with 2-ventricle CHD (3.8; 95% CI, 0.3 to 7.3). Maternal stress and anxiety were associated with smaller left hippocampal (stress: −0.003 cm3; 95% CI, −0.005 to −0.001 cm3), right hippocampal (stress: −0.004; 95% CI, −0.007 to −0.002; trait anxiety: −0.003; 95% CI, −0.005 to −0.001), and cerebellar (stress: −0.06; 95% CI, −0.09 to −0.02) volumes only among women with fetal CHD. Impaired hippocampal regions were noted in the medial aspect of left hippocampal head and inferior aspect of right hippocampal head and body. Impaired cerebellar regions were noted in the anterior superior aspect of vermal and paravermal regions and the left cerebellar lobe.
Conclusions and Relevance
These findings suggested that psychological distress among women carrying fetuses with CHD is prevalent and is associated with impaired fetal cerebellar and hippocampal development. These data underscore the importance of universal screening for maternal psychological distress, integrated prenatal mental health support, and targeted early cognitive-behavioral interventions given that stress is a potentially modifiable risk factor in this high-risk population.
Impaired brain development in newborns with congenital heart disease (CHD) is increasingly recognized to have in utero origins.1-4 Members of our group have previously shown that fetuses with CHD exhibit disturbances in brain growth, cortical folding, and biochemistry in utero.1,4 To complicate matters, a prenatal diagnosis of CHD increases maternal stress.5,6 Mental health problems are now recognized as the most common complications of pregnancy, affecting up to 22% of women in the prenatal period or first postpartum year.7 This percentage nearly doubles in high-risk pregnancies, including among women who receive a diagnosis of fetal CHD.5,8 It is well established that prenatal maternal stress has adverse effects on pregnancy outcomes (ie, preeclampsia, spontaneous abortion, and preterm delivery)9-11 and is also increasingly linked to altered fetal programming and long-term cognitive, emotional, and behavioral dysfunction in children and adults.12,13 Moreover, a growing body of literature suggests that prenatal maternal depression, anxiety, or stress is associated with cortical thinning,14 altered amygdala15 and hippocampal growth,16 and altered brain microstructure17 and functional connectivity18 in the offspring. Despite this growing body of evidence to support the adverse effects of intrauterine exposure to stress on child brain structure and function, the association of prenatal maternal stress exposure with brain development in fetuses with CHD is unknown to date.
The objectives of the present study were 3-fold: (1) to determine the prevalence of stress, anxiety, and depression among women who had received a diagnosis of fetal CHD; (2) to use quantitative in vivo fetal magnetic resonance imaging (MRI) to examine the association of maternal stress, anxiety, and depression with global, regional, and local measures of brain development in fetuses with CHD; (3) to determine whether any association of maternal stress, anxiety, and depression with fetal brain development differ among women pregnant with fetuses with CHD vs without CHD. The terms psychological or mental distress are used in this study to denote stress, anxiety, depression, or a combination of these conditions that did not reach the severity of mental illness or mental disorder.19
Participants were consecutively recruited into a longitudinal prospective case-control study. Pregnant women who underwent a fetal echocardiogram at Children’s National Health System located in Washington, DC between January 2016 and September 2018 for suspected fetal CHD were eligible for enrollment. Included cases were pregnant women with confirmed fetal CHD, whereas included controls were healthy volunteers with low-risk pregnancies. Exclusion criteria included women with fetuses with extracardiac anomalies or chromosomal abnormalities or pregnant women with the following: pregnancy-related complications such as preeclampsia or gestational diabetes; known genetic, metabolic, or psychiatric disorders; maternal medications or illicit drug use; contraindications to MRI (eg, claustrophobia or metal implants); and multiple pregnancy. Eligible participants were scheduled to undergo 2 fetal MRIs (MRI 1 and MRI 2) between 21 and 40 weeks of gestational age. The institutional review board at Children’s National Health System approved the protocol for this study. Informed written consent was obtained from all participants in a manner consistent with the Common Rule requirements. Each participant received $75 at each visit as compensation.
Maternal psychological distress was measured using psychometrically sound questionnaires, namely, the Perceived Stress Scale (PSS),20 Edinburgh Postnatal Depression Scale (EPDS),21 and Spielberger State-Trait Anxiety Inventory (SSAI, state anxiety; STAI, trait anxiety).22 All questionnaires were completed by participants on the same day as their MRI. The PSS assesses the degree to which people perceive their lives as stressful during the preceding month. Higher PSS scores (range, 0-40) indicate higher stress, and a PSS score of 15 or higher indicates that stress is higher than average.23 The EPDS (10-item test; range, 0-30) measures the severity of perinatal depression symptoms in the previous 7 days. It has been validated for use in both the prenatal and postnatal periods.24 Higher EPDS scores indicate more depressive symptoms, and a score of 10 or higher indicates the presence of depression symptoms during pregnancy.25 The Spielberger Anxiety Inventory measures state (SSAI, current feeling) and trait (STAI; usual feeling) anxiety. A higher SSAI or STAI score (range, 20-80) suggests increased anxiety, with scores of 40 or higher indicating symptoms of anxiety.26
Multiplanar T2-weighted single-shot fast spin echo were acquired on a GE Healthcare 1.5-T MRI scanner using an 8-channel receiver coil (repetition time: 1100 milliseconds; echo time: 160 milliseconds; flip angle: 90°; field-of-view: 32 cm; matrix: 256 × 192; in-plane resolution: 1.25 × 1.66 mm2; slice thickness: 2 mm) with an acquisition time of 2 or 3 minutes per plane. Images of axial, coronal, and sagittal planes were reconstructed into a high-quality volumetric image using a previously validated method,27 which uses flexible point-spread functions and fast slice-to-volume reconstruction of high-resolution isotropic data from motion-corrupted stacks of 2-dimensional slices (Figure 1). The image resolution was 0.86 × 0.86 × 0.86 mm3 after reconstruction and reorientation. All MRI studies were reviewed by an experienced fetal neuroradiologist (G.V.) who was blinded to CHD status and maternal questionnaire scores.
We quantified volumes (in cubic centimeters) of the total brain (total brain volume [TBV]), cerebrum, cerebellum, brainstem, left hippocampus (LH), and right hippocampus (RH). All structures were automatically segmented using Draw-EM (Developing Brain Region Annotation With Expectation-Maximization) pipelines28 and manually corrected using ITK-SNAP software (eFigure 1 in the Supplement). The hippocampi were manually segmented based on previously reported anatomical criteria.29,30 All brain structures were manually segmented by the same rater (Y.W.). Of the total scans, 20% (15 CHD, 30 controls) were randomly selected and segmented by another trained rater (K.K.). Interrater reliability was estimated using Bland-Altman analysis (limits of agreement plots). For all measures, more than 95% of points were within 2 SD limits of agreement. Raters were blinded to questionnaire scores and CHD status.
Spherical harmonic description31 was used to compare local differences in brain structures between psychological distress–positive and psychological distress–negative groups. The object size was scaled inversely to TBV to correct for overall brain size differences between groups. Object segmentations were converted into a corresponding spherical harmonic description and sampled into triangulated surfaces. Correspondences (2562 vertices) across object surfaces were obtained using a uniform icosahedron subdivision of the spherical parameterization.32 An atlas was constructed by iteratively registering the objects of the compared groups.33 Thereafter, all individual objects were rigidly registered to that atlas. Differences between groups of object surfaces were measured by vertex-wise displacement between each individual and atlas surfaces.
All participants underwent standardized echocardiographic studies on the same day of the MRI. Echocardiographic data were obtained on GE Vivid 7 or Philips IE33 devices. In addition to standard Doppler ultrasonographic measurements, this study analyzed middle cerebral artery and umbilical artery resistance indexes and the cerebral to placental resistance ratio.
Statistical analyses were performed using MATLAB, version 2018b, and SAS, version 9.3 software (SAS Institute Inc). Characteristics of participants with fetal CHD and controls were compared using the Fisher exact test for categorical variables and t tests for continuous variables. Associations between maternal psychological distress and fetal brain volumes and shape differences between groups were evaluated using linear generalized estimating equation models with sandwich variance estimators. An unstructured covariance structure was assumed based on quasi-information criterion values suggesting best fit. To account for known associations with brain volume, generalized estimating equation models were adjusted for gestational age at MRI and fetal sex; outcomes with significant gestational age by sex interactions were additionally adjusted. Models evaluating regional differences were further adjusted for TBV to remove the influence of brain volume overall. Finally, to evaluate any differential association between maternal psychological distress and fetal brain volumes in CHD vs controls, group by distress scale interaction terms were included in generalized estimating equations models. Additional adjustments for days between diagnosis and fetal MRI, ventricle physiology, maternal age, employment, educational level, and race/ethnicity were considered, with no material effect on estimates. Possible interactions between distress scales and CHD types were evaluated, with no significant influence on the outcome. The P values were adjusted for multiple testing based on the false discovery rate according to the Benjamini-Hochberg method,34 and 2-sided adjusted P ≤ .05 were considered statistically significant.
A flow diagram summarizing enrollment in the present study is shown in eFigure 2 in the Supplement. We excluded 6 participants with genetic syndromes and 2 participants with abnormal MRI results. Twenty-six MRI scans were excluded because of fetal motion (18 scans [8%]) or missing maternal questionnaires (8 scans [4%]). Our final sample consisted of 223 fetal MRI scans from 140 pregnant women, including 48 women with fetal CHD (29 males, 19 females) with 74 MRI scans (42 MRI 1; 32 MRI 2) and 92 healthy fetuses (50 males, 42 females) with 149 MRI scans (75 MRI 1; 74 MRI 2). The mean (SD) gestational age at MRI was 32.1 (4.2) weeks (range, 21.6-38.3 weeks) for fetuses with CHD and 32.4 (4.5) weeks (range, 24.1-39.4 weeks) for healthy fetuses. The demographics of our cohort are given in Table 1. Most participants (46 [96%]) had received a diagnosis of fetal CHD in the second trimester. For the fetal CHD cohort, 39 pregnant women (81%) had a high educational level (attended college) and 36 (75%) had professional occupations. All control fetuses and fetuses with CHD had a structurally normal brain without evidence of any injury on conventional MRI.
Among 48 pregnant women carrying fetuses with CHD, 31 (65%) tested positive for stress, 21 (44%) were positive for anxiety (20 [42%] for state anxiety and 11 [23%] for trait anxiety), and 14 (29%) were positive for depression. Among 92 pregnant women carrying healthy fetuses, 25 (27%) tested positive for stress, 24 (26%) for anxiety (19 [21%] for state anxiety and 15 [16%] for trait anxiety), and 8 (9%) for depression. None of the participants had been previously screened for prenatal depression or anxiety, and none were taking medications or receiving mental health intervention. Prenatal stress (PSS score difference, 5.3; 95% CI, 3.0-7.5), depression (EPDS score difference, 3.3; 95% CI, 1.7-4.9), and anxiety (SSAI score difference, 7.5; 95% CI, 3.5-11.4) were significantly higher among women carrying fetuses with CHD vs those with healthy fetuses (eTable 1 in the Supplement). Among women with fetal CHD, all but 1 woman who tested positive for anxiety and depression also scored positive for stress, 20 women (42%) tested positive for both stress and anxiety, 13 women (27%) for both stress and depression, 11 women (23%) for both anxiety and depression, and 10 women (21%) for anxiety, stress, and depression. Among 17 women carrying fetuses with single-ventricle CHD vs 31 women carrying fetuses with 2-ventricle CHD, there was no significant difference in the proportion of women who tested positive for stress (12 [71%] vs 19 [62%]; P = .75), anxiety (state: 10 [59%] vs 10 [32%]; P = .12); trait: 6 [35%] vs 5 [16%]; P = .16), or depression (6 [35%] vs 8 [26%]; P = .74). However, the mean depression score was significantly higher (3.8; 95% CI, 0.3-7.3) among women with single-ventricle vs 2-ventricle fetal CHD. In the subset with a second follow-up visit, women reported significantly lower trait anxiety (−1.2; 95% CI, −2.1 to −0.3) and stress (−2.2; 95% CI, −3.1 to −1.2) at the second visit compared with those at the first visit (eTable 1 in the Supplement).
The CHD diagnostic categories of our cohort are summarized in eTable 2 in the Supplement. Of 48 fetuses with CHD, 17 (35%) had single-ventricle CHD, 19 (40%) had CHD with aortic obstruction, and 8 (17%) had transposition of the great arteries. The association between maternal psychological distress and fetal echocardiographic measures was examined. Maternal depression was associated with lower middle cerebral artery resistance index (−0.003; 95% CI, −0.005 to −0.001) in fetuses with CHD.
Fetal brain volumes increased linearly with increased gestational age in both CHD and controls. In fetuses with CHD, males showed faster growth rates than females for TBV (18.6 vs 16.1 cm3/week), cerebrum volume (17.1 vs 14.8 cm3/week), and cerebellum volume (1.3 vs 1.1 cm3/week). This phenomenon was also observed in controls for TBV (19.8 vs 18.9 cm3/week), cerebrum volume (18.2 vs 17.3 cm3/week), and brainstem volume (0.28 vs 0.26 cm3/week). In fetuses with CHD and in healthy fetuses, RH volumes were significantly larger than LH volumes (CHD: 0.03 cm3; 95% CI, 0.02-0.05 cm3; controls: 0.04 cm3; 95% CI, 0.03-0.05 cm3). In fetuses with CHD, brain volumes did not significantly differ in single-ventricle vs 2-ventricle CHD or in those with vs without aortic obstruction (eTable 3 in the Supplement). However, TBV and cerebrum volume were smaller in single-ventricle vs 2-ventricle CHD excluding the great arteries (TBV: −11.9 cm3; 95% CI, −23.5 to −0.4 cm3; cerebrum: −11.9 cm3; 95% CI, −22.7 to −1.1 cm3), but this difference was not significant after adjusting for multiple testing.
In the CHD cohort, elevated maternal stress was significantly associated with smaller fetal cerebellum (−0.06 cm3; 95% CI, −0.09 to −0.02 cm3), LH (−0.003 cm3; 95% CI, −0.005 to −0.001 cm3), and RH (−0.004 cm3; 95% CI, −0.007 to −0.002 cm3) volumes. Maternal trait anxiety was significantly associated with smaller RH volume (−0.003 cm3; 95% CI, −0.005 to −0.001 cm3). In controls, maternal trait anxiety was negatively associated with fetal LH and RH volumes although these associations were not significant after adjusting for multiple testing.
The associations between maternal psychological distress and fetal brain volumes in CHD vs controls were compared (Table 2). We found that the cerebellum was differentially associated with maternal trait anxiety and stress in CHD vs controls and with maternal trait anxiety and stress negatively associated with cerebellum volumes in fetuses with CHD, but these associations were relatively unchanged in controls (eFigure 3 in the Supplement). However, these differential associations were not significant after adjusting for multiple testing.
Because more than 21% of our CHD cohort tested positive on 2 or 3 measures of psychological distress (ie, anxiety, stress, or depression), we further estimated the association of testing positive for more than 1 of these maternal stressors with CHD fetal brain volumes (eTable 4 in the Supplement). The results showed that there was no significant difference between fetal brain volumes in women with 1 vs more stressors. Given that almost all participants with fetal CHD who were positive for anxiety and depression were also positive for stress (except for 1 woman), we combined the participants with any (1 or more) positive stressor and referred to them as positive for maternal psychological distress.
Local impairments of the hippocampus and cerebellum in fetuses with CHD and with positive maternal psychological distress are shown in Figure 2 and Figure 3. Impaired hippocampal regions were noted in the medial aspect of left hippocampal head and inferior aspect of right hippocampal head and body (Figure 2). In the cerebellum, impaired development was evident in the anterior superior aspect of vermal and paravermal regions and the left cerebellar lobe (Figure 3). For control fetuses, hippocampal and cerebellar local regions did not significantly differ between groups negative for maternal psychological distress vs those positive for maternal psychological distress.
We report a high prevalence of psychological distress in pregnant women carrying fetuses with CHD. To our knowledge, we showed for the first time that maternal psychological distress was associated with decreased cerebellar and hippocampal volumes in fetuses with CHD during the second half of gestation. We further detected local growth impairments of fetal hippocampi and cerebellum associated with increased maternal psychological distress. Our data also suggested that fetal cerebellar volumetric growth may be differentially associated with maternal stress and anxiety in CHD vs controls. Finally, we showed that maternal depression was associated with a lower middle cerebral artery resistance index in fetuses with CHD.
Maternal stress and depression symptoms have been previously reported following receipt of a prenatal or postnatal diagnosis of CHD.5,6 In the present CHD cohort, 81% of pregnant women had a high educational level (attended college) and 75% of them had professional occupations. Although psychological distress has been reported to be more common among pregnant women of low socioeconomic status,35 we reported a high prevalence of psychological distress in the present cohort, in which nearly two-thirds (65%) of these women tested positive for stress, 44% for anxiety, and just less than 30% for depression despite their socioeconomic advantage. None of the participants had been formally screened for prenatal depression or anxiety, and none were taking medications nor were receiving any mental health intervention. Our data suggested the need for universal screening for prenatal mental health problems as part of routine clinical care in pregnant women with fetal CHD across gestation. Our data also underscored a critical need for providing pregnant women with access to perinatal interventions that focus on stress reduction and behavioral health to optimize maternal and fetal well-being.
To our knowledge, this is the first report indicating that maternal psychological distress is associated with impaired brain development in fetuses with CHD. The mechanisms underlying impaired fetal brain development in the setting of maternal stress are complex and multifactorial. Studies have reported that maternal mental distress increases uterine artery resistance, limiting blood flow, oxygen, and nutrients to the fetus.36 Maternal mental distress has also been associated with altered placental functions, including decreased placental expression of neurotropic precursors, such as 11β-hydroxysteroid dehydrogenase type 237 and monoamine oxidase A.38 Decreased expression of the former may increase exposure of the fetus to cortisol, and decreased exposure to the latter may increase exposure of the fetus to serotonin. Increased cortisol exposure affects gene expression in fetal brain cells.39 Serotonin affects cell neurogenesis, migration, and differentiation in the fetal brain.40 In addition, our data showed that higher maternal depression was associated with a lower middle cerebral artery resistance index, which reflects redistribution of the combined fetal cardiac output to the brain.41 Ongoing studies are needed to better address the causal (and modifiable) pathways of maternal psychological distress on the human fetal brain.
Our findings point to an association between impaired hippocampal development in fetuses with CHD and increased maternal psychological distress. Hippocampal impairments were noted on the hippocampal head (primary) and body (Figure 2). The hippocampus contains high levels of glucocorticoid receptors. Studies in adults have reported selective impairments of the hippocampal head with emotional abuse and posttraumatic stress disorder,42,43 whereas impairments of the hippocampal head and body have been reported in adults with mild cognitive impairment.44 The cornu ammonis region 3 (CA3) subfield is particularly sensitive to glucocorticoids.45 The hippocampal head has the largest part of the CA1 through CA3 subfields and has reciprocal connections with the prefrontal cortex, whereas the hippocampal body includes the biggest part of dentate gyrus.46,47 Volume losses in the hippocampal CA3 and dentate gyrus have been shown in veterans with posttraumatic stress disorder.45 In animal models, prenatal maternal stress caused apical dendritic atrophy of pyramidal neurons in hippocampal CA3 of offspring rats48 and altered hippocampal CA1 through CA3 under experimental stress.49 Prenatal stress also disrupts functional connectivity in the hippocampal–prefrontal cortex axis in offspring rats.50 Our data point to a selective regional susceptibility of the hippocampal head (and body) associated with maternal mental distress in fetuses with CHD.
We also showed that increased levels of maternal stress and anxiety were associated with disturbances in cerebellar growth in fetuses with CHD. Prenatal maternal anxiety has shown associations with altered neural structures in the neonatal cerebellum17 and reduced cerebellar gray matter density in children.51 An experimental study showed that increased exposure to glucocorticoid levels in the immature cerebellum may play a role in impaired proliferation of the external granular layer and thereby contribute to cerebellar underdevelopment.52 Moreover, our shape analysis revealed that maternal psychological distress was associated with impairment of the anterior superior aspect of vermal and paravermal regions and the left cerebellar lobe (Figure 3). Although the functional topography of immature cerebellum is unclear, impaired vermian and paravermian regions have been associated with sensorimotor, social-behavioral, and affective disorders in children and adults.53,54 Furthermore, injury to the left cerebellar lobe has been associated with nonverbal and visuospatial deficits.55 Many of these functional disturbances have been reported in children with CHD in the years following successful cardiac repair and may in part be mediated by prenatal maternal psychological distress. Future studies, which are currently under way, are needed to correlate these intriguing prenatal findings to long-term neurodevelopment.
Finally, we report an important differential association between maternal stress and anxiety and fetal cerebellar volumes in CHD vs controls. The fetal cerebellum grows faster than other brain structures during the second half of gestation, with its highest growth rate reported in the third trimester.56 In our CHD cohort, most participants (96%) had received a diagnosis of fetal CHD in the second trimester. Thus, the highly distressed period for the mother since receiving a diagnosis of fetal CHD covered the fastest growth period of the fetal cerebellum. This may in part explain the selective vulnerability of the cerebellum associated with our CHD cohort. Higher maternal stress and anxiety levels in CHD vs controls may also contribute to this differential association. Whether the cerebellum is more sensitive to high levels of psychological distress or whether CHD has potential effects on this differential association requires further studies.
Although our study has a number of strengths, the limitations deserve mention. Individual maternal questionnaires at select time points may not fully reflect the cumulative influence of maternal mental health on fetal brain development. The longitudinal assessments currently under way by members of our team are needed to better understand the timing of maternal mental well-being associated with fetal brain development. In addition, we did not report the potential associations of maternal psychological distress on long-term neurodevelopment in infants and adolescents, which is also currently under investigation by members of our team. The MRI segmentation of the fetal hippocampus is challenging; however, we used previously reported criteria and fetal and preterm brain atlases to guide segmentation.29,30,57-59 An experienced neuroradiologist (G.V.) also assisted with hippocampal anatomical localization, and our interrater reliability suggested good agreement from 2 raters’ segmentation results. Fetal motion caused the exclusion of 8% of the MRI scans. However, adding the distress scores of these participants did not influence our reported prevalence of maternal psychological distress. Finally, our sample size for each CHD type was likely too small to show different associations between maternal mental distress and fetal brain region volumes or echocardiographic measures among CHD subgroups. Ongoing work by members of our team includes increasing the fetal CHD sample size and examining the timing of the association of maternal psychological distress with fetal brain development and the association of maternal psychological distress with child long-term neurodevelopmental outcomes.
Our findings suggested that stress, anxiety, and depression are prevalent, and clinically underappreciated in pregnant women carrying fetuses with CHD and appears to be associated with impaired fetal hippocampal and cerebellar development. These findings support the need for further study of the potential association of fetal brain development with psychological distress in pregnant women who receive a diagnosis of fetal anomalies. If confirmed, these findings would suggest the need for universal screening for maternal distress, clinical trials of strategies during pregnancy to minimize these maternal complications, and targeted early cognitive-behavioral interventions to optimize neuropsychological outcome in offspring.
Accepted for Publication: October 30, 2019.
Corresponding Author: Catherine Limperopoulos, PhD, Center for the Developing Brain, Children’s National Health System, 111 Michigan Ave NW, Washington, DC 20010 (firstname.lastname@example.org).
Published Online: January 13, 2020. doi:10.1001/jamapediatrics.2019.5316
Author Contributions: Dr Wu had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: Wu, Jacobs, Wessel, du Plessis, Limperopoulos.
Acquisition, analysis, or interpretation of data: Wu, Kapse, Jacobs, Niforatos-Andescavage, Donofrio, Krishnan, Vezina, du Plessis, Limperopoulos.
Drafting of the manuscript: Wu, Kapse, Jacobs.
Critical revision of the manuscript for important intellectual content: All authors.
Statistical analysis: Wu, Jacobs.
Obtained funding: Wu, Wessel, Limperopoulos.
Administrative, technical, or material support: Wu, Kapse, Jacobs, Niforatos-Andescavage, Limperopoulos.
Supervision: Vezina, Wessel, du Plessis, Limperopoulos.
Conflict of Interest Disclosures: None reported.
Funding/Support: This study was funded by grant R01 HL116585-01 from the National Heart, Lung, and Blood Institute and by Early Career award number 14764 from the Thrasher Research Fund.
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 in the study. We also thank our laboratory members for their contributions and the study coordinators Catherine Lopez, MS, Kathryn Lee Bannantine, BSN, and Jessica Lynn Quistorff, MPH, for their assistance with patient enrollment and study coordination.
Create a personal account or sign in to: